chrisphan.com

Happy (belated) π Day!

Wed, 18 Mar 2026 16:15:00 -0500

Happy belated π Day! π Day is the 14th of March, i.e. 03-14 both the U.S. or ISO 8601 date formats. Two times in the past, I have posted some kind of method (not necessarily efficient) to approximate π as a π Day celebration:

2021: Used a Monte Carlo method: randomly selecting points $(x, y) \in [0, 1] \times [0, 1]$, and seeing what proportion have $x^2 + y^2 < 1$. Implemented in JavaScript.
2024: Used a power series approximation and an argument from Papa Rudin¹. Implemented in Julia

This year's entry, admittedly a little late, involves WebAssembly and differential equations.

A system of differential equations

This year, I decided to estimate π by approximating $\sin t$ for values of $t > 0$, using differential equations and Euler's method. Specifically, note that $x = \sin t, y = \cos t$ is the only solution to the system of differential equations \[\begin{aligned} \frac{dx}{dt} &= y, \\ \frac{dy}{dt} &= -x \end{aligned}\] that satisfies the initial conditions $x(0) = 0,\; y(0) = 1$.

Euler's method

Suppose we have a system of differential equations of the form \[\begin{aligned} \frac{dx}{dt} &= a x + b y,\\ \frac{dy}{dt} &= c x + d y. \end{aligned}\] We can use Euler's method to find numerical solutions (that is to say, approximations of the solutions) to this system. For any real number $t_0$, and small real number $\Delta t$, we can estimate the quantities \[\begin{aligned} \Delta x &= x(t_0 + \Delta t) - x(t_0),\\ \Delta y &= y(t_0 + \Delta t) - y(t_0), \end{aligned}\] with the approximations \[\begin{aligned} \Delta x &\approx \left(a x(t_0) + b y(t_0)\right) \Delta t,\\ \Delta y &\approx \left(c x(t_0) + d y(t_0)\right) \Delta t. \end{aligned}\]

Given initial values $x(0) = x_0$ and $y(0) = y_0$, we can come up with approximations $x_k$ and $y_k$ for $x(k \Delta t)$ and $y(k \Delta t)$, respectively, by recursively defining \[\begin{aligned} x_{k + 1} &= x_k + \left(a x_k + b y_k\right) \Delta t,\\ y_{k + 1} &= y_k + \left(c x_k + d y_k\right) \Delta t. \end{aligned}\]

Enter WASM

Lately, I have been learning WebAssembly (WASM), and thought it would be fun to implement this year's π Day demo in WASM. While traditional assembly language is used to specify processor instructions in order to program a computer directly, WebAssembly instructions are run on a stack-based virtual machine inside the web browser.² WASM in created by writing the instructions in a human-readable text format, WebAssembly text or WAT.

For example, here is some WAT code for my WASM implementation of Euler's method:

pi_day_2026.wat WAT


        ;; set delta_x = (a*x + b*y) * delta_t
        local.get $a
        local.get $x
        f64.mul
        local.get $b
        local.get $y
        f64.mul
        f64.add
        local.tee $dxdt
        local.get $delta_t
        f64.mul
        local.set $delta_x

Lines starting with ;; are comments. The virtual machine that executes WASM is a stack machine. In the above snippet³:

Line	Stack before	WAT Instruction	Description	Stack after
44	(empty)	`local.get $a`	Copy the value stored in the variable `$a` and push it on top of the stack	`$a`
45	`$a`	`local.get $x`	Copy the value stored in the variable `$x` and push it on top of the stack	`$a`, `$x`
46	`$a`, `$x`	`f64.mul`	Pop the top two values off the stack, multiply them, and push the result on the stack	`$a` × `$x`
47	`$a` × `$x`	`local.get $b`	Copy the value stored in the variable `$b` and push it on top of the stack	`$a` × `$x`, `$b`
48	`$a` × `$x`, `$b`	`local.get $y`	Copy the value stored in the variable `$y` and push it on top of the stack	`$a` × `$x`, `$b`, `$y`
49	`$a` × `$x`, `$b`, `$y`	`f64.mul`	Pop the top two values off the stack, multiply them, and push the result on the stack	`$a` × `$x`, `$b` × `$y`
50	`$a` × `$x`, `$b` × `$y`	`f64.add`	Pop the top two values off the stack, add them, and push the result on the stack	`$a` × `$x` + `$b` × `$y`
51	`$a` × `$x` + `$b` × `$y`	`local.tee $dxdt`	Copy the top value from the stack (without removing it) and store in the variable `$dxdt`	`$a` × `$x` + `$b` × `$y`
52	`$a` × `$x` + `$b` × `$y`	`local.get $delta_t`	Copy the value stored in the variable `$delta_t` and push it on top of the stack	`$a` × `$x` + `$b` × `$y`, `$delta_t`
53	`$a` × `$x` + `$b` × `$y`, `$delta_t`	`f64.mul`	Pop the top two values off the stack, multiply them, and push the result on the stack	(`$a` × `$x` + `$b` × `$y`) × `$delta_t`
54	(`$a` × `$x` + `$b` × `$y`) × `$delta_t`	`local.set $delta_x`	Pop the top value off the stack and store in the variable `$delta_x`	(empty)

The overall effect of lines 44-54 is to:

Compute $a × $x + $b × $y and store in $dxdt
Compute $dxdt × $delta_t and store it in $delta_x.

This illustrates the downside to WASM: it takes 11 lines of code to accomplish what most programming languages can do in two.

Euler step demo

The total function to do one step of Euler's method takes about 40 lines of WAT code. This JavaScript on this page is able to access the Euler step function. To the browser, this function is called eulerStep. It takes 7 numbers as arguments ($a, b, c, d, x(t), y(t),$ and $\Delta T$) and returns four numbers (estimates for $x(t), y(t), x'(t),$ and $y'(t)$. You can play with this function by opening your browser's JavaScript console and evaluating the function there, once the WASM is done loading:

JavaScript (console)
eulerStep(1, 2, 3, 4, 1, 1, 0.0625)

[1.1875, 1.4375, 3, 7]

On the other hand, if mucking around in your browser's JavaScript console isn't your thing, you can also access eulerStep via the following demo:

Demo is not ready yet. Loading WASM...

a =

b =

c =

d =

System:

dx/dt = a x + b y

dy/dt = c x + d y

x(0) =

y(0) =

Δ t =

t	x	y	dx/dt	dy/dt

Approximating π

One definition of π is that $t = \pi$ smallest positive real solution to the equation $\sin t = 0$. Earlier I mentioned that $x(t) = \sin t, y(t) = \cos t$ is the unique solution to the system of differential equations \[\begin{aligned} \frac{dx}{dt} & = y, \\ \frac{dy}{dt} &= -x \end{aligned}\] that satisfies the initial conditions $x(0) = 0, y(0) = 1$. My plan is to use Euler's method to approximate a solution to this system (with initial conditions), which means approximating $\sin t$ and $\cos t$, and looking for the first positive value of $t$ for which $\cos t \leq 0$, which will be our approximation of π. The smaller $\Delta t$ used, the better approximation of π we get.

Once the WASM has loaded successfully, click on the Run button to come up with different estimates for π using smaller and smaller values of $\Delta t$.

Our estimates are produced as IEEE 754 double-precision floating-point numbers. This 64-bit format represents numbers⁴ in the form \[(-1)^s \times \left(1 + f/2^{52}\right)\times 2^{e - 1023},\] using one bit for $s$, 11 bits for $e$, and 52 bits for $f$. For example, the number 3.25 is represented as

\[\begin{aligned} (-1)^0 \times \left(1 + \frac{2814749767106560}{2^{52}}\right) \times 2^{1024 - 1023} &= \left(1 + \frac{10 \times 2^{48}}{2^{52}} \right) \times 2 \\ &= \left(1 + 10/16\right) \times 2 \\ &= 3.25. \end{aligned}\]

For each approximation produced, click on the Show me the IEEE! to see the approximation as it's represented internally.

Loading the WASM...

k	Δt = 2^-k	Approx. for π	Error from `Math.PI`	Time (ms)	Num. values	Plot

Real and Complex Analysis, 3rd Edition by Walter Rudin (McGraw Hill 1987) ↩
Or other runtimes, but the browser is my focus here. ↩
In this table, the stacks are written from bottom to top. For example, the stack after line 45 is executed, written as $a, $x, has $a on the bottom and $x on the top. ↩
I'm ignoring values such as NaN (not a number) and $\pm\infty$, which can also be represented as IEEE double-precision floating-point numbers. ↩

6 7

Wed, 31 Dec 2025 12:30:00 -0600

Yeah, if you have or frequently interact with children, you are probably familiar with the nonsense saying six seven. Is this a ridiculously silly meme? Of course it is, but so is (for example) trying to trick people into watching a particular music video. Watching my first-grader experience her first meme is wonderful and I choose to celebrate it!

Therefore, without further ado, I present:

A version on its own page

Source code

In case you are interested, below is the source code for this animation.

Standalone HTML

This is the page linked above.

index.html HTML

doctype html>
<html lang="en">
  
  <head>
    <title>Six seventitle>
    <meta charset="utf-8" />
    <link href="reset.css" rel="stylesheet" />
    <link href="animate.css" rel="stylesheet" />
  head>

  <body>
    <svg
      width="800"
      height="500"
      xmlns="http://www.w3.org/2000/svg"
      id="67_animation"
      aria-label="The number 67 and the words 'six seven' in various fonts and colors move horizontally from right to left at differing speeds."
    >
      >
      <defs>
        <path
          d="M 0,0 v -40 a 50 50 180 0 0 -100 0 v 120
                a 50 50 180 0 0 100 0 v -19 h -20 v 14
                a 30 30 180 0 1 -60 0 v -20
                a 30 30 180 0 1 60 0 h 20 v -5 a 40 40 90 0 0 -40 -40
                q -40 0 -40 10 v -60 a 30 30 180 0 1 60 0 v 30 z"
          id="six"
        />
        <path
          d="M 0,0 h 150 v 25 c -30,0  -70,30 -70,70 v 30 h 50 v 20 h -50
                v 75 h -25 v -75 h -50 v -20 h 50 v -35 c 0,-30 30,-50 40,-60
                h -110 v -25 z  "
          id="seven"
        />
        <g id="six-seven">
          <use href="#six" x="100" y="90" style="transform: scale(0.2)" />
          <use href="#seven" x="135" y="2" style="transform: scale(0.2)" />
        g>

        <path
          id="disp_seg_vert"
          d="M 0,0 l 20,20 v 50 l -20,20 l -20,-20
                    v -50 l 20,-20 "
        />
        <g id="disp_seg_horiz">
          <use href="#disp_seg_vert" style="transform: rotateZ(-90deg)" />
        g>

        <g id="6_lcd">
          <use href="#disp_seg_horiz" x="5" y="-5" />
          <use href="#disp_seg_vert" />
          <use href="#disp_seg_vert" y="100" />
          <use href="#disp_seg_vert" x="100" y="100" />
          <use href="#disp_seg_horiz" x="5" y="95" />
          <use href="#disp_seg_horiz" x="5" y="195" />
        g>

        <g id="7_lcd">
          <use href="#disp_seg_horiz" />
          <use href="#disp_seg_vert" x="95" y="5" />
          <use href="#disp_seg_vert" x="95" y="100" />
        g>

        <g id="67_lcd">
          <use href="#6_lcd" />
          <use href="#7_lcd" x="150" />
        g>

        <pattern
          id="background_6_7"
          width="80"
          height="80"
          patternUnits="userSpaceOnUse"
          patternContentUnits="userSpaceOnUse"
        >
          <rect fill="black" x="0" y="0" width="80" height="80" />
          <use href="#six-seven" fill="#333" style="transform: scale(0.5)" />
          <use
            href="#67_lcd"
            fill="#333"
            x="400"
            y="25"
            style="transform: scale(0.1)"
          />
          <use
            href="#67_lcd"
            fill="#333"
            y="400"
            x="25"
            style="transform: scale(0.1)"
          />
          <use
            href="#six-seven"
            fill="#333"
            x="80"
            y="80"
            style="transform: scale(0.5)"
          />
        pattern>
        <text class="label1" id="67_1">67text>
        <text class="label2" id="67_2">67text>
        <text class="label3" id="67_3">67text>
        <text class="label1" id="67_4">six seventext>
        <text class="label2" id="67_5">six seventext>
        <text class="label3" id="67_6">six seventext>
        <text class="label4" id="67_7">67text>
        <text class="label5" id="67_8">67text>
        <text class="label4" id="67_9">six seventext>
        <text class="label5" id="67_10">six seventext>
        <text class="label6" id="67_11">67text>
        <text class="label6" id="67_12">six seventext>
        <rect
          class="background"
          id="bg"
          x="-80"
          y="-80"
          width="960"
          height="660"
          fill="url(#background_6_7)"
        />
      defs>
      <style>
        .label1 {
          font: 60px sans-serif;
        }
        .label2 {
          font: 120px serif;
        }
        .label3 {
          font: 90px monospace;
        }
        .label4 {
          font: 100px cursive;
        }
        .label5 {
          font: 75px fantasy;
        }
        .label6 {
          font: 130px system-ui;
        }
      style>
    svg>

    <script src="animation_67.js">script>
  body>
html>

Typescript

I wrote the main script in TypeScript. It needs to be transpiled down to JavaScript before being used.

animation_67.ts TypeScript

/* @filename: animation_67.ts
 *
 * Silly* six-seven animation
 *
 * Christopher Phan 
 * 2026-W01
 *
 * For Emmy
 *
 * SPDX-License-Identifier: MIT
 *
 * *This whole six-seven thing is silly, but so is finding 1000 creative ways to
 *  trick someone into watching the same Rick Astley video :-)
 *
 */


class AnimatedElt {
    // Represents a `use` element in an SVG that we will be moving around
    elt: SVGUseElement
    rootSvg: SVGElement

    constructor(elt: SVGUseElement, rootSvg: SVGElement) {
        this.elt = elt
        this.rootSvg = rootSvg
    }

    create(
        targetSVG: SVGElement,
        x: number,
        y: number,
        href: string,
        id: string,
        fill: string
    ): AnimatedElt {
        const elt = document.createElementNS('http://www.w3.org/2000/svg', 'use')
        elt.setAttribute('x', `${x}`)
        elt.setAttribute('y', `${y}`)
        elt.setAttribute('id', id)
        elt.setAttribute('fill', fill)
        elt.setAttribute('href', `#${href}`)
        targetSVG.appendChild(elt)
        return new AnimatedElt(elt, targetSVG)
    }


    move(dx: number, dy: number) {
        if (dx !== 0) {
            const currentX = this.elt.getAttribute('x')
        let newX = dx
            if (currentX !== null) {
                newX += parseFloat(currentX)
            }
            this.elt.setAttribute('x', `${newX}`)
        }
        const currentY = this.elt.getAttribute('y')
        if (dy !== 0) {
            let newY = dy
            if (currentY !== null) {
                newY += parseFloat(currentY)
            }
            this.elt.setAttribute('y', `${newY}`)
        }
    }
}

class Mover {
    velocityX: number
    velocityY: number
    item: AnimatedElt
    lastUpdate: Date

    constructor(velocityX: number, velocityY: number, item: AnimatedElt) {
        this.velocityX = velocityX
        this.velocityY = velocityY
        this.item = item
        this.lastUpdate = new Date()
    }

    create(
        targetSVG: SVGElement,
        x: number,
        y: number,
        href: string,
        id: string,
        fill: string,
        velocityX: number,
        velocityY: number
    ): Mover {
            const item = AnimatedElt.prototype.create(
                targetSVG, x, y, href, id, fill
            )
            return new Mover(velocityX, velocityY, item)
    }

    update() {
        const now = new Date()
        const elapsedTime = (now.valueOf() - this.lastUpdate.valueOf()) / 1000
        this.lastUpdate = now
        this.item.move(this.velocityX * elapsedTime, this.velocityY * elapsedTime)
    }
}

function randomColor(): string {
    const hue = Math.floor(Math.random() * 360)
    const sat = Math.random() * 100
    const light = Math.random() * 75 + 25
    return `hsl(${hue}, ${sat}%, ${light}%)`
}

/* The SVG should have the following elements defined:
 * - 67_lcd
 * - background_6_7 (pattern)
 * - 67_n for n = 1 to 12
 *
 * For example: https://chrisphan.com/67_animation/
 */

function createNew(n: number, targetSVG: SVGElement): Mover {
    const y = Math.random() * 500
    let c = n
    if (n > 13) {
        c = Math.floor(Math.random() * 14)
    }
    let href: string = ""
    if (c === 0) {
        href = "six-seven"
    } else if (c === 13) {
        href = "67_lcd"
    } else  {
        href = `67_${c}`
    }
    const id = `mover_${n}`
    const fill = randomColor()
    const velocityX = - (Math.random() * 50 + 50)
    return Mover.prototype.create(targetSVG, 850, y, href, id, fill, velocityX, 0)
}

function update(
    targetSVG: SVGElement,
    objects: Array<Mover>,
    active: Array<number>,
    startTime: Date,
    count: number,
    lastAdded: Date | null,
    inactive: Array<number>) {
    const now = new Date()
    const elapsedTime = (now.valueOf() - startTime.valueOf()) / 1000
    let sinceLastAdded = 1000000
    if (lastAdded !== null) {
        sinceLastAdded = (now.valueOf() - lastAdded.valueOf()) / 1000
    }
    let newCount = count
    let addedOne = false
    if (elapsedTime > newCount && sinceLastAdded > 0.5) {
        if (objects.length < 60) {
            const newThing = createNew(count, targetSVG)
            objects.push(newThing)
            active.push(objects.length - 1)
            addedOne = true
        } else if (inactive.length > 0) {
            const recycledIndex = inactive.shift()
            const recycled = objects[recycledIndex]
            recycled.item.elt.setAttribute('x', `${850}`)
            recycled.lastUpdate = new Date()
            active.push(recycledIndex)
            addedOne = true
        }
        if (addedOne) {
            newCount++
            lastAdded = new Date()
        }

    }
    let newActive: Array<number> = []
    active.forEach((idx) => {
        objects[idx].update()
        if (objects[idx].item.elt.getAttribute('id') === 'back') {
            if (parseFloat(objects[idx].item.elt.getAttribute('x')) > 80) {
                objects[idx].item.elt.setAttribute('x', '0')
            }
            if (parseFloat(objects[idx].item.elt.getAttribute('y')) > 80) {
                objects[idx].item.elt.setAttribute('y', '0')
            }
            newActive.push(idx)
        }
        else if (parseFloat(objects[idx].item.elt.getAttribute('x')) > -1000) {
            newActive.push(idx)
        } else {
            inactive.push(idx)
        }
    })

    window.requestAnimationFrame(
        () =>
        update(
            targetSVG,
            objects,
            newActive,
            startTime,
            newCount,
            lastAdded,
            inactive)
    )
}


const mainSvg = document.querySelector("svg[id='67_animation']") as SVGElement
let background = Mover.prototype.create(mainSvg, -80, -80, "bg", "back", "#333", 5, 5)
let objects: Array<Mover> = [background]
let active: Array<number> = [0]
const startTime = new Date()
update(mainSvg, objects, [0], startTime, 0, null, [])

Happy ISO Week Date New Year!

Mon, 29 Dec 2025 09:00:00 -0600

Happy New Year—under the ISO week date calendar, that is!

As I explained exactly 52 weeks ago, a year in the ISO week date calendar is made up of exactly 52 or 53 weeks, so that every week (Monday through Sunday) is in the same year. So even though today is 2025-12-29 in the Gregorian calendar (in ISO 8601), under the ISO week date calendar, today is 2026-W01-1 , i.e., the first day (Monday) of the first week of 2025.

Also, 2026 has 53 weeks in the ISO week date calendar. Most years only get 52, but any Gregorian year that beings on a Thursday, and any Gregorian leap year that begins on a Wednesday, will have 53 weeks in the ISO week date calendar.

In general, the year that a week falls in is determined by the Gregorian year of its Thursday. This week is assigned to 2026 because Thursday is in 2026; indeed, Thursday is the Gregorian New Year's Day.

Here is a simple Python script I wrote to make a table showing the start of each week in the ISO week date calendar:

iso_week_date_table.py Python

#!/usr/bin/env python3
"""Produce a Markdown table for the ISO week-date calendar."""

# Christopher Phan
# 2025-W52
# SPDX-License-Identifier: MIT

import datetime
import string
import sys


def iso_table(year: int) -> str:
    """Produce a Markdown table for the ISO week-date calendar."""
    outstr = "| ISO week date | ISO Gregorian |\n"
    outstr += "|:--------------|:--------------|\n"
    jan_1 = datetime.date(year, 1, 1)
    # If Jan 1 is a Thursday, or is Wednesday of a leap year, there is 53 weeks.
    # Otherwise 52 weeks:
    if jan_1.isoweekday() == 4 or (
        jan_1.isoweekday() == 3 and (year % 400 == 0 or year % 4 == 0 and year % 100)
    ):
        num_weeks = 53
    else:
        num_weeks = 52
    for week in range(num_weeks):
        d = datetime.date.fromisocalendar(year, week + 1, 1)
        outstr += f"| {year}-W{week+1:02d}-1    | {d:%Y-%m-%d}    |\n"
    return outstr


if __name__ == "__main__":
    if len(sys.argv) > 1 and (
        # strip out any non-digit characters from the first argument
        year_str := "".join(k if k in string.digits else "" for k in sys.argv[1])
    ):
        year = int(year_str)
    else:
        year = datetime.date.today().year
    print(iso_table(year))

And here is the resulting table (after the Markdown is rendered as HTML):

ISO week date	ISO Gregorian
2026-W01-1	2025-12-29
2026-W02-1	2026-01-05
2026-W03-1	2026-01-12
2026-W04-1	2026-01-19
2026-W05-1	2026-01-26
2026-W06-1	2026-02-02
2026-W07-1	2026-02-09
2026-W08-1	2026-02-16
2026-W09-1	2026-02-23
2026-W10-1	2026-03-02
2026-W11-1	2026-03-09
2026-W12-1	2026-03-16
2026-W13-1	2026-03-23
2026-W14-1	2026-03-30
2026-W15-1	2026-04-06
2026-W16-1	2026-04-13
2026-W17-1	2026-04-20
2026-W18-1	2026-04-27
2026-W19-1	2026-05-04
2026-W20-1	2026-05-11
2026-W21-1	2026-05-18
2026-W22-1	2026-05-25
2026-W23-1	2026-06-01
2026-W24-1	2026-06-08
2026-W25-1	2026-06-15
2026-W26-1	2026-06-22
2026-W27-1	2026-06-29
2026-W28-1	2026-07-06
2026-W29-1	2026-07-13
2026-W30-1	2026-07-20
2026-W31-1	2026-07-27
2026-W32-1	2026-08-03
2026-W33-1	2026-08-10
2026-W34-1	2026-08-17
2026-W35-1	2026-08-24
2026-W36-1	2026-08-31
2026-W37-1	2026-09-07
2026-W38-1	2026-09-14
2026-W39-1	2026-09-21
2026-W40-1	2026-09-28
2026-W41-1	2026-10-05
2026-W42-1	2026-10-12
2026-W43-1	2026-10-19
2026-W44-1	2026-10-26
2026-W45-1	2026-11-02
2026-W46-1	2026-11-09
2026-W47-1	2026-11-16
2026-W48-1	2026-11-23
2026-W49-1	2026-11-30
2026-W50-1	2026-12-07
2026-W51-1	2026-12-14
2026-W52-1	2026-12-21
2026-W53-1	2026-12-28

I wish you health, happiness, and prosperity in the next 53 weeks, and beyond.

Improved color contrast tool

Sat, 27 Dec 2025 12:15:00 -0600

I'm now indisputably middle-aged, and my eyesight isn't what it used to be. Specifically, I get very frustrated when text on screens contrasts poorly with background colors.

It's hard to read this, especially after 40.

A few years ago, I made an JavaScript color contract checker. You specify two colors and the widget will tell you how well the colors contrast, and whether the contrast meets accessibility guidelines.

The default mode of the tool is the original mode, which evaluates the contrast between two colors.

I just released an improved version of my tool. The upgrades include:

Permanent links to your comparison (in case you find a particular color combo amusing.)
A new mode allowing you to evaluate the contrast between one color and shades of another.
Some help pages: instructions for using the tool and a gentle explanation of colors in CSS.

A new mode compares a fixed color against a variety of shades of a different color.

On the technical side, I rewrote the code in TypeScript, a language that I have been meaning to learn.

Oops

Sat, 27 Dec 2025 12:15:00 -0600

This was supposed to go to this URL: /posts/2025-12-27_color_contrast_improved.html

Mastodon posts for 2025-W49

Mon, 8 Dec 2025 13:30:00 -0600

Quotes a profane word for fecal matter

Mastodon posts for 2025-W46 through W48

Sun, 30 Nov 2025 22:10:00 -0600

Anyone else keep accidentally going to "Missouri Digital News" when trying to get to MDN (Mozilla Developer Network) web docs? #programming #webdesign
— Chris Phan, " alt="An analog clock reading 8:48"class="small_clock" /> 2025-11-12 / 2025-W46-3T08:48:43-06:00 / 0x69149e4b

I made a thing: https://chrisphan.com/posts/2025-11-19_pwn_checker.html
It's a #TUI app queries the #API for @troyhunt 's Pwned Passwords service.
I wrote it in #python using @willmcgugan 's wonderful #Textual package.
Available on @codeberg, under an MIT license: https://codeberg.org/christopherphan/pwn_checker
— Chris Phan, " alt="An analog clock reading 12:18"class="small_clock" /> 2025-11-19 / 2025-W47-3T12:18:12-06:00 / 0x691e09e4

I spent way too much time making this diagram (#SVG by hand!) for my blog post.
— Chris Phan, " alt="An analog clock reading 12:31"class="small_clock" /> 2025-11-19 / 2025-W47-3T12:31:01-06:00 / 0x691e0ce5

On the bright side, I think I have a better (but still imperfect) handle on how #patterns work in #SVG now.
— Chris Phan, " alt="An analog clock reading 12:37"class="small_clock" /> 2025-11-19 / 2025-W47-3T12:37:54-06:00 / 0x691e0e82

Any other #Jeopardy fans out there?
WOW, Harrison Whitaker is a machine! The way he dominated Friday’s game was just amazing.
(I watch Jeopardy on Peacock so I have to wait a day longer than everyone else to watch episodes.)
— Chris Phan, " alt="An analog clock reading 12:23"class="small_clock" /> 2025-11-22 / 2025-W47-6T12:23:32-06:00 / 0x6921ffa4

Made this #holiday (and #probability) themed #JavaScript doohickey the other day. https://chrisphan.com/posts/2025-11-27_merry_christmas.html
— Chris Phan, " alt="An analog clock reading 4:21"class="small_clock" /> 2025-11-29 / 2025-W48-6T16:21:55-06:00 / 0x692b7203

Merry Christmas (probably)

Thu, 27 Nov 2025 23:37:00 -0600

Now that we are (almost, in my time zone) past Thanksgiving, Christmas decorations are appropriate. I will be putting up a Christmas tree in my physical home soon. Here (in my virtual home) is a probability-themed Christmas Tree for your enjoyment:

Fun with Unix time

Fri, 21 Nov 2025 22:40:00 -0600

1763786400

I think Unix time can be a lot of fun!

If you don't know what Unix time is, it's the way a lot of computers (anything running Unix or Linux, which includes modern Macs, iPads, iPhones, Android devices, and so much more) keep track of time, which is by counting the number of seconds since the start of the year 1970 UTC. The current Unix time is supposed to be written here, but maybe there's something wrong with the JavaScript.

However, I prefer to write Unix times in hexadecimal (base-16). The current Unix time in hexadecimal is —this page requires JavaScript, sorry. (Numbers written in hex are written with the letters a through f as well as the traditional decimal digits 0 through 9. Often times, hex numbers are prefixed with 0x to distinguish them from decimal numbers.)

By the way, my blog's current setup shows each post's time as Unix time hexadecimal, as well as the ISO Gregorian and week-date formats.

Some dates of note

To get us started, here are some dates of note in Unix time. At the time I am posting this, we have a ways to go until we need another hex digit (except we might have more trouble before that—more about that later).

Division by powers of 4

Dividing by 16 (and rounding down) a number in hexadecimal is equivalent to shifting all the digits one space to the right, similar to dividing by 10 when working in decimal. However, because 16 is a perfect square, we can do a kind of half-shift by dividing by four.

Rounding

In this table, we round down and up to the nearest multiple of a power of 16.

Adding/subtracting powers of 4

This is like changing one digit by 1 or 4.

Adding/subtracting multiples of something

Bascially, I'm changing the first digit here.

How long until an epoch fail?

Okay, the sad truth is that older programs store the Unix time as a 32-bit signed integer. Storing as a signed integer makes sense, as it allows dates before 1970 to be stored. But storing as a 32-bit signed integer means that we overflow into the negative a second after we hit 0x7ffffff. This is called the year 2038 problem.

Fortunately, people tend to design things to store Unix time as a 64-bit integer nowadays, which will last us roughly 300 billion years.

As of the writing of this page, about 12 years before the epochalypse, the heading above reads "How long until an epoch fail?" After 2038-01-19 03:14:07 UTC, the word "until" will be replaced with "since", thanks to the magic of JavaScript.

Pwn Checker

Wed, 19 Nov 2025 12:06:00 -0600

I made a new thing: Pwn Checker, a TUI app to check whether a password appears in Pwned Passwords' database of breached passwords.

You type a password into the app, and it queries the Pwned Passwords service for you.

Of course, most people would be justifiably reluctant to send their passwords to a third party. However, that is now how the Pwned Passwords service works. Rather than sending your password over the Internet, you first compute the SHA1 hash (a kind of digital fingerprint) of your password and then send only the first five hex digits of that hash to the API. The API then replies with all hashes in the database that have those same first five hex digits.

The user types in their password (in this case, secret) into their computer running the Pwn Checker app. The program calculates the SHA1 hash of the password, and then sends the first five of 40 hex digits of the hash to the Pwned Passwords API. The API responds with a list of every hash in the database with the same five initial digits, as well as the number of times the corresponding password appears in the database. The program then checks the returned hashes and determines whether the password was stolen.

I wrote Pwn Checker in Python, using Textual, a cool package for making TUI apps. Pwn Checker is now on Codeberg, and is offered under the MIT license.

Mastodon posts for 2025-W45

Wed, 12 Nov 2025 08:45:00 -0600

Spending my morning helping out with a @thecarpentries workshop. It's a great organization that teaches people to use tools such as #Python, #git, and the #Unix #shell
— Chris Phan, " alt="An analog clock reading 10:44"class="small_clock" /> 2025-11-06 / 2025-W45-4T10:44:31-06:00 / 0x690cd06f

Today in silly #python scripts
#unicode
str: Transform all ASCII letters to the corresponding outlined letters. if len(x) == 1: # Transform of a single character if x in string.ascii_uppercase: return chr(ord(x) + DIFFERENCE) elif x in string.ascii_lowercase: return chr(ord(x.capitalize()) + DIFFERENCE) else: return x else: # Call this function for every character in x return .join(transform(c) for c in x) if __name__ == __main__ : try: for line in fileinput.input(): print(transform(line), end= ) except KeyboardInterrupt: # Supress printing of traceback if user quits with ^C pass " id="pic_115503873815105571_0"/>
— Chris Phan, " alt="An analog clock reading 11:14"class="small_clock" /> 2025-11-06 / 2025-W45-4T11:14:06-06:00 / 0x690cd75e

To be clear, I don't think you should actually post this kind of output anywhere, because of the potential problems it poses to screen readers. #a11y
— Chris Phan, " alt="An analog clock reading 11:17"class="small_clock" /> 2025-11-06 / 2025-W45-4T11:17:52-06:00 / 0x690cd840

It's called trim in Rust. 🤷‍♂️
https://docs.python.org/3/library/stdtypes.html#str.strip
https://doc.rust-lang.org/std/primitive.str.html#method.trim
code = parts[1].trim() AttributeError: 'str' object has no attribute 'trim'. Did you mean: 'strip'?" id="pic_115511804514123617_0"/>
— Chris Phan, " alt="An analog clock reading 8:50"class="small_clock" /> 2025-11-07 / 2025-W45-5T20:50:59-06:00 / 0x690eb013

I assumed this was a conscious decision, since I imagine I am not the only person who gets trim/strip confused. But looking at the source code last night, I could not find an explicit reference to “trim” or “strip” in the suggestions code. Rather, I think it’s probably just a (really nice) coincidence, since they have the same three middle letters.
#python
— Chris Phan, " alt="An analog clock reading 11:54"class="small_clock" /> 2025-11-08 / 2025-W45-6T11:54:33-06:00 / 0x690f83d9

I’m going to learn how to ferment stuff this morning!
https://www.bluff.coop/event/fermentation-class-101/
I’ve been trying to ferment food recently but have had mixed success. I successfully made some sauerkraut and pickled some celery, but my attempts with carrots or mung bean sprouts have ended up moldy.
#WinonaMN #food #cooking
— Chris Phan, " alt="An analog clock reading 9:12"class="small_clock" /> 2025-11-08 / 2025-W45-6T09:12:30-06:00 / 0x690f5dde

With any luck, this will be a jar of #sauerkraut in a little while!
#WinonaMN #food
— Chris Phan, " alt="An analog clock reading 11:47"class="small_clock" /> 2025-11-08 / 2025-W45-6T11:47:34-06:00 / 0x690f8236

Mastodon posts for 2025-W44

Thu, 6 Nov 2025 13:00:00 -0600

I agree with @emilydoesastro : uv is really awesome for #python environment and package management. I have been using it a lot recently.
https://emily.space/posts/251023-uv
— Chris Phan, " alt="An analog clock reading 11:00"class="small_clock" /> 2025-10-31 / 2025-W44-5T23:00:17-05:00 / 0x690585d1

@emilydoesastro One neat thing that uv allows is for you to make standalone scripts that specify the external packages needed.
https://docs.astral.sh/uv/guides/scripts/
=3.11 # dependencies = [ requests , psycopg[binary] ] # /// # # SPDX-License-Identifier: MIT Build the U.S. Population Chains database. from __future__ import annotations __author__ = Christopher L. Phan, Ph.D. __copyright__ = Copyright 2025, Christopher L. Phan, Ph.D. __credits__ = {__author__} __date__ = 2025-10-29 __license__ = Copyright 2025 Christopher L. Phan, Ph.D. Screenshot cuts of there." id="pic_115472464475699837_0"/>
— Chris Phan, " alt="An analog clock reading 11:06"class="small_clock" /> 2025-10-31 / 2025-W44-5T23:06:17-05:00 / 0x69058739

Timestamp stored in microseconds, not seconds. Oops! 🤦🏽‍♂️ #python
timer, table = dump_pg(config, db, len(ident_list), timer) File /Users/chris/Documents/projects/sandbox_3/pop_chains_reproducible/./build.py , line 1172, in dump_pg + f INSERT INTO db_info VALUES ('{db.creation.isoformat()}'); File /Users/chris/Documents/projects/sandbox_3/pop_chains_reproducible/./build.py , line 572, in creation return datetime.datetime.fromtimestamp(self.timestamp) ValueError: year must be in 1..9999, not 55842094" id="pic_115483968908095307_0"/>
— Chris Phan, " alt="An analog clock reading 10:52"class="small_clock" /> 2025-11-02 / 2025-W44-7T22:52:01-06:00 / 0x690834f1

Let's hope they remove this limitation before the year 10000 🤣 #python
— Chris Phan, " alt="An analog clock reading 10:57"class="small_clock" /> 2025-11-02 / 2025-W44-7T22:57:16-06:00 / 0x6908362c

Ten most populous Census-designated places in the USA. (Data from 2020 U.S. Census) #census #demographics #SQL
— Chris Phan, " alt="An analog clock reading 11:50"class="small_clock" /> 2025-11-02 / 2025-W44-7T23:50:25-06:00 / 0x690842a1

44° North

Wed, 8 Oct 2025 23:15:00 -0500

There is a hotel being built along the Mississippi River in downtown Winona, Minnesota. It appears to be close to complete. When I saw the name, it reminded me about Winona and Eugene being at the same latitude!

The new Hotel 44° North being completed in downtown Winona.

ISO week-date calendar terminal app

Sat, 30 Aug 2025 15:30:00 -0500

As I've stated before, I love the ISO week-date calendar. One indication is that (in its current incarnation), this site shows dates in the ISO week-date format as well as the Gregorian ISO 8601/RFC 3339 format.

A few years ago, I wrote a simple Python script to produce ISO week-date calendars from the command line:

$ isocal --year 2021


                                     2021                                      
                            ISO week-date calendar                             
Wk Mo Tu We Th Fr Sa Su   │Wk Mo Tu We Th Fr Sa Su   │Wk Mo Tu We Th Fr Sa Su   
53 28 29 30 31│ 1  2  3   │18  3  4  5  6  7  8  9   │   ─────┘                 
   ───────────┘           │19 10 11 12 13 14 15 16 M │36  6  7  8  9 10 11 12   
01  4  5  6  7  8  9 10   │20 17 18 19 20 21 22 23   │37 13 14 15 16 17 18 19 S 
02 11 12 13 14 15 16 17 J │21 24 25 26 27 28 29 30   │38 20 21 22 23 24 25 26   
03 18 19 20 21 22 23 24   │     ┌──────────────────  │              ┌─────────  
04 25 26 27 28 29 30 31   │22 31│ 1  2  3  4  5  6   │39 27 28 29 30│ 1  2  3   
   ─────────────────────  │   ──┘                    │   ───────────┘           
05  1  2  3  4  5  6  7   │23  7  8  9 10 11 12 13   │40  4  5  6  7  8  9 10   
06  8  9 10 11 12 13 14   │24 14 15 16 17 18 19 20 J │41 11 12 13 14 15 16 17 O 
07 15 16 17 18 19 20 21 F │25 21 22 23 24 25 26 27   │42 18 19 20 21 22 23 24   
08 22 23 24 25 26 27 28   │           ┌────────────  │43 25 26 27 28 29 30 31   
   ─────────────────────  │26 28 29 30│ 1  2  3  4   │   ─────────────────────  
09  1  2  3  4  5  6  7   │   ────────┘              │44  1  2  3  4  5  6  7   
10  8  9 10 11 12 13 14   │27  5  6  7  8  9 10 11   │45  8  9 10 11 12 13 14   
11 15 16 17 18 19 20 21 M │28 12 13 14 15 16 17 18 J │46 15 16 17 18 19 20 21 N 
12 22 23 24 25 26 27 28   │29 19 20 21 22 23 24 25   │47 22 23 24 25 26 27 28   
           ┌────────────  │                    ┌───  │        ┌───────────────  
13 29 30 31│ 1  2  3  4   │30 26 27 28 29 30 31│ 1   │48 29 30│ 1  2  3  4  5   
   ────────┘              │   ─────────────────┘     │   ─────┘                 
14  5  6  7  8  9 10 11   │31  2  3  4  5  6  7  8   │49  6  7  8  9 10 11 12   
15 12 13 14 15 16 17 18 A │32  9 10 11 12 13 14 15 A │50 13 14 15 16 17 18 19 D 
16 19 20 21 22 23 24 25   │33 16 17 18 19 20 21 22   │51 20 21 22 23 24 25 26   
                 ┌──────  │34 23 24 25 26 27 28 29   │                 ┌──────  
17 26 27 28 29 30│ 1  2   │        ┌───────────────  │52 27 28 29 30 31│ 1  2   
   ──────────────┘        │35 30 31│ 1  2  3  4  5   │   ──────────────┘        
Wk Mo Tu We Th Fr Sa Su   │Wk Mo Tu We Th Fr Sa Su   │Wk Mo Tu We Th Fr Sa Su

I had intended to release this under an open-source license, but never got around to it. Well, today I updated the code and posted it to Codeberg.

The program can produce a color output.

Producing booklets from the command line

Mon, 28 Jul 2025 22:00:00 -0500

I’m writing this mostly for my own future reference, but perhaps it will be useful to someone else.

Today, I wanted to print out a PDF so that I could turn it into a booklet by folding the pages in half and stapling in the middle. The pages need to be in a different order. Suppose you have a PDF with $n$ pages, numbered 0 through $n - 1$. Then the first sheet out of your printer should have pages $n - 1$ and 0 on one side, and pages 1 and $n - 2$ on the other. The next sheet should have pages $n - 3$ and 2 on one side and pages 3 and $n - 4$ on the other, and so on.

At first, I considered manually rearranging the pages in the macOS Preview app. But this would be time-consuming and I wanted to have a script I could use in the future. I briefly considered writing my own program to achieve this, before thinking to myself “This has to be a solved problem.”

Indeed, it was. After a few seconds using a search engine¹, I found a some instructions from a Philadelphia-based zine publisher named Iffy Books. The guide points to the program pdfbook2. This program is included in many TeX distributions and was already on my machine.

When you go to print your booklet, be sure to print double-sided with short-edge binding.

I tend to use the same search engine as alternate-timeline James T. Kirk. ↩

Same latitude

Tue, 15 Jul 2025 18:20:00 -0500

The community where I lived the longest is Sandy, Oregon, where I grew up. In second and third place are, respectively, Winona, Minnesota and Eugene, Oregon. Winona is the town where I have made my home, having moved here almost 13 years ago. I lived in Eugene for nearly six years, between 2003 and 2009, when I attended graduate school at the University of Oregon.

Eugene and Winona are both at the same latitude, just north of 44° N. State boundary data from the U.S. Census Bureau’s TIGER/Line shapefiles. Great Lakes shoreline data from Great Lakes Environmental Research Laboratory

Winona and Eugene are very different places. Although both college towns, they have significant differences in size, climate, and vibe. However, one eerie similarity, which I did not notice until several years after moving to Winona, is that they are essentially at the same latitude, just above 44°N.

Maps of Eugene and Winona centered at the same latitude. Maps provided by OpenStreetMap.

What I found even more surprising was that my (former) home in Winona (where I lived from the summer of 2013 until late last year), and Fenton Hall (one of the two buildings that house the University of Oregon’s math department) are at the same latitude. That is to say, if you stood in my former front yard and headed due west for several thousand kilometers, you would eventually find yourself in front of Fenton Hall.

Left: Fenton Hall, University of Oregon, October 2005. Right: My former home in Winona, Minnesota. Both these buildings are at about 44.0457°N latitude.

Maps of University of Oregon campus and my former Winona neighborhood, centered at the same latitude. Maps provided by OpenStreetMap.

System time before Unix epoch?

Mon, 21 Apr 2025 15:00:00 -0500

On what operating systems is it possible to set the system time before 1970?

That's a question I had after reading this example from the Rust standard library documentation:

src/main.rs Rust


  use std::time::{SystemTime, UNIX_EPOCH};

  match SystemTime::now().duration_since(UNIX_EPOCH) {
      Ok(n) => println!("1970-01-01 00:00:00 UTC was {} seconds ago!", n.as_secs()),
      Err(_) => panic!("SystemTime before UNIX EPOCH!"),
  }

Here's an explanation of what's going on:

SystemTime is a struct that stores timestamp information.
SystemTime::now() returns the current reading of your computer's clock as a SystemTime.
If $x$ and $y$ are SystemTimes, then $y$.duration_since($x$) returns a Result:
- If $y$ occurs at the same time or after $x$, then duration_since will return Ok($n$), where $n$ is a Duration representing the amount of time between $x$ and $y$.
- If $y$ occurs before $x$, then duration_since will return Err($z$), where $z$ is a SystemTimeError.
UNIX_EPOCH is a constant of type SystemTime that represents the Unix epoch, that is, 1970-01-01T00:00:00 UTC, the exact start of the year 1970 in Coordinated Universal Time (which is roughly the same as Greenwich Mean Time).
So, SystemTime::now().duration_since(UNIX_EPOCH) returns either:
- Ok($n$), where $n$ is the duration between now and the Unix epoch, if the system clock is set to the Unix epoch or afterwards, or
- Err($z$), where $z$ is a SystemTimeError, if (for some bizarre reason) the system clock is set to before the Unix epoch.
If $n$ is a Duration, then $n$.as_secs() will return the number of seconds in $n$, rounded down to the nearest integer.

So the overall behavior of this code is:

If the system clock is set to the Unix epoch or afterwards, output to the standard output something like "1970-01-01 00:00:00 UTC was 1745265600 seconds ago!"
If the system clock is set before the Unix epoch (for some bizarre reason), panic (i.e., crash) with the error message "SystemTime before UNIX EPOCH!"

So, my question was, after reading this: On which operating systems is it possible to set the system clock to before the Unix epoch?

The reason I asked is that on Unix and Unix-like systems (such as MacOS, iOS, Android, Linux), the system time is actually represented internally as the time since the Unix epoch—this is usually referred to as "Unix time". For example, this post is timestamped to 2025-04-21T15:00:00-05:00 (April 21, 2015 at 3:00 p.m., in a timezone five hours behind UTC, which in my case is North America's Central Daylight Time), but that can also be represented in Unix time as 1745265600 seconds (or, in hexadecimal, 0x6806a3c0 seconds, which my blog is currently configured to display at the top of this post).

Now, Unix time is usually represented (in seconds¹) as a signed integer², making it possible to represent dates before 1970 (e.g., the birthday of anyone currently aged 56 and higher). However, a number of systems use the (reasonable) assumption that the current date is in 1970 or later. See, for example, this question from someone who is trying (and failing) to set their computer to a date before 1970.

A hint to the answer to my question is in a table on Wikipedia, which lists the epochs used by various operating systems and computer languages. For example, modern Windows systems use the beginning of the year 1601. I knew Windows used a different epoch, but I didn't realize it was before 1970 until I read this table.

Still, it would be interesting to see which of these operating systems actually allow the system time to be set that far in the past.

Leap seconds are ignored; if leap seconds were taken into account, it would be impossible to precisely specify future dates in Unix time more than about six months in advance. ↩
Early systems stored the Unix time as a signed 32-bit integer, which can represent the values between -2³¹ and 2³¹ - 1 (i.e., roughly ±2.1 billion), a timespan of roughly ±68 years from the Unix epoch. Unfortunately, this leads to the Year 2038 problem (the binary cousin of the Year 2000 problem), in which the latest datetimes you can represent are in January of 2038. Quoth Wikipedia,

Modern systems and software updates to legacy systems address this problem by using signed 64-bit integers instead of 32-bit integers, which will take 292 billion years to overflow—approximately 21 times the estimated age of the universe.

↩

About me

Tue, 14 Jan 2025 18:30:00 -0600

Mailing address: Chris Phan, PO Box 244, Winona MN 55987
I live in Winona, Minnesota, a beautiful college town on the Mississippi River, surrounded by bluffs, in southeastern Minnesota.
I have two wonderful children.
I spent the first 28 years of my life living in Oregon. Then I spent short amounts of time living in Illinois, Scotland, and Pennsylvania, before making my home in Minnesota.
I have a Ph.D. in mathematics from the University of Oregon. For a while, I was a full-time mathematics professor in a series of contingent academic jobs, most recently at a state university. Then I spent six years as a stay-at-home dad. Recently, I have been teaching part time. See my math page for information about my work in mathematics.

See also my disclaimer.
I volunteer for the Minnesota Democratic-Farmer-Labor Party, including service as a party officer. Again, see also my disclaimer.
Pronouns: he/him/his

Turbofish

Mon, 13 Jan 2025 16:00:00 -0600

Today, I found this really cool site: turbo.fish

The turbofish is a strange-looking piece of syntax in the Rust programming language. It takes the form ::, and is used to convey type information to the compiler. Consider line 559 of the following (taken from the program I currently use to produce this site):

src/code_highlight.rs Rust


                let lineno_start: Option<usize> = match specs.get(2) {
                    Some(&"") => None,
                    Some(w) => w.parse::<usize>().ok(),
                    None => None,
                };

The turbofish :: on line 559 tells the compiler that we want to attempt to parse the string slice w as a usize (pointer-sized unsigned integer).

I was aware of this terminology, but not its origin. As explained on the turbo.fish site, it was the whimsical coinage of a software developer named Anna Harren, who used it in a Reddit thread. The term caught on and now it's in official documentation and such. Alas, Harren passed away in 2021, but in the Rust source code, in a file named bastion-of-the-turbofish.rs, there remains a memorial to Harren.

Happy ISO Week-Date New Year!

Mon, 30 Dec 2024 14:20:00 -0600

Happy New Year!

Of course, under the Gregorian calendar, the year doesn't start until Wednesday. But under the ISO week date calendar, today is the first day of 2025.

In our familiar Gregorian calendar, a year either has 365 or 366 days, split into 12 months, each of which has between 28 and 31 days. In the ISO week date calendar, a year is split into 52 or 53 weeks, each of which starts on Monday and has seven days.

Today's date in the ISO week date calendar is 2025-W01-1, i.e., the first day (Monday) of the first week of 2025.

Here are some other example dates in the ISO week date calendar:

Typical American rendering	ISO Gregorian	ISO week date
Monday, December 30, 2024	2024-12-30	2025-W01-1
Tuesday, December 29, 1998	1998-12-29	1998-W53-2
Wednesday, October 21, 2015	2015-10-21	2015-W43-3
Thursday, February 3, 1983	1983-02-03	1983-W05-4
Friday, April 22, 2005	2005-04-22	2005-W16-5
Saturday, July 11, 2020	2020-07-11	2020-W28-6
Sunday, January 3, 2010	2010-01-03	2009-W53-7

In the ISO week date calendar, every week (starting on Monday and ending on Sunday) is in the same year. For most weeks, the entire week is in the same Gregorian year, so it is assigned to that year in the ISO week date calendar. However, any week containing both December 31 and January 1 (such as the current week) will be split between two Gregorian years. These weeks are assigned to the year with the most days in that week. For example, our current week has two days in 2024 (Monday and Tuesday), and five days in 2025 (Wednesday through Sunday), so in the ISO week date calendar, this is the first week of 2025. (Equivalently, each week is assigned to the same ISO week date year as the Gregorian year of its Thursday.)

Today is an example of a day whose ISO week date year is ahead of its Gregorian year. On the other hand, on the table above, we see that January 3, 2010 is assigned to 2009.

There are many ways to get the current ISO week date, or convert between Gregorian and week date. For example, at any Unix or Unix-like shell, you can do this:

Bash
date +%G-W%V-%u

2025-W01-1

For most interfaces that understand strptime/strftime codes, like the version of date on my Mac, %G is the ISO week date year, %V is the ISO week date week, and %u is the day of the week. For example, in Python:

Python (REPL)
from datetime import date
date.strftime(date.today(), "%G-W%V-%u")

'2025-W01-1'

Python's datetime.date objects have methods to convert between Gregorian and ISO week date:

Python (REPL)
from datetime import date
date(1983, 2, 3).isocalendar()

datetime.IsoCalendarDate(year=1983, week=5, weekday=4)

date.fromisocalendar(2025, 23, 7)

datetime.date(2025, 6, 8)

The Unix command ncal can be used to print a calendar with week numbers:

Bash
ncal -w 5 2025

    May 2025
Mo     5 12 19 26
Tu     6 13 20 27
We     7 14 21 28
Th  1  8 15 22 29
Fr  2  9 16 23 30
Sa  3 10 17 24 31
Su  4 11 18 25
   18 19 20 21 22

And I wrote a Python program that prints nice year calendars with ISO week numbers:

Bash
isocal -y 2025


                                     2025
                            ISO week-date calendar
Wk Mo Tu We Th Fr Sa Su   │Wk Mo Tu We Th Fr Sa Su   │Wk Mo Tu We Th Fr Sa Su
01 30 31│ 1  2  3  4  5   │   ────────┘              │37  8  9 10 11 12 13 14
   ─────┘                 │19  5  6  7  8  9 10 11   │38 15 16 17 18 19 20 21 S
02  6  7  8  9 10 11 12   │20 12 13 14 15 16 17 18 M │39 22 23 24 25 26 27 28
03 13 14 15 16 17 18 19 J │21 19 20 21 22 23 24 25   │        ┌───────────────
04 20 21 22 23 24 25 26   │                    ┌───  │40 29 30│ 1  2  3  4  5
                 ┌──────  │22 26 27 28 29 30 31│ 1   │   ─────┘
05 27 28 29 30 31│ 1  2   │   ─────────────────┘     │41  6  7  8  9 10 11 12
   ──────────────┘        │23  2  3  4  5  6  7  8   │42 13 14 15 16 17 18 19 O
06  3  4  5  6  7  8  9   │24  9 10 11 12 13 14 15 J │43 20 21 22 23 24 25 26
07 10 11 12 13 14 15 16 F │25 16 17 18 19 20 21 22   │                 ┌──────
08 17 18 19 20 21 22 23   │26 23 24 25 26 27 28 29   │44 27 28 29 30 31│ 1  2
                 ┌──────  │     ┌──────────────────  │   ──────────────┘
09 24 25 26 27 28│ 1  2   │27 30│ 1  2  3  4  5  6   │45  3  4  5  6  7  8  9
   ──────────────┘        │   ──┘                    │46 10 11 12 13 14 15 16 N
10  3  4  5  6  7  8  9   │28  7  8  9 10 11 12 13   │47 17 18 19 20 21 22 23
11 10 11 12 13 14 15 16 M │29 14 15 16 17 18 19 20 J │48 24 25 26 27 28 29 30
12 17 18 19 20 21 22 23   │30 21 22 23 24 25 26 27   │   ─────────────────────
13 24 25 26 27 28 29 30   │              ┌─────────  │49  1  2  3  4  5  6  7
     ┌──────────────────  │31 28 29 30 31│ 1  2  3   │50  8  9 10 11 12 13 14
14 31│ 1  2  3  4  5  6   │   ───────────┘           │51 15 16 17 18 19 20 21 D
   ──┘                    │32  4  5  6  7  8  9 10   │52 22 23 24 25 26 27 28
15  7  8  9 10 11 12 13   │33 11 12 13 14 15 16 17 A │           ┌────────────
16 14 15 16 17 18 19 20 A │34 18 19 20 21 22 23 24   │01 29 30 31│ 1  2  3  4
17 21 22 23 24 25 26 27   │35 25 26 27 28 29 30 31   │   ────────┘
           ┌────────────  │   ─────────────────────  │
18 28 29 30│ 1  2  3  4   │36  1  2  3  4  5  6  7   │
Wk Mo Tu We Th Fr Sa Su   │Wk Mo Tu We Th Fr Sa Su   │Wk Mo Tu We Th Fr Sa Su

I really should release the source code for this.

Happy New Year!

Some personal news

Sun, 8 Sep 2024 14:35:00 -0500

This is probably only pertinent if you know me personally: about a month ago, my wife of over 13 years served me with a divorce petition. We are now going through the legal process of dissolving our marriage.

IMPORTANT message for MathJax users

Tue, 25 Jun 2024 22:20:00 -0500

If you have ever used MathJax (an awesome tool that makes it easy to display mathematics on your website), you may have copied the following into your HTML, following the directions on their website:

HTML

<script src="https://polyfill.io/v3/polyfill.min.js?features=es6">script>
<script id="MathJax-script" async src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-mml-chtml.js">script>

If you did that, you are going to want to delete that first line immediately. This loads a piece of JavaScript from the domain polyfill.io, and as Simon Willison put it

if you are loading assets from the cdn.polyfill.io domain you need to remove that right now: the new owners of the domain (as of a few months ago) appear to be using it to serve malicious JavaScript.

(Yes, the code from the MathJax site loads polyfill.min.js from polyfill.io, not cdn.polyfill.io, but the warning still applies.)

Willison further explains that the purpose of the polyfill.io service

was to serve up a set of JavaScript polyfills - pieces of code that implemented missing web platform features for older browsers - dynamically, based on the incoming user-agent. This required a CDN that varied its output dynamically based on the user-agent, hence the popularity of the single hosted service.

Andrew Betts, the original author of the service, has been warning people to move off it since February 20[2]4 [...]

He now works for Fastly, which started offering a free polyfill-fastly.io alternative in February. Andrew says you probably don't need that either, given that modern browsers have much better compatibility than when the service was first introduced over a decade ago.

I use MathJax a lot on this site, which allows me to show mathematical content such as \[\int_{-1}^1 \sqrt{1 - x^2}\; dx\] or \[\mu: V^{\otimes j} \otimes V^{\otimes k} \rightarrow V^{\otimes (k + j)}.\]

Every page with MathJax content on this site (and there are quite a few) contained the two lines of HTML displayed at the start of this post. I removed the first line (without specifying an alternative) and everything seems to be working fine.

Minnesota's new flag: from scratch

Tue, 30 Apr 2024 12:40:00 -0500

A new flag

Minnesota is soon getting a new state flag.

New design for Minnesota's flag, to be adopted 2024-05-11.

In May 2023, the Minnesota Legislature passed, and the governor signed, HF 1830, a state government finance bill that also created a commission to redesign Minnesota's state flag and state seal. The previous flag design uses the classic "seal on a bedsheet" arrangement decried by vexillologists.¹ Seals and flags have opposite purposes: seals are supposed to be complicated while flags are supposed to be simple and recognizable from a distance.²

The previous flag design (Public Domain, from Wikipedia)

In the second half of 2023, the State Emblems Redesign Commission invited members of the public to submit proposed designs for Minnesota's new flag and seal. Many designs were proposed. The winning proposal was proposal F1953, designed and submitted by Andrew Prekker of Luverne, Minnesota. The commission adopted a simplified version of Prekker's design this past December. This design will be adopted on May 11, the 166th anniversary of Minnesota statehood. The new design is thought to be one of the best in the nation.

In this post, I will explain how to implement the design of this flag as a short manually-coded SVG file.

The specifications

The new flag as a very simple design, specified by the State Emblems Redesign Commission in its official report (p. 21). You can make the flag out of three shapes:

Official specification from the State Emblems Redesign Commission

A water blue (#52c9e8) rectangle. Two sides of the rectangle have length $n$ and the other two sides have length $5n/3$.
A night sky blue (#002d5d) pentagon with three sides lying on the edges of the rectangle: one side that is the entire edge of the rectangle with length $n$, and the two neighboring perpendicular sides that have a length of $14n/15$. The last two sides meet at a vertex halfway between the long edges of the rectangle and a distance $13n/20$ away from the opposite edge of the pentagon.

Specifications for the water blue and night blue portions of the flag.
A white eight-pointed star, which is a regular octogram. The center of the star is halfway between the long edges of the rectangle, and a distance of $21n/60$ from the short edge of the rectangle that is also an edge of the pentagon. The points of the star are eight evenly spaced points along a circle of radius $11n/60$, with two points on a line perpendicular to the long sides of the rectangle and two points on a line perpendicular to the short sides. An outline of this polygon can be constructed by connecting each of these eight points with the points three away.

the eight points of the star are equally spaced around the circle of radius $11n/60$.

The center of the star has a distance of $21n/60$ from the left edge and is centered vertically.

Coding the flag

We can use these descriptions to write our SVG file.

Outside element

We begin with the outside svg element, which defines the viewport. Since the flag sits inside of a $n \times 5n/3$ rectangle, we'll set $n = 300$, so that the flag's dimensions are 500 × 300.

Typically, in computer graphics, the $y$ coordinates increase as you go down, which is the opposite of the convention used in mathematics. So, our viewport's upper-left corner will be at $(0, 0)$, and the lower right corner will be at $(500, 300)$. This is expressed in SVG with the attribute viewbox="0 0 500 300" in the svg element.

mn_flag.svg SVG

<svg viewbox="0 0 500 300" xmlns='http://www.w3.org/2000/svg'>
svg>

Water blue rectangle

The next element to add is a water blue (#52c9e8) rectangle with the same dimensions as the viewport. We'll use the rect element for this. We insert the rect inside the svg element.

mn_flag.svg SVG

<svg viewbox="0 0 500 300" xmlns='http://www.w3.org/2000/svg'>
  <rect x="0" y="0" width="500" height="300" stroke="none" fill="#52c9e8" />
svg>

Night sky blue pentagon

The next element to add is the night-sky blue (#002d5d) pentagon. For this, we'll use the polygon element.³ Setting $n = 300$, we see that the points of the vertices of the pentagon are at the following coordinates: \[(0, 0), (280, 0), (195, 150), (280, 300), (0, 300).\] So, we insert after the rect:

mn_flag.svg SVG


  <polygon points="0,0 280,0 195,150 280,300 0,300"
    stroke="none" fill="#002d5d" />

White star

Last, and most tricky, is the eight-pointed star that goes inside the pentagon. To get these points, we are going to use some trigonometry. If a circle has radius $r$ and center at the origin, then each point $P$ on the circle has the form \[P = (r \cos \theta, r \sin \theta),\] where $\theta$ is the angle between the positive $x$-axis and the ray starting at the center of the circle and passing through $P$. For a circle with arbitrary center $(h, k),$ we translate to get the parameterization \[P = (h + r \cos \theta, k + r \sin \theta).\]

Parameterization of an arbitrary point on a circle

In our case, the points of the star sit on a circle with center $105, 150)$ and radius 55. There are eight points equally spaced around the circle, with two points on the same horizontal line as the center, so the appropriate angles are 0°, 45°, 90°, 135°, and so on—or, in radians, $0, \pi/4, \pi/2, 3\pi/4, \dots$. However, we create the star by attaching each point to a point three away, so the first point has an angle of $0$, the next an angle of $3\pi/4$, the next an angle of $3\pi/2$, etc. Specifically, the $j$th point will have an angle of $3j\pi/4$, and so the $j$th point will have coordinates \[(105 + 55 \cos(3j\pi/4), 150 + 55 \sin(3j\pi/4)).\]

We could compute these by hand (using knowledge of trigonometry and a decimal approximation for $\sqrt{2}/2$), but a short Python script is perfect for this.

star_points.py Python

"""Calculate the points of the star on the Minnesota flag"""
import math

center_x = 105
center_y = 150
r = 55

theta = [3 * k * math.pi / 4 for k in range(0, 9)]
p = [(center_x + r * math.cos(k), center_y + r * math.sin(k)) for k in theta]

for pt in p:
    print(f"{pt[0]:08.8f},{pt[1]:08.8f}")

We run this script to get the coordinates:

Bash
python3 star_points.py

00000000,150.00000000
10912703,188.89087297
00000000,95.00000000
89087297,188.89087297
00000000,150.00000000
89087297,111.10912703
00000000,205.00000000
10912703,111.10912703
00000000,150.00000000

Now, we use the polyline SVG element, inserting the following after the pentagon in our SVG file:

mn_flag.svg SVG


  <polyline points="
    160.00000000,150.00000000
    66.10912703,188.89087297
    105.00000000,95.00000000
    143.89087297,188.89087297
    50.00000000,150.00000000
    143.89087297,111.10912703
    105.00000000,205.00000000
    66.10912703,111.10912703
    160.00000000,150.00000000"
    stroke="none" fill="white" />

Final result

The final result is this short SVG file:

mn_flag.svg SVG

<svg viewbox="0 0 500 300" xmlns='http://www.w3.org/2000/svg'>
  <rect x="0" y="0" width="500" height="300" stroke="none" fill="#52c9e8" />
  <polygon points="0,0 280,0 195,150 280,300 0,300"
    stroke="none" fill="#002d5d" />
  <polyline points="
    160.00000000,150.00000000
    66.10912703,188.89087297
    105.00000000,95.00000000
    143.89087297,188.89087297
    50.00000000,150.00000000
    143.89087297,111.10912703
    105.00000000,205.00000000
    66.10912703,111.10912703
    160.00000000,150.00000000"
    stroke="none" fill="white" />
svg>

And it looks like this:

What a beautiful flag!

Footnotes

The seal in question was also considered distasteful by many for its celebration of settler-colonialism, and so the same commission that redesigned the flag also came up with a new seal. But even if the seal had been perfectly fine, as the new seal is, putting your state seal on a flag is a terrible flag design. ↩
The North American Vexillological Association has identified five principles of flag design, several of which the old design clearly violates. ↩
We could have used a polygon for the sky blue rectangle as well. ↩

New look!

Thu, 25 Apr 2024 22:05:00 -0500

This site has a new look.

Also, I'm using a new static site generator: a custom one I wrote for myself. I was inspired to write my own SSG by Arne Bahlo. It was a nontrivial amount of work.

Note about tweets

Thu, 25 Apr 2024 14:48:00 -0500

Some of the posts here include reposts from the social media site previously known as Twitter, many from a now-deleted account of mine. I have tried to replace these with links to the Wayback Machine, when the material in question is available there. Alas, in some cases the material is not (in which case, I will link to this page).

Happy π Day!

Wed, 13 Mar 2024 23:20:00 -0500

Check out my celebration of π Day, 2024!

(Also available as a GitHub gist.)

PGP key

Tue, 27 Feb 2024 13:26:00 -0600

Made a link page

Sun, 16 Jul 2023 14:12:00 -0500

I made a link page, to add as a link to social media, etc.

Link page

Color contrast checker

Sat, 15 Jul 2023 11:05:00 -0500

I made this color contrast checker. You can specify colors either by entering hex triplets (e.g. #00ffdd) or using your browser's built-in color-selecting widget. It will compute the contrast ratio and determine if the color contrast meets the WCAG 2.1 accessibility success criterion 1.4.3 (AA) or 1.4.6 (AAA).

Now on Hackyderm

Tue, 11 Jul 2023 21:30:00 -0500

I have an account on the Mastodon instance Hackyderm now!

@chrisphan@hackyderm.io

Code to make treemap

Thu, 6 Jul 2023 18:00:00 -0500

Earlier this week I posted about this treemap I made.

(Interactive version)

Today, I have placed the code to produce this treemap in a Codeberg repo. It's not the most beautiful code in the world. (If I were really ambitious, I'd turn it into a proper library.)

This is America

Tue, 4 Jul 2023 23:54:00 -0500

This is a treemap of every county (or county equivalent) in the United States, where the area of each rectangle is proportional to the population of the corresponding county.

Counties that are part of the same core-based statistical area are grouped together. (These have a thicker green border.) Moreover, core-based statistical areas that belong to the same combined statistical area are grouped together. (CSAs have an even-thicker red border.)

For example, the red rectangle in the upper-left corner represents the New York-Newark, NY-NJ-CT-PA CSA:

Check out the interactive version, where you can mouse over each rectangle and get information about that county and any statistical area of which it is part.

Later, I will post the code I used to make this treemap.

Silly Decorator #2: @plus_one

Thu, 14 Apr 2022 14:22:00 -0500

It's like a off-by-one error on your wedding day.

Python (REPL)
from plusonedecorator import plus_one
@plus_one
def cost(quantity: int, marginal_cost: float) -> float:
    """Compute the cost of production."""
    return quantity * marginal_cost

cost(1000, 0.25)

251.0

@plus_one
def invite(name: str, pronoun: str) -> str:
    """Decide to invite someone to your wedding."""
    return f"Let's be sure to invite {name} and {pronoun}."

invite("Emilia", "her")

"Let's be sure to invite Emilia and her plus one."

invite("Maxwell", "his")

"Let's be sure to invite Maxwell and his plus one."

invite("Taylor", "their")

"Let's be sure to invite Taylor and their plus one."

(Source)

Silly Decorator #1: @a_hole

Wed, 23 Mar 2022 17:00:00 -0500

I'm like a pâtissier, except I decorate functions instead of cakes, and I decorate with U+1F4A9 instead of frosting.

Python (REPL)
from aholedecorator import a_hole
@a_hole
def hello(name: str = "friend") -> str:
    return f"Hello {name}, I hope you are well."

hello()

'Hello friend, I hope you are well.'

hello()

Traceback (most recent call last):
  File "", line 1, in 
  File "/Users/chris/Documents/projects/gists/a_hole_decorator/aholedecorator.py", line 70, in new_func
    raise AHoleError(choice(_MESSAGES))
aholedecorator.AHoleError: I'm like a pâtissier, except I decorate functions
  instead of cakes, and I decorate with U+1F4A9 instead of frosting.

(Source)

Hexagonal snowflakes

Mon, 13 Dec 2021 23:54:00 -0600

A hexagonal paper snowflake

Remember back in elementary school, where you made paper snowflakes by folding a square piece of paper into a small triangle and then cut notches into it? One thing that always bothered me is that you end up with a square snowflake, whereas real snowflakes are hexagonal. Recently, when I was doing craft projects with my daughter, I wondered if I could come up with a way to make hexagonal snowflakes.

The classic square paper snowflake. Real snowflakes are hexagonal, not quadrilateral.

After some trial and error, here’s what I came up with. (I suspect that this construction is already known, e.g., by people who have studied origami.)

In these diagrams, green lines represent the folds to make in that step, white lines represent pre-existing creases, and purple lines represent the cuts to make in that step.

First, fold a piece of paper in half lengthwise.

Next, fold the paper in half lengthwise, again.

Unfold the paper. The paper should now be divided into four equal strips lengthwise. Fold the paper in half widthwise.

Next, fold the paper in half again, along the first crease you made. Orient the resulting twice-folded paper so that the folded edges are on the top and right, and the paper's original edge are on the bottom and left.

Now, fold over the top left corner, so that the resulting crease starts at the top-right corner, and so the corner of the resulting fold touches the middle crease.

The corner of the fold should touch the crease.

Flip the paper around and repeat the same move. Then unfold once, so that you have a half-sheet with two corners down, as in the diagram. Note that it's important that the two creases meet in the middle.

When you fold the second flap, its top should line up with the top of the first flap.

Now, make two new folds. Each of these folds will go through the corner of the previous folds, and the new crease should be perpendicular to the diagonal folded edge. Fold these flaps back. You want the top edge of the flaps on the back to line up with the top diagonal folded edge of the rest of the paper.

The fold should go through the corner of the flap.

The top edges of the two sides of the fold should line up.

Now, make a crease between the two bottom corners of the flaps (parallel to the bottom edge of the paper and perpendicular to the middle crease). Then cut along that crease.

When you make this crease, the part you fold down will fit snugly.

Next cut along the creases formed by folding the two flaps back.

Unfold the two flaps. At this point, you should have the bottom half of a hexagon. (If you unfolded completely, you would have a regular hexagon.) Fold the resulting shape into an equilateral triangle, by making new folds from each bottom corner to the top of the middle crease.

You don't need to unfold the entire paper at this point, but if you do, it will be a regular hexagon—or at least a close approximation.

Fold the resulting triangle in half along the middle crease to make a right triangle. Cut out notches from the sides and corners as desired.

Unfold and you will have a hexagonal snowflake!

I'm sure some chemist will come along shortly and point out how my snowflake is still unrealistic. 🤣

mypy sum error

Thu, 28 Oct 2021 17:40:00 -0500

Recently, I've been exploring the type hints functionality in Python. The other day, I ran across what I think is a bug (already known) in mypy.

Here is a minimal working example of the issue I came across:

example.py Python

from fractions import Fraction

my_value = 1 + sum([Fraction(k + 1, 5 ** k) for k in range(5)])
print(f"The result is {my_value}.")

This program runs as you would expect.

Bash
python example.py

The result is 64/25.

However, the type-check fails!

Bash
mypy example.py

example.py:3: error: List comprehension has incompatible type List[Fraction]; expected List[int]
Found 1 error in 1 file (checked 1 source file)

At this point, I was really confused, but after searching for the error, I came across an issue on GitHub that reported this. In the discussion, someone explained that the problem is that mypy isn't taking into account the __radd__ method. (In general, x + y is shorthand for x.__add__(y). However, if x.__add__ doesn't know how to deal with y, then Python tries to use y.__radd__(x) instead.) Following the example in the discussion there, I modified the program as follows:

example.py Python


my_value = Fraction(1) + sum([Fraction(k + 1, 5 ** k) for k in range(5)])

The modified version type-checked okay.

At this point, I decided to come up with what my doctoral advisor would call a "Toy Example":

example.py Python

from __future__ import annotations 
# Allow self-referential type hints
from typing import Union

class A:
    def __init__(self: A, value: int) -> None:
        self.value = value

    def __repr__(self: A) -> str:
        return f"{self.__class__.__name__}({self.value!r})"

    def __add__(self: A, other: Union[A, int]) -> A:
        print(f"Calling {self!r}.__add__({other!r})")
        if isinstance(other, A):
            return self.__class__(self.value + other.value)
        elif isinstance(other, int):
            return self.__class__(self.value + other)
        else:
            return NotImplemented

    def __radd__(self: A, other: int) -> A:
        print(f"Calling {self!r}.__radd__({other!r})")
        if isinstance(other, int):
            return self.__class__(self.value + other)
        else:
            return NotImplemented

if __name__ == "__main__":
    values_to_sum = [A(k) for k in range(1, 5)]
    print(f"Computation #1: {sum(values_to_sum)=}")
    print(f"Computation #2: {A(5) + sum(values_to_sum)=}")
    print(f"Computation #3: {5 + sum(values_to_sum)=}")
    print(f"Computation #4: {sum(values_to_sum) + A(5)=}")
    print(f"Computation #5: {sum(values_to_sum) + 5=}")
    print(f"Computation #6: {A(5) + A(6)=}")
    print(f"Computation #7 {5 + A(6)=}")
    print(f"Computation #8 {A(5) + 6=}")

Then it executes just fine, but mypy is upset about line 33, in which A(10).__radd__(5) is called. (In the output below, the Calling ... text is output before the result of the computation.)

Bash
python3 example.py

Calling A(1).__radd__(0)
Calling A(1).__add__(A(2))
Calling A(3).__add__(A(3))
Calling A(6).__add__(A(4))
Computation #1: sum(values_to_sum)=A(10)
Calling A(1).__radd__(0)
Calling A(1).__add__(A(2))
Calling A(3).__add__(A(3))
Calling A(6).__add__(A(4))
Calling A(5).__add__(A(10))
Computation #2: A(5) + sum(values_to_sum)=A(15)
Calling A(1).__radd__(0)
Calling A(1).__add__(A(2))
Calling A(3).__add__(A(3))
Calling A(6).__add__(A(4))
Calling A(10).__radd__(5)
Computation #3: 5 + sum(values_to_sum)=A(15)
Calling A(1).__radd__(0)
Calling A(1).__add__(A(2))
Calling A(3).__add__(A(3))
Calling A(6).__add__(A(4))
Calling A(10).__add__(A(5))
Computation #4: sum(values_to_sum) + A(5)=A(15)
Calling A(1).__radd__(0)
Calling A(1).__add__(A(2))
Calling A(3).__add__(A(3))
Calling A(6).__add__(A(4))
Calling A(10).__add__(5)
Computation #5: sum(values_to_sum) + 5=A(15)
Calling A(5).__add__(A(6))
Computation #6: A(5) + A(6)=A(11)
Calling A(6).__radd__(5)
Computation #7 5 + A(6)=A(11)
Calling A(5).__add__(6)
Computation #8 A(5) + 6=A(11)

mypy example.py

example.py:33: error: Argument 1 to "sum" has incompatible type "List[A]"; expected "Iterable[int]"
Found 1 error in 1 file (checked 1 source file)

If you remove line 33 in the script (the one that evaluates 5 + sum(values_to_sum)), then mypy has no problem!

Bash
mypy example.py

Success: no issues found in 1 source file

Note that line 33 isn't the only place that __radd__ ends up being called. It's called every time sum(values_to_sum) is evaluated (since you start each sum with a value of the integer 0). It's also called in line 37, in the computation of 5 + A(6). These other invocations of __radd__ do not mess up mypy. It's only messed up when you __radd__ the result of a sum.

SArCAsTic cApiTalIZatioN GeNerATOr

Fri, 26 Mar 2021 14:54:07 -0500

Here's another fun thing I made.

GitHub repo

Happy π Day!

Sun, 14 Mar 2021 19:27:05 -0500

I spent way too much time working on this. Enjoy!

What size are national legislatures?

Wed, 29 Apr 2020 13:45:00 -0500

Introduction

The other day, I came across something called the "cube root rule", which is a proposed reform to the size of the U.S. House of Representative. Since 1913, the size of the U.S. House has remained constant at 435 seats (except for a few years when two seats were temporarily added to accommodate the addition of Alaska and Hawaii). Under the cube root rule, the size of the U.S. House would be increased to the cube root of the U.S. population (suitably rounded), adjusted every census. As I write this, the U.S. Census bureau estimates that the population is currently about 329.6 million, and so adopting the cube root rule would expand the U.S. House to 691 seats.

The cube root rule was first proposed by political scientist Rein Taagepera in a 1972 paper. In this paper, he examines the size of national legislative bodies and concludes that $n = \sqrt[3]{P}$ is a reasonable for the size $n$ of a nation's legislature as a function of population $P$. Then he gives a theoretical argument for why $\sqrt[3]{P}$ is an optimal value for $n$.¹

I do believe that we would benefit from having a larger U.S. House, for many of the reasons articulated by the New York Times in a 2018 editorial, and perhaps the Cube Root Rule would be a good way to accomplish that. However, the point of this post is not to make that argument, but rather, see what kind of models we can make for the actual observed sizes of legislative bodies.

Disclaimer

I want to be clear here that I don't think I'm doing anything novel here. I am pretty sure that many actual political scientists (which I am absolutely not) have tread this ground before. If I were writing an academic paper, rather than just messing around on my blog, I would do a proper literature review and put it here. This post is mostly about having fun programming than anything else.

Regression for all countries

In order to investigate this question, I needed a data set that contained the size of legislative bodies and populations for a number of countries. If I were a real political scientist, I would probably use some carefully curated data set from some international organization, but because I'm just some guy messing around on his blog, I turned to Wikipedia, namely the article "List of legislatures by number of members".² Fortunately, Pandas has the capacity to scrape web pages and extract data frames from their tables.

For this analysis, I considered the size of the lower house of each legislature (in cases where the legislature has more than one chamber). Wikipedia's table doesn't just include sovereign nations, but also external territories of various degrees of autonomy (e.g. Puerto Rico and Hong Kong), as well as the European Union (which has an elected parliament). Note that countries are included even if their legislatures are not elected using what we would consider free and fair elections (such as North Korea).

I threw out countries where the size of the lower house was listed as "maximum" or "minimum" (because who knows what the actual size is). If it said "usually" or "normally", I used the number listed. For Timor-Leste it gave a range ("52 to 65"), so I took the midpoint.

I took the logarithm³ of the population and lower house size and performed a least-squares linear regression, obtaining: \[\log(n) = 0.3482\log(P) - 0.5369, ; r^2 = 0.8482.\] This corresponds to $n = 0.5846P^{0.3482}$.

Here is a scatterplot of the data along with $n = 0.5846P^{0.3482}$ and $n = \sqrt[3]{P}$:

And here is the same, but plotted in log scale (which makes more sense for this data):

(It should be noted that a very similar plot appears in Taagepera's paper.)

Regression for other subsets of countries

As mentioned before, the data set includes countries that are not democracies (e.g. North Korea) as well as non-countries (territories and the European Union). We might restrict our attention to various subsets of countries, such as the members of:

The North Atlantic Treaty Organization
The Organisation for Economic Co-operation and Development
The European Union

The results are in this table:

Group of countries	Regression⁴	$r^2$	Plots
all	$\log(n) = 0.3482\log(P) - 0.5369 \Rightarrow n = 0.5846P^{0.3482}$	$r^2 = 0.8482$	linear log
NATO	$\log(n) = 0.3990\log(P) - 1.0466 \Rightarrow n = 0.3511P^{0.3990}$	$r^2 = 0.8764$	linear log
OECD	$\log(n) = 0.3850\log(P) - 0.8940 \Rightarrow n = 0.4090P^{0.3850}$	$r^2 = 0.8026$	linear log
EU	$\log(n) = 0.4477\log(P) - 1.8032 \Rightarrow n = 0.1648P^{0.4477}$	$r^2 = 0.8817$	linear log
OECD, NATO, or EU	$\log(n) = 0.3796\log(P) - 0.7875 \Rightarrow n = 0.4550P^{0.3796}$	$r^2 = 0.8339$	linear log

For example, the aforementioned New York Times editorial has a plot similar to this one:

Source code

The Python script to parse the Wikipedia article, run the regressions, and produce the plots can be downloaded here: leg_sizes.py

It is also shown below:

leg_sizes.py Python

#!/usr/bin/env python3

# leg_sizes.py
# Christopher Phan
# cphan@chrisphan.com
# https://chrisphan.com/
#
# Takes data from the Wikipedia article "List of legislatures by number of
# members", and peforms a linear regression on n = kP^alpha
# where n is the number of members of the lower house of a nation's legislature
# and P is the population.
#
# This script was written for Python 3.6.
#
# To run this script, save
# https://en.wikipedia.org/w/index.php?title=List_of_legislatures_by_number_of_members&oldid=951703531
# to legislatures.html
#
################################################################################
#
# Copyright 2020 Christopher Phan
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
#

import pandas as pd
import scipy.stats as stats
import numpy as np
import matplotlib
import matplotlib.pyplot as plt

# See note above
data = pd.read_html("legislatures.html")[0]

def contains_min_or_max(x):
    if ("minimum" in x) or ("maximum" in x):
        return True
    else:
        return False

# Get rid of entries with only a minimum or maximum leg size listed
# (since we have no idea the actual size)
data = data[data["Lowerhouse[1]"].apply(contains_min_or_max) == False]

def number_parse(x):
    if isinstance(x, int):
        return x
    else:
        num = x.split("[")[0] # Get rid of footnotes
        if "to" in num:
            # Take the mean if it says "a to b"
            splits = num.split(" to ")
            num = (int(splits[0]) + int(splits[1]))/2
        else:
            # Get rid of words "usually" or "normally"
            num = int(num.split(" ")[0])
    return num


countries = data['Country']
leg_size = data["Lowerhouse[1]"].apply(number_parse)
pop_size = data["Population[2]"].apply(number_parse)

# List of OECD countries
# From http://www.oecd.org/about/members-and-partners/

oecd_countries = [
    "Australia",
    "Austria",
    "Belgium",
    "Canada",
    "Chile",
    "Czech Republic",
    "Denmark",
    "Estonia",
    "Finland",
    "France",
    "Germany",
    "Greece",
    "Hungary",
    "Iceland",
    "Ireland",
    "Israel",
    "Italy",
    "Japan",
    "Korea, South", # Changed from "Korea" to match Wikipedia
    "Latvia",
    "Lithuania",
    "Luxembourg",
    "Mexico",
    "Netherlands",
    "New Zealand",
    "Norway",
    "Poland",
    "Portugal",
    "Slovakia", # Changed from "Slovak Republic" to match Wikipedia
    "Slovenia",
    "Spain",
    "Sweden",
    "Switzerland",
    "Turkey",
    "United Kingdom",
    "United States"
]

# List of NATO countries
# From https://www.nato.int/cps/en/natohq/nato_countries.htm

nato_countries = [
    "Albania",
    "Belgium",
    "Bulgaria",
    "Canada",
    "Croatia",
    "Czech Republic",
    "Denmark",
    "Estonia",
    "France",
    "Germany",
    "Greece",
    "Hungary",
    "Iceland",
    "Italy",
    "Latvia",
    "Lithuania",
    "Luxembourg",
    "Montenegro",
    "Netherlands",
    "North Macedonia",
    "Norway",
    "Poland",
    "Portugal",
    "Romania",
    "Slovakia",
    "Slovenia",
    "Spain",
    "Turkey",
    "United Kingdom",
    "United States"
]

# List of European Union countries
# From https://europa.eu/european-union/about-eu/countries_en

eu_countries = [
    "Austria", "Italy",
    "Belgium", "Latvia",
    "Bulgaria", "Lithuania",
    "Croatia", "Luxembourg",
    "Cyprus", "Malta",
    "Czech Republic", "Netherlands", # Changed from "Czechia" to match Wikipedia
    "Denmark", "Poland",
    "Estonia", "Portugal",
    "Finland", "Romania",
    "France", "Slovakia",
    "Germany", "Slovenia",
    "Greece", "Spain",
    "Hungary", "Sweden",
    "Ireland"
]

# Output the parsed data to a new CSV file

newdata = pd.DataFrame({
    'Country': countries,
    'Lower house size': leg_size,
    'Population': pop_size}
)

newdata["OECD"] = newdata["Country"].apply(
    lambda x: x in oecd_countries
)

newdata["NATO"] = newdata["Country"].apply(
    lambda x: x in nato_countries
)

newdata["EU"] = newdata["Country"].apply(
    lambda x: x in eu_countries
)

newdata["OECD, NATO, or EU"] = newdata["Country"].apply(
    lambda x: (x in oecd_countries) or
        (x in nato_countries) or (x in eu_countries)
)

newdata.to_csv("Parsed_data.csv")


all_table = """
| Group of countries  | Regression[^4]               | $r^2$          | Plots         |
| :------------------ | :--------------------------- |:-------------- |:--------------|
"""

for condition in ["all", "NATO", "OECD", "EU", "OECD, NATO, or EU"]:

    # Calculate the regression
    if condition == "all":
        current_data = newdata
    else:
        current_data = newdata[newdata[condition] == True]
    log_leg_size = np.log(current_data["Lower house size"])
    log_pop_size = np.log(current_data["Population"])

    m, b, r, p, stderr = stats.linregress(log_pop_size, log_leg_size)

    def signstr(x):
        if x < 0:
            return "-"
        else:
            return "+"

    filename_str = condition.replace(" ", "_")
    filename_str = filename_str.replace(",", "")

    bracket_to_brace = lambda x: x.replace("[", "{").replace("]", "}")

    with open("regression_data_{}.txt".format(
        filename_str), "wt") as outfile:
            outfile.write("{} countries:\n".format(condition))
            outfile.write("log(n) = {:0.4f}*log(P) {} {:0.4f}\n".format(
                m, signstr(b), abs(b)))
            outfile.write("n = {:0.4f}P^{:0.4f}\n".format(np.exp(b), m))
            outfile.write("r^2 = {:0.4f}\n".format(r**2))

    regression_text = "$n = {:0.4f}P^".format(np.exp(b))
    regression_text += bracket_to_brace("[{:0.4f}]$".format(m))

    # Add to markdown table

    all_table += r"| {} | $\log(n) = {:0.4f}\log(P) {} {:0.4f}".format(
        condition, m, signstr(b), abs(b))
    all_table += r" \Rightarrow {}".format(
        regression_text[1:])
    all_table += r"| $r^2 = {:0.4f}$ |".format(r**2)
    all_table += "[linear](/leg_sizes_2020-04/legsizes_{}_linear.png) ".format(
        filename_str)
    all_table += "[log](/leg_sizes_2020-04/legsizes_{}_log.png) |\n".format(
        filename_str)

    # Make plots

    source_txt = 'Sources listed at: https://chrisphan.com/2020-04-29_what_size_are_national_legislatures/#sources'

    maxpop = max(current_data["Population"])
    maxsize = max(current_data["Lower house size"])
    minpop = min(current_data["Population"])
    minsize = min(current_data["Lower house size"])


    for scale in ["linear", "log"]:
        if scale == "linear":
            xmin = 0
            xmax = 1.1*maxpop
            ymin = 0
            ymax = 1.1*maxsize
        else:
            xmin = 10**(int(np.log(minpop)/np.log(10)))
            xmax = 10**(int(np.log(maxpop)/np.log(10) + 1))
            ymin = 10**(int(np.log(minsize)/np.log(10)))
            ymax = 10**(int(np.log(maxsize)/np.log(10) + 1))
        fig = plt.figure(figsize=(8, 6))
        ax = fig.add_subplot(111)
        if scale == "linear":
            xvals = np.arange(xmin/10**6, xmax/10**6)*10**6
        else:
            xvals = np.exp(np.arange(np.log(xmin), np.log(xmax) + 1))
        yvals = np.exp(b)*(xvals)**m
        yvals2 = xvals**(1/3)
        plt.xscale(scale)
        plt.yscale(scale)
        if scale == 'linear':
            fmtx = matplotlib.ticker.FuncFormatter(
                lambda x, f: '{:,.0f} m'.format(x/10**6))
            fmty = matplotlib.ticker.FuncFormatter(
                lambda y, f: '{:,.0f}'.format(y))
            ax.yaxis.set_major_formatter(fmty)
            ax.xaxis.set_major_formatter(fmtx)
        plt.xlim(xmin, xmax)
        plt.ylim(ymin, ymax)
        plt.plot(current_data["Population"],
            current_data["Lower house size"], "b.")
        plt.plot(xvals, yvals, 'r', label=regression_text)
        plt.plot(xvals, yvals2, 'g--', label= r'$n = \sqrt[3]{P}$')
        plt.grid(True, which="both")
        plt.legend()
        title = "Population and lower legislative house size"
        title += " ({} scale, {} countries)".format(
            scale, condition)
        plt.title(title)
        plt.xlabel("Population ($P$)\n" + source_txt)
        plt.ylabel("Lower house size ($n$)")
        plt.tight_layout()
        fig.savefig("legsizes_{}_{}.png".format(filename_str, scale))
        fig.savefig("legsizes_{}_{}.pdf".format(filename_str, scale))
        plt.close()

with open("regression_table.md", "wt") as markout:
    markout.write(all_table)

CSV file

My script produces a CSV file of the data after its processed (but before the regression is run): Parsed_data.csv

It is also shown below:

Parsed_data.csv CSV

,Country,Lower house size,Population,OECD,NATO,EU,"OECD, NATO, or EU"
0,"China, People's Republic of",2980.0,1355692576,False,False,False,False
1,United Kingdom,650.0,63742977,True,True,False,True
2,Italy,630.0,61680122,True,True,True,True
3,France,577.0,66259012,True,True,True,True
4,India,545.0,1236344631,False,False,False,False
5,European Union,705.0,511434812,False,False,False,False
6,Germany,709.0,80996685,True,True,True,True
7,Japan,465.0,126451398,True,False,False,True
8,Indonesia,575.0,271407446,False,False,False,False
9,"Korea, North",687.0,24851627,False,False,False,False
10,Morocco,395.0,32987206,False,False,False,False
11,Burma (Myanmar),440.0,55746253,False,False,False,False
12,Ethiopia,547.0,96633458,False,False,False,False
13,Thailand,500.0,67741401,False,False,False,False
14,Mexico,500.0,120286655,True,False,False,True
15,Russia,450.0,142470272,False,False,False,False
16,Cuba,612.0,11047251,False,False,False,False
17,"Congo, Democratic Republic of the",500.0,77433744,False,False,False,False
18,Spain,350.0,47737941,True,True,True,True
19,Algeria,462.0,38813722,False,False,False,False
20,Brazil,513.0,202656788,False,False,False,False
21,Poland,460.0,38346279,True,True,True,True
22,Turkey,600.0,84339067,True,True,False,True
23,United States,435.0,318892103,True,True,False,True
24,Sudan,450.0,35482233,False,False,False,False
25,Vietnam,500.0,93421835,False,False,False,False
26,South Africa,400.0,48375645,False,False,False,False
27,Nigeria,360.0,177155754,False,False,False,False
28,Romania,329.0,20121641,False,True,True,True
29,Egypt,596.0,100388073,False,False,False,False
30,Ukraine,450.0,44291413,False,False,False,False
31,Pakistan,342.0,196174380,False,False,False,False
32,Kenya,349.0,45010056,False,False,False,False
33,Canada,338.0,34834841,True,True,False,True
34,Yemen,301.0,26052966,False,False,False,False
35,South Sudan,332.0,11562695,False,False,False,False
36,Uganda,375.0,35918915,False,False,False,False
37,Tanzania,357.0,49639138,False,False,False,False
38,CÃ´te d'Ivoire,255.0,22848945,False,False,False,False
40,Zimbabwe,270.0,13771721,False,False,False,False
41,Sweden,349.0,9723809,True,False,True,True
42,Nepal,275.0,28982771,False,False,False,False
43,Argentina,257.0,43024374,False,False,False,False
44,Somalia,275.0,10428043,False,False,False,False
45,Iraq,325.0,32585692,False,False,False,False
46,Philippines,304.0,107668231,False,False,False,False
47,Bangladesh,300.0,166280712,False,False,False,False
48,Greece,300.0,10775557,True,True,True,True
49,"Korea, South",300.0,49039986,True,False,False,True
50,Malaysia,222.0,30073353,False,False,False,False
51,Iran,290.0,80840713,False,False,False,False
52,Czech Republic,200.0,10627448,True,True,True,True
53,Cameroon,180.0,23130708,False,False,False,False
54,Ghana,275.0,25758108,False,False,False,False
55,Colombia,166.0,46245297,False,False,False,False
56,Mozambique,250.0,24692144,False,False,False,False
57,Serbia,250.0,7209764,False,False,False,False
58,Syria,250.0,17951639,False,False,False,False
59,Uzbekistan,150.0,28929716,False,False,False,False
60,Switzerland,200.0,8061516,True,False,False,True
61,Austria,183.0,8223062,True,False,True,True
62,Bulgaria,240.0,6924716,False,True,True,True
63,Portugal,230.0,10813834,True,True,True,True
64,Australia,150.0,22507617,True,False,False,True
65,Ireland,166.0,4832765,True,False,True,True
66,Netherlands,150.0,16877351,True,True,True,True
67,Sri Lanka,225.0,21866445,False,False,False,False
68,Gabon,120.0,1672597,False,False,False,False
69,Belgium,150.0,11420163,True,True,True,True
70,Angola,220.0,19088106,False,False,False,False
71,Tunisia,217.0,10937521,False,False,False,False
72,Dominican Republic,183.0,10349741,False,False,False,False
73,"Congo, Republic of the",139.0,4662446,False,False,False,False
74,Jordan,150.0,7930491,False,False,False,False
75,Finland,200.0,5268799,True,False,True,True
76,Libya,200.0,6244174,False,False,False,False
77,Hungary,199.0,9919128,True,True,True,True
78,Malawi,193.0,17377468,False,False,False,False
79,Chad,188.0,11412107,False,False,False,False
80,Cambodia,123.0,15458332,False,False,False,False
81,Madagascar,151.0,23201926,False,False,False,False
82,Denmark,179.0,5569077,True,True,True,True
83,Belarus,110.0,9608058,False,False,False,False
84,Equatorial Guinea,100.0,722254,False,False,False,False
85,Norway,169.0,5147792,True,True,False,True
86,Bolivia,130.0,10631486,False,False,False,False
87,Venezuela,165.0,28868486,False,False,False,False
88,Mali,160.0,16455903,False,False,False,False
89,Chile,155.0,17574003,True,False,False,True
90,Guatemala,158.0,14647083,False,False,False,False
91,Zambia,158.0,14638505,False,False,False,False
92,Oman,84.0,3219775,False,False,False,False
94,Kazakhstan,107.0,17948816,False,False,False,False
95,Lesotho,120.0,1942008,False,False,False,False
96,Croatia,151.0,4470534,False,True,True,True
97,Eritrea,150.0,6380803,False,False,False,False
98,Georgia,150.0,4935880,False,False,False,False
99,Saudi Arabia,150.0,27345986,False,False,False,False
100,Senegal,150.0,13635927,False,False,False,False
101,Slovakia,150.0,5443583,True,True,True,True
102,Mauritania,146.0,3516806,False,False,False,False
103,Lithuania,141.0,2848000,True,True,True,True
104,Albania,140.0,3020209,False,True,False,True
105,Ecuador,137.0,15654411,False,False,False,False
106,Laos,149.0,6803699,False,False,False,False
108,Palestine,132.0,4816503,False,False,False,False
109,Peru,130.0,30147935,False,False,False,False
110,Slovenia,90.0,2070050,True,True,True,True
111,Haiti,99.0,9996731,False,False,False,False
112,Uruguay,99.0,3332972,False,False,False,False
113,Honduras,128.0,8598561,False,False,False,False
114,Lebanon,128.0,5882562,False,False,False,False
115,Burkina Faso,127.0,18365123,False,False,False,False
116,Azerbaijan,125.0,9686210,False,False,False,False
117,Paraguay,80.0,6703860,False,False,False,False
118,Turkmenistan,125.0,5171943,False,False,False,False
119,Sierra Leone,124.0,5743725,False,False,False,False
120,North Macedonia,123.0,2091719,False,True,False,True
121,Israel,120.0,7821850,True,False,False,True
122,Kosovo,120.0,1859203,False,False,False,False
123,Kyrgyzstan,120.0,5604212,False,False,False,False
124,New Zealand,120.0,4401916,True,False,False,True
125,Guinea,114.0,11474383,False,False,False,False
126,Niger,113.0,17466172,False,False,False,False
127,"China, Republic of",113.0,23359928,False,False,False,False
128,Papua New Guinea,111.0,6552730,False,False,False,False
129,Rwanda,80.0,12337138,False,False,False,False
130,Central African Republic,105.0,5277959,False,False,False,False
131,Liberia,73.0,4092310,False,False,False,False
132,Guinea-Bissau,102.0,1693398,False,False,False,False
133,Estonia,101.0,1257921,True,True,True,True
134,Moldova,101.0,3583288,False,False,False,False
135,Latvia,100.0,2165165,True,True,True,True
136,Namibia,72.0,2198406,False,False,False,False
137,Tajikistan,63.0,8051512,False,False,False,False
138,Swaziland,65.0,1419623,False,False,False,False
139,Nicaragua,92.0,5848641,False,False,False,False
140,Togo,91.0,7351374,False,False,False,False
141,Singapore,87.0,5567301,False,False,False,False
142,Maldives,85.0,393595,False,False,False,False
143,El Salvador,84.0,6125512,False,False,False,False
144,Jamaica,63.0,2930050,False,False,False,False
145,Benin,83.0,10160556,False,False,False,False
146,Montenegro,81.0,650036,False,True,False,True
147,Bahrain,40.0,1314089,False,False,False,False
148,Cyprus,80.0,1172458,False,False,True,True
149,Puerto Rico,51.0,3620897,False,False,False,False
150,Mongolia,76.0,3251587,False,False,False,False
151,Bhutan,47.0,733643,False,False,False,False
152,Cabo Verde,72.0,538535,False,False,False,False
153,Trinidad and Tobago,41.0,1223916,False,False,False,False
154,Panama,71.0,3608431,False,False,False,False
155,Hong Kong,70.0,7112688,False,False,False,False
156,Mauritius,70.0,1331155,False,False,False,False
157,Djibouti,65.0,810179,False,False,False,False
158,Guyana,65.0,735554,False,False,False,False
159,Kuwait,65.0,2742711,False,False,False,False
160,Malta,65.0,412655,False,False,True,True
161,Timor-Leste,58.5,1201542,False,False,False,False
162,Botswana,63.0,2155784,False,False,False,False
163,Iceland,63.0,317351,True,True,False,True
164,Luxembourg,60.0,520672,True,True,True,True
165,San Marino,60.0,32742,False,False,False,False
166,Jersey,58.0,96513,False,False,False,False
167,Bosnia and Herzegovina,42.0,3871643,False,False,False,False
168,Costa Rica,57.0,4755234,False,False,False,False
169,French Polynesia,57.0,280026,False,False,False,False
170,SÃ£o TomÃ© and PrÃncipe,55.0,190428,False,False,False,False
171,"Bahamas, The",38.0,321834,False,False,False,False
172,New Caledonia,54.0,267840,False,False,False,False
173,"Gambia, The",53.0,1925527,False,False,False,False
174,Vanuatu,52.0,266937,False,False,False,False
175,Barbados,30.0,289680,False,False,False,False
176,Suriname,51.0,573311,False,False,False,False
177,Fiji,50.0,903207,False,False,False,False
178,Solomon Islands,50.0,609883,False,False,False,False
179,Samoa,49.0,196628,False,False,False,False
180,Bermuda,36.0,69839,False,False,False,False
181,Kiribati,46.0,104488,False,False,False,False
182,Guernsey,45.0,65849,False,False,False,False
183,Qatar,45.0,2123160,False,False,False,False
184,Transnistria,45.0,469000,False,False,False,False
185,Belize,31.0,340844,False,False,False,False
186,United Arab Emirates,40.0,5628805,False,False,False,False
187,American Samoa,21.0,54517,False,False,False,False
188,Isle of Man,24.0,86866,False,False,False,False
189,Abkhazia,35.0,243206,False,False,False,False
190,Antigua and Barbuda,17.0,91295,False,False,False,False
191,Seychelles,34.0,91650,False,False,False,False
192,South Ossetia,34.0,53532,False,False,False,False
193,Artsakh,33.0,150932,False,False,False,False
194,Brunei,33.0,422675,False,False,False,False
195,Comoros,33.0,766865,False,False,False,False
196,Faroe Islands,33.0,49947,False,False,False,False
197,Macau,33.0,587914,False,False,False,False
198,Marshall Islands,33.0,70983,False,False,False,False
199,Dominica,32.0,73449,False,False,False,False
200,Greenland,31.0,57728,False,False,False,False
201,Northern Mariana Islands,20.0,51483,False,False,False,False
203,Grenada,15.0,110152,False,False,False,False
204,Saint Lucia,17.0,163362,False,False,False,False
205,Tonga,26.0,106440,False,False,False,False
206,Liechtenstein,25.0,37313,False,False,False,False
207,Palau,16.0,21186,False,False,False,False
208,Cook Islands,24.0,10134,False,False,False,False
209,Monaco,24.0,30508,False,False,False,False
210,Saint Martin,23.0,31530,False,False,False,False
211,Aruba,21.0,110663,False,False,False,False
212,Cayman Islands,21.0,54914,False,False,False,False
213,CuraÃ§ao,21.0,146836,False,False,False,False
214,Saint Vincent and the Grenadines,21.0,102918,False,False,False,False
215,Niue,20.0,1190,False,False,False,False
216,Tokelau,20.0,1337,False,False,False,False
217,Wallis and Futuna,20.0,15561,False,False,False,False
218,Nauru,19.0,9488,False,False,False,False
219,Saint Barthelemy,19.0,7267,False,False,False,False
220,Saint Pierre and Miquelon,19.0,5716,False,False,False,False
221,Turks and Caicos Islands,19.0,49070,False,False,False,False
222,Gibraltar,18.0,29185,False,False,False,False
223,"Saint Helena, Ascension and Tristan da Cunha",17.0,7776,False,False,False,False
224,British Virgin Islands,15.0,32680,False,False,False,False
225,Guam,15.0,161001,False,False,False,False
226,Sint Maarten,15.0,39689,False,False,False,False
227,Tuvalu,15.0,10782,False,False,False,False
228,United States Virgin Islands,15.0,104170,False,False,False,False
229,"Micronesia, Federated States of",14.0,105681,False,False,False,False
230,Saint Kitts and Nevis,14.0,51538,False,False,False,False
231,Anguilla,11.0,16086,False,False,False,False
232,Montserrat,11.0,5215,False,False,False,False
233,Falkland Islands,10.0,2840,False,False,False,False
234,Pitcairn Islands,10.0,48,False,False,False,False
235,Christmas Island,9.0,1530,False,False,False,False
236,Norfolk Island,9.0,2210,False,False,False,False
237,Cocos (Keeling) Islands,7.0,596,False,False,False,False
238,Vatican City,7.0,842,False,False,False,False

Sources

Wikipedia article "List of legislatures by number of members": https://en.wikipedia.org/w/index.php?title=List_of_legislatures_by_number_of_members&oldid=951703531
List of NATO countries: https://www.nato.int/cps/en/natohq/nato_countries.htm
List of OECD countries: http://www.oecd.org/about/members-and-partners/
List of EU countries: https://europa.eu/european-union/about-eu/countries_en

Footnotes

Taagepera's argument is, basically, that a member of the legislature should spend an equal amount of energy paying attention to conversations between members (of which there are $n(n-1)/2$) and their own discussions with politically-interested constituents (of which there are $kP/n - 1$, where $k$ is the proportion of the population that cares about politics). Setting these equal and solving for $P$, we get \[P = \frac{n^3 - n^2 + 2n}{2k},\] i.e. $n$ should be asymptotically proportional to $\sqrt[3]{P}$. Making what Taagepera calls the "rather reasonable assumption that about one half of the population is politically active", i.e. that $k = 1/2$, we have $n \approx \sqrt[3]{P}$ for $n$ sufficiently large. ↩
This link goes to the specific revision of the Wikipedia article that I used in my analysis. ↩
Natural logarithm. ↩
Linear regression on $\log(n) = \alpha\log(P) + k$. ↩

On (not) voting at a polling place

Thu, 23 Apr 2020 14:06:00 -0500

Because of the potential for COVID-19 to disrupt elections this fall, mail voting is getting a lot of attention. It's not a good idea to make people choose between their safety and exercising the franchise.

I grew up in Oregon, and spent the first decade of my adult life there, but now I have spent more of my adult life living outside Oregon than within it. Since moving from Oregon, I have shed many of my Oregonian habits: I have become used to pumping my own gas. I have become used to paying sales tax. I haven't been to Powell's in years. I no longer consider an outdoor temperature of 35°F to be "cold", or three inches of snow "excessive". I don't feel like I'm losing money whenever I put a pop can in a recycling bin.

But I still prefer to vote early rather than at a polling place on Election Day.

In November 1998—weeks before my 18th birthday—Oregonians passed a citizen initiative which eliminated polling places and switched the state to all-mail balloting. Under this system, each Oregon voter receives a ballot in the mail weeks before Election Day, fills out the ballot out at home (or at a microbrewery, or wherever), and then returns the ballot to elections officials. To be counted, the ballot must be received by a certain time on the evening of Election Day. A voter has the option of mailing back their completed ballot (in which case, they must send it early enough to arrive at the elections office by Election Day, as postmarks do not count). A voter may also return their ballot to an official drop box (located at many public places, such libraries). Sometimes, campaign workers or volunteers will offer to collect and return ballots for their supporters. My preferred method was to use an official drop box.

Licking the return envelope for my 2008 primary ballot. Yes, I filled out the spot for the return address despite using a drop box, rather than the mail.

An official Lane County, Oregon ballot drop box, located on the University of Oregon campus, in 2008.

My first election living outside Oregon was in the spring of 2010. For that election, I was living overseas and hence voted absentee. (I remember mailing my absentee ballot from a British Post Office.) It wasn't until November 2010—weeks before my 30th birthday—that I first voted at a polling place on Election Day. At the time, I was living in Pennsylvania, which did not have no-excuse absentee voting at the time. Although it was not inconvenient (my polling place was located at the university where I worked at the time, and the line was not very long), I did not really enjoy having to block out time on a specific day to vote.

Before March 3 of this year, my last time voting at a polling place on Election Day was for the 2012 General Election, a few months after my wife and I had moved to Minnesota. In that election, voters put the Minnesota Democratic-Farmer-Labor (DFL) Party in control of both chambers of the Minnesota Legislature, The next year, the DFL-controlled legislature passed a law bringing no-excuse absentee voting to Minnesota, which was subsequently signed by then-Gov. Mark Dayton, also a DFLer.

This law brought the kind of voting I was used to—and prefer—to the state I have made my home. For every election between the 2012 General Election and this year's presidential primary, I have voted early in every election. Sometimes I request a ballot to be mailed to my house, but more often I just go to my county auditor-treasurer's office and vote early in person. (I have signed up for a mail ballot this year.)

(It should be noted that some rural precincts in Minnesota conduct their elections like Oregon, having everyone vote by mail rather than operating polling places.)

My 2020 presidential primary ballot, on the table inside a voting booth, moments before I marked it.

I did vote at my polling place on March 3 (which seems like years ago), in Minnesota's Super Tuesday presidential primary. The presidential race was very fluid this year, and I was worried that my preferred candidate could drop out right before Super Tuesday. (That did not happen, but my concern was not unwarrented, given that two other candidates did drop out within days of the vote.) So, I waited until Election Day to vote. As I have encouraged countless voters while door-knocking on behalf of DFL candidates, I made a voting plan, blocking out a specific time in my calendar to go to my polling place (the location of which I confirmed at mnvotes.org) and cast my ballot. And I did.

But most elections are not that fluid, and I typically know how I am going to vote weeks before Election Day. So, I vote early. (Campaigns like it when you vote early, because they know that you will turn out, and they don't have to use volunteer time and energy to contact you and remind you to vote.) It's possible that many people will be voting early for the first time this fall.

Giving up Celsius

Mon, 20 Apr 2020 14:12:00 -0500

When I was in elementary school, I came across a series of books in the school library about the metric system. I don't remember the title or the author, but I do remember reading each of them and falling in love with the measurement system used by almost every country in the world other than my own. The advantages of the metric system, especially ease of conversion between units, are well-known, and I won't belabor them here. Let's just say (and this will not come as a surprise to those who know me well) I became a passionate fan of the metric system from a young age.¹

That said, I have always felt that the metric unit with the weakest case is degrees Celsius.² Yes, it's nice that water freezes and boils at 0 and 100, rather than 32 and 212, but frankly it's really not that hard to remember the numbers 32 and 212. The biggest advantage of the metric system, ease of conversion, really does not apply to temperature. It's a pain to remember that there are 5,280 feet in a mile or that there's 128 fluid ounces in a gallon³, compared to remembering that there are 1,000 meters in a kilometer or 1,000 milliliters in a liter, but we don't have to convert degrees Fahrenheit into any smaller or bigger units.

Some passionate advocates of the Fahrenheit scale will point out that the scale is more granular, allowing finer measurements for the air temperatures encountered in everyday life. For example, the National Weather Service reports that the warmest high temperature in the contiguous United States on Christmas was 84°F in Falfurrias, Texas, and the lowest low temperature was -13°F in Copper Basin, Idaho, a difference of 97 Fahrenheit degrees. In Celsius, those temperatures would be reported as 29°C and -25°C, for a range of 54 Celsius degrees. Frankly, I think not a very strong argument, in terms of everday applications such as weather or cooking: most people can hardly detect a 1 Fahrenheit degree difference in air temperature, and most ovens are not so precisely calibrated. In these cases, the increased granularity of Fahrenheit is unnecessary.

By far the strongest argument for America adopting Celsius is its universality outside the U.S. Standardization of units makes commerce and communication between nations easier.

After earning my PhD in mathematics in June 2009, my first job was a short stint as an adjunct instructor at a culinary school, where I taught my students (among other things) how to convert between different customary units as well as between customary and metric. My second job was as a temporary lecturer for an academic year at an ancient university⁴ in Scotland. In the U.K., my dream of a Celsius life came true, as weather reports were given in that scale. (On the other hand, speed limits, driving distances, and the volume of beer sold in pubs, are still given Imperial⁵ units.) Upon repatriation, I was getting most of my weather information from the Internet, and I realized I could set website an weather apps to Celsius, allowing me to continue my Celsius paradise in America.

However, my own private Celsius paradise couldn't last. Some cracks developed:

While I could set apps and websites to report temperatures in Celsius, the thermometer and thermostat in our car, the thermostat in our house, and the dial on our oven (as well as the American cookbooks we use) give temperatures in Fahrenheit. (I could configure my home thermostat to use Celsius, but when your spouse does not share your metric fanaticism, consideration requires your home thermostat to remain set for American units.)

Then there’s the fact that I now live in Minnesota, a place with serious weather and hence frequent weather-related conversations. Several times a week, I read the Updraft blog, which provides detailed discussions of weather forecasts, weather records, and comparisons of different weather computer models (and gives temperatures in Fahrenheit). I consult the weather report before getting dressed, to determine if I want to wear (depending on the season) long underwear, a sweater, short sleeves, pants, or shorts. I am often talking about the weather with other Minnesotans, and of course those conversations use Fahrenheit.

But the biggest problem with operating in Celsius was that I was not developing a Fahrenheit intuition for cold temperatures. I spent the first 28 years of my life living in temperate places: just outside of the Portland metro area, inside the Portland metro area, and in Eugene. Usually, the temperature stayed above 30°F, and so I developed a Fahrenheit intuition for temperatures above 30°F. Later, living in Scotland (also a relatively temperate place) allowed me to develop a Celsius intuition for the same range of temperatures. But Minnesota experiences much, much colder temperatures, sometimes down to -31°F or -35°C, and in my Celsius paradise I was only developing a Celsius intuition for winter temperatures. I was developing Celsius-based heuristics, such as not wearing long underwear or fleece-lined pants unless the temperature was below -15°C (5°F). When someone told me that the high temperature was going to be 10°F, I would have to make the conversion to Celsius in my head.

It turns out, for the same reasons that a nation should adopt the same units of measurement as the other nations in the global community, an individual should probably use the same units of measurements as the other individuals in their community. As most of my weather conversations are with other Americans, who think and use Fahrenheit, it makes sense for me to do so as well.

But I'm still not going to change the navigation app on my phone to miles.

While I would still consider myself a metrication advocate, it's safe to say that metrication is very far from my highest priority, politically. Besides, it's happening slowly anyway. ↩
Yes, technically, the SI unit for temperature is Kelvin, but that's just Celsius shifted by 273.15 degrees. ↩
As an aside, I recently realized that the secret to remembering the U.S. customary liquid volume measures is to memorize the following list: 1 1/2 teaspoons, tablespoon, fluid ounce, 1/4 cup, 1/2 cup, cup, pint, quart, 1/2 gallon, gallon. In this list, each measurement is twice the size of the previous measurement. ↩
That is to say, one founded before 1600. ↩
While they have much in common, U.K. Imperial system and U.S. customary system are different systems, with different volume units. For example, Imperial pints are 568 mL, while customary pints are 473 mL. ↩

Bugs in the iOS calendar app

Mon, 6 Apr 2020 14:58:00 -0500

Stand-up mathematician¹ Matt Parker found some interesting bugs in the iOS calendar app:

I encourage you to watch the video, but the short summary is that if you scroll back far enough (centuries) in the year view of the iOS Calendar app, the year labels and some months disappear. I reproduced this bug on my iPhone:

My iPhone's calendar app year view for 1582 is messed up.

I don't really believe Parker's suggested explanation for this bug, although to be fair, I don't think he really does, either. It should be noted that 1582 (the year where my iPhone first shows this problem) is the year that the Gregorian calendar was introduced.²

Speaking of the Gregorian calendar, while trying to reproduce Parker's bug, I found another bug. Check out the year view for 1900:

My iPhone's calendar app year view for 1900

Notice two things:

February is shown as having 29 days.
March 1 and the nonexistent February 29 are both shown as being Thursdays.

When you zoom in to February 1900, it shows that month correctly as having only 28 days.³

My iPhone's calendar app month view for February 1900

However, the year view for 2100 shows the correct number of days for February 2100.

My iPhone's calendar app year view for 2100

In terms of severity, these bugs are extremely minor. Very few people are using their iPhones to view calendar events in the 16th century, or indeed probably before the 1990s.

Coolest job description EVER! ↩
Although different countries adopted the Gregorian calendar in different years.

For example, Great Britain and its colonies adopted the Gregorian calendar in September 1752, by skipping 11 days of that month. You can see this in many versions of the Unix cal command-line utility, including the one on my MacBook:
```
Christophers-MacBook-Pro:~ chris$ cal 9 1752
   September 1752
Su Mo Tu We Th Fr Sa
       1  2 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
```
↩
If you are feeling confused, remember that in the Gregorian calendar years that are divisible by 100 are not leap years, unless they are also divisible by 400. For example, 1800 and 1900 were not leap years, 2100 will not be a leap year, but 2000 was a leap year. The means if you were born in the 20th century, every year divisible by 4 in your lifetime will probably be a leap year, just like it would have been under the previous Julian calendar. ↩

Repository of files related to the NYT COVID-19 dataset

Sat, 4 Apr 2020 23:36:00 -0500

I've created a GitHub repository based on the New York Times' COVID-19 database.

The New York Times has been compiling COVID-19 data from various state health departments. They have published the data as a GitHub repository. There are two files of interest in their repository: us-states.csv and us-counties.csv.

In these CSV files, each row represents the number of (confirmed) cases and deaths in a particular geographical area as of a particular date. For example, the first ten rows of us-states.csv are:

us-states.csv

date,state,fips,cases,deaths
2020-01-21,Washington,53,1,0
2020-01-22,Washington,53,1,0
2020-01-23,Washington,53,1,0
2020-01-24,Illinois,17,1,0
2020-01-24,Washington,53,1,0
2020-01-25,California,06,1,0
2020-01-25,Illinois,17,1,0
2020-01-25,Washington,53,1,0
2020-01-26,Arizona,04,1,0

These are easy enough to analyze using a tool like Pandas. However, this kind of file might be hard to analyze if you want to use a spreadsheet program.

So, I wrote a Python script that uses Pandas to output a bunch of CSV files, so people can analyze the data using spreadsheet software. The script also creates a bunch of plots, including one for each state.

The script and all the output are available in my new repo. I have licensed the script under an MIT-style license. Feel free to fork and improve! (The outputs of the script I haven't so licensed since they are subject to the license restrictions imposed by the Times.)

A big disclaimer: I'm not an epidemiologist (my expertise being in noncommutative algebra, not communicable disease). So, please don't make any policy decisions based on my plots or any other amateurs who don't know what they are doing.

Stay safe and healthy!

U.S. COVID-19 cases and deaths, based on data from the New York Times

What I believe

Fri, 11 Oct 2019 12:31:28 -0500

I strongly believe:

Diversity enriches a society.
Immigrants enrich a society.
Minnesota is a better state because of the refugees who have made it their home.
The Somali community's presence in Minnesota is worth celebrating.
The ideologies that pose the greatest danger to America are white nationalism and toxic masculinity.
Anyone who seeks political power by demonizing immigrants and minorities is not fit to hold office.

Book recommendation: Palaces for the people

Mon, 3 Jun 2019 13:10:00 -0500

In Palaces for the People: How Social Infrastructure Can Help Fight Inequality, Polarization, and the Decline of Civic Life, sociologist Eric Klinenberg discusses the role of social infrastructure in maintaining the health of a community and its people. Klinenberg argues that well-designed social infrastructure can help reduce crime, improve people's lives, bring different people into contact with one-another, and even protect lives during natural disasters. He examines a variety of social infrastructures, including libraries, public housing courtyards, schools, barbershops, bookstores, and community gardens. For example, he explains that when abandoned inner-city lots are turned into community gardens, crime is reduced, which also decreases the stress felt by residents. He explains how having common meeting spaces also decrease political polarization.

I first learned about this book from an episode of the podcast 99% Invisible. That episode serves as a good executive summary of the book, and I would recommend it also.

This book made me more grateful for the social infrastructure present in Winona, Minnesota, the charming college town on the Mississippi River where I make my home. Occasionally, while reading Klinenberg's examples of the benefits of robust social infrastructure in various communities, I would find comparable examples in my own community.

For example, Klinenberg became interested in the concept of social infrastructure when he was a graduate student studying a deadly 1995 heat wave in Chicago. He learned that robust social infrastructure in a neighborhood could prevent people from dying in the heat.

As I read this story, I thought about something that happened in my town during the brutal cold snap we experienced in late January, when nighttime temperatures in Winona plunged as low as -31°F (-35°C), and even daytime temperatures were dangerous. To keep Winona's homeless population safe at night, there is a volunteer-run warming center in a church basement during the winter months, open every evening between 9 p.m. and 7 a.m. During the cold snap, the director of our public library decided to keep the library open late, until the warming center opened, so that homeless people wouldn't have to wait outside in the dangerous cold. Local businesses donated food, and the city's fire department gave rides from the library to the warming center at 9 p.m. The library and warming center are a vital part of my town's social infrastructure that kept people safe.

Klinenberg also explains how physical infrastructure can be designed in a such way to double as social infrastructure. For example, he talks about how the physical measures to control flooding (which will become increasingly important with climate change) can be designed to function as community spaces. In my town, the levee that protects us from the Mississippi's floods also functions as a park. Indeed, Levee Park was recently redesigned to create an even more inviting space for gatherings, such as the farmer's market where my family buys fresh produce in the summer.

Finally, understanding the importance of social infrastructure reaffirms the importance of state and local government. Our political attention is often focused on national politics, especially presidential elections. Of course, presidential elections are very important, and people should pay attention to them, but state and local government are very important, too, and often do not get enough attention. (This is not helped by the recent business difficulties of many communities' local media, who traditionally took on the responsibility of informing citizens about their local government.) Decisions by your state legislature and local government have a large impact on the quality of social infrastructure in your community. They are the ones deciding how much funding should go to parks and libraries, and those decisions have a large impact on the health of communities.

(Image above: Low temperatures on the evening of January 31, 2019, as reported by the National Weather Service)

Consequences of YouTube's algorithm

Thu, 30 May 2019 11:03:00 -0500

As a follow-up to the previous post, consider this video by science educator Derek Muller, recently posted to his YouTube channel Veritasium:

In the video, Muller explains why he thinks a previous video, about the use of shade balls in a Los Angeles-area reservoir, went viral. Basically, he explains, when deciding which videos to promote, YouTube's algorithm seems to be attempting to maximize two outcomes:

Watch time: How long the user will spend watching the video
Click-through rate: How likely the user will click that particular thumbnail when it's displayed

As a result, Muller says that going forward, he will:

"Keep making high-quality videos"
"Choose topics that are more clickable"
"Use clickbaity titles and thumbnails"

Because, frankly, if I'm going to work for days and weeks and months on a video, I would like that video to get, say, 10 million views instead of half-a-million.

Now, my heart sunk a little when I first watched this video, because I have very much enjoyed Muller's previous videos, and I am not a big fan of clickbait. In the past, some of his topics haven't been as "clickable" but were very interesting (to me) nonetheless. But I absolutely cannot fault him, either, since part of his livelihood depends on people continuing to watch his videos.¹

On a related note, recently on YouTube, I have noticed a trend to longer videos, a trend driven by the algorithm's favoring of videos with longer watch time. I find this a bit unfortunate, because short YouTube videos are great for taking a short break from work, and because longer videos require a larger investment of time. I'm reluctant to watch a video that's over 10 minutes, unless the video comes highly recommended or the creator has made content I've enjoyed before.

For that matter, I absolutely don't fault news outlets for posting clickbait, either, given the changes to the media landscape wrought by the Internet in general and social media in particular. ↩

Life in the time of paperclip maximizers

Thu, 16 May 2019 10:08:00 -0500

In an influential 2003 essay, philosopher Nick Bostrom explored the ethical implications of developing a "superintelligence", that is, an "intellect that is vastly outperforms the best human brains in practically every field, including scientific creativity, general wisdom, and social skills." He argued that any such intelligence should be given "philanthropic values", and designed to be ultimately motivated with improving human lives. Without such a "supergoal of philanthropy", a superintelligence could be dangerous. For example, Bostrom imagines a powerful artificial intelligence that was tasked with making as many paperclips as possible, "with the [unintended] consequence that it starts transforming first all of earth and then increasing portions of space into paperclip manufacturing facilities."

A similar example was given by Rob Miles in a Computerphile video. Instead of being asked to make paperclips, Miles imagines a superintelligent machine tasked with an ultimate goal of collecting stamps. The machine's designer imagines that the machine would buy stamps from eBay, but instead the machine starts finding unforeseen ways to accumulate stamps, ways that are creative yet extremely undesirable. The machine starts by fraudulently convincing stamp collectors to send it their collections, and the undesirability of the tactics escalate from there:

There comes a point when the stamp-collecting device is thinking [. . .] "Paper is made of carbon and hydrogen and oxygen. I'm gonna need all of that I can get to make stamps." And it's gonna notice that people are made of carbon, hydrogen, and oxygen, right?

There comes a point where the stamp collecting device becomes extremely dangerous and that point is as soon as you switch it on.

We're probably a long way from being able to create the kind of superintelligence envisioned by Bostrom and Miles in their thought experiments. Compared to the hypothetical superintelligences envisioned by above, today's nascent artificial intelligences are rather rudimentary and are far from having the capacity to even understand "philanthropic values". But it's becoming increasingly clear that even rudimentary AIs can pose risks to our well-being and society when tasked with maximizing some outcome.

I have been a social media user for a long time. In 2001, I joined LiveJournal, a blogging platform that was also one of the first social media sites. One of LiveJournal's great innovations, sometimes mistakenly credited to Mark Zuckerberg and friends, was the "friends page", which displayed your "friends'" posts in reverse-chronological order.

LiveJournal is still around, but it never achieved the same market penetration as Facebook. Today's top services, such as Facebook and YouTube, now use computer programs to determine what shows up in your feed, and in what order, rather than just showing you posts in reverse-chronological order. These programs are designed to maximize "engagement", that is, how much time you spend on their sites (and hence how many ad impressions they can sell). These computer programs are based on machine learning, meaning that rather than having a human programer using human judgment to determine how to choose what content is shown, the computer chooses based on what content has lead to the most engagement in the past. Unfortunately, the most engaging content (and hence shown by an algorithm trying to maximize engagement) is often content that is extremist, misleading, or polarizing.

One of the first to raise the alarm was sociologist Zeynep Tufekci. Her TED talk "We're building a dystopia just to make people click on ads" is required viewing for anyone who uses these services (or cares about the future of society).

So in 2016, I attended rallies of then-candidate Donald Trump to study as a scholar the movement supporting him. I study social movements, so I was studying it, too. And then I wanted to write something about one of his rallies, so I watched it a few times on YouTube. YouTube started recommending to me and autoplaying to me white supremacist videos in increasing order of extremism. If I watched one, it served up one even more extreme and autoplayed that one, too. If you watch Hillary Clinton or Bernie Sanders content, YouTube recommends and autoplays conspiracy left, and it goes downhill from there.

Well, you might be thinking, this is politics, but it's not. This isn't about politics. This is just the algorithm figuring out human behavior. I once watched a video about vegetarianism on YouTube and YouTube recommended and autoplayed a video about being vegan. It's like you're never hardcore enough for YouTube.

Since then, many others have joined in sounding this alarm. In April 2018, Matt Yglesias wrote:

The association between Facebook and fake news is by now well-known, but the stark facts are worth repeating — according to Craig Silverman’s path-breaking analysis for BuzzFeed, the 20 highest-performing fake news stories of the closing days of the 2016 campaign did better on Facebook than the 20 highest-performing real ones.

Rumors, misinformation, and bad reporting can and do exist in any medium. But Facebook created a medium that is optimized for fakeness, not as an algorithmic quirk but due to the core conception of the platform. By turning news consumption and news discovery into a performative social process, Facebook turns itself into a confirmation bias machine — a machine that can best be fed through deliberate engineering. [. . .] Facebook’s imperative to maximize engagement [. . .] lands it in an endless cycle of sensationalism and nonsense.

To be clear, I'm not suggesting that these companies are deliberately promoting extremism or misinformation. It's probably an emergent property of the machine-learning algorithms that use trial and error to maximize users' engagement. It's an unintended side-effect, like the homicidal actions of Miles' hypothetical stamp-collecting machine.

But the probably inadvertent promotion of inaccurate information and extremism isn't the only problem with Facebook's engagement-maximizing algorithms. A bigger issue, to me, is that you have a nascent artificial intelligence trying to get you to waste more time on their web site, distracting you from other priorities. Just like Rob Miles' hypothetical stamp-collecting machine ends up trying to turn all available hydrocarbons into stamps, the computer programs used by social media companies try to convert as many of our waking hours at they can into time spent on their sites. Even the co-founder of Facebook has come to see the danger of this:

I was on the original News Feed team (my name is on the patent), and that product now gets billions of hours of attention and pulls in unknowable amounts of data each year. The average Facebook user spends an hour a day on the platform; Instagram users spend 53 minutes a day scrolling through pictures and videos. They create immense amounts of data — not just likes and dislikes, but how many seconds they watch a particular video — that Facebook uses to refine its targeted advertising. Facebook also collects data from partner companies and apps, without most users knowing about it, according to testing by The Wall Street Journal.

Some days, lying on the floor next to my 1-year-old son as he plays with his dinosaurs, I catch myself scrolling through Instagram, waiting to see if the next image will be more beautiful than the last. What am I doing? I know it’s not good for me, or for my son, and yet I do it anyway.

The choice is mine, but it doesn’t feel like a choice. Facebook seeps into every corner of our lives to capture as much of our attention and data as possible and, without any alternative, we make the trade.

What can we do to counteract these effects? I'm not sure, to be honest. But one small step I've taken, in the wake of the 2016 election, is to take back control of what I read. Instead of waiting for people to post content on Facebook, I actively seek it out, visiting a variety of news sites and reading the physical newspaper. I've also started using an RSS reader again: I read things from my feed when I have a spare moment, and exhausting my feed is a good sign to quit surfing the Internet and re-engage with people who are physically proximate.

I also heartily recommend the book Bored and Brilliant: How Spacing Out Can Unlock Your Most Productive and Creative Self by Manoush Zomorodi.

Plotting with TikZ, Part III: Plotting a function defined by a formula, and plotting functions that go through certain points

Mon, 13 May 2019 15:41:00 -0500

In the last post, I described how I use TikZ to create a graph that might otherwise be created freehand:

In this post, I will describe one method for using TikZ to to plot a function defined by a formula, such as \[y = \left(2x^2 - x - 1\right)e^{-x}.\] Then I will show two ways to make a function go through a certain point.

Plotting from a formula

Simple example

There are a number of ways to achieve this, and PGF actually includes the functionality to perform calculations in TeX. (For example, I have used the PGF pseudorandom number generator in graphics. Someone even made TikZ code to generate a random city skyline.) However, for efficiency sake, I prefer to have the calculations done outside TeX, in a more efficient computer language.

The trick is to use the gnuplot table import format. You simply have to generate a text file with a list of coordinates in the following form:

-1.250000 11.779907
-1.240000 11.456050
-1.230000 11.138839

This file contains the coordinates $(-1.25, 11.779907)$, $(-1.24, 11.456050)$, $(-1.23, 11.138839)$.

There are a number of ways to generate such a file. Since I like Python, I used the following (simple) Python script:

Python

#! /usr/bin/env python3

import numpy as np

f = lambda x: (2*x**2 - x - 1)*np.exp(-x)

xvals = np.arange(-1.25, 10.25, 0.01)

with open("plot1.table", "wt") as outfile:
    for x in xvals:
        outfile.write("{:f} {:f}\n".format(x, f(x)))

Now, this curve can be imported into your TikZ illustration using \draw [. . .] plot; in this case:

LaTeX


\draw[thick, blue, <->] plot[smooth] file {plot1.table};

Here is a complete example (with coordinate axes and everything):

LaTeX

\documentclass[border=0.25cm]{standalone}

\usepackage{tikz}
\usepackage{fouriernc}

\begin{document}

\begin{tikzpicture}[yscale=0.5]

      \draw[gray] (-1.25, -1.25) grid (10.25, 12.25);

      % x-axis
      \draw[thick, black, ->] (-1.25, 0) -- (10.25, 0)
        node[anchor=south west] {$x$};

      % x-axis tick marks
      \foreach \x in {-1, 10}
        \draw[thick] (\x, 0.2) -- (\x, -0.2)
          node[anchor=north] {\(\x\)};

      % y-axis
      \draw[thick, black, ->] (0, -1.25) -- (0, 12.25)
        node[anchor=south west] {$y$};

      % y-axis tick marks
      \foreach \y in {-1, 12}
        \draw[thick] (-0.1, \y) -- (0.1, \y)
          node[anchor=west] {\(\y\)};

      % graph of function
      \draw[thick, blue, <->] plot[smooth] file {plot1.table};

\end{tikzpicture}
\end{document}

And here is the output:

In my Python script above, I carefully chose the set of inputs $x$ (np.arange(-1.25, 10.25, 0.01), i.e., every real number of the form $-1.25 + 0.01k, \; k \in \mathbb{Z}$ in the interval $[-1.25, 10.25)$) to keep $f(x) < 12$. This can also be handled computationally.

Example involving asymptotes

For example, suppose I want to plot $y = r(x)$, where \[r(x) = \frac{6x^3 + 21x^2 - 21x - 36}{3x^3 + 3x^2 - 12x - 12},\] with $-15.5 \leq x \leq 15.5$. Let's additionally suppose I want to only see $y$-values with $-15.5 \leq y \leq 15.5$. (In other words, the "viewing rectangle" will be $[-15.5, 15.5] \times [-15.5, 15.5]$.)

Factoring the numerator and denominator, we have:

\[r(x) = \frac{6x^3 + 21x^2 - 21x - 36}{3x^3 + 3x^2 - 12x - 12}\] \[\phantom{r(x)}= \frac{\left(x + 1\right) \left(2 x - 3\right) \left(3 x + 12\right)}{\left(x + 1\right) \left(x + 2\right) \left(3 x - 6\right)}\] \[\phantom{r(x)}= \frac{\left(2 x - 3\right) \left(3 x + 12\right)}{\left(x + 2\right) \left(3 x - 6\right)}, \; x \not = -1.\]

As you can see, we will have vertical asymptotes at $x = -2$ and $x = 2$, and a hole at $x = -1$. There will also be a horizontal asymptote at $y = 2$.

To handle the asymptotes, we plot the function \[\hat r(x) = \frac{\left(2 x - 3\right) \left(3 x + 12\right)}{\left(x + 2\right) \left(3 x - 6\right)}\] along three intervals:

$[-15.5, -2)$,
$(-2, 2)$, and
$(2, 15.5]$

(Actually, for each of the above intervals $I$, we plot something approximating $\left\{(x, \hat r(x)): x \in I \text{ and } |\hat r(x)|\leq 16\right\}$.)

Here is the Python script:

Python

#! /usr/bin/env python3

import numpy as np

r = lambda x: (2*x - 3)*(3*x + 12)/((x + 2)*(3*x - 6))

xvals = [
    np.arange(-15.5, -2, 0.01),
    np.arange(-2.01, 2, 0.01),
    np.arange(2.01, 15.6, 0.01)
    ]

for idx, interval in enumerate(xvals):
    with open("plot2_{}.table".format(idx), "wt") as outfile:
        for x in interval:
            if abs(r(x)) <= 16:
                outfile.write("{:f} {:f}\n".format(x, r(x)))

Here is the corresponding LaTeX code:

LaTeX

\documentclass[border=0.25cm]{standalone}

\usepackage{tikz}
\usepackage{fouriernc}

\begin{document}

  \begin{tikzpicture}[scale=0.5] % Scale keeps the graphic small

    \draw[gray] (-15.5, -15.5) grid (15.5, 15.5);

    % x-axis
    \draw[thick, black, ->] (-15.5, 0) -- (15.5, 0)
      node[anchor=south west] {$x$};

    % x-axis tick marks
    \foreach \x in {-15, -10, -5, 5, 10, 15}
      \draw[thick] (\x, 0.5) -- (\x, -0.5)
        node[anchor=north] {\(\x\)};

    % y-axis
    \draw[thick, black, ->] (0, -15.5) -- (0, 15.5)
      node[anchor=south west] {$y$};

    % y-axis tick marks
    \foreach \y in {-15, -10, -5, 5, 10, 15}
      \draw[thick] (-0.5, \y) -- (0.5, \y)
        node[anchor=west] {\(\y\)};

    % graph of function
    \draw[thick, blue, <->] plot[smooth] file {plot2_0.table};
    \draw[thick, blue, <->] plot[smooth] file {plot2_1.table};
    \draw[thick, blue, <->] plot[smooth] file {plot2_2.table};

    % hole at (-1, 5)
    \draw[thick, blue, fill=white] (-1, 5) circle (2.5mm);

    % vertical asymptotes
    \draw[thick, dashed, red] (-2, -15.5) -- (-2, 15.5);
    \draw[thick, dashed, red] (2, -15.5) -- (2, 15.5);

    % horizontal asymptote
    \draw[thick, dashed, red] (-15.5, 2) -- (15.5, 2);

  \end{tikzpicture}
\end{document}

And here is the result:

Plotting functions that go through certain points: two ways

Suppose I want to plot a function that go through certain points, say:

$(-4, 2)$,
$(-1, 6)$,
$(2, 3)$, and
$(4, 5)$.

I want the function to be differentiable.

Method one: Bézier curves

Recall from the last post that a (cubic) Bézier curve is specified by four points:

A start point ($(x_0, y_0)$ in the diagram below)
Two control points ($(x_1, y_1)$ and $(x_2, y_2)$ in the diagram below)
An end point ($(x_3, y_3)$ in the diagram below)

The syntax is: \draw ($x_0, y_0$) .. controls ($x_1, y_1$) and ($x_2, y_2$) .. ($x_3, y_3$);

Now, we can combine multiple Bézier curves into a single \draw. For example, \draw ($x_0, y_0$) .. controls ($x_1, y_1$) and ($x_2, y_2$) .. ($x_3, y_3$) .. controls ($x_4, y_4$) and ($x_5, y_5$) .. ($x_6, y_6$); will produce:

In general, this curve will not necessarily be differentiable. We can get differentiability by ensuring that $(x_2, y_2)$, $(x_3, y_3)$ and $(x_4, y_4)$ are colinear (i.e., the end of the first Bézier curve has the same derivative as the beginning of the second) with $x_2 \not = x_3$:

Keeping these principles in mind, we can plot a function going through the desired points with:

LaTeX

\documentclass[border=0.25cm]{standalone}

\usepackage{tikz}
\usepackage{fouriernc}

\begin{document}

  \begin{tikzpicture}[scale=0.5] % Scale keeps the graphic small

    \draw[gray] (-5.25, 0) grid (5.25, 8.25);

    % x-axis
    \draw[thick, black, ->] (-5.25, 0) -- (5.25, 0)
      node[anchor=south west] {$x$};

    % x-axis tick marks
    \foreach \x in {-5, 5}
      \draw[thick] (\x, 0.25) -- (\x, -0.25)
        node[anchor=north] {\(\x\)};

    % y-axis
    \draw[thick, black, ->] (0, 0) -- (0, 8.25)
      node[anchor=south west] {$y$};

    % y-axis tick marks
    \foreach \y in {4, 8}
      \draw[thick] (-0.25, \y) -- (0.25, \y)
        node[anchor=west] {\(\y\)};

    % graph of function
    \draw[thick, blue, <->] (-5.25, -0.5)
      .. controls (-4, 0) and (-5, 0) .. (-4, 2)
      .. controls (-3.5, 3) and (-3, 6) .. (-1, 6)
      .. controls (1, 6) and (1.75, 4) .. (2, 3)
      .. controls (2.25, 2) and (3, 2) .. (4, 5)
      .. controls (4.5, 6.5) and (4.5, 7) .. (5.25, 8);

  \end{tikzpicture}
\end{document}

Here is the output:

Method two: Polynomial interpolation

SciPy's interpolation functionality can find a smooth function that passes through an arbitrary list of points. Here is an example of a Python script to produce such a function and create a table in the necessary format:

Python

#! /usr/bin/env python3

import numpy as np
import scipy.interpolate

coords = [(-10, 8), (-4, 2), (-1, 6), (2, 3), (4, 5), (10, 1)]

# Since our goal is to plot from x = -5.25 to x = 5.25,

# we need to specify some points beyond that interval on both sides.

xcoords, ycoords = zip(*coords)

# This produces a seperate list of x-coordinates and y-coordinates

f = scipy.interpolate.interp1d(xcoords, ycoords, kind="cubic")

xvals = np.arange(-5.25, 5.25, 0.01)
yvals = f(xvals)

outcoords = zip(xvals, yvals)

# This produces a list of coordinate pairs

with open("plot3.table", "wt") as outfile:
    for pair in outcoords:
        outfile.write("{:f} {:f}\n".format(pair[0], pair[1]))

Note that if we had set coords = [(-4, 2), (-1, 6), (2, 3), (4, 5)], then the resulting function would not have had a defined value outside of the interval $[-4, 4]$.

To plot, we use the following tikzpicture:

LaTeX

\documentclass[border=0.25cm]{standalone}

\usepackage{tikz}
\usepackage{fouriernc}

\begin{document}

  \begin{tikzpicture}[scale=0.5] % Scale keeps the graphic small

    \draw[gray] (-5.25, 0) grid (5.25, 8.25);

    % x-axis
    \draw[thick, black, ->] (-5.25, 0) -- (5.25, 0)
      node[anchor=south west] {$x$};

    % x-axis tick marks
    \foreach \x in {-5, 5}
      \draw[thick] (\x, 0.25) -- (\x, -0.25)
        node[anchor=north] {\(\x\)};

    % y-axis
    \draw[thick, black, ->] (0, 0) -- (0, 8.25)
      node[anchor=south west] {$y$};

    % y-axis tick marks
    \foreach \y in {4, 8}
      \draw[thick] (-0.25, \y) -- (0.25, \y)
        node[anchor=west] {\(\y\)};

    % graph of function
    \draw[thick, blue, <->] plot[smooth] file {plot3.table};

  \end{tikzpicture}
\end{document}

Here is the result:

As we have seen, creating plots with TikZ allows you a deep amount of control over the final result, and can be especially powerful when combined with a programming language such as Python.

View of the Mississippi from Levee Park

Sat, 11 May 2019 22:23:52 -0500

(Levee Park, Winona, Minnesota, this morning at 11:02 a.m. CDT.)

Plotting with TikZ, Part II: Functions without formulæ

Tue, 7 May 2019 10:10:00 -0500

In the last post, I explained why TikZ is awesome for making plots. One good use case for handmade TikZ plots is to typeset a question like this:

(2 points each) A plot of the graph of the function $f$ is below:

For each of the following expressions, either evaluate the expression or state that it is undefined:

a. $\lim_{x \rightarrow 2^{-}} f(x)$

b. $\lim_{x \rightarrow 2^{+}} f(x)$

c. $\lim_{x \rightarrow 2} f(x)$

d. $f(2)$

[. . .]

It would be a pain to come up with a formula for a function with the features shown. Before I made these kinds of graphs in TikZ, I would either draw it by hand, or use the drawing features in word-processing software. Drawing by hand is suboptimal, since it requires you to be good at drawing (which I'm not) and the resulting files don't play well with version control. The output from a word-processing software won't play well with version control, either. And using TikZ will make the output automatically have the same font settings as the rest of your document.

How to create such a plot

Here is the source code for the plot above:

LaTeX

\documentclass[border=0.25cm]{standalone}

\usepackage{tikz}
\usepackage{fouriernc}

\begin{document}

  \begin{tikzpicture}[scale=0.75]

    \draw[gray] (-5.25, -5.25) grid (5.25, 5.25);

    % x-axis
    \draw[thick, black, ->] (-5.25, 0) -- (5.25, 0)
      node[anchor=south west] {$x$};

    % x-axis tick marks
    \foreach \x in {-5, 5}
      \draw[thick] (\x, 0.2) -- (\x, -0.2)
        node[anchor=north] {\(\x\)};

    % y-axis
    \draw[thick, black, ->] (0, -5.25) -- (0, 5.25)
      node[anchor=south west] {$y$};

    % y-axis tick marks
    \foreach \y in {-5, 5}
      \draw[thick] (-0.2, \y) -- (0.2, \y)
        node[anchor=west] {\(\y\)};

    % graph of function

    \draw[thick, blue, ->] (-5, 4) .. controls (-5, 2) and (-4, 2) .. (-3, 2)
      .. controls (-2.1, 2) and (-2.1, 3) ..
      (-2.1, 5.25);

    \draw[thick, dashed, red] (-2, -5.25) -- (-2, 5.25);

    \draw[thick, blue, <-] (-1.9, 5.25) ..
    controls (-1.9, 0) and (0, -4) .. (2, -4);

    \draw[thick, blue] (2, 4) -- (5, -2);

    \draw[thick, blue, fill=blue] (-5, 4) circle (1mm);

    \draw[thick, blue, fill=white] (-3, 2) circle (1mm);

    \draw[thick, blue, fill=blue] (-3, -3) circle (1mm);

    \draw[thick, blue, fill=white] (2, -4) circle (1mm);

    \draw[thick, blue, fill=blue] (2, 4) circle (1mm);

    \draw[thick, blue, fill=blue] (5, -2) circle (1mm);

  \end{tikzpicture}
\end{document}

Let's go through the construction of this plot. This will also explain the basic syntax of TikZ.

Grid

LaTeX


\draw[gray] (-5.25, -5.25) grid (5.25, 5.25);

This creates a grid inside the rectangle with opposite corners at $(-5.25, -5.25)$ and $(5.25, 5.25)$. By default, grids in TikZ have their gridlines one unit apart, at integer coordinates. By having the edge of my grid 0.25 units off an integer, I give the subtle impression that the grid extends beyond the graphic. Compare these two grids:

The grid on the left was created with \draw[gray] (-5.25, -5.25) grid (5.25, 5.25); (off-integer edges), while the code on the right was created with \draw[gray] (6, -5) grid (16, 5); (integer edges).

Notice that the grid drawing code is placed first in the tikzpicture. This ensures the grid is drawn first, placing all other elements on top.

Axes

My preference is to have each axis be an arrow with the arrowhead in the positive direction. This is accomplished with:

LaTeX


% x-axis
\draw[thick, black, ->] (-5.25, 0) -- (5.25, 0)
  node[anchor=south west] {$x$};

LaTeX


% y-axis
\draw[thick, black, ->] (0, -5.25) -- (0, 5.25)
  node[anchor=south west] {$y$};

The \draw command is used to draw, unsurprisingly. To produce a line segment from $(-5.25, 0)$ to $(5.25, 0)$, one could just do:

LaTeX


\draw (-5.25, 0) -- (5.25, 0);

You can actually chain line segments together by providing multiple coordinates. For example:

LaTeX


\draw (3, 0) -- (4, 2) -- (5, 0) -- (3, 4);

Back to the $x$-axis, the options [thick, black, ->] produce a thick black line segment with an arrowhead at the end (the last coordinate). But after I draw the line segment, I add some text, using node[anchor=south west] {$x$}. This tells TikZ to place the LaTeX text $x$ just to the right and above the arrowhead. (It's a bit confusing, but anchor=south west refers to the point of view of the node with the text $x$, not the point of view of the arrowhead! That is, the arrowhead is southwest of the $x$.)

Tick marks

LaTeX


% x-axis tick marks
\foreach \x in {-5, 5}
  \draw[thick] (\x, 0.2) -- (\x, -0.2)
    node[anchor=north] {\(\x\)};

LaTeX


% y-axis tick marks
\foreach \y in {-5, 5}
  \draw[thick] (-0.2, \y) -- (0.2, \y)
    node[anchor=west] {\(\y\)};

On each axis are two tick marks. For example, on the $x$-axis, I want a tick mark at $(-5, 0)$ and another at $(5, 0)$. Each tick mark is a tick line segment from $(x, 0.2)$ to $(x, -0.2)$, where $x$ is the horizontal coordinate of the tick mark. I use a for (\foreach) loop to create the tick marks. Below each tick mark is a node with the horizontal coordinate; anchor=north tells TikZ to position the label so that the bottom end of the tick mark (at $(x, -0.2)$) is north of the label.

Bézier curves

We now start making the plot of the function itself. Two parts are made from Bézier curves, a go-to for making curvy parts.

LaTeX


\draw[thick, blue, ->] (-5, 4) .. controls (-5, 2) and (-4, 2) .. (-3, 2)
  .. controls (-2.1, 2) and (-2.1, 3) ..
  (-2.1, 5.25);

LaTeX


\draw[thick, blue, <-] (-1.9, 5.25) ..
  controls (-1.9, 0) and (0, -4) .. (2, -4);

The basic syntax for a Bézier curve in TikZ is \draw ($x_0, y_0$) .. controls ($x_1, y_1$) and ($x_2, y_2$) .. ($x_3, y_3$); A Bézier curve is specified by four points:

A start point ($(x_0, y_0)$ in the diagram below)
Two control points ($(x_1, y_1)$ and $(x_2, y_2)$ in the diagram below)
An end point ($(x_3, y_3)$ in the diagram below)

This diagram gives you a sense for how the control points "pull" the curve.

(Side note: Technically, this is a cubic Bézier curve. The curve is parametrized by \[B(t) = (1-t)^3(x_0, y_0) + 3 t(1-t)^2(x_1, y_1) + 3t^2(1-t)(x_2, y_2) + t^3(x_3, y_3), \; 0 \leq t \leq 1.\] For more details, see this article by Bill Casselman, a great expositor on the mathematics of computer graphics.)

You can even put your Bézier curve in the middle of a draw command, e.g. the following code:

LaTeX


\draw[thick, blue] (0, 0) -- (2, 2) .. controls (-3, 1) and (4, 7) .. (5, 8) -- (5, 5);

Back to our drawing, consider the left-most Bézier curve, given by:

LaTeX


\draw[thick, blue, ->] (-5, 4) .. controls (-5, 2) and (-4, 2) .. (-3, 2)
  .. controls (-2.1, 2) and (-2.1, 3) ..
  (-2.1, 5.25);

I want a vertical asymptote at $x=-2$. For readability sake, I actually make the curve end at $(-2.1, 5.25)$. The control point at $(-2.1, 3)$ pulls the curve back and gives the appearance of a vertical asymptote at $x = -2$. The option -> puts an arrowhead on the end of the curve.

Sidenote: Ensuring that a Bézier curve makes $y$ a function of $x$

It's possible to create a Bézier curve in which $y$ is not a function of $x$ (i.e., fails the vertical line test).

With some calculus, one can show that $x_0 \leq x_1 \leq x_2 \leq x_3$ will ensure this does not happen.¹

Vertical asymptote

I use a thick, dashed, red line segment from $(-2, -5.25)$ to $(-2, 5.25)$ to represent the vertical asymptote at $x = -2$.

LaTeX


\draw[thick, dashed, red] (-2, -5.25) -- (-2, 5.25);

Line segment part of graph

Part of the graph is the line segment from $(2, 4)$ to $(5, -2)$.

LaTeX


\draw[thick, blue] (2, 4) -- (5, -2);

Explicit points

A key advantage for making these graphs using TikZ is to incorporate my visual vocabulary for graphs. I follow the common convention where a filled-in circle indicates a point on the graph, while a hollow circle indicates a point which is not on the graph, e.g. $(-3, 2)$ is not on the graph, while $(-3, -3)$ is.

To accommodate this, we use circles, either filled with the same color as rest of the graph (for points on the graph) or filled with white (to make them hollow). These commands are last, so the white fill punches a hole in the graph.

LaTeX


\draw[thick, blue, fill=blue] (-5, 4) circle (1mm);

\draw[thick, blue, fill=white] (-3, 2) circle (1mm);

\draw[thick, blue, fill=blue] (-3, -3) circle (1mm);

\draw[thick, blue, fill=white] (2, -4) circle (1mm);

\draw[thick, blue, fill=blue] (2, 4) circle (1mm);

\draw[thick, blue, fill=blue] (5, -2) circle (1mm);

And with that, our graph is complete!

In the next post, I will explain how to use TikZ to plot graphs of functions defined by a mathematical formula.

This is because \[\frac{dx}{dt} = 3 \left((1 - t)^2 \left(x_1 - x_0\right) + 2t(1-t)\left(x_2 - x_1\right) + t^2 \left(x_3 - x_2\right) \right),\] and $0 \leq t \leq 1$. ↩

Plotting with TikZ, Part I: Why?

Wed, 1 May 2019 08:58:30 -0500

This is the first part of a three-part series of posts on generating plots of graphs with TikZ. Last year, I left my position as a mathematics professor, after teaching mathematics at the college level for 15 years (nine years as a faculty member, and six years as a graduate teaching fellow). In that time, I picked up a lot of tricks using LaTeX to produce teaching materials (handouts and slides). I decided it might be good to record some of that knowledge somewhere, and my blog seems like a natural place.

Outline

I'm planning a three-part series:

Part I will explain what TikZ is and why you might want to use it to produce plots for classroom use.
Part II will show how you can use TikZ to produce graphs for graph-reading and limit questions without having to make up functions.
Part III will show how I prefer to plot functions from a formula, using Python.

What is TikZ?

TikZ is a powerful LaTeX package for producing illustrations. TikZ is short for "TikZ ist kein Zeichenprogramm", which I understand means "TikZ is not a drawing program" in German. (Technically, TikZ comes bundled on top of another package called PGF, and the actual package is PGF/TikZ. I will refer to the package as just "TikZ" in this series of posts.) Unlike traditional illustration software such as Illustrator, in which the user would use a graphical interface create graphics, one creates graphics in TikZ with drawing commands. Here is an example of a short LaTeX document in which TikZ is used to produce a few shapes:

LaTeX

\documentclass[border=0.25cm]{standalone}

\usepackage{tikz}

\begin{document}
  \begin{tikzpicture}
    % Square
    \draw[dashed, blue] (0, 0) -- (0, 2)
    -- (2, 2) -- (2, 0) -- cycle;

    % Triangle
    \draw[thick, green, fill=green] (3, 0)
      -- (4, 2) -- (5, 0) -- cycle;

    % Circle
    \draw[thick, red, fill=purple]
      (6, 0) circle (0.5 cm);

    % Arrow
    \draw[->, thick, brown] (0, -1) -- (6.5, -1);

  \end{tikzpicture}
\end{document}

And here is the result:

Here's an example of an illustration I created for a calculus exam, using TikZ:

You can find a number of beautiful TikZ examples on TeXample.net and on the TeX StackExchange under the "TikZ-PGF" tag.

Why use TikZ to make plots?

To be frank, creating illustrations in TikZ can sometimes be time-consuming. For example, to create the water trough illustration above, it would not be unreasonable to prefer using illustration software or drawing the diagram freehand. For plots, you can make great illustration using SageMath, Desmos, or Matplotlib. But I still sometimes prefer TikZ to create plots.

Complete control means plots fit the look and feel of your class: When teaching, it's important that plots have a consistent look and feel, to avoid confusing students, and to communicate expectations for their hand-drawn plots. By creating plots from scratch with TikZ allows you to have complete control over elements such as axis placement and tick mark labeling.
Better integration with the document: Plots created with TikZ will use the same fonts as your core document. For example, if you decide to replace \usepackage{times} with \usepackage{fouriernc}, the plots will change too.
It's compatible with version control: I'm a bit of a version control fanatic, and unlike scans of hand-drawn sketches of plots or screen shots from other software, TikZ code plays very well with version control.
Incorporate your symbols for holes and asymptotes: With TikZ, you can incorporate your graph notation for features, such as a dotted line to illustrate an asymptote, or an unfilled circle to indicate a hole.
Make graphs of functions without having to actually come up with an explicit formula for the rule of the function: Sometimes it's nice to make up a function for illustration purposes without actually coming up with the formula for the rule of the function, for a problem such as:
1. (2 points each) A plot of the graph of the function $f$ is below:
For each of the following expressions, either evaluate the expression or state that it is undefined:

a. $\lim_{x \rightarrow 2^{-}} f(x)$

b. $\lim_{x \rightarrow 2^{+}} f(x)$

c. $\lim_{x \rightarrow 2} f(x)$

d. $f(2)$

[. . .]
With TikZ, you can make a plot like this using features such as Bezier curves.

In the next post, I will explain how to create a plot like the one above.

Migrated to Hugo

Thu, 25 Apr 2019 10:13:00 -0500

When I set up this blog in 2012, WordPress was a natural choice for content management system. However, I recently decided to investigate using Hugo to generate this blog, and after some playing around, I made the leap. The biggest reason: I wanted the benefits of a statically-generated site. (Hugo is also more compatible with version control systems.)

For the actual migration, I used Cyrill Schumacher's WordPress to Hugo Exporter plug-in. I did have to do some manual futzing.

While I was at it, I merged my main site (chrisphan.com) with the blog (blog.chrisphan.com). I also made self-hosted copies of the photos that were on Flickr and posted here.

I also found these instructions by Euan T. to be very helpful!

Disclaimer

Tue, 23 Apr 2019 12:00:00 +0000

chrisphan.com is my personal web site, and content provided here is provided by me in my personal capacity.

The opinions expressed on this site are my own and do not necessarily represent the opinion of any of my employers (former or current) or any organization with which I am or was affiliated.

Math page for Chistopher Phan

Mon, 22 Apr 2019 21:46:53 -0500

Academic Background

Education

Ph.D. 2009, mathematics, University of Oregon
- Advisor: Professor Brad Shelton
- Dissertation: Koszul and generalized Koszul properties for noncommutative graded algebras
M.S. 2005, mathematics, University of Oregon
B.A. 2003, mathematics and communication, Lewis & Clark College

Academic positions

Fixed-term Assistant Professor, Department of Mathematics and Statistics, Winona State University, Winona, Minnesota, August 2012-May 2018.
Visiting Assistant Professor, Department of Mathematics, Bucknell University, Lewisburg, Pennsylvania, August 2010-May 2012.
Temporary Lecturer, Department of Mathematics, University of Glasgow, Glasgow, Scotland, September 2009-August 2010.
Adjunct Instructor, General Education Department, Cooking & Hospitality Institute of Chicago, Chicago, Illinois, August-September 2009.
Graduate Teaching Fellow, Department of Mathematics, University of Oregon, Eugene, Oregon, September 2003-June 2009.

Full CV

My full curriculum vitae (last updated August 2018)

Teaching demos

These are some materials that I created for use in the classroom.

Interactive demos

The derivative as a function
Behavior of exponential functions
ODEs of the form $y'' + by' + cy = 0$
Forced vibrations (ODEs of the form $y'' + 0.125y' + y = 3 \cos(\omega t)$.)
Implicit differentiation

Non-interactive materials

Ring Theory Research

I was a noncommutative ring theorist whose primary research interest was homological results involving graded algebras. For example, I studied the Koszul and $\mathcal{K}_2$ properties.

Here is my statement of research interests, from back when I was on the academic job market (circa 2012).

Research articles

Quotients of Koszul algebras and 2-$d$-determined algebras (with T. Cassidy), Communications in Algebra, 42 (2014), 3742–3752. Preprint available at arXiv:1210.3847 [math.RA]. MR3200055

Vatne and Green & Marcos have independently studied the Koszul-like homological properties of graded algebras that have defining relations in degree 2 and exactly one other degree. We contrast these two approaches, answer two questions posed by Green & Marcos, and find conditions that imply the corresponding Yoneda algebras are generated in the lowest possible degrees.
The Yoneda algebra of a graded Ore extension, Communications in Algebra, 40 (2012) 834–844. Preprint available at arXiv:1002.2318 [math.RA]. MR2899911

Let $A$ be a connected-graded algebra with trivial module $k$, and let $B$ be a graded Ore extension of $A$. We relate the structure of the Yoneda algebra $\mathrm{E}(A):= \mathrm{Ext}_A(k,k)$ to $\mathrm{E}(B)$. Cassidy and Shelton have shown that when $A$ satisfies their $\mathcal{K}_2$ property, $B$ will also be $\mathcal{K}_2$. We prove the converse of this result.
Localization algebras and deformations of Koszul algebras (with T. Braden, A. Licata, N. Proudfoot, and B. Webster), Selecta Mathematica, 17 (2011) 533–572. Preprint available at arXiv:0905.1335 [math.RA]. MR2827176

We show that the center of a flat graded deformation of a standard Koszul algebra $A$ behaves in many ways like the torus-equivariant cohomology ring of an algebraic variety with finite fixed point set. In particular, the center of $A$ acts by characters on the deformed standard modules, providing a “localization map”. We construct a universal graded deformation of $A$ and show that the spectrum of its center is supported on a certain arrangement of hyperplanes which is orthogonal to the arrangement coming from the algebra Koszul dual to $A$. This is an algebraic version of a duality discovered by Goresky and MacPherson between the equivariant cohomology rings of partial flag varieties and Springer fibers; we recover and generalize their result by showing that the center of the universal deformation for the ring governing a block of parabolic category $\mathcal{O}$ for is isomorphic to the equivariant cohomology of a Spaltenstein variety. We also identify the center of the deformed version of the “category $\mathcal{O}$” of a hyperplane arrangement (defined by the authors in a previous paper) with the equivariant cohomology of a hypertoric variety.
Noncommutative Koszul algebras from combinatorial topology (with T. Cassidy and B. Shelton), Journal für die reine und angewandte Mathematik (Crelle’s Journal), 646 (2010) 45–63. Preprint available at arXiv:0811:3450 [math.RA]. MR2719555

Associated to any uniform finite layered graph $\Gamma$ there is a noncommutative graded quadratic algebra $A(\Gamma)$ given by a construction due to Gelfand, Retakh, Serconek and Wilson. It is natural to ask when these algebras are Koszul. Unfortunately, a mistake in the literature states that all such algebras are Koszul. That is not the case and the theorem was recently retracted. We analyze the Koszul property of these algebras for two large classes of graphs associated to finite regular CW-complexes, $X$. Our methods are primarily topological. We solve the Koszul problem by introducing new cohomology groups $H_X(n,k)$, generalizing the usual cohomology groups $H_n(X)$. Along with several other results, our methods give a new and primarily topological proof of the main result of [Serconek and Wilson, J. Algebra 278: 473–493, 2004] and [Piontkovski, J. Alg. Comput. 15, 643–648, 2005].
The Yoneda algebra of a $\mathcal{K}_2$ algebra need not be another $\mathcal{K}_2$ algebra (with T. Cassidy and B. Shelton), Communications in Algebra, 38 (2010) 46–48. Preprint available at arXiv:0810.4656 [math.RA]. MR2597480

The Yoneda algebra of a Koszul algebra or a $D$-Koszul algebra is Koszul. $\mathcal{K}_2$ algebras are a natural generalization of Koszul algebras, and one would hope that the Yoneda algebra of a $\mathcal{K}_2$ algebra would be another $\mathcal{K}_2$ algebra. We show that this is not necessarily the case by constructing a monomial $\mathcal{K}_2$ algebra for which the corresponding Yoneda algebra is not $\mathcal{K}_2$.
Generalized Koszul properties for augmented algebras, Journal of Algebra, 321 (2009) 1522–1537. Preprint available at arXiv:0711.3480 [math.RA]. MR2494406

Under certain conditions, a filtration on an augmented algebra $A$ admits a related filtration on the Yoneda algebra $\mathrm{E}(A) := \mathrm{Ext}_A(K,K)$. We show that there exists a bigraded algebra monomorphism

\[\mathrm{gr}\;\mathrm{E}(A) \hookrightarrow \mathrm{E}_{\mathrm{Gr}}(\mathrm{gr}\;A),\]

where $\mathrm{E}_{\mathrm{Gr}}(\mathrm{gr}\;A)$ is the graded Yoneda algebra of $\mathrm{gr}\;A$. This monomorphism can be applied in the case where A is connected graded to determine that $A$ has the $\mathcal{K}_2$ property recently introduced by Cassidy and Shelton.

Dissertation

Koszul and generalized Koszul properties for noncommutative graded algebras, University of Oregon, Department of Mathematics, 2009.
- Advisor: Professor Brad Shelton
We investigate some homological properties of graded algebras. If $A$ is an $R$-algebra, then $\mathrm{E}(A):= \mathrm{Ext}_A(R,R)$ is an $R$-algebra under the cup product and is called the Yoneda algebra. (In most cases, we assume $R$ is a field.) A well-known and widely-studied condition on $\mathrm{E}(A)$ is the Koszul property. We study a class of deformations of Koszul algebras that arises from the study of equivariant cohomology and algebraic groups and show that under certain circumstances these deformations are Poincaré–Birkhoff–Witt deformations.

Some of our results involve the $\mathcal{K}_2$ property, recently introduced by Cassidy and Shelton, which is a generalization of the Koszul property. While a Koszul algebra must be quadratic, a $\mathcal{K}_2$ algebra may have its ideal of relations generated in different degrees. We study the structure of the Yoneda algebra corresponding to a monomial $\mathcal{K}_2$ algebra and provide an example of a monomial $\mathcal{K}_2$ algebra whose Yoneda algebra is not also $\mathcal{K}_2$. This example illustrates the difficulty of finding a $\mathcal{K}_2$ analogue of the classical theory of Koszul duality.

It is well-known that Poincaré–Birkhoff–Witt algebras are Koszul. We find a $\mathcal{K}_2$ analogue of this theory. If $V$ is a finite-dimensional vector space with an ordered basis, and $A:=\mathbb{T}(V)/I$ is a connected-graded algebra, we can place a filtration $F$ on $A$ as well as $\mathrm{E}(A)$. We show there is a bigraded algebra embedding

\[\Lambda: \mathrm{gr}\;\mathrm{E}(A) \hookrightarrow \mathrm{E}_{\mathrm{Gr}}(\mathrm{gr}\;A).\]

If $I$ has a Gröbner basis meeting certain conditions and $\mathrm{gr}_F\;A$ is $\mathcal{K}_2$, then $\Lambda$ can be used to show that $A$ is also $\mathcal{K}_2$.

This dissertation contains both previously published and co-authored materials.

“10 PRINT” in PostScript

Mon, 18 Feb 2019 22:00:40 -0600

Back in May, I posted a LaTeX document using TikZ to implement the Commodore 64 BASIC program:

BASIC

10 PRINT CHR$(205.5+RND(1)); : GOTO 10

The amusement factor (for me) is due to LaTeX's ability to generate pseudorandom numbers, which may be surprising to some, especially those who think of LaTeX documents as static (deterministic) documents and not computer programs.

PostScript can also generate pseudorandom numbers, and hence I present "10 PRINT" in PostScript:

PostScript

%!

% Implementation of "10 PRINT" in PostScript
% Christopher Phan, cphan@chrisphan.com
% 2019-02-17

% Draw frame around page

newpath
  36 36 moveto % 36 pt = 0.5 in
  36 756 lineto % 756 pt = 10.5 in
  576 756 lineto % 576 pt = 8 in
  576 36 lineto
  closepath
stroke

% each slash will occupy a 0.25 in x 0.25 in square

36 18 558 { % 18 pt = 0.25 in
  /x exch def
  36 18 738 {
    /y exch def
    newpath
      /u rand 2 mod 18 mul def
      x y u add moveto
      x 18 add y 18 u sub add lineto
    closepath stroke
  } for
} for

showpage

Example output:

“10 PRINT” in TikZ

Mon, 28 May 2018 16:42:51 -0500

The book 10 PRINT CHR$(205.5+RND(1)); : GOTO 10, by Nick Montfort, et. al., uses a one-line Commodore 64 BASIC program “as a gateway into a deeper understanding of how computing works in society and what the writing, reading, and execution of computer code mean” (p. 4). The focus is on the titular program, which the authors call 10 PRINT for short:

BASIC

10 PRINT CHR$(205.5+RND(1)); : GOTO 10

This program prints an infinite random sequence made up of the box-drawing characters ╱ and ╲ (that’s U+2571 and U+2572, respectively, encoded in the Commodore character set PETSCII as 205 and 206.), in order to make a random maze. This produces an output similar to the following, except that the characters on a Commodore screen are more square:

(The book’s web site gives a better idea how the output on a Commodore would look.)

It’s not hard to reproduce 10 PRINT in Python, albeit using more than one line:

Python

#! /usr/bin/env python3

import random

for j in range(0, 24):
    print("".join([random.choice(["\u2571","\u2572"])
      for k in range(0, 40)]))

(Instead of an infinite string, my program only outputs a 40 × 24 block of characters.) To get the output to look right, you might need to mess with the font settings in your Terminal. I found that that Melno Regular font on my Mac works.

But I really thought it would be more fun to obtain a 10 PRINT-like output using TikZ, as TikZ is capable of generating random numbers! Instead of using the characters ╱ and ╲, I will draw the lines directly:

LaTeX

\documentclass{standalone}

\usepackage{tikz}

\begin{document}

  \begin{tikzpicture}

    % Draw background
    \fill[blue!75!black] (0, 0) -- (0, 24) --
    (40, 24) -- (40, 0) -- cycle;

    % Draw maze
    \foreach \y in {0, ..., 23}{
      \foreach \x in {0, ..., 39}{

        % Randomly choose a = 0 or a = 1
        \pgfmathrandominteger{\a}{0}{1};

        % If a = 0, then will draw SW-NE line
        % If a = 1, then will draw NW-SE line
        \draw[very thick, white] (\x, \y + \a)
          -- (\x + 1, \y + 1 - \a);
      }
    }

  \end{tikzpicture}

\end{document}

I really like the output:

LaTeX mailmerge package

Sun, 27 May 2018 21:40:57 -0500

The LaTeX mailmerge package is super useful, especially for creating multiple versions of tests.

Before I discovered mailmerge, my workflow for writing a test was basically:

Write one version of the test.
Make a duplicate of the file.
Edit the file to change all the constants.

At least, that was my intended workflow. But typically I would also:

Realize that there was some change I wanted to make to the test.
Painstakingly make the same change to both versions of the test, double-checking that the wording remained consistent between versions.

Sometimes I would repeat the last two steps multiple times. This was obnoxiously inefficient.

Then I discovered the mailmerge package.

Here is a basic (silly) example illustrating how the package works:

LaTeX

\documentclass[letterpaper, 12pt]{amsart}

%% packages
\usepackage{mailmerge}
\usepackage{fouriernc}
\usepackage{url}
\usepackage{tikz}

\title{Some U.S.~state capital facts}

\begin{document}

  \maketitle

  \mailrepeat{
    The capital of \field{state} is \field{capital}.
    The U.S. Census Bureau estimates that in mid-2017,
    there were \field{population} people living in
    \field{capital}\footnote{\field{cite}}.

  }

  \mailfields{state,capital,population,cite}

  \mailentry{Illinois,Springfield,%
    {167,376},\url{https://goo.gl/YGDdE6}}

  %% Draw a square:
  \vskip 1 em
  \begin{tikzpicture}
    \draw (0, 0) -- (1, 0)
      -- (1, 1) -- (0, 1) -- cycle;
  \end{tikzpicture}
  \vskip 1 em

  \mailentry{Minnesota,St.~Paul,%
    {306,621},\url{https://goo.gl/F6rXZ7}}

  \mailentry{Oregon,Salem,%
    {169,798},\url{https://goo.gl/Zc4tDC}}

  \mailentry{Pennsylvania,Harrisburg,%
    {49,192},\url{https://goo.gl/NQypQG}}

\end{document}

This produces:

SOME U.S. STATE CAPITAL FACTS

The capital of Illinois is Springfield. The U.S. Census Bureau estimates that in mid-2017, there were 167,376 people living in Springfield¹.

The capital of Minnesota is St. Paul. The U.S. Census Bureau estimates that in mid-2017, there were 306,621 people living in St. Paul².

The capital of Oregon is Salem. The U.S. Census Bureau estimates that in mid-2017, there were 169,798 people living in Salem³.

The capital of Pennsylvania is Harrisburg. The U.S. Census Bureau estimates that in mid-2017, there were 49,192 people living in Harrisburg⁴.

This is followed by the appropriate footnotes.

Notice:

The text you want repeated is specified as the argument of the \mailrepeat macro.
Within the \mailrepeat, we specify the mail-merged info with \field{[field name]}
The names of the mail merge fields are given as a comma-delimited list with the \mailfields macro.
The actual mail-merge is accomplished with the \mailentry macro. The argument to this macro is a comma-delimited list of the material to place into the fields.

I put the silly TikZ square in this example to illustrate that the mail-merged text is placed where the \mailentry occurs in the code.

Spacing is significant in the \mailentry macro. If there is a space next to \field{[field name]} in the \mailrepeat and after the comma the corresponding field in the \mailentry, this will produce an awkward amount of space in the output. For example,

LaTeX


\mailentry{Minnesota, St.~Paul,
{306,621},\url{https://goo.gl/F6rXZ7}}

will produce:

Notice the extra space before the two instances of “St. Paul” and before the population. Compare to what is produced when there is no space before the comma and there is a % before the line break (which prevents the line break from being treated as a space):

LaTeX


\mailentry{Minnesota,St.~Paul,%
{306,621},\url{https://goo.gl/F6rXZ7}}

It’s very subtle, but I am a perfectionist.

If you have many fields, the comma-delimited list argument to the \mailentry can get unwieldy. One trick I use is to end each entry with a comment containing the field name:

LaTeX


\mailentry{Minnesota%state
,St.~Paul%capital
,{306,621}%population
,\url{https://goo.gl/F6rXZ7}%cite
}

Especially when the fields are difficult to tell apart based on content (e.g., when they are different constants for math problems), this makes it easier to know which field you are changing. Also, the comments prevent any space or line break before the comma from affecting the output.

For a test, each version would correspond to a single \mailentry. Consider this made-up quiz, which has two versions (Form A and B):

LaTeX

\documentclass[letterpaper, 11pt]{article}

%% packages
\usepackage{amsmath}
\usepackage{mailmerge}
\usepackage{fouriernc}
\usepackage{enumitem}
\usepackage{fullpage}

\begin{document}

  \mailrepeat{
    \setcounter{page}{1}
    \centerline{\textbf{Quiz \#1}, Form \field{form}}
    \vskip 1 em

    \begin{enumerate}
      \item (2 points) Expand
      \(\left(x^{\field{exponent1}} + 2x\right)^2\).


      \vfill

      \item (3 points) Suppose
      \(f(x) =
        \field{coeff1}x^2 + \field{coeff2}x + 1\).
      Find \(f(5)\).
      \vfill

    \end{enumerate}

    \newpage

    \begin{enumerate}[resume]
      \item (5 points) Solve for \(x\):
      \[\field{coeff3}x^4 - \field{coeff4}x^3
        = 0\]

      \vfill

    \end{enumerate}
    \newpage

  }

  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
  %%% Mail merge info:

  \mailfields{form, coeff1, coeff2,
  coeff3, coeff4, exponent1}

  \mailentry{A%form
  ,6%coeff1
  ,8%coeff2
  ,2%coeff3
  ,7%coeff4
  ,3%exponent1
  }

  \mailentry{B%form
  ,5%coeff1
  ,9%coeff2
  ,3%coeff3
  ,5%coeff4
  ,5%exponent1
  }
\end{document}

At the start of the \mailrepeated code is the macro \setcounter{page}{1}, which resets the page number to 1; otherwise, Form B would start on Page 3. Also, I ended the \mailrepeated code with a \newpage, ensuring that each version is on a separate page.

If you print $\lceil N/2\rceil$ copies of the resulting PDF (where $N$ is the number of students in your course), two-sided, then the pile of quizzes will be alternated by version as they come out of the printer, ready to hand out to your class.

LaTeX menukeys package

Fri, 27 Apr 2018 10:52:22 -0500

I’d like to put in a quick word in favor the LaTeX menukeys package. This package makes really nice menu sequences, which is useful when you have to explain how to use software.

For example, \menu[>]{Tools > Web Developer > Page Source} produces:

Web Developer > Page Source" class="white_bk" />

Why I use version control

Fri, 13 Apr 2018 13:07:23 -0500

Bash
cp t_test3.pg inclass_t_test2.pg

Oh, crap! I wanted to copy that to inclass_t_test3.pg, not overwrite inclass_t_test2.pg!

Bash
git status

On branch master
Your branch is up-to-date with 'origin/master'.
Changes not staged for commit:
  (use "git add ..." to update what will be committed)
  (use "git checkout -- ..." to discard changes in working directory)

        modified: inclass_t_test2.pg

no changes added to commit (use "git add" and/or "git commit -a")

git checkout -- inclass_t_test2.pg
git status

On branch master
Your branch is up-to-date with 'origin/master'.
nothing to commit, working directory clean

Whew!!

Bash
cp t_test3.pg inclass_t_test3.pg

I’ve (mostly) stopped using Facebook

Wed, 2 Aug 2017 20:32:34 -0500

I have mostly stopped using Facebook. I’ve only made two posts to my own timeline since the beginning of the year, and one was to update my bio with a message explaining that I’ve stopped using Facebook (which I will probably edit to include a link to this post). I’m not a “Facebook vegan”: I still log in (occasionally) and comment or like others’ posts (even less occasionally). I also occasionally make posts for various organizations with which I’m involved (but I post them to those organizations’ pages). But I don’t log in daily (or even weekly), as I used to do.

At this moment, I’m not going to write a long essay about why I’ve (mostly) stopped using Facebook, and I’m not going to tell everyone else they should, either. It’s true that I have some serious concerns about Facebook and its impact on the world, many of which have been explained by Cory Doctorow better than I could. However, my main reason for (mostly) eschewing Facebook is personal: I became convinced late last year that my use of Facebook was harmful to my mental health, and drastically reducing my participation with Facebook has been a positive change for me.

So, please don’t take it personally if I don’t respond to your friend request, comment, or post in which you tagged me. (It’s likely I didn’t see it) And if you want to communicate with me, you’ll have much better luck if you use email or phone, rather than Facebook.

Some politics observations, 2017-02-28

Tue, 28 Feb 2017 21:54:49 -0600

Some observations:

Refugees are risking frostbite to flee the U.S. for Canada. This is a very sad commentary on the lack of compassion from our current administration.
I signed the Unitarian Universalist Association’s declaration of conscience, and encourage you to consider signing it as well.
I agree with Congressman Ellison:

To all my supporters: we may have come up short, but we need to be united. I look forward to continuing to work for the people in MN-05. pic.twitter.com/sab2fUmDZu

— Ellison for Congress (@EllisonCampaign) February 25, 2017

Democrats need to stay united! We need to work together to regain as many seats in Congress as we can in 2018.

Adventures in TikZ: tkz-graph

Wed, 22 Feb 2017 22:32:15 -0600

The other day, I was writing some lecture notes for my linear algebra class, and wanted to create the following diagram (to illustrate the concept of a Markov chain):

I had a very limited time in which to finish these notes. Fortunately, I found the tkz-graph package, which made this a snap:

LaTeX

\documentclass{standalone}
\usepackage{tikz}
\usepackage{fouriernc}
\usepackage{tkz-graph}

\begin{document}
\begin{tikzpicture}
\SetGraphUnit{5}

\Vertex[x=0, y=10]{0 points};
\Vertex[x=0, y=5]{1 point};
\Vertex[x=0, y=0]{Win};
\Vertex[x=5, y=5]{Lose};

\Edge[style ={->}, label={$1/3$}]({0 points})({1 point});
\Edge[style ={->}, label={$1/3$}]({1 point})({Win});
\Edge[style ={->}, label={$1/6$}]({0 points})({Lose});
\Edge[style ={->}, label={$1/6$}]({1 point})({Lose});

\Loop[style ={->}, label={$1/2$}, labelstyle={fill=white}]({0 points});
\Loop[style ={->}, label={$1/2$}, labelstyle={fill=white}]({1 point});
\Loop[style ={->}, label={$1$}, dir=EA, labelstyle={fill=white}]({Lose});
\Loop[style ={->}, label={$1$}, labelstyle={fill=white}]({Win});

\end{tikzpicture}
\end{document}

You don’t even have to specify the locations of the vertices; you can throw caution to the wind and have LaTeX decide where to place them! (I am a bit too much of a perfectionist for that.)

One slight issue I had was that the documentation for this package (at least on my computer, as retrieved by texdoc) was in French. Fortunately, I seem to have retained enough knowledge since I took the French language exam as a grad student that I could read most of the documentation.

It’s been a long time

Sun, 12 Feb 2017 17:26:53 -0600

It’s been a long time (over a year) since I’ve posted on this blog, because I have (to put it mildly) been very busy with other responsibilities and passions that have taken me away from blogging. Also, I serve as a (low-level, volunteer-basis) officer in a political party, and as a result, I am sometimes reluctant to post my opinions in public, for fear that they might be taken (or portrayed) as official statements, despite my disclaimer (which, to be clear, says that everything written here is my personal opinion and does not reflect the position of my employer or any organization of which I am a member).

However, we are now facing a national emergency, and it is important for people to speak out. And I’ve decided that I distrust Twitter and Facebook as platforms for doing so (a topic on which I will elaborate later), leading to my desire to start writing again here. I certainly don’t have time for this, but I am going to try to make the time, hopefully posting here more frequently than once every two years.

I have also added https/SSL to this blog, using Let’s Encrypt. I took this step a few months ago, right as the national emergency began, and promptly could not log into the interface for this blog. Because I was (and continue to be) so busy, I put off fixing the problem, only to discover that the problem seems to have fixed itself. Go figure.

While I do plan on talking about politics on this blog, I also have other interests (mathematics, for example), and so I will be posting on these as well.

By the way, I don’t have time for comment moderation, so I don’t plan to enable comments on my posts.

Book recommendation: How Not to Be Wrong

Sat, 26 Sep 2015 22:14:39 -0500

Today, I finished reading How Not to Be Wrong: The Power of Mathematical Thinking by Jordan Ellenberg. This is a very enjoyable, very well-written, general-audience book about mathematics, which I recommend whole-heartedly.

Ellenberg, a math professor at the University of Wisconsin, does a great job weaving together a plethora of mathematical topics, including non-Euclidean geometry, probability, statistics, and mathematical analysis of voting systems. He writes in a way that someone who only vaguely remembers—or never really understood—high school algebra would be able to follow and enjoy. His exposition is made more lively by a cast of historical and contemporary characters, some famous and some primarily known only to mathematicians, including Abraham Wald, Bernhard Riemann, Teddy Roosevelt, Francis Galton, Voltaire, Nicolas de Condorcet, David Hilbert, Ronald Fisher, Antonin Scalia, and Nate Silver. (My favorite line in the book is the one in which Ellenberg describes Silver as a “Kurt Cobain of probability.”)

The best part of the book is about how the Massachusetts Lottery ran a game in which it was occasionally profitable to play (that is, there were some drawings in which the expected value of a ticket’s winnings was higher than the price of a ticket). Of course, some smart people (e.g. an MIT student) figured this out and recruited investors to buy absurd numbers of tickets for the profitable drawings. In the course of telling this story, Ellenberg weaves in discussions of finite geometries and error-correcting codes, both of which are relevant in describing how one buys thousands of lottery tickets without accidentally having to split winnings with yourself.

I think it would be awesome to use this book in a gen-ed math class.

In defense of FERPA

Sat, 12 Sep 2015 21:25:16 -0500

In an op-ed piece titled “College kids have too much privacy“, Michele Willens criticizes the Family Educational Rights and Privacy Act (FERPA) for making it difficult for families to monitor their college student children’s academic performance.

I can understand why some families are frustrated by FERPA. Many families spend a great deal of money to send their young to college. Consider the parents in Willens’s opening anecdote, who learned their daughter had not actually graduated and skipped class for the last two years. I think anyone would be angry to be in their shoes.

However, it’s just not true that FERPA means, as Willens puts it, that “you have to take [your kid’s] word for it when they say ‘everything’s fine.'” FERPA allows a student to voluntarily release their records to another party. I recall having at least one scholarship in college whose sponsors required me to send an official transcript every year. I don’t see why a parent couldn’t insist on this as a condition for continued financial support.

In addition, a student can sign a FERPA waiver allowing a third party to get their records directly from the school. Although Willens describes this as a “laborious process”, it really amounts to having a student sign a form or check a box on a web site.

Part of being in college is learning how to become an independent adult. The transfer of FERPA rights from parents to student is part of that process. So is students learning to take responsibility for their learning. All FERPA requires is for parents to deal with their adult children directly, as adults, rather than being able to go behind their back.

Some thoughts about online discourse

Fri, 15 May 2015 09:41:32 -0500

I hate it when people reply or comment to something you posted, in a way that makes it clear they didn't read what you wrote or posted, (1/)

— Chris Phan (@functoruser) May 15, 2015

and that they were just looking for an argument. It's disrespectful and rude. Also, I'm not looking for arguments on the Internet. (2/)

— Chris Phan (@functoruser) May 15, 2015

I'm looking for dialogue, discussion, and reasonable disagreement. Debate is good. Glib, unnecessarily aggressive responses are not. (3/3)

— Chris Phan (@functoruser) May 15, 2015

Busy recently…

Sun, 19 Oct 2014 17:40:24 -0500

I’ve been very busy recently with work and various other things recently.

pic.twitter.com/o0PnRFbCq9

— Chris Phan (@functoruser) October 19, 2014

Sum question!

Wed, 24 Sep 2014 19:51:21 -0500

The American Mathematical Society made the following amusing post to their Facebook page today:

It's National Punctuation Day (at least in the U,S,)! To honor the occasion, for positive integers n, let n? = 1 + 2 + ... + n. What is 3???? (Hint: < 1000)

You can calculate 3???? = 26796, which of course is greater than 1000.

The joke is that they aren’t asking you to calculate 3????. They are asking you to calculate 3???. (This is 231.) The last question mark in the sentence “What is 3????” is the punctuation that ends the sentence, not the postfix notation for adding the first n positive integers. You could rewrite the sentence as “Find 3???.” Of course, that wouldn’t be as funny.

MVPB and me, yesterday morning

Thu, 14 Aug 2014 22:07:46 -0500

Good morning! pic.twitter.com/qR3phXAszo

— Chris Phan (@functoruser) August 13, 2014

He’s really growing!

Not all cars

Mon, 23 Jun 2014 09:55:44 -0500

Imagine there was a serious manufacturing defect present in about six percent of cars. This defect caused a malfunction that could seriously injure the car’s passengers. Of course, consumers would demand action to eradicate this defect. Suppose auto industry representatives dismissed these concerns by protesting “Not all cars blow up because of this defect.”

This joke is awesome and deserves more than just a retweet

Wed, 11 Jun 2014 16:20:00 -0500

This joke is awesome and deserves more than just a retweet:

I'd wager that right now, Eric Cantor’s sorrows are…

(•_•)
( •_•)>⌐■-■
(⌐■_■)

uncountable.

— “patrick” (@importantshock) June 11, 2014

(Georg Cantor’s diagonal argument shows that the real numbers are uncountable.)

Um, some conservatives DO dispute E=mc^2

Sun, 8 Jun 2014 15:20:03 -0500

A few times on my Facebook news feed, I’ve noticed a graphic with a photo of astrophysicist and (awesome) science popularizer Neil deGrasse Tyson and this quotation of him (emphasis and ellipsis in original):

Climate change has taken on political dimensions… That’s odd because I don’t see people choosing sides over $E=mc^2$ or other fundamental facts of science!

It appears that Dr. Tyson has not checked out Conservapedia, a right-wing competitor to Wikiepdia. (I certainly don’t blame him for this—he’s got to have much better things to do than read a web site full of stupid garbage.) Conservapedia takes issue with relativity overall, and says of $E=mc^2$ (emphasis in original):

Political pressure, however, has since made it impossible for anyone pursuing an academic career in science to even question the validity of this nonsensical equation. Simply put, $E=mc^2$ is liberal claptrap.

(The whole $E=mc^2$ Conservapedia article is actually pretty funny. My wife, who is a physicist, thought it was a joke.)

Conservapedia is a project of Andrew Schlafly, son of prominent conservative activist Phyllis Schlafly. Indeed, you can see from my link above (adorned with rel="nofollow", of course!) that the most recent edit to the $E=mc^2$ article was by Mr. Schlafly himself.

Stop punching down at millennials

Fri, 6 Jun 2014 11:49:56 -0500

Um, complaining loudly in a coffee shop about how much you dislike millennials is pretty rude.

— Chris Phan (@functoruser) June 6, 2014

Also, it's hilarious how your complaint boils down to "They were raised poorly." And whose fault would that be?

— Chris Phan (@functoruser) June 6, 2014

At least my parents raised me to know that complaining about people loudly in coffee shops (and making gross generalizations) is rude.

— Chris Phan (@functoruser) June 6, 2014

(FWIW, I don't think millennials were raised poorly.)

— Chris Phan (@functoruser) June 6, 2014

I absolutely hate millennial-bashing. While I don’t consider myself a millennial—I’m just a tad bit too old—I have a great deal of sympathy and affinity for that generation. That generation has been dealt a bad hand. Many of the manufacturing jobs that provide a path to prosperity for those without college degrees have been moved overseas, and as a result millennials desiring prosperity were told they had to go to college. Alas, millennials did not enjoy the same level of government subsidy for their college education as their parents, so many of them have been saddled with crippling amounts of student debt. Furthermore, many millennials came out of college right as the economy tanked due to the financial sins of their parents’ generation. Yet, it’s common to hear people blame millennials for their inability to launch.

Annoyingly, much of the millennial-bashing you hear boils down to complaints about how they were raised. Millennials, many baby boomers will tell you, were praised too much as children. There was too much emphasis on self-esteem. They were given too high of expectations by their parents. For whatever it’s worth, I think these complaints are over-wrought. In my experience, millennials don’t suffer from too much self-esteem and many have adjusted expectations—I think more than they should have to—to today’s economic realities. But even if these complaints were accurate, it takes quite a bit of chutzpah for someone to say, essentially, “Ha ha, your generation sucks because we did a poor job raising you!”

Millennial-bashing by older people is good example of “punching down”, that is, it’s people using their power to attack someone with less power—which I find very distasteful. And criticism of millennials is used as a convenient excuse to avoid questioning the policy decisions (such as financial deregulation, globalization, and defunding of post-secondary education) that have put that generation at a disadvantage.

Tech support scam

Tue, 3 Jun 2014 20:24:34 -0500

Today, I was the target of a tech support scam.

I received a phone call from someone purporting to work in tech support. The caller said they were calling about my Windows computer. Thinking they might have the wrong number, I replied that I didn’t have a Windows computer. The caller then hung up, without even apologizing or saying goodbye.

The abrupt ending of the call made me suspicious, and so I did some research. It turns out that this is known scam. I found information from both the Federal Trade Commission and Microsoft about this scam:

According to these sites, the scammers sometimes try to convince you that your computer has a virus or malware, and then trick you into installing malware or divulging your password or credit card number. Sometimes the scammers claim to be working for Microsoft or tech support providers.

I ran the caller’s number through a search engine, and found reports from others that are consistent with Microsoft’s and the FTC’s descriptions of the scam.

I decided to post about my experience in order to spread the word about these scams. In closing, here’s some of the FTC’s advice for those cold-called about computer security issues:

If you get a call from someone who claims to be a tech support person, hang up and call the company yourself on a phone number you know to be genuine. A caller who creates a sense of urgency or uses high-pressure tactics is probably a scam artist.

Brains over calculators, part 2.

Thu, 22 May 2014 16:21:47 -0500

I recently found another calculation that I would consider reasonable for my students to perform that their calculators cannot. On my Calc II students’ final exam, I asked them to evaluate the infinite series $\displaystyle \sum_{n = 1}^\infty \frac{4}{(n + 3)(n+5)}$. As a hint, I told them to rewrite in telescoping form using partial fractions. However, the TI-89 cannot evaluate this series. Since it’s hard to take a good photo of the calculator screen, I’ve instead included a screen shot of the TI-Nsipre CAS iPad app.

Congrats, Oregon! Welcome to marriage equality!

Wed, 21 May 2014 12:21:05 -0500

Oregon Teacher of the Year Brett Bigham and his spouse Mike Turay just married by our own Jeana Frazzini pic.twitter.com/GPSE7S0Dmv

— Basic Rights Oregon (@basicrights) May 19, 2014

1st marriage in Eugene: Bruce and Matt together for 40 years! pic.twitter.com/pIoBxHtwJ8

— Basic Rights Oregon (@basicrights) May 19, 2014

On Monday, a federal judge in Eugene struck down Oregon’s state constitutional ban on same-sex marriage, which was enacted by citizen initiative nearly a decade ago. This ruling was unsurprising, given that the defendants in the lawsuit, such as Oregon’s attorney general, agreed with the plaintiffs’ contention that the ban violated the federal constitution, leaving no one with standing willing to defend the ban’s constitutionality.

I cannot tell you how happy I was about this ruling. I grew up in Oregon, and lived there through the first decade of my adult life. I voted against Measure 36, the ballot measure that added discrimination to the Oregon Constitution, and was quite upset when it passed in 2004. Seeing pictures of Oregonian couples finally able to be married was really heartwarming to me.

Furthermore, my wife and I were married in Oregon; a framed copy of our marriage record from the Oregon Department of Health Services hangs on the wall near my desk in my home office. We chose our officiant, a senior judge, because she was on record as enthusiastic to marry same-sex couples during the brief window in 2004, before Measure 36 passed, when Multnomah County was issuing marriage licenses to same-sex couples. We encouraged our wedding guests to donate to Basic Rights Oregon, in the hope that same-sex couples could someday be married under the same laws we were. I am very happy to see that day finally come.

Yet, I must echo the sentiments expressed by a friend, who wrote on Twitter:

Oh, Oregon. It took a federal judge to tell you to play nice, but that bit of love for you that I was withholding is still gushing forth.

— qousqous (@qousqous) May 19, 2014

I am sad that marriage equality didn’t come to Oregon by the voters or their elected representatives, but through the actions of a federal judge. Now, please don’t get me wrong, I think it’s completely appropriate that a judge strike down a law if it violates the constitution. I’m not saying it’s bad that a judge overturned the will of the people—judges absolutely should overturn the will of the people when the people want to violate the constitution. But I am much less happy when a state does the right thing because it’s forced to, rather than because it chose to.

Fate has decided that my wife and I make our home in the wonderful state of Minnesota. I have discovered a number of reasons to love and be proud of my new home state. One reason (out of many) I am proud is that, a months after we moved here, Minnesotans rejected at the ballot box a state constitutional ban on same-sex marriage. Furthermore, within a year after that, Minnesota enacted marriage equality through a legislative process. I am proud that Minnesotans did the right thing without needing a judge to force them to.

There is some talk of a ballot measure in 2016 in Oregon to repeal Measure 36. I do hope Oregon voters have a chance to formally repudiate Measure 36 at the ballot box at some point.

By the way, I was inspired by the beautiful language in Judge McShane’s ruling. I think it’s worth quoting at length:

Generations of Americans, my own included, were raised in a world in which homosexuality was believed to be a moral perversion, a mental disorder, or a mortal sin. I remember that one of the more popular playground games of my childhood was called “smear the queer” and it was played with great zeal and without a moment’s thought to today’s political correctness. On a darker level, that same worldview led to an environment of cruelty, violence, and self-loathing. It was but 1986 when the United States Supreme Court justified, on the basis of a “millennia of moral teaching,” the imprisonment of gay men and lesbian women who engaged in consensual sexual acts. [. . .] Even today I am reminded of the legacy that we have bequeathed today’ s generation when my son looks dismissively at the sweater I bought him for Christmas and, with a roll of his eyes, says “dad … that is so gay.”

It is not surprising then that many of us raised with such a world view would wish to protect our beliefs and our families by turning to the ballot box to enshrine in law those traditions we have come to value. But just as the Constitution protects the expression of these moral viewpoints, it equally protects the minority from being diminished by them.

[. . .]

My decision will not be the final word on this subject, but on this issue of marriage I am struck more by our similarities than our differences. I believe that if we can look for a moment past gender and sexuality, we can see in these plaintiffs nothing more or less than our own families. Families who we would expect our Constitution to protect, if not exalt, in equal measure. With discernment we see not shadows lurking in closets or the stereotypes of what was once believed; rather, we see families committed to the common purpose of love, devotion, and service to the greater community.

Where will this all lead? I know that many suggest we are going down a slippery slope that will have no moral boundaries. To those who truly harbor such fears, I can only say this: Let us look less to the sky to see what might fall; rather, let us look to each other … and rise.

Brains over calculators!

Mon, 5 May 2014 18:36:45 -0500

I enjoy finding calculations that I would consider reasonable for my students to perform that their calculators cannot.

Using a double-angle identity and the pythagorean identity, it’s pretty straightforward to show that $\sin\left(2\sin^{-1}\left(-\frac{4}{5}\right)\right) = -\frac{24}{25}.$ However, my TI-89 returns an unhelpful $-\sin(2\sin^{-1}(4/5))$ instead.

Amusingly, the calculator does know that $2\sin\left(\sin^{-1}\left(-\frac{4}{5}\right)\right)\cos\left(\sin^{-1}\left(-\frac{4}{5}\right)\right) = -\frac{24}{25}.$

(My calculator is running the 2005 version of the software. It’s possible this has been fixed since then.)

Rejoining the tweet-o-sphere

Sun, 6 Apr 2014 20:43:10 -0500

After a few years away, I am back on Twitter.

I'm back on Twitter!

— Chris Phan (@functoruser) April 7, 2014

Re the Mozilla thing

Thu, 3 Apr 2014 21:27:42 -0500

Re the Mozilla thing: Given that wealthy people now have a right to unlimited political donations, why shouldn’t people take CEO politics into account in their consumer decisions?

(Yes, I am aware people don’t pay money to use Firefox. But Google pays the Mozilla Corporation $300 million per year to be the default search engine in Firefox, which presumably reflects Firefox’s market share.)

hanging with MVPB

Wed, 19 Mar 2014 19:45:22 -0500

Arizona State Senator opposes Common Core, in part, because ALGEBRA!

Sat, 22 Feb 2014 13:53:53 -0600

According to the Arizona Daily Star, the Arizona Senate Education Committee passed a bill that would prohibit the state from implementing the Common Core standards. Quoth the Daily Star (emphasis mine):

[Gubernatorial candidate and state Senator Al] Melvin said he understands “some of the reading material is borderline pornographic.” And he said the program uses “fuzzy math,” substituting letters for numbers in some examples.

Later on in the same article, Senator Melvin is quoted as expressing concern about the rigor of American academic standards, arguing that “We have cheated several generations of Americans out of a decent education.” I might argue that the person who has been cheated out of a decent education is Senator Melvin, who is apparently unfamiliar with fifth-grade mathematics.

Of course, it’s entirely possible Senator Melvin is fully aware that this “fuzzy math” line is completely ignorant poppycock and is only saying it in a cynical attempt to appeal to ignorant people in his gubernatorial campaign. (This possibility, to be frank, would be even more distressing.)

Intel’s second largest U.S. site is in Arizona. I wonder how Intel executives feel about their company’s investment in a state where members of the state senate education committee oppose the teaching of algebra.

(To be absolutely clear, I’m not saying that all Common Core skeptics and opponents are as ridiculous as Senator Melvin.)

HT: Matthew Arbo

Yes, this.

Sat, 15 Feb 2014 09:58:47 -0600

Related note, arguments about whether "piracy" is "stealing" are irrelevant to debates about piracy's ethics, you can safely ignore them.

— gia manry (@giapet) February 15, 2014

Edited to add:

If you pirate media, you are choosing your *personal entertainment* over the rights of the owners of said media. Don't pretend otherwise.

— gia manry (@giapet) February 15, 2014

MVPB and me

Sun, 9 Feb 2014 11:33:12 -0600

One thing I love about the new NYT website redesign

Mon, 20 Jan 2014 11:17:32 -0600

Many people are complaining about the redesign of the New York Times website. Some of these complaints are from people having trouble finding things on the site, which is prone to happen anytime you change a user interface. Other complaints deal with the smaller font size, which I find annoying as well. (I am a frequent user of the ⌘+ key combo.)

But one thing I love, love, love about the redesign, which I hope is copied by other news sites, is that articles appear on a single page, and I don’t have to click “Next” repeatedly. It really bothers me to have to have to click “Next” to read the remainder of an article.

Lake Winona on a frozen winter day

Sun, 19 Jan 2014 20:07:40 -0600

Full size

It’s fun!

Sat, 18 Jan 2014 15:19:32 -0600

I didn’t think this would be true, but I actually quite enjoy using the snowblower.

2013 Year in Review

Fri, 27 Dec 2013 18:58:51 -0600

It was another eventful year for the Phan-Budd family, and you can read all about it in our year-in-review letter!

Winter is here

Sun, 22 Dec 2013 18:46:39 -0600

Right after sunset on the solstice, I snapped this picture of Main Channel Bridge, which connects Winona with Wisconsin to the north. As its name suggests, it crosses the main channel of the Mississippi River, which is currently frozen.

Today, I captured this video of a BNSF train passing through Trempealeau, Wisconsin:

Ha ha nice try

Sat, 26 Oct 2013 11:20:12 -0500

From: Qatar Foundation

To: Recipients

You Have Been Awarded $1,000,000.00 USD

October 24, 2013 at 16:36

Dear Beneficiary,

This is to inform you that you were among the lucky beneficiary selected to receive this donations award sum of $1,000,000.00 USD, as charity donations/aid from the Qatar Foundation held in Doha, Qatar, 24th of October 2013, to promote your business and personal Interest. Kindly get back for more details on how to claims your award via email:

Chromecast: Pretty neat

Mon, 16 Sep 2013 22:24:47 -0500

A few weeks ago, I received my Chromecast in the mail. This is a small $35 device sold by Google that vaguely resembles a USB flash drive and fits into the HDMI port on your TV, and connects to your WiFi. For what it does, it is a very nice device.

My television comes with built-in software to watch Netflix and YouTube. Alas, that software is slow and is a pain to use. Entering a search term using the arrow keys on my remote control, one letter at a time, is absolutely frustrating. However, Chomecast has much nicer software for watching Netflix and YouTube, which you can control using your computer or mobile device. For example, I can launch the Netflix app on my iPad, pick a show off of my instant queue (or whatever they are calling it these days), and send the info on my selection to the Chromecast, which starts streaming the show to my TV. (Chormecast uses your home WiFi network to stream video and get commands from your devices.) The same is true of YouTube, and I guess Google Play movies, which I have never used. I love watching YouTube videos on my TV, selecting them from my subscriptions using my iPad. (I can also use my laptop or Droid 4 smartphone.)

Chormecast also allows you to send any Chrome browser tab on your computer to your TV. This doesn’t work quite as well as the built-in apps. For one thing, there is a delay of a few seconds between something happening in the browser tab on your computer and the display of the tab on the TV. (Fortunately, the sound also gets delayed, so watching a video in a Chromecasted tab will mean the sound and video will be in sync on your TV.) Second, this is bandwidth-intensive when watching a video, since the computer has to stream the video from the Internet and then restream it to your TV over WiFi. So, it’s better to use the built-in apps to watch Netflix and YouTube.

Nevertheless, casting tabs still works good enough. The other night, I Chromecasted a NYT article on the president’s Syria speech, and played the embedded video of his speech. It worked fine.

Now, this device is not really that necessary. My TV has built-in apps for Netflix and YouTube. Likewise, as you can see from the photo above, it’s perfectly possible for me to hook my computer up directly to the HDMI input on my TV, which would allow me to send the whole screen to the TV, not just what I can get in a Chrome tab. But Chromecast is much easier to use than my TV’s built-in apps, and less awkward than stringing HDMI cables back to the couch. For $35, this nice little device is a good help.

Facebook’s news feed was not a novel idea

Thu, 12 Sep 2013 12:24:57 +0000

Today in Slate, Farhad Manjoo celebrates the 7th birthday of Facebook’s news feed feature.

Get this: Before news feed, which launched seven years ago this month, you could post a picture or some other personal detail somewhere—your Facebook or MySpace or Friendster page, Flickr, Blogger, LiveJournal—and be reasonably sure that it would remain just there, unseen by pretty much everyone you knew. The only way someone might find it is by checking your page. Sure, some people would do that—but everyone had scores of connections online, so no one was checking each of their friends’ pages. The net effect was solitude.

Sorry, but this is complete bunk. Every LiveJournal user has a friends page, which shows all recent entries by their LJ friends. When I joined LiveJournal in July 2001, over 12 years ago and before Facebook even existed, LiveJournal had this feature already. Granted, as LiveJournal never attained the market penetration that Facebook has, posting something to your LiveJournal was never as effective at spreading your news as posting to Facebook today. But that’s due to Facebook benefiting from a network effect on a scale that LiveJournal never did.

I wouldn’t dispute that Facebook’s news feed is an important feature that has allowed for those network effects. Without the news feed, the site would just be a glorified directory and messaging service. But it’s an absurd revision of Internet history to say that Mark Zuckerberg and his friends, as smart as they are, came up with the idea from scratch.

What (not) to do if you don’t like sports

Wed, 4 Sep 2013 21:58:23 -0500

Let’s say you don’t like watching sports. You have no interest in the World Series, the BCS, or the NBA Finals. I think that’s okay. It’s okay to not pay attention to sports. It’s okay to flip past ESPN on your TV. But here’s what’s not okay to do:

Make fun of people for liking to watch sports.
Derisively dismiss sports as “sportsball”.
Assume people are dumb because they like to watch sports.

If you are a geek about my age, you may have not-so-fond memories of being ridiculed when you were a teenager for having geeky interests. You may remember people teasing you for socializing on the Internet (something just about every teenager does today), or for being interested in Star Trek. For someone with those memories, it’s very tempting, now that geek interests are more mainstream, to turn the tables and dish out ridicule for being into sports. But you should resist the temptation because it makes you as much as a jerk as the kids who teased you. Actually, it makes you worse, because you are some decade-and-a-half older and should know better.

Next adventure: homeownership

Wed, 7 Aug 2013 20:18:42 -0500

Late last month, my wife and I went down to our local credit union, signed a very large stack of legal documents, and joined the ranks of millions of Americans who own their own home.

And then, about a week or so later, I bought a cordless electric lawnmower and mowed the lawn.

Ha ha nice try

Wed, 7 Aug 2013 13:33:15 -0500

Important Notification

Patrick Nso [REDACTED]@gmail.com

Important Notification,

This is to inform you that we (International Financial Monitoring Agency) have been instructed to handle the completion of your transaction worth the sum of $950,000 (nine hundred and fifty thousand United States Dollars). We have also been instructed to make sure that we take care of the paper works needed from our end to transfer your fund while you take care of your end to receive your fund once it is transfered out to you.

Your response is highly anticipated if you are willing to finalise this, for us to proceed with the processing by letting you know the information needed from you.

Thank you.
Patrick Nso.
Manager,
International Financial Monitoring Agency.

PGP key

Wed, 17 Jul 2013 20:31:41 -0500

In reaction to recent events, I have created a PGP key, and have posted the public key on my web site. The fingerprint to this key is:

15A1 D221 79D1 2FEB 8BA4 4F47 33AB 9AD2 C4A9 A097

Personally, I have found GPG tools, a Mac implementation of GNU Privacy Guard, pretty easy to use.

My hope in the wake of the Zimmerman verdict

Mon, 15 Jul 2013 14:19:55 -0500

My hope, in the wake of the Zimmerman verdict, is that people don’t interpret the “not guilty” verdict as an exoneration of George Zimmerman’s actions leading up to Trayvon Martin’s death. The jury concluded, perhaps justifiably, that the state failed to prove that Zimmerman committed a crime. But that doesn’t mean that what Zimmerman did that night was honorable or ethical. It wasn’t.

It’s just not a good idea for people to act like police officers when they don’t have the training and authority. It’s dangerous for both you and the people you accost. The dispatcher was absolutely correct to attempt to dissuade Zimmerman from following Martin. I agree with Jay Smooth that “The fundamental danger of an acquittal is not more riots, it is more George Zimmermans.”

I’ve read a number of blog posts and opinion articles in the wake of the verdict, but here are two observations I found especially insightful:

First, Scott Lemieux argues that the Zimmerman case shows the deficiency of our self-defense laws:

Carrying a deadly weapon in public should carry unique responsibilities. In most cases someone with a gun should not be able to escape culpability if he initiates a conflict with someone unarmed and the other party ends up getting shot and killed. Under the current law in many states, people threatened by armed people have few good options, because fighting back might create a license to kill. As the New Yorker’s Amy Davidson puts it, “I still don’t understand what Trayvon was supposed to do.” Unless the law is changed to deal with the large number of people carrying concealed guns, there will be more tragic and unnecessary deaths of innocent people like Trayvon Martin for which nobody is legally culpable. And to make claims of self-defense easier to bring, as Florida and more than 20 other states have done, is moving in precisely the wrong direction. And, even more importantly, no matter how self-defense laws are structured the extremely unusual American practice of allowing large number of citizens to carry concealed weapons leads to many unecessary deaths.

Second, Ta-Nehisi Coates (perhaps the best blogger writing today) argues that the law—not the verdict, but the underlying statute— is the result of structural racism “forces we set in motion years ago and have done very little to arrest.”

We have spent much of this year outlining the ways in which American policy has placed black people outside of the law. We are now being told that after having pursued such policies for 200 years, after codifying violence in slavery, after a people conceived in mass rape, after permitting the disenfranchisement of black people through violence, after Draft riots, after white-lines, white leagues, and red shirts, after terrorism, after standing aside for the better reduction of Rosewood and the improvement of Tulsa, after the coup d’etat in Wilmington, after Airport Homes and Cicero, after Ossian Sweet, after Arthur Lee McDuffie, after Anthony Baez, Amadou Diallo and Eleanor Bumpers, after Kathryn Johnston and the Danziger Bridge, that there are no ill effects, that we are pure, that we are just, that we are clean. Our sense of self is incredible. We believe ourselves to have inherited all of Jefferson’s love of freedom, but none of his affection for white supremacy.

You should not be troubled that George Zimmerman “got away” with the killing of Trayvon Martin, you should be troubled that you live in a country that ensures that Trayvon Martin will happen.

Both these articles are worth reading in full.

Finally, this morning I read a Washington Post story about Zimmerman’s lawsuit against NBC News for their uncouth edits to the 911 tape. I do think NBC’s edits were unfair and don’t blame Zimmerman for suing. I’m not going to criticize Zimmerman’s lawyers for doing their job by engaging in full-throated advocacy on his behalf, even if that advocacy is sometimes full of B.S. However, dear reader, you should recognize some of their arguments as B.S., especially this:

When asked how the not-guilty verdict affects the civil case against NBC News, Beasley responded, “This verdict of not guilty is just that, and shows that at least this jury didn’t believe that George was a racist, profiling, or anything that the press accused George of being. That probably doesn’t get you that much but it’s simply time for us to start the case and hold accountable anyone who was irresponsible in their journalism.”

The jury wasn’t asked to determine if Zimmerman was a racist or profiling, just if there was enough evidence to convict him of homicide. The rest of us are quite entitled to form our own opinions on the matter.

Making a “Places I have been” map

Sat, 13 Jul 2013 11:49:07 -0500

For a while, I have maintained a list of places I’ve lived or visited as a note on Facebook. Recently, I decided to repatriate that content to this blog as a page. I also decided it would be nice to have a color-coded map of the states with the same information.

Here’s how I made the map:

I started with a blank SVG map of the states from the Wikimedia Commons, available under a Creative Commons license. (Here’s the actual version I used.) This map is very nice because you can edit it in a text editor. That makes it easy to change the colors of the state.

I wanted four colors for the states, showing the states in which I have lived, visited, had airport layovers, or never stepped foot in. In the original file, there was only one “style” of state, as shown in the CSS style sheet embedded in the file:

US_Map_chris_visit.svg SVG


<style type="text/css">
.state {
	fill:#d3d3d3;
	stroke:#fff;
	stroke-width:0.75;
	stroke-opacity:1;
}
style>

To this, I inserted three more styles. The resulting stylesheet is:

US_Map_chris_visit.svg SVG


<style type="text/css">
.state {
	fill:#d3d3d3;
	stroke:#fff;
	stroke-width:0.75;
	stroke-opacity:1;
}

.statel {
	fill:#669966;
	stroke:#fff;
	stroke-width:0.75;
	stroke-opacity:1;
}

.statev {
	fill:#ff7f00;
	stroke:#fff;
	stroke-width:0.75;
	stroke-opacity:1;
}

.statea {
	fill:#ffcc33;
	stroke:#fff;
	stroke-width:0.75;
	stroke-opacity:1;
}
style>

The colors are given by the fill attributes, and are specified as RGB hex triplets (e.g. #669966 is the green color used in the image).

Once I established the styles, it was straightforward to modify the states to assign the different styles. Each state is drawn in the SVG file with a path element; you just need to change the class attribute in the corresponding path element. For example, here is the original code to draw Colorado:

US_Map_chris_visit.svg SVG


<path d="m 380.03242,320.96457 4.90324,-86.32496 -113.38856,-12.64396 -12.21382,87.93916 120.69914,11.02976 z" id="CO" class="state" />

And here is how I modified the code:

US_Map_chris_visit.svg SVG


<path d="m 380.03242,320.96457 4.90324,-86.32496 -113.38856,-12.64396 -12.21382,87.93916 120.69914,11.02976 z" id="CO" class="statea" />

Note that the id attribute is used to identify the state.

Because browser support for SVG is imperfect at this point, I converted the SVG file to a PNG before uploading. Unfortunately, the version of the Imagemagick convert utility on my computer didn’t do the conversion correctly, so I used Inkscape instead.

Credit: Map of the US adapted from a file from the Wikimedia Commons that was originally created by Wikimedia Commons user Theshibboleth and modified by users Fibonacci, NuclearVacuum, and Wylve. Used under the Creative Commons Attribution-ShareAlike 3.0 Unported (CC-BY-SA 3.0) license. SVG source code of my modified version of the file. Both my modified SVG and derived PNGs are offered under the Creative Commons Attribution-ShareAlike 3.0 Unported (CC-BY-SA 3.0) license.

Places I have been

Sat, 13 Jul 2013 11:12:27 -0500

Last updated: 13 July 2013

For the purposes of this page, I have only been to a place if I have been on the ground; being in a jurisdiction’s airspace does not count.

States I have lived in:

Illinois
Minnesota
Oregon
Pennsylvania

States I have not lived in, but have visited beyond just the airport:

Arkansas
California
Connecticut
District of Columbia*
Idaho
Indiana
Iowa
Kentucky
Louisiana
Maryland
Massachusetts
Michigan
Mississippi
Missouri
Montana
Nebraska
New Jersey
New York
North Carolina
North Dakota
Ohio
South Dakota
Tennessee
Utah
Virginia
Washington
West Virginia
Wisconsin
Wyoming

* For the purpose of this exercise, I am counting the District of Columbia as a state.

States I have been to only for airport layovers:

Arizona
Colorado
Georgia
New Mexico

States I have never been to, not even for airport layovers:

Alabama
Alaska
Deleware
Florida
Hawaii
Kansas
Maine
Nevada
New Hampshire
Oklahoma
Rhode Island
South Carolina
Texas
Vermont

Canadian provinces I have visited beyond just the airport:

Alberta
British Columbia
Ontario

(There is no Canadian province that I have only visited for an airport layover.)

Countries I have lived in:

United Kingdom
United States

Countries I have not lived in, but have visited beyond just the airport:

Canada
South Korea
Vietnam

Countries I have been to only for airport layovers:

Ireland
The Netherlands

(Technically, I visited South Korea during a long airport layover, but I left Incheon International Airport and went into Seoul.)

How I made the map.

Credit

Map of the US adapted from a file from the Wikimedia Commons that was originally created by Wikimedia Commons user Theshibboleth and modified by users Fibonacci, NuclearVacuum, and Wylve. Used under the Creative Commons Attribution-ShareAlike 3.0 Unported (CC-BY-SA 3.0) license. SVG source code of my modified version of the file. Both my modified SVG and derived PNGs are offered under the Creative Commons Attribution-ShareAlike 3.0 Unported (CC-BY-SA 3.0) license.

Deluge of comment spam

Tue, 11 Jun 2013 17:59:23 -0500

This blog has received a deluge of attempted comment spam recently. Spammers attempted to post over 150 comments since the beginning of June. I say “attempted” because every comment made to this blog by a new user is sent to moderation; I have had to manually delete every spam comment made so far. I don’t know what the spammers are hoping to achieve, given I am not interested in illegally acquiring prescription pain medication, and nobody else has seen their missives on this topic they have tried to post here. But going forward, I guess I’m going to have to leave comments off.

Facebook, “free speech”, and a monopoly on town squares

Thu, 30 May 2013 19:42:59 -0500

As you may know, there was a recent campaign by activists to pressure Facebook to remove misogynistic, in some cases graphic, posts glorifying rape and domestic violence. (If you can stomach it, Buzzfeed has some screen shots of the misogynistic posts.) Part of the motivation of this campaign was to highlight Facebook’s ridiculous standards: Facebook had been slow to remove the misogynistic posts, in some cases saying they weren’t hate speech, while removing some pictures that featured breastfeeding, on the basis that the breastfeeding pictures violated Facebook’s prohibitions on nudity. As a result of the campaign, in which advertisers were pressured by activists, Facebook has pledged to review and revise its policies, admitting that they failed to remove posts that constitute gender-based hate speech:

In some cases, content is not being removed as quickly as we want. In other cases, content that should be removed has not been or has been evaluated using outdated criteria. [. . .T]he guidelines used by these systems have failed to capture all the content that violates our standards. We need to do better – and we will.

In considering this controversy, it’s important to remember, as Mary Gardiner wrote last September (in the context of the Reddit/creepshots controversy), the phrase “freedom of speech”

isn’t a very specific term on the Internet: it can mean anything from “I believe governments should not restrict expression” to “I believe that never deleting comments* from a forum improves the quality of discussion” to “I believe that never deleting comments from a forum is the only ethically correct way to run a forum.” (Or the disingenuous version: “I believe that I personally should be able to say what I want in any forum.”)

(Gardiner goes on to point out that “No one seems to believe this about spam.”)

In response to Facebook’s announcement, Jillian C. York, Director for International Freedom of Expression for the Electronic Frontier Foundation–a cyberspace civil liberties group whose work I generally respect and admire–wrote an article in Slate questioning whether Facebook should be in the business of policing “hate speech” at all. Conceding that Facebook has a right as a private business to remove content it sees fit, York argues that the Facebook is ill-equipped to handle the hate speech complaints that go along with serving a billion users worldwide, citing examples where Facebook has erred. But then she goes on argue that Facebook shouldn’t engage in censorship at all, given its status as “the new town square”:

Should private companies be determining what constitutes “hate speech”? In the United States (where Facebook is based), most of the speech flagged by Women, Action & the Media as offensive is, while abhorrent, protected by law. And while Facebook may be private, many of its users treat it like the new town square, making it more of a quasi-public sphere. While the campaigners on this issue are to be commended for raising awareness of such awful speech on Facebook’s platform, their proposed solution is ultimately futile and sets a dangerous precedent for special interest groups looking to bring their pet issue to the attention of Facebook’s censors.

Although I am very sympathetic to York’s argument, ultimately I must respectfully disagree. First of all, it’s just not feasible to expect Facebook to operate the way York suggests. Facebook’s business model involves posting advertisements on nearly every page served by the site, but a company will understandably be reluctant to advertise on Facebook if the ad might give the impression that the company supports misogyny (an impression that would not be wholly incorrect when the company’s ad is paying for the server and bandwidth costs for a misogynistic post). Facebook has to remove gender-based hate speech for the same reason MTV bleeps out the word “hash” in the Weezer song “Hash Pipe”. Expecting an advertisement-supported site to allow all constitutionally-protected content is a pipe dream.

More importantly, I think the greater issue is one company’s extent of control over Internet discourse. York’s article is accompanied by a photo of Mark Zuckerberg, with the caption “Should we give this guy the power to decide what is and isn’t hate speech?” My answer, if we’re talking about the majority of public discourse on the Internet, is a resounding “no”. Indeed, if we’re talking about the majority of public discourse on the Internet, I don’t want Mark Zuckerberg–or anyone else–to have power to decide anything. And that’s why the problem isn’t that Facebook bans hate speech; the problem is that Facebook has become the de facto default “town square”. It’s dangerous to allow any one company or entity to have this much power.

If you are worried about Facebook having the power to determine what is “hate speech”, then the proper solution is to encourage people to stop viewing Facebook as the de facto method of interaction on the Internet. Encourage people to use other social networks in addition to Facebook. Encourage people to use forums outside of Facebook. Share things on your own site instead of just Facebook. (My web host, NearlyFreeSpeech.net, has relatively few restrictions on content and is pretty affordable.) I’m not suggesting that people stop using Facebook, but merely that dissuade people from only using Facebook. If we work to ensure that there is a variety of “town squares”, then we won’t have to worry as much about the restrictions at any particular one.

Yay for Minnesota!

Tue, 14 May 2013 21:53:03 -0500

Earlier today, Gov. Mark Dayton signed into law a bill to recognize same-sex marriages! The law will take effect in August. Minnesota will become the 12th state to recognize same-sex marriage.

Yay for Minnesota! I am very proud of my new state!

A moving speech on marriage equality

Fri, 10 May 2013 22:41:02 -0500

I’m a newcomer to Minnesota, having lived here only 9 months. But one thing that has made me proud to be a new Minnesotan is how this state has handled the issue of marriage equality. In the November 2012 election, Minnesotans defeated at the polls a proposed state constitutional amendment banning on same-sex marriage. And yesterday, the Minnesota House of Representatives passed a bill to legalize same-sex marriage.

Today, on Minnesota Public Radio, they were replaying speeches made in debate the Minnesota house yesterday during the floor debate before the historic vote. One speech, made Rep. Tim Faust, who is a Lutheran pastor, was particularly moving:

The Minnesota Senate will vote on the bill Monday, and should it pass (which is expected), our governor has promised to sign it. While I’m not going to celebrate too much until the bill passes the Senate, it’s appearing more and more likely that next week Minnesota will become the 12th state in the union to have marriage equality.

Record snow!

Fri, 3 May 2013 07:15:51 -0500

We had some nice spring weather this week in Southeastern Minnesota. On Tuesday, the temperature got up to 27°C (80°F) and we had our apartment windows open in the evening.

Then on Wednesday, the temperature dropped precipitously, and we awoke on Thursday to this:

That picture was taken in the morning. The snow came down most of the day, piling up to 3.5 inches (89 mm).

But we didn’t get it as bad as Rochester, which is about an hour west of here. As the National Weather Service points out, Rochester didn’t break the snow record for May 2nd yesterday, they got more snow yesterday than all other days in May combined since record-keeping began in 1886. Here is the public information statement released by the La Crosse office of the NWS last night:

000

NOUS43 KARX 030107

PNSARX

PUBLIC INFORMATION STATEMENT

NATIONAL WEATHER SERVICE LA CROSSE WI

807 PM CDT THU MAY 2 2013

…UNPRECEDENTED SNOWFALL FOR ROCHESTER MINNESOTA…

WHAT A STORM FOR THE RECORD BOOKS. THE INFORMATION BELOW SHOWCASES HOW THIS SNOWSTORM RANKS FROM A MAY PERSPECTIVE…AND ALSO THE ENTIRE CALENDAR YEAR.

THE 13.9 INCHES…THROUGH 730 PM…BREAKS THE FOLLOWING RECORDS…

…DAILY RECORD SNOWFALL FOR MAY 2ND / 0.7 INCHES IN 1954 /

…DAILY RECORD SNOWFALL FOR MONTH OF MAY / 1.2 INCHES MAY 5 1944/

…MAY MONTHLY SNOWFALL TOTAL / 2.0 INCHES MAY 1944 /

IT CURRENTLY RANKS AS SIXTH GREATEST ALL-TIME DAILY SNOWFALL. THIS WILL LIKELY RISE ONCE THE OFFICIAL SNOW TOTAL IS REPORTED.
ALL-TIME 1-DAY SNOWFALL RECORDS
             ROCHESTER MINNESOTA
                 1886-2013

   RANK          SNOWFALL         YEAR
   ----       -------------      -------
     1         19.8 INCHES      3/18/2005
     2         15.4 INCHES      1/22/1982
     3         15.0 INCHES     12/11/2010
     4         14.0 INCHES      3/30/1934
                                4/20/1893
     6         13.9 INCHES      5/02/2013
     7         13.5 INCHES      2/27/1893
     8         13.0 INCHES      4/26/1988
     9         12.0 INCHES     11/30/1934
OTHER RECORDS BROKEN TODAY INCLUDE…

THE LOWEST MAXIMUM TEMPERATURE FOR MAY 2ND…33 DEGREES. THIS ALSO MAKES IT THE LOWEST MAXIMUM TEMPERATURE FOR ANY DAY IN THE MONTH OF MAY. THE PREVIOUS LOWEST MAXIMUM TEMPERATURE IN MAY WAS 36 DEGREES…ON MAY 1ST 1940.

TO PUT THIS SNOWSTORM IN ANOTHER PERSPECTIVE…TODAYS SNOWFALL IN ROCHESTER SHATTERS THE COMBINED TOTAL FOR ALL PREVIOUS MAY MEASURABLE SNOW EVENTS. SINCE 1886…THERE WERE 10 MEASURABLE MAY EVENTS…FOR A TOTAL OF 4.3 INCHES. TODAYS SNOWFALL IS OVER 3 TIMES THAT.

FURTHER CLIMATE AND RECORD INFORMATION WILL BE ADDED ONCE TOTALS ARE CONFIRMED AND THE SNOW ENDS.

$$

ZT

Still some ice in the river

Sun, 14 Apr 2013 00:23:41 +0000

Taken at the Verchota Landing, just outside Winona in the Upper Mississippi River National Wildlife & Fish Refuge:

Yikes! Zebra mussels are pretty terrible:

Duke of Wellington statue on Google treasure map

Mon, 1 Apr 2013 21:16:09 -0500

For April Fool’s Day, Google has added a fun “treasure map” mode on Google Maps. If you zoom into various cities, you can see hand-drawn landmarks, such as the Sears Tower or Washington Monument. Or, if you zoom into Glasgow, you see this:

This is the Duke of Wellington statue in front Glasgow’s Gallery of Modern Art. Here’s a real-life picture (albeit from the other side):

late March in Winona

Sat, 30 Mar 2013 11:07:47 -0500

It’s starting to melt!

SageTeX is awesome!

Fri, 8 Mar 2013 21:39:36 -0600

For a long time, I have been a user of the math software system Sage, and for a longer time, I have been a user of LaTeX. So, it’s with some embarrassment that I report that I only recently discovered the awesomeness that is SageTeX, a LaTeX package that allows your LaTeX document to run Sage code and include the results.

I often use Sage to create plots for use in my lectures or printed class materials. Previously, my workflow was as follows:

Create the graphic by running commands in the Sage notebook.
Use the save command to save the graphic as a PDF, usually with a filename based on my LaTeX source file’s name (e.g. m160-sp13-mid2-practice-1.pdf, for use in the LaTeX document with source file m160-sp13-mid2-practice.tex).
Save the file in the directory where my LaTeX source file lives.
\includegraphics, FTW!

Like most teachers, every time I teach a course I’ve taught before, I like to reuse old materials, often after improving them or adjusting them to suit the new class. Therefore, it’s good to record the exact Sage commands somewhere, as opposed to just keeping the graphic file, in case I want to modify the graphic. One way you can do this is to paste the Sage commands as a comment in the LaTeX file. This has the advantage of allowing future tweaks to be recorded in your version control system. (And you really should be using a version control system, so that way you won’t be afraid to make changes to your document.) Unfortunately, modifications then mean a lot of copy-and-pasting: you have to modify the code in the Sage notebook, and then copy-and-paste the modifications to the comments in the LaTeX document, or vice-versa. And then you have to save another version of your graphic from the browser. What a headache!

SageTeX makes this a lot easier. When I want to put a Sage plot in my document, I can put the Sage commands right in the document:

LaTeX


\begin{sagesilent}
var('x,y')
gfx = implicit_plot((x + 1)*(x^2 + y^2) == 3*x^2, (x, -2, 2),(y, -2, 2), axes=true, frame=false)
gfx = gfx + list_plot([(1/2, -1/2)], size=50)
save(gfx, "m160-sp13-mid2-practice-1.pdf", figsize=[3, 3])
\end{sagesilent}

\item (15 points)
Consider the curve with the equation
$(x + 1)(x^2 + y^2) = 3x^2$ (shown below).
Use implicit differentation to find $\tfrac{dy}{dx}$,
and then find an equation for the tangent line
of the curve at the point
$\left(\tfrac{1}{2},-\tfrac{1}{2}\right)$.\\
\includegraphics[width=3 in]{m160-sp13-mid2-practice-1.pdf}

When you compile your document with pdflatex, it creates a Sage script file titled (in this case) m160-sp13-mid2-practice.sagetex.py. Compiling this with Sage will create (or update) the file m160-sp13-mid2-practice-1.pdf, which you can include (or update) by running pdflatex again!

That’s pretty awesome.

Ha ha nice try

Fri, 1 Mar 2013 21:51:58 -0600

Transaction is completed

Payment notification system [Redacted]@gmail.com

To: phan03

Cc: [Redacted]

ACH transaction is completed. $8047 has been successfully transferred. If the transaction was made by mistake please contact our customer service. Payment receipt is attached.

*** This is an automatically generated email, please do not reply ***

Executable Attachment Removed

Winona in late February

Sun, 24 Feb 2013 22:53:33 -0600

Sarah and I took a break from grading to enjoy the lake today.

Here’s a device they use to keep part of the lake from freezing over:

“On very thin ice, very thin ice, very thin ice, very thin ice…”

Winona in February

Wed, 6 Feb 2013 20:11:04 -0600

(That’s a lake, BTW.)

Thanks

Mon, 21 Jan 2013 16:49:51 -0600

Winona in January

Sat, 19 Jan 2013 18:40:36 -0600

Up goer five dissertation summary

Sat, 19 Jan 2013 14:34:05 -0600

I learned about this Tumblr blog where people try to summarize their PhD dissertations in the style of xkcd’s “Up Goer Five” comic—that is, “using only the ten hundred words people use most often”. There’s an online text editor that checks whether you’ve strayed from these thousand most-common words.

Inspired by this blog, here’s my attempt to summarize my dissertation (Koszul and generalized Koszul properties for noncommutative graded algebras) in “Up Goer Five” style:

You can add and times numbers. There are ways make up a set other things you can add and times like you can with numbers. We call these sets “rings”. Sometimes, with these new things, the order in which you times things changes what you get.

Some of these rings have floors, like a building. When you times two things in the ring, you have to add the floor numbers to figure out which floor the thing you get is on.

There is a way to study these rings with floors with strange add and times that comes from the study of forms in space. Given a ring with floors, you can make a new ring with floors that is sort of like it, but with some things about it inside out.

I studied stuff about these new rings with floors coming from the old, especially in cases where the order in which you times things changes what you get from times.

And here’s Sarah’s “Up Goer Five” summary of her dissertation (Search for single top quark production with the CDF Run II detector using a multivariate likelihood method):

We made very very very small but very heavy things (called top things) in a new way. These top things are very small and are like a point, but are very heavy. They also live for a very short time. Usually you see two top things made at the same time. We found one top thing made with another thing. This one top thing looks a lot like other small heavy things that are made more easily. So, we had to tell our one top thing from all the other small heavy things that we made as well. It was hard. We used computers to help with the numbers. I worked with many people. But, I found the one top thing turning into three other things, while other people found the one top thing turning into two other things. We added my finding with my friends to make a strong finding.

Our finding was so strong we thought for sure it was true. But, we like to know for sure for sure for sure, not just for sure. So, the next year another friend made another finding of the same thing. With our finding and the new finding and we thought for sure for sure for sure it was true. So, now every person thinks our finding is true as well. Now our finding is in the way of seeing other new very small very heavy things, but that is life.

Lake Winona, frozen over

Wed, 16 Jan 2013 20:59:36 -0600

2012 Year in Review

Tue, 25 Dec 2012 23:27:33 -0800

It’s the end of the year, and hence time for another year in review letter.

Policies aren’t the only thing that matter in elections

Wed, 24 Oct 2012 23:03:44 -0500

Much of the talk about “values” and “character” that surrounds elections is phony, not uncommon for campaign rhetoric, but it’s a mistake for voters to disregard these issues.

Every once in a while, I’ll run across some online quiz where you answer some questions about your policy preferences—e.g. should we spend more or less on war—and it identifies the presidential candidate you should support. I regard these results with some skepticism. It’s not just because I’m worried the issues might be cherry-picked, and it’s not just because such questionnaires usually fail to ask about the strengths of my beliefs. It’s because such metrics disregard the candidates’ competence and trustworthiness.

I moved away from Portland in 2003, haven’t lived there since, and doubt I’ll live there again. So, let me disclose up front that I don’t have much of a stake in the upcoming mayoral election there. But I have been following it, because I find it interesting.

Portland uses a nonpartisan two-round system for municipal elections. Advancing to the second round are Charlie Hales, a former city commissioner, and Jefferson Smith, a state legislator and founder of the Bus Project, a group that gets young people involved in progressive politics in Oregon. (Here’s a picture of me wearing a button reading “I’m on the Bus”. I’m posing with the guy who ran this awesome ad.) Both are Democrats, which is unsurprising given that Democrats outnumber Republicans by over a 3-to-1 ratio in Multnomah County.

Smith seemed like a rising star in Oregon politics, but some unpleasant background surfaced in the mayoral race. First, there was his “embarrassing” driving record (to use his word), which included a citation for driving with a suspended license.

More troubling were two incidents of violence. He punched a woman while in college, back in 1993, hard enough that she needed stitches. The victim told police that he had spent the evening leading up to the punch harassing her. In a stunningly idiotic move, Smith turned up uninvited at the home of the victim on the day the story would break, ostensibly to assure her that he wouldn’t identify her publicly. This and his failure to disclose the incident earlier are distressing.

More recently, last year, Smith was kicked out of a pick-up basketball game for fighting. Other witnesses say Smith punched a man in the groin; Smith denies he hit the groin specifically. Before that, in January 2011, he was banned from a co-ed soccer league for a season after pushing another player.

(In the interest of fairness, I will also point out that there are allegations that Hales, Smith’s opponent, claimed Washington residency for tax purposes while continuing to vote in Oregon.)

Unsurprisingly, Smith’s candidacy has been hurt by this. Some organizations have withdrawn their endorsement. In response, Smith’s campaign blamed the media for “character assassination”. Quoth the Oregonian,

Jefferson Smith supporters are trying to raise money to buy ads that would blame the candidate’s recent troubles on “gotcha” journalism.

An email circulating this week says coverage of the Portland mayoral candidate in The Oregonian and Willamette Week newspapers amounts to “character assassination.” The email lists other supporters and asks for money to buy ads in the two newpapers.

“Their coverage has bypassed the issues and the positions of the candidates in favor of ‘gotcha’ journalism,” the email reads. “Portland deserves better.” The group is also planning a news conference this week, the email says.

This is absurd. Portland does deserve a debate on policy issues, but a candidate’s character and competency is also important. It’s impossible to predict exactly what issues an elected official will face—I don’t think most Portland voters predicted that water fluoridation would later become an issue when they voted in the last election—and so Smith is asking voters to trust his judgement to make decisions on unforeseen issues. His past decisions, the good and the bad, is all voters have to go on. And so it’s absolutely appropriate for the media to shed light on a candidate’s past, and what the candidate has done to learn from their past, especially in the wake of the candidate’s failure to disclose major incidents.

Likewise, competency is an important issue. Even if a candidate is on the right side of all the issues, they might still be terrible at doing the job. Temperament comes into play as well; I believe the current dysfunction in Congress has less to do with policy positions and more to do with the poor temperament: the unwillingness to compromise, the preference starting fights over finding solutions. Likewise, electing a mayor that has a problem with punching people probably doesn’t encourage a functioning city government.

Smith blames the media for his political problems, instead of his own poor handling of his past misbehavior. Smith seems to think that character issues are unimportant. That, alone, is probably a good reason for Portland voters to be reject him.

The only explanation that makes sense to me is misogyny

Mon, 15 Oct 2012 22:15:09 -0500

Let’s consider two morally problematic actions:

Surreptitiously taking photographs of women, or stealing them from the women’s Facebook pages, and posting them on a forum with the explicit intent of having you and your buddies leer at said women.
Unmasking the identity of a person who participates in a forum pseudonymously, against that person’s wishes.

The last few days, since a reporter at Gawker did action (2) to the moderator of a forum dedicated to (1), I have been trying to imagine a reasonable system of ethics under which (1) is acceptable but (2) is not. Indeed, I can’t even come up with a reasonable argument why (2) is worse than (1). Or an explanation, other than misogyny, why a person would think (2) is worse than (1).

Nevertheless, the reaction of much of the Reddit user community has been anger at Gawker for the unmasking, and solidarity with the moderator of a forum titled “Creepshots”.

Personally, I’m inclined to agree with Scott Lemieux: while outing pseudonymous people online is generally a bad thing, this case is probably a reasonable exception.

Also, one of the most stunning revelations in the Gawker story, in my mind, is the relationship between Reddit’s administrators and the guy who was running the “Creepshots” forum.

But Violentacrez has historically had a close relationship with Reddit’s staff, a fact far less well-known than his controversial behavior. [. . .] A few years ago, while Jailbait [a forum dedicated to leering a photos of teenage girls] was still going strong, Reddit’s administrators gave him a special one-of-a-kind “pimp hat” badge to honor his contributions to the site, which he proudly displayed on his profile. Brutsch said he was even in the final running for a job as a customer support representative at Reddit last year.

This is beyond the laissez-faire “we let people post whatever legal content they want” attitude. Seriously, if someone you know runs a forum dedicated to leering at teenage girls, or leering over photos of women taken surreptitiously, you don’t give that person an award. You shun them.

How can this be for real?

Thu, 11 Oct 2012 22:34:45 -0500

Someone tried to leave this comment on my blog:

Aug 21st 2012 – Interior decoration is a matter of pleasure or stress, depending on how you look at it. This technology is brand new and has only been around for the past 6 months. The flight is scheduled according to the needs of the passenger at his or her own pace complete with any stops or side-trips that may be requested. How can this be for real? Something is going wrong. It is 2012, and this is the United States of America

Does that make any sense? Not really. I posted it because I found it particularly amusing. This was, indeed, the 167th spam comment I’ve received. (In the URL field was the address for some site selling cheap handbags.) Something is going wrong, at least for this commenter. It is 2012, this is the United States of America, and I moderate my comments, bozo!

socialized pot

Fri, 5 Oct 2012 23:17:13 -0500

A few months ago, I made a half-joking post about the government monopoly on the retail sale of liquor in many states, including my native Oregon, calling it “socialized booze”. Part of the point of that post is that I am tired of people calling things (like certain recent laws to improve health care) “socialism” that don’t involve public or government ownership of capital. When people misuse the word “socialism” to mean “government subsidy or social welfare program”, I have an urge to hit them over the head with a political science textbook.

Anyway, if you think socialized booze is weird, wait until you learn about socialized pot. This fall, Oregonians will vote on a ballot measure to legalize marijuana. But this measure wouldn’t just legalize marijuana, but set up a system where the state government would buy pot from growers and sell it to consumers at state stores.

Mind you, I don’t have a strong opinion on this measure, and as I don’t live in Oregon anymore I don’t need to decide how vote on it. I am sympathetic to legalizing marijuana and to changing our drug laws generally. But I’m not quite sure of the wisdom of having the government in the pot-selling business. (Or the booze-selling business, for that matter.)

a beautiful fall day along the Upper Misssissippi

Sun, 30 Sep 2012 21:40:03 -0500

One neat thing about Apple’s new maps

Sat, 29 Sep 2012 22:54:00 -0500

I don’t have any iThings, so I haven’t tried the new Apple Maps in iOS 6, which appear to have been met with almost uniform disappointment from those who have tried them. But while browsing this incredibly amusing blog dedicated to documenting errors in the new Apple Maps, I noticed one really neat feature: specific state route shields! Most maps (paper or online) use a generic oval shield for state routes, but Apple Maps appears to use the proper route shield, as seen here for Pennsylvania, New York, Louisiana, and Oregon.

This is a nice touch that Google should copy immediately.

Some state route shields, reviewed

Thu, 27 Sep 2012 08:01:12 -0500

I have lived in four states:

Of these four states, I believe Pennsylvania has the best route shield. Pennsylvania is the keystone state (if you don’t understand why, look at a map of the original 13 states), and so it’s fitting that the route shield looks like a keystone. Distinctive and legible—good qualities in a route shield.

I like the Minnesota route shield. The colors are nice. I also like the small outline of the state in the corner. One issue is that it’s kind of hard to read the “Minnesota” from far away, but you don’t really need to, given that this shield is so distinctive.

The route shield of my home state of Oregon leaves something to be desired. It’s not very distinctive: it looks very much like all the other states that use a circular or vaguely oval-like pattern. When I was a child, I thought this route shield was some sort of deformed oval, like the default oval used for state highways on maps, until I realized it’s supposed to be the state seal. Except, with three-digit highways, they have to distort the seal to make it wide enough. I’m not saying Oregon should change its route shield—that would be a waste of money—but it’s really sad that such a beautiful state has such a drab route shield.

Sorry, but I don’t like Illinios’s route shield. It’s extremely boring and undistinctive. Worse, it looks too much like a speed limit sign!

(Route shields by Wikipedia users TwinMetsFan, Master_son, Fredddie, and SPUI, respectively. The Oregon Route shield image is used under a CC-BY-SA license.)

61/14 or 14/61?

Wed, 26 Sep 2012 21:10:47 -0500

The major highway through this town is a concurrency of U.S. Route 14 and U.S. Route 61. Some of the signage looks like this:

And some of the signage looks like this:

I still haven’t figured out what determines the order of the route shields.

(Route shields by Wikipedia user SPUI: 14 source, 61 source.)

Alternate side

Fri, 21 Sep 2012 19:27:47 -0500

sharrow

Fri, 21 Sep 2012 19:27:18 -0500

Eleven years…

Wed, 12 Sep 2012 20:20:44 -0500

It’s been eleven years since that horrible day the world changed, and I don’t know that I have anything substantial to say about it. Maybe instead I’ll post a photo of this memorial I found in the Glasgow Botanic Gardens. When I came across this memorial in 2010, it moved me, and made me hopeful that we might make a world with more global understanding.

train cars

Mon, 10 Sep 2012 20:52:37 -0500

The demise of LiveJournal

Sat, 8 Sep 2012 14:10:35 -0500

Aja Romano chronicles the demise of LiveJournal. (It’s still around, but far less busy than during its glory days.)

I joined LiveJournal in 2001 and used it pretty heavily until the mid- to late-noughties. It makes you wonder if Facebook will still be as popular as today in, say, 2020. (Of course, the Eff Bee has a lot higher market penetration than LJ ever did.)

About a year or so ago, I took my journal down from LJ’s servers. This is where my LiveJournal lives today:

(HT TNH.)

barge on the Mississippi

Wed, 5 Sep 2012 21:52:28 -0500

“Good general theory does not search for the maximum generality, but for the right generality.”

Wed, 5 Sep 2012 20:08:55 -0500

This comes from Saunders Mac Lane in his textbook Categories for the Working Mathematician (p. 108, emphasis mine), but I contend it applies to more than just mathematics:

One may also speculate as to why the discovery of adjoint functors was so delayed. Ideas about Hilbert space or universal constructions in general topology might have suggested adjoints, but they did not; perhaps the 1939–1945 war interrupted this development. During the next decade 1945–55 there were very few studies of categories, category theory was just a language, and possible workers may have been discouraged by the widespread pragmatic distrust of “general abstract nonsense” (category theory). Bourbaki just missed [. . .] Bourbaki’s idea of universal construction was devised to be so general as to include more–and in partcular, to include the ideas of multilinear algebra which were important to French Mathematical traditions. In retrospect, this added generality seems mistaken; Bourbaki’s construction problem emphasized representable functors, and asked “Find $F,x$ so that $W(x, a) \cong A(F,x, a)$”. This formulation lacks the symmetry of the adjunction problem, “Find $F,x$ so that $X(x, G,a) \cong A(F,x, a)$”—and so missed a basic discovery; this discovery was left to a younger man, perhaps one less beholden to tradition or to fashion. Put differently, good general theory does not search for the maximum generality, but for the right generality.

NOT BAD

Wed, 29 Aug 2012 21:42:00 -0500

By the way, if you want to know what I think about this whole reddit experience – NOT BAD!

I’d like to point out, not only did the president of the United States do an AMA on Reddit today, he threw in an allusion to a rage face he inspired.

Lake Winona at dusk (#2)

Mon, 27 Aug 2012 20:57:01 -0500

“The real satisfaction from mathematics is in learning from others and sharing with others.”

Sun, 26 Aug 2012 11:16:33 -0500

[M]athematics only exists in a living community of mathematicians that spreads understanding and breaths life into ideas both old and new. The real satisfaction from mathematics is in learning from others and sharing with others. All of us have clear understanding of a few things and murky concepts of many more. There is no way to run out of ideas in need of clarification. The question of who is the first person to ever set foot on some square meter of land is really secondary. Revolutionary change does matter, but revolutions are few, and they are not self-sustaining — they depend very heavily on the community of mathematicians.

—Bill Thurston, mathematics professor at Cornell and a Fields medalist, who passed away on Tuesday.

I also recommend Thurston’s 1994 essay “On proof and progress in mathematics”.

Peace, love, and neutrinos

Tue, 14 Aug 2012 17:49:53 -0500

Edited to add: I had meant to add this caption: My wife is a physicist who studies neutrinos. She was given these buttons at a physics event by a physicist who studies how to use neutrinos in counterproliferation.

bánh mì sandwich

Thu, 2 Aug 2012 14:36:28 -0500

Yum!

blog name change

Thu, 2 Aug 2012 10:46:10 -0500

Administrative note: I’ve changed the name of this blog, reusing the name of my (now defunct and largely deleted) LiveJournal. I’m not sure I’m going to keep this name, either.

Update: Okay, I changed it again. I might tinker more later. (I’m sure this is the most interesting post ever.)

Humans are not to be trusted!

Wed, 1 Aug 2012 11:05:47 -0500

Of course, given the increasing sophistication of ATM skimmers, I’m not sure machines are too trustworthy, either.

future Twin Cities light rail (#2)

Wed, 1 Aug 2012 11:03:03 -0500

Fact checking Mitt Romney

Fri, 27 Jul 2012 18:27:53 -0500

Let’s wade into some poly-ticks.

The British media is having a field day making fun of Mitt Romney’s recent stumbles in London, where he insulted his hosts by criticizing their preparation for the 2012 Summer Olympics. (The Guardian‘s headline yesterday: “Mitt Romney’s Olympics blunder stuns No 10 and hands gift to Obama”.) This has lead some bloggers and journalists to reprint this passage from Romney’s book, No Apology: The Case for American Greatness:

England is just a small island. Its roads and houses are small. With few exceptions, it doesn’t make things that people in the rest of the world want to buy. And if it hadn’t been separated from the continent by water, it almost certainly would have been lost to Hitler’s ambitions.

That’s pretty insulting, but is it true? Let’s take these assertions one by one:

#8220;England is just a small island.” Of course, on the face of it, this is false. England isn’t any kind of island. We’ll assume the governor meant Great Britain (which is an island), or the United Kingdom (which is a nation-state situated on a number of islands, the largest of which is Great Britain). Now, Wikipedia has a list of the world’s largest islands (by area), and Great Britain is the 9th largest in the world, so I wouldn’t personally consider it a small island. Sure, the U.K. is small (in area) compared to the United States—the U.K.’s area is about 243,610 km², compared to 255,026 km² for my native state of Oregon—but the U.K. has some 63 million people, and a GDP of US$2.29 trillion, giving it the 9th largest economy in the world. It’s not an insignificant place.
“Its roads and houses are small.” Matthew Ygelsais at Slate points out that it is true that the U.K. has smaller houses. Of course, given the large increase in house sizes America has seen in the past few decades, I’m not sure this is a bad thing. (Remember, the U.K. didn’t have a subprime mortgage crisis, either—although some of their banks did need to be bailed out, because they got involved in American subprime mortgages.) As far as the roads go, it’s true that you will find narrow streets in the older parts of European cities, because such streets predate the invention of the automobile. Nevertheless, you will also find highways in the U.K. that look just like the ones in America:

Well, okay: modulo the side of the road the cars drive on.
“With few exceptions, it doesn’t make things that people in the rest of the world want to buy.” To fact-check this claim, I made a small table:

United States United Kingdom

Exports (2011) US$1.511 trillion US$495.4 billion

Population (2011) 313,847,465 63,047,162

Exports per capita US$4814 US$7858

For whatever it’s worth, the U.K. is the world’s 11th largest exporter.
“And if it hadn’t been separated from the continent by water, it almost certainly would have been lost to Hitler’s ambitions.” I don’t know enough about World War II to evaluate the merit of this claim, but it does come across as a jerky thing to say. Look, the U.K. was fighting Nazi Germany as early as September 1939; the U.S. didn’t get involved in direct fighting until December 1941. By the way, the Battle for Britain, in which the British successfully resisted German invasion, occurred in 1940—while we were sitting on the sidelines.

	United States	United Kingdom
Exports (2011)	US$1.511 trillion	US$495.4 billion
Population (2011)	313,847,465	63,047,162
Exports per capita	US$4814	US$7858

America is a great country. We shouldn’t have to say incorrect and jerky things about other countries to feel good about ourselves.

(Information about GDPs, country size, population, and exports were taken from the CIA World Factbook.)

System Administrator Appreciation Day

Fri, 27 Jul 2012 16:45:54 -0500

I have learned (from BoingBoing, TeXblog, and a few other places) that today is System Administrator Appreciation Day.

I don’t know who the system administrators are for this site. I pay a company, NearlyFreeSpeech, to host this site, ChrisPhan.com, and PhanBudd.com. I have not had any communication with NearlyFreeSpeech, save a few automated emails reminding me to renew domains, et cetra, so I haven’t had an opportunity to get to know my sysadmins. But I will say this: in the four or so years since I started using NearlyFreeSpeech to host my stuff, I have not, to my knowledge, experienced a second of downtime. So, NearlyFreeSpeech sysadmins, even though I don’t know who you are, you all rock, and you are very much appreciated!

Michigan and Ontario photos

Tue, 10 Jul 2012 13:25:24 -0500

The Mackinac Bridge, connecting Michigan’s Upper Peninsula with the Mitten, is the third-longest suspension bridge in the world.

Bikes parked on Mackinac Island, where cars are banned:

The International Bridge, connecting Michigan and Ontario (from the Ontario side):

In Sault Ste. Marie, you can see two sets of fireworks in early July, for Canada Day and Independence Day:

The Soo Locks (allowing ship traffic between Lake Superior and Lake Huron):

I got this awesome souvenir from Michigan:

Downers Grove Fire Station 5

Sat, 30 Jun 2012 17:29:22 -0500

Is there a Ducks fan working for the Village of Downers Grove?

Sat, 30 Jun 2012 15:31:10 -0500

If so, then awesome! Go Ducks!

Facebook is not a good forum for political arguments

Thu, 28 Jun 2012 16:12:32 -0500

Every once so often, I get sucked into a political argument on Facebook. Sometimes someone will post one of those “share this if you agree” pictures, and I don’t agree with it. Sometimes one of my liberal friends’ conservative friends will make a comment, and because (I’m not proud to admit) I sometimes suffer from the tendency illustrated by xkcd #386, I will feel moved to respond. Every one of these encounters has left me increasingly convinced that Facebook is not a good forum for political arguments. I’ve identified two reasons:

The way Facebook displays comments is poorly suited for discussion.
People who post political content don’t necessarily want to spark a discussion, and it’s hard to tell.

The problems with Facebook’s commenting system

A productive, respectful, and thoughtful exchange of ideas is most likely if every participant has carefully read and considered the previous contributions to the discussion. The design of Facebook’s commenting system discourages people from doing this.

That’s because:

Facebook hides all but the two most recent comments.
After comments reach a certain length, Facebook hides the remainder of the comment behind a “See More” link.
Facebook will even truncate the original post, if it’s beyond a certain length, by hiding most of the post behind a “See More” link.

The overall effect is to require additional effort for people to read everything that’s been said before. As a result, it’s easy for people to repeat points that have already been made, or to criticize others’ views without actually having read them in full, or without understanding the reasoning. People end up talking past each other.

Furthermore, Facebook encourages shorter comments, by having the [Enter] key post the comment, and requiring [Shift]-[Enter] to make a line break. This further discourages substantive discussion.

Don’t get me wrong: I don’t think these are poor design decisions on the part of Facebook’s engineers. Indeed, these design decisions are perfectly fine for comments congratulating people on a new job, wishing someone a happy birthday, or commiserating with someone who has had a poor day at work. They are just poorly suited to political arguments. I imagine this is because Mark Zuckerberg and his friends did not have political discussions in mind when they designed Facebook.

It’s hard to tell when people actually want to have a discussion

People don’t just use social networking sites to stay in touch with friends and family, but also as a way to express their identity. Political posts are no exception to that: for many of us, our political ideology and beliefs are an important part of our identity. And not everyone who posts political views is interested in having those view challenged. Sometimes they are looking for affirmations from like-minded friends. Sometimes they just want to share information or lines of argumentation with like-minded friends.

By the way, there’s not necessarily anything wrong with this. I do believe that honest, frank debate of political issues makes for a better society, but such discussion can be draining. It’s reasonable for people to not want to engage in such discussion at all times. A good analogy would be someone who has an Obama ’12 bumper sticker on their car. Such a person doesn’t necessarily want to discuss the president’s policies with every driver they meet on the highway.

The problem is that’s it can be difficult to tell whether a poster is open to having their views challenged or not. So, when you respond to such posts, you run the risk of bothering someone.

Happy 100th Birthday, Alan Turing!

Sat, 23 Jun 2012 09:43:45 -0500

Today is the 100th birthday of one of the 20th century’s most under-appreciated people, British mathematician, computer scientist, and cryptoanalyst Alan Turing.

It’s impossible to live in modern society without coming into the consequences of Turing’s work. Alan Turing was a pioneer computer scientist, laying the theoretical framework for the information age. He also made key contributions to the Allies’ code-breaking efforts during World War II. It’s been estimated that his contributions sped up the defeat of Hitler by as much as two years.

Unfortunately, Alan Turing was a gay man in an age when it was illegal for men to have sex with other men.* Despite his contributions to winning the war, Turing was chemically castrated by the British goverment in the 1950s, and committed suicide as a result.

For a general nontechnical introduction to this great man, you might want to check out this episode of the public radio program Radiolab.
Gordon Brown’s 2009 apology, on behalf of the British government, for the government’s poor treatment of Alan Turing.
How Alan Turing Invented the Computer Age, by computer scientist Ian Watson.

(Pictured is the Enigma machine, used by the German military to encrypt messages during the Second World War.)

Edited to add (21:18 CDT):

Here’s a neat video summarizing Turing’s work.

*It’s worth noting that as late as 2003, it was illegal in some states in the U.S. for a man to have sex with another man.

Mixed-case street sign

Thu, 21 Jun 2012 16:57:43 -0500

A recent change to the MUTCD (Manual on Uniform Traffic Control Devices) requires all new street-name signs to be in mixed case, as opposed to the more traditional all-uppercase format. This is probably why we’ve seen some mixed-case street-name signs (like this one) pop up in Downers Grove.

View of downtown Naperville from the Millennium Carillon

Tue, 19 Jun 2012 17:15:34 -0500

You can actually see some Chicago skyline if you zoom in. (The Loop is about 45 km from the location of this photo.)

I’m not sure what this means

Tue, 19 Jun 2012 15:18:32 -0500

This clearly-unofficial hand-painted sign was affixed to a signpost in Downtown Downers Grove.

ACHIEVEMENT UNLOCKED: round-trip Empire Builder, Chicago to Portland

Fri, 15 Jun 2012 14:37:50 -0700

A friend asked me (via the Eff Bee) about my recent trip from Chicago to Portland, and back, on Amtrak’s Empire Builder. (I just got back yesterday.) I wrote a fairly long comment on Facebook, and I thought it would be good to reproduce it here (lightly edited):

It cost me $376.20 for a round-trip ticket from Portland to Chicago. I got a reserved coach seat instead of a sleeper car compartment. A sleeper car compartment is substantially more expensive.

The seat you get is sort of similar to an airplane seat except with considerably more legroom. You can recline the seat without even slightly inconveniencing the person behind you, and there is a leg rest you can pull out to help be more comfy. That said, it’s not as comfy as an actual bed.

There was electrical outlets on my train but no WiFi. I got 3G pretty consistently in Illinois, Wisconsin, Minnesota, and North Dakota. In Montana, you don’t get 3G except when you are close to towns. The rest of the time was hit-or-miss. (My carrier is Verizon.)

The trip takes about two days. On my return trip, I left Portland at 4:45 p.m. Tuesday and got into Chicago at about 7:00 p.m. Thursday. My train was 3 hours late—which is common—and was scheduled to arrive at 3:55 p.m. Thursday. (On the other hand, on the trip out to Portland, my train arrived slightly early.)

On the train, there is a observation car with tables, seats, and large windows. You can hang out there when you are tired of sitting in your coach seat.

I would recommend bringing some of your own food. (I brought crackers, bagels, a jar of peanut butter, dried fruit, tuna in packets, etc.) You can buy food from the cafe in the observation car, but it’s a bit overpriced. (A 12-ounce cup of coffee or a 12-ounce can of soda costs $2.) There is also a dining car which is basically a medium-priced sit-down restaurant. (The entrees are between $15 to $25.) If you get the sleeper car compartment, the dining car meals are included.

One very nice thing about the train is you see the landscape you can’t really see while flying or driving. My father-in-law pointed out to me that the sleeping car accommodation, while expensive compared to a plane ticket, is not a bad deal compared to driving, staying in a hotel every night, and buying restaurant food for every meal.

Another thing I should mention is that half the train goes to (or comes from) Portland, and half of it goes to (or comes Seattle); they split apart (or join together) in Spokane. The practical significance of this is that the dining car goes to (or comes from) Seattle and the observation car goes to (or comes from) Portland. So, if you are coming from or to Portland, you won’t have the dining car west of Spokane (i.e. the first night on a eastbound trip, or the last morning on a westbound trip).

Overall, I think it’s worth taking the train if you aren’t in a hurry, want to see some beautiful landscape along the way, and don’t mind sleeping in a seat for two nights. I enjoyed my trip.

from the train #16 (Portland)

Sat, 2 Jun 2012 10:01:05 -0700

from the train #15 (Vancouver, Washington)

Sat, 2 Jun 2012 09:19:36 -0700

from the train #14 (east of The Dalles)

Sat, 2 Jun 2012 07:54:24 -0700

from the train #13 (east of Umatilla)

Sat, 2 Jun 2012 06:13:52 -0700

from the train #12 (east of Glacier NP)

Fri, 1 Jun 2012 21:05:15 -0600

from the train #11 (Cut Bank)

Fri, 1 Jun 2012 19:05:21 -0600

from the train #10 (east of Malta)

Fri, 1 Jun 2012 13:42:05 -0600

from the train #9 (east of Williston)

Fri, 1 Jun 2012 11:02:21 -0500

from the train #8 (west of Des Lacs)

Fri, 1 Jun 2012 09:47:46 -0500

from the train #7 (Lake Irvine)

Fri, 1 Jun 2012 07:26:29 -0500

from the train #6 (good morning North Dakota)

Fri, 1 Jun 2012 05:06:53 -0500

from the train #5 (north of Winona)

Thu, 31 May 2012 20:43:07 -0500

from the train #4 (La Crosse)

Thu, 31 May 2012 19:48:28 -0500

And now I am in Minnesota, not far from where we will be moving later this summer.

from the train #3 (Wisconsin)

Thu, 31 May 2012 18:29:11 -0500

from the train #2 (Milwaukee)

Thu, 31 May 2012 15:45:05 -0500

from the train #1 (Chicago)

Thu, 31 May 2012 15:05:20 -0500

to whom I paid tolls

Tue, 29 May 2012 20:49:27 -0500

Mounted to the windshield of my car is an I-Pass toll transponder, which is Illinois’s version of the E-ZPass toll transponder used to collect tolls throughout much of the eastern half of the country. I find it both fascinating and creepy that using a toll transponder creates an electronic record of your whereabouts:

Some observations:

Both the transponder in my car and the transponder in my wife’s car are in her name. (This report shows only tolls incurred by the transponder in my car.)
It costs more to cross the George Washington Bridge ($7.50) than it does to drive the entire 156-mile length of the Indiana Toll Road ($4.18). (Although, to be fair, the trip across the George Washington Bridge is much nicer than the trip across northern Indiana.)
In contrast to the George Washington Bridge, the Delaware Water Gap Toll Bridge is a bargain at $1.
Both the Indiana Toll Road and Ohio Turnpike use a ticket system, where you take a ticket when you enter the toll road, and then pay tolls when you leave the toll road, based on how far you’ve travelled. (Since I have the I-Pass, I don’t need to take a ticket—it’s handled electronically.) That’s why you see only one tool recorded for each trip.
On the Illinois Tollway, by contrast, you simply pay a toll every so many miles along the highway (as well as at certain on- and off-ramps). The last two tolls, for example, represent my cost for using I-355.

socialized booze

Mon, 28 May 2012 18:42:15 -0500

The word “socialism” has been frequently misused recently, mostly in hyperbole about the recent health care reforms. However, socialism is alive and well in another sector of the economy, in a number states in which the government has a monopoly on retail sales of spirits. The New York Times describes the liquor control system in the state of Washington, a system which will be eliminated on Friday:

State control, in turn, made generations of civil servants tastemaking critics — their decisions on what to stock dictating what people could order in bars or buy in the stores.

In 2010, for example, a tiny distiller here in Cashmere, called It’s Five O’Clock Somewhere, made a grape brandy that the owner, Colin Levi, was quite proud of. Liquor Control Board officials came by for a tasting and did not much care for it, Mr. Levi said, and that was that — it never went into distribution.

Until 2009, I lived in Oregon, another state with a government monopoly on the sale of liquor. In 2002, Oregonians rejected a ballot measure that would have created a single-payer health care system in that state. At the time, I remarked that people in Oregon were afraid of “socialized medicine” but had no problem with “socialized booze”.

Someone needs to take a stats class

Mon, 21 May 2012 19:19:53 -0400

“[I]n the end this is not a scientific survey. It’s a random survey.”—Rep. Daniel Webster (R-Florida), explaining his opposition to the Census Bureau’s American Community Survey.

Rep. Webster should hope that his campaign advisors have a better understanding of sound polling methodology than he does.

(Via Matthew Yglesias.)

Visit to the 9/11 Memorial

Thu, 17 May 2012 23:37:10 -0400

Today, Sarah and I visited the National September 11 Memorial, which sits on Ground Zero.

Visiting the memorial requires a visitor pass. The passes are free, but have to be reserved ahead of time on their web site. (I reserved our passes in late April, and there were already no slots available on Friday, Saturday, or Sunday.) There was about a 15 minute wait to enter the memorial, which included passing through metal detectors.

The memorial consists of a plaza full of trees, and two recessed pools in the footprints of the twin towers. Water flows down the sides of the pools, and the names of 9/11 victims line the rims of the pools.

Looking forward to the 14th b’ak’tun

Sat, 12 May 2012 23:45:47 -0400

The media has been discussing the recent discovery of Mayan astronomical calculations of events well past 2012. They are saying that this debunks the notion that the Maya predicted the end of the world on 21 December. Here’s a video news summary from Slate:

While this is really cool as an archaeological find, it doesn’t debunk anything that hasn’t been debunked a many, many times before. Indeed, the notion that the Maya ascribed any eschatological meaning to the end of the 13th b’ak’tun has been throughly discredited. For example, here’s an expert on the ancient Maya quoted in 2009:

“For the ancient Maya, it was a huge celebration to make it to the end of a whole cycle,” says Sandra Noble, executive director of the Foundation for the Advancement of Mesoamerican Studies in Crystal River, Fla. To render Dec. 21, 2012, as a doomsday or moment of cosmic shifting, she says, is “a complete fabrication and a chance for a lot of people to cash in.”

And here’s a Mayan elder quoted by the Associated Press in 2009:

Apolinario Chile Pixtun is tired of being bombarded with frantic questions about the Mayan calendar supposedly “running out” on Dec. 21, 2012. After all, it’s not the end of the world.

Or is it?

Definitely not, the Mayan elder insists. “I came back from England last year and, man, they had me fed up with this stuff.”

Let me tell you why I find the 2012 hype obnoxious: One of the key tenets of this 2012 stuff is that the ancient Mayan civilization had important insights that are now being ignored by modern Western society. Its believers hold themselves as more “enlightened” for recognizing the value of this native culture. Except, in reality, the hype is based on ignoring what the Maya said and misappropriating their voice to push silly new-age ideas.

Hopefully, these latest archaeological finds will tamp down this 2012 nonsense.

PhanBlog: now with blogroll!

Wed, 9 May 2012 00:14:58 -0400

I’ve added a blogroll, which you should see on the left of the main page of this blog. (Of course, some of these blogs regularly express opinions with which I disagree, so you shouldn’t construe inclusion on my blogroll as an endorsement of that author’s political views.)

Agricultural area cautions

Sun, 29 Apr 2012 21:31:27 -0400

A bowl full of candy bars is another example of “a breakfast they’ll want to eat”

Sat, 28 Apr 2012 14:55:39 -0400

Recently, the makers of Nutella settled a class-action lawsuit from a San Diego mother who claimed advertising mislead her into thinking the stuff was healthy.

I found a YouTube video of one of the ads embedded in Carly Rothman’s article about the lawsuit:

Here’s a question for you: Do you think the health claims in this ad are misleading?

That’s a trick question, because the ad makes no health claims. The mother in the ad says “it’s perfect on multigrain toast, and even whole wheat waffles”—in this context, “perfect” could mean “healthy”, but it could also mean “really yummy”. She says it’s made from “simple, quality ingredients”, which is also true of a stick of butter. The mother also says that after feeding her kids Nutella, “I feel good that they’re ready to tackle the day”. None of these statements imply the stuff is healthy. In fact, the words “healthy”, “balanced breakfast”, or “low-fat” appear nowhere in the ad.

(Mind you, I haven’t seen any of the other Nutella ads, so I’m not saying Ferrero, the makers of Nutella, isn’t guilty of making false health claims, just that this particular ad doesn’t make any.)

This commercial is a great illustration of how an advertiser can leave an impression (in this case, that Nutella is healthy) without explicitly saying it. Of course, this is done by advertisers all the time: a textbook example is the use of attractive women to give the impression that using the product (often some extremely mediocre beer) will attract such women. This is why we need more media literacy education. It’s not enough to ban advertising that’s blatantly false—you have to make people aware of advertisers’ subtle tricks.

Also, I agree with Carly Rothman: if you are going to start feeding some food to your kids every morning, perhaps it would be wise to glance at the nutrition label beforehand.

My photography featured on the BBC News website!

Wed, 25 Apr 2012 08:35:00 -0400

Yesterday, I found a photo I took in 2003 used on the BBC News site! It’s pretty awesome. Here’s what happened:

Back in 2003, I received a digital camera as a graduation gift. I used it to take a photo of one of my favorite signs: the sign for the Boring/Oregon City exit on U.S. 26. I was relatively new to photography; I don’t think it’s my best work.

The next year, I uploaded a copy of this photo to Wikipedia. I didn’t know any better, so I reduced the image to 512 × 384, which you are not supposed to do when putting photos on Wikipedia. If I had access to the original full-size file, I’d replace the one on Wikipedia. The closest I have is this Flickr photo, taken in the same session, which lacks the truck.

Yesterday, one of my friends from high school posted a BBC News article on her Facebook wall. According to the BBC, the village of Dull in the Scottish Highlands wants to become the sister community of Boring, Oregon. I followed the link because I, like almost anyone who attended Boring Middle School, was excited to see Boring featured in the international media. But then I saw the photo on the article, which they presumably took from Wikipedia, and said “HOLY CRAP! That picture of the ‘Boring, Oregon City’ sign? It’s my picture!”

I am extremely honored to find my work featured by the BBC!

Since many of my Flickr photos are available under Creative Commons licenses, I am used to seeing them occasionally appear on blogs. I believe this is the first time, however, that my work has been used in an international news story!

(anti)social reading apps

Tue, 24 Apr 2012 19:47:37 -0400

THIS.

Except: when I click on a link on the Eff Bee to an article and I’m asked to add an app to follow the link, I don’t Google for the article—I refuse to read it at all. I refuse to reward antisocial behavior.

First post

Mon, 23 Apr 2012 21:03:30 -0400

Welcome to PhanBlog!

I’ve created this blog to share my valuable insights with the rest of the world. I am planning to post about funny things found on the Intertubes, politics, mathematics, and perhaps West Wing episodes (because more online discussion of six-year-old TV shows can’t hurt). Anyway, in the mean time, here is a video of my parents’ cat:

Group of countries	Regression⁴	\(r^2\)	Plots
all	\(\log(n) = 0.3482\log(P) - 0.5369 \Rightarrow n = 0.5846P^{0.3482}\)	\(r^2 = 0.8482\)	linear log
NATO	\(\log(n) = 0.3990\log(P) - 1.0466 \Rightarrow n = 0.3511P^{0.3990}\)	\(r^2 = 0.8764\)	linear log
OECD	\(\log(n) = 0.3850\log(P) - 0.8940 \Rightarrow n = 0.4090P^{0.3850}\)	\(r^2 = 0.8026\)	linear log
EU	\(\log(n) = 0.4477\log(P) - 1.8032 \Rightarrow n = 0.1648P^{0.4477}\)	\(r^2 = 0.8817\)	linear log
OECD, NATO, or EU	\(\log(n) = 0.3796\log(P) - 0.7875 \Rightarrow n = 0.4550P^{0.3796}\)	\(r^2 = 0.8339\)	linear log