A few days ago, ooh.directory showed up on Hacker News with collections of blogs on all sorts of topics. This reminded me how fun the internet was in the early 2000s and provided me the encouragement I needed to start a blog of my own. I'm not sure entirely what content will show up here, but probably a mix of old and new side projects, being a new father, games I'm playing, and other similar sorts of things.

If you're reading this and want to talk more about anything I've written, feel free to reach out at blog@ariblumenthal.com.

Several years ago, I created the side project I still use the most.

booggle is a fictional ghost inspired web implementation of a popular word game involving a 4x4 grid of letters. Designed to be played by me and my then girlfriend (now wife), it helps kill a few minutes pretty much whenever we need.

The application was built using a minimal nodejs server (with express) that tracks users coming and going, handles state for games that are in progress, checks words against a dictionary as users are playing, and provides scoring at the end.

On the client, I used mostly vanilla javascript with handlebars. Simple json messages are passed between the client and server via a websocket.

Want to try it out? It runs over at ari.blumenthal.dev/booggle. It's a little buggy, only one game can be played at a time, the dictionary isn't perfect, but hey, anyone can play.

Source for booggle is now available over on my github at github.com/zkhr/booggle.

When I set out on creating this new blog, I knew I wanted it to have a novel aspect. While the high level content is pretty familiar across personal websites on the web (about me, blog, projects, etc), the structure puts a fun spin on the "single page application" concept.

As you may have seen by the time you got to this blog post, the site is an infinite grid. Each page on this site is a panel in that grid.

The prototype for this is relatively straightforward, but it required a couple interesting pieces across my nginx config, CSS, and JS.

For the prototype implementation, I load the entire website, no matter which page you hit. This allows exploring without having to load anything additional from the server. Since this site is all text, it loads quickly (and will for the foreseeable future).

To support this in my nginx configuration, I do two important things:

1. I enable Server Side Includes with ssi on. This allows me to create a new file per panel and then include them individually in the main index.html.

2. I use the rewrite directive so that all paths that start with /! will get sent to that index.html file.

location /! {
  rewrite ^.*$ /;
}

This allows me to track the current position of the user in the grid in the URL. For example, this page's path starts with /!/1/-3, which defines the x and y coordinates of 1 and -3 respectively. When the user refreshes the page, we correctly load the index.html file (instead of throwing a 404 not found error).

The grid is a position: relative element that holds all of the panels. To move from panel to panel, we change the grid's left and top properties.

As an analogy, you can think of your monitor as a microscope and the grid as a slide under the lens. To move around, we move around the slide, while the scope remains stationary.

Within the grid are several position: absolute panels. Each panel has distinct x and y coordinates, which are used to set appropriate left and top properties when initializing the page. Because of the relative position attribute on the grid, these are absolute positions relative to the grid.

For the prototype, the JS code does three main things:

1. Placement of the grid panels at their appropriate locations in the grid.

2. Enabling the grid links so that we can easily navigate between panels.

3. Supporting keyboard navigation (asdf, hjkl, arrow keys) as well as touchscreen navigation as alternative mechanisms to move between panels.

And that's pretty much it. As mentioned, I load the whole site on every page load which isn't ideal. If I stick at this blogging thing for a while, I will probably write a small server to better handle loading panels as needed.

Early on in the pandemic, I left my job at Verily which was hitting some growing pains while at the same time expecting everyone to work long hours on its new covid testing program. At my new job, a coworker asked what I did for fun. Since my wife and I had been doing a lot of jigsaw puzzles I mentioned, "Puzzles?" and they said something like, "oh you mean like puzzle hunts?".

And with that, I was nerd sniped! The rest is history.

A puzzle hunt is a collection of puzzles, often with certain shared characteristics like metapuzzles (that require the answers to previous puzzles to solve) and answers that are short phrases or words.

Brian Chen's Introduction to Puzzlehunts provides an excellent background.

I've done several hunts so far with my wife and friends, typically under the caffiends alias. I would consider myself to be squarely in the "novice" category. I can do the intro puzzles and maybe get a meta, but I still have a ways to go.

• 2022-08-26 - Galactic Puzzle Hunt 2022 (solves)
• 2022-07-02 - 2022 Truzzle Hunt
• 2022-06-18 - Huntinality 2.0 (solves)
• 2022-03-11 - glyph 2
• 2022-01-14 - MIT Mystery Hunt 2022
• 2021-12-11 - Silph Puzzle Hunt (solves)
• 2021-10-01 - Teammate Hunt 2021 (solves)
• 2021-08-21 - Puzzle Potluck 4 (solves)
• 2021-07-23 - Galactic Puzzle Hunt 20/20 Vision (solves)
• 2021-01-30 - Inexact Puzzle Hunt
• 2021-01-15 - MIT Mystery Hunt 2021

puzzlehuntcalendar.com is the definitive place to find upcoming puzzle hunts.

Puzzled Pint also has a monthly release of themed puzzles.

The Puzzle Hunt Toolkit is a collection of tools I've found useful while solving hunts.

The last year or so has been great for the pokémon franchise with several games hitting that nostalgic itch and innovating in interesting directions.

In April 2021, New Pokémon Snap revisited the best N64 rails shooter and in Jan 2022, Pokémon Legends: Arceus took a spin on the standard formula.

Pokémon Scarlet and Violet continue this with another solid entry in the franchise. While many reviews rightfully criticize the performance and graphical issues, having grown up with the graphics of n64 and gamecube games, that didn't really bother me too much. Instead, I could enjoy a really engaging gameplay loop of just wandering around an open world, catching pokémon, and taking down gyms.

This time around, I tried to have a balance of pokémon from different generations with good type coverage. Here is the lineup:

I started with the fire croc Fuecoco, so its evolution Skeledirge was the first pick. Running into Pawmi early on meant that of course Pawmmot would take spot #2. I have a sweet spot for any mon that can learn false swipe, so Scizor, Drednaw, and Garchomp all made the cut. Finally, to round out the team with some fairy and psychic coverage, Gardevoir took spot #6.

As mentioned in the new site, who dis. blog post, this site is an infinite grid, with each page on the site being a panel on that grid. This raises the question, how do I decide where on the grid to place new pages?

Being an infinite grid, there are an infinite number of options! For example, we could just go off in an arbitrary direction:

··············
··█───────────
··············

But that doesn't really take advantage of the dimensionality of the grid. Instead, we could spiral around a point:

┌─────┐
│┌───┐│
││┌─┐││
│││█│││
│││└┘││
││└──┘│
│└────┘

Or we could use the Cantor Pairing Function or even pick randomly!

So, what did I go with? Well, for blog posts, I decided to use the Hilbert curve, a space filling fractal, since it has some nice properties:

█·······
│┌─x┌─┐┌
└┘┌┘└┐└┘
┌┐└┐┌┘┌┐
│└─┘└─┘│
└┐┌──┐┌┘
┌┘└┐┌┘└┐
│┌┐││┌┐│
└┘└┘└┘└┘

This blog post is at 'x' on the curve.

If you get lost navigating the grid, you can always grab a map to check your position.

Congratulations to the folks over at LLNL for their news yesterday on achieving fusion ignition. From the linked announcement: So, why is this milestone exciting? Let's dive in.

Looking at the U.S. Energy Information Administration's electricity generation by energy source in the United States for 2021, roughly 4.11 petawatt hours were generated with ~60% coming from fossil fuels, ~20% from nuclear energy, and ~20% from renewables.

Fusion energy, often relegated to the realm of science fiction, has significant advantages over the nuclear and fossil fuel categories.

In comparison to nuclear energy, there is no risk of catastrophic meltdown, the raw materials are abundant, and the nuclear waste is minimal and decays relatively quickly.

When considering the looming effects of climate change, traditional fossil fuels like oil, natural gas, and coal become a lot less attractive.

So fusion energy, when available, could easily fit inside a portfolio of cleaner energy solutions like wind, hydropower, solar, biomass, and geothermal.

One of the more popular approaches (with a few dozen around the world), tokamak devices work to study fusion by generating a powerful magnetic field to confine the fusion fuel.

On the other hand, LLNL's National Ignition Facility (NIF) pursues fusion by means of inertial confinement (i.e. using energy from lasers instead). I'm not a nuclear physicist, so I won't go into more details than magnets vs lasers, but check out LLNL's article on How NIF Works for some interesting background.

In the announcement yesterday, NIF successfully created 3.15 MJ of energy by delivering only 2.05 MJ to the target. This is an awesome milestone, with the big caveat that in order to deliver 2.05 MJ of energy, the laser system needed ~300 MJ from the grid.

So, what's next? Like a tech tree in a video game, achieving fusion energy as a long-term power source requires prerequisite research advances in many areas. How do we fire shots at a higher frequency? How can we increase the yield from the lasers to require less energy from the grid? How do we further increase the energy yield? And so on.

The joke has been that fusion energy is always a couple decades away, but yesterday's milestone truly feels like we've moved one step closer. Who knows... ask me again in 2040.

The last few decades have seen a board game renaissance, with a tremendous number of amazing games coming out. Even better, they have made their way online where you can play against folks from around the world.

When I was younger, there wasn't the same variety of board games we have today, especially in digital form. I remember learning a bunch of Risk strategy by playing TurboRisk bots and then using that later to work my way up through the ranks of conquerclub.com.

Now, with sites like board game arena, there are all sorts of games available to play! I'm currently playing seven different games at seven virtual tables.

In the order of when the games were released, here they are:

2004 - A simple game with three parts: (A) a stock market where you can buy and sell shares in stocks, (B) a Risk-style map where you can expand and take over other companies, and (C) a deck of cards that determines which companies you can expand each round.

2010 - . A card drafting game where you build up your civilization. While taking simultaneous turns to draft (great for online games), you get points based on your military, science, and more. Cards can only be drafted if you have the right resources or have drafted prerequisite cards.

2013 - Another drafting game with a similar playstyle to 7 Wonders. Instead of building a civilization (and needing to buy drafted cards), you instead earn points based on the types and combinations of sushi you draft.

2015 - What if you could buy patents for the letters of the alphabet? In this game, you draw letters from a deck of cards to make words and get bonuses based on the letter patents that you've bought.

2016 - This is a game of drafting dominoes to build up a 5x5 grid. Points are earned based on contiguous types of land (lakes, forests, etc) and the number of crowns contained on those tiles.

2017 - This is a game of drafting patterned ceramic tiles and using them to build a wall, scoring points based on matching colors and building rows/columns.

2019 - Tableau building game with birds. Probably my favorite new board game of 2022.

Learning about typography can lead you to noticing things that you might have otherwise missed before. This can suck, like seeing bad kerning everywhere, but it can also be fun to see the patterns in fonts used by restaurants, websites, movies, and more.

I'm not an expert on this topic, but I love seeing new and interesting fonts and wish more sites on the internet took advantage of the variety out there. So to that end, I've gone ahead and updated this site to use a few that I like.

Most text content is written using a font based on Edsger W. Dijkstra's handwriting that I grabbed from Luca Cardelli.

For titles, section headers, and links around the blog, I use Carl Krull's Traveling Typewriter font.

For code, I use Ubuntu Mono, which is my favorite monospace font and the one I've used for my terminal for many years (even though Ubuntu is no longer my linux distro of choice). Canonical funded Dalton Maag to create this font family for the Maverick Meerkat release in 2010.

After beating the main story in pokémon scarlet a few weeks ago, I decided to keep the creature catching going with monster sanctuary on switch.

This game is a cross between a typical monster taming adventure and a side-scrolling metroidvania. Combat in the game is in the form of 3v3 turn-based monster battles with your typical elemental resistances, buffs, debuffs, etc, but there is enough variance between the different monsters and some interesting combat mechanics which keep things interesting and engaging.

For my first playthrough, I started with the and ended up keeping my same team of early monsters throughout: and That said, all the other monsters I caught for their abilities and to fill out the journal seemed pretty viable and had interesting skill combinations of their own.

I think this is one of the games where you kind of know ahead of time if you'll like it or not. If a monster catching side-scrolling metroidvania sounds fun to you, give it a whirl. The combat mechanics are a lot of fun and it won't disappoint.

2022 saw many issues, what with the ongoing covid pandemic, tech layoffs, the conflict in Ukraine, and more. But for me personally, it has ended on a high being able to spend the last couple months on paternity leave with my first kid.

The internet loves lists of things, so here are three lists of random categories for the year.

In 2022, I played plenty of games from past years, whether that was trying out Seablock in Factorio, thinking I could replay FFXV before my son was born (spoiler, he got here 2 weeks early), playing the 6.1 and 6.2 FFXIV patches, continuing a very-long-running Gloomhaven campaign with my friends, or trying another run of Don't Starve Together with my wife.

But, for this category we'll look specifically at my top five games that were released in 2022.

First, in no particular order, are some runner ups. Games I still enjoyed, but didn't make the cut: pokémon scarlet the latest entry in the long-running pokémon franchise, the tactical rpg Triangle Strategy with the Octopath-like HD-2D visual style, the short innovative solitaire village builder Stacklands, and the donut-based puzzle game Freshly Frosted.

Starting at #5 is APICO, the indie beekeeping simulator. It was simultaneously super relaxing, while remaining engaging collecting bees and mastering the mechanics of breeding and gathering resources. Not to mention an awesome soundtrack by Mothense.

At #4, described by Yahtzee in Zero Punctuation as a "post-dad game", Hardspace: Shipbreaker has you dismantling spaceships for parts in a dystopian capitalist scifi future. Like APICO, both relaxing and engaging, but rather than breeding bees, the joy came in learning the most efficient way to scrap ships and score those sweet credits to slowly chip away at your crippling debt.

Coming in at #3 is Strange Horticulture, an occult puzzle game of identifying plants and uncovering plots. I loved the user interface of this game that left it to the player to decide how to organize their stuff, the overall aesthetic that oozed with charm, and the thought that clearly went into designing the different plants and their associated puzzles.

At #2 is a game that actually came out in South Korea in 2019, but made its global debut in 2022. Lost Ark is an Action MMORPG with super fun combat, gorgeous locales, interesting characters, and plenty of end-game content. It had different modes that were reminiscent of playing Diablo, Monster Hunter, and FF14 all with the right mix of challenge and rewards. And not to mention all the mokoko seeds and island souls to collect!

Last at #1 is Pokémon Legends: Arceus which took a spin on the standard formula of catching pokemon. Running around an open world and throwing pokemon balls at unsuspecting monsters never got old. Here is the squad that I went with:

And that's it for games! Up next is TV shows!

For this category, rather than shows that debuted in 2022, I'll consider any show that released a new season in 2022. This is purely out of convenience, since I didn't watch that many new shows this year.

In 2022, I watched a of panel shows, quiz shows, and the like, including the newcomer The 1% Club, as well as classics like Jeopardy, Only Connect, QI, 8 out of 10 Cats Does Countdown, Would I Lie To You, and more. But in honor of Jeremy Paxman's last season of hosting, my #5 slot goes to University Challenge.

At #4, I honor a relative newcomer to the online streaming scene with College Humor's Game Changer. With a recurring cast of amazing comedians and incredibly clever games, this show never disappoints. Close runners up are their game shows and

Another dropout show, Dimension 20 takes the #3 slot, with their 2022 release of with Aabria Iyengar as GM. I was introduced to this show with but between and this show has hit after hit.

At #2 is Netflix's comedy murder mystery Murderville, hosted by Will Arnett. Each episode sees a different celebrity try and solve a case, having to improvise as they go along. With guests like Conan and Kumail Nanjiani, it was a joy to watch. It also introduced me to Murder in Successville, which this show was based on.

Last but not least, is my #1 pick, the new comedy action anime SPYxFAMILY, which sees a spy, an assassin, a telepath, and a dog who can see the future try and avert war. Hilarious, well animated, interesting characters. Definitely a fun new show, and with music by (who also did the amazing tracks for Dorohedoro), I can't wait to see where it goes next.

And that's it for TV. Last list is food!

Speed round! With a kid on the way in 2022, we moved to have a little more space after 3.5 years in San Mateo, CA. In honor of the town's amazing food scene, here's a quick list of our top 10 places to grab a bite.

To leave room for the amazing places unique to San Mateo, I've left out larger chains like Bonchon, Teaspoon, Mochinut, Curry Up Now, and more amazing spots that are more broadly available. Also special shout out to Suruki Supermarket, which could steal the #1 spot, but I'll keep this list to restaurants.

1. Jeffrey's Hamburgers
2. Apple Fritter
3. Ramen Dojo / Ramen Parlor
4. Pancho Villa Taqueria
5. Izakaya Ginji
6. Wursthall
7. Sichuan Chong Qing Cuisine
8. Avenida Restaurant
9. Pausa Bar & Cookery
10. Backhaus

Anyways, here's to a healthy and happy 2023!

Over the pandemic, I've played a few asynchronous games of Civ 6 with friends, using discord to keep track of whose turn it is. We wanted a way to automate @mention'ing the next player, so we looked around.

There are a few github repos that have this functionality that you can clone and run with, and there is an often cited reddit post that explains how to set up IFTTT for this, but there wasn't an easy way to add a bot to your server and have it "just work".

Until now!

Introducing, roze.run, the self-proclaimed "premier way to get turn notifications in Discord for Sid Meier's Civilization VI". This was a fun weekend project, in that there are only a few moving parts, but it gets the job done! Well mostly, Civ's webhook system has a bunch of issues that can actually make it kinda flaky, but under certain circumstances it works well.

This project is named after Datatech Sinder Roze from the Alien Crossfire expansion of one of my favorite games of all time: Sid Meier's Alpha Centauri. As an aside, if you like 4x games and haven't played it, do yourself a favor, close this blog, and buy it now. It is often on sale on gog and at time of writing it is only $1.49. I've easily spent hundreds of hours over the last two decades on this one game.

roze's site above has more details and her source is available on github in the zkhr/roze repo.

With a nearly 3 month old at home, portable pausable gaming has allowed me to relax during my downtime when the little guy naps or feeds, but still lets me jump in on a moment's notice. After a couple of months doing this on the switch (with games like pokemon scarlet and monster sanctuary) or on my phone with board game arena, I decided to unlock my steam library and pick up a deck.

And I enjoy using it immensely! The biggest technical issue I've encountered involves the screen not turning back on after sleeping, but with its short boot times, that issue hasn't bothered me too much.

So far, I've played a few games on it. I tried out Desk Job, replayed some Hades and Cozy Grove, delved into the newer acclaimed Vampire Survivors, and took my Civ VI turns in my play by cloud games with friends.

I even got ff14 working on it to catch up on the 6.3 main story! While reluctant to move from M+KB to controller gaming for an MMO, I've picked it up quicker than I thought I would, and actually prefer it for a few things already. Maybe not for extreme trials, but I'm mostly doing solo content anyway right now (so I can pause whenever I need to).

Not sure what I'll play next, but studios have packed February with new games up my alley. Hogwarts Legacy comes out on Feb 10, Theatrhythm Final Bar Line on Feb 16, and Octopath Traveler 2 on Feb 24. And in the latest Friday Facts, Twinsen announced controller support for Factorio is on its way, so I've got that to look forward to at some point, as well.

Suppose we wanted a blue block:

How would you specify its color in CSS? Well, let's start by rolling the clock back to 1996. Independence Day started airing in theaters, Game Freak launched Pokemon Red and Blue, Sid Meier released Civ II, and W3C published the first specification for CSS.

Per the CSS1 specs, we could write this using a color name, such as background: blue. The 16 original named colors are black, silver, gray, white, maroon, red, purple, fuchsia, green, lime, olive, yellow, navy, blue, teal, and aqua.

But folks typically want to use more than a few predefined colors. Well, you can also specify one of the ~16.7 million colors in the sRGB color space, which allows defining a color by its red, green, and blue channels. Instead of using the named color, you can use one of the following:

• background: #00f - a 3-digit hex code
• background: #0000ff - a 6-digit hex code
• background: rgb(0, 0, 255) - an integer rgb() value (from 0 to 255)
• background: rgb(0%, 0%, 100%) - a float rgb() from 0.0% to 100.0%

The CSS2 and CSS2.1 standards didn't change too much in the way that developers could specify colors, except the addition of orange as a new named color. So let's move on.

After CSS2 (which was one large spec for all of CSS), W3C switched to using smaller documents for different modules. The CSS Color Module Level 3 introduced a few things.

130 new named colors were added (with some overlaps like dimgray and dimgrey).

To specify an opacity for an element, rgba() was added. It takes a 4th parameter for an alpha channel: background: rgba(0, 0, 255, 0.5).

CSS Color 3 also added hsl() and hsla(), which use hue, saturation, and lightness. If you haven't seen this before, let's take a look at how this works.

Without invoking physics or just shouting the word color over and over, is pretty hard to define, but really easy to understand. It's where in that ROYGBIV wheel the color lands:

tells how vibrant the color is:

tells how light or dark the shade of the color is:

Putting these together, we get something like the following grid, with saturation on the x-axis and lightness on the y-axis (if we pick a single hue, like blue):

CSS Color Module Level 4 introduces several new and interesting things.

For starters, rgb() and hsl() syntax has been updated so that they no longer use commas and can optionally provide an alpha value after a forward slash: background: rgb(0 0 255 / 50%).

Hex codes also support alpha channels with 4 digit and 8 digit hex codes: #0000ff80.

As an alternative to specifying color via rgb() or hsl(), there are 5 additional color functions introduced: hwb(), lab(), lch(), oklab(), oklch().

The functional notation expresses a given color according to its hue, whiteness, and blackness. Hue is the same as the hue used in hsl(). The whiteness and blackness refer to how much white or black get mixed in.

For example, we can look at the square below and see how adding whiteness or blackness changes the color, with blackness on the x-axis and whiteness on the y-axis (again picking a single hue).

The last 4 ( and ) are similar to hsl(), but better map to human perception of lightness (versus using the same scales for all hues like in hsl()). These aren't available in all browsers yet, but will be worth checking out (especially when CSS Color Module Level 5 brings native relative colors to CSS).

I finished reading The Three-Body Problem earlier today, and decided to throw together a quick simulation over at three body simulation.

It kinda works, but mostly it just let me know how rusty my math and physics knowledge has gotten after not using it for 10+ years.

Busy couple of months. Went back to work, moved near family, and the kid is almost 8 months old.

In honor of returning to the Midwest, here are the rules I documented a few years ago for three-handed euchre:

The game starts with all 3 players at 20 points and is played over a sequence of rounds using a deck with 10,J,Q,K,A of every suit and the 9 of clubs. The goal of the game is to be the first to get to 0 points.

Each round is played as follows. The dealer deals 4 hands of 5 cards (1 to each player and 1 face down) and flips the remaining card face up. Starting left of the dealer and moving clockwise, players will decide if they want to pick it up and move the card into their hand and discard back down to 5 cards.

If a player chooses for the dealer to draw that card, that is now the trump suit for the hand. If no player calls it up, the card is flipped over and you go around the table in the same order picking one of the remaining 3 suits. The dealer will be the last to pick and is required to pick a suit (this is the screw the dealer variation in the original Euchre, that I've always played with in cutthroat).

Starting with the dealer and going clockwise, each player can decide if they want to , , or (if the dealer did not take the extra hand). If they stay, they will play with their current hand. If they drop, they will not take part in the round. If they take the other hand, they must play the current round with the hand that was face down. Note that they are playing with that hand sight unseen; they can't look ahead of time.

All remaining players will play the hand with standard Euchre rules (Jack of the chosen trump suit is highest, Jack of the opposite suit but same color is second highest, trump beats nontrump, high beats low, etc). Each trick taken is worth -1 point to your score. Taking 0 tricks is worth +5 points. If you called a suit (or told the dealer to pick the card up) you must take at least 3 tricks, or the hand is worth +5 points (and you get no points for the tricks you took).

3 players are playing in the first round; Ari dealt the hand and Chad had Ari pick the 10 of clubs up. Ari and Greg both stayed and ended up taking 1 trick each, Chad took 3 tricks. So Ari and Greg are both at 19 and Chad is at 17

1 player is playing in the first round. Ari and Chad both dropped and Greg was the only one who stayed in. Greg takes all 5 tricks and is at 15.

2 players are playing in the first round. Ari called a suit, Chad stayed in, and Greg dropped. Ari only took 2 tricks, and Chad took the remaining 3, so Ari is now at 25 and Chad is at 17.

There are 3 remaining special rules:

When the 9 of clubs is the face up card on the table, the bidding is skipped. Nobody draws the 9 of clubs, nobody can switch hands, clubs are immediately the trump suit, everybody must play, and everybody needs to take 1 trick to avoid getting 5 points (nobody is required to take 3 tricks)

9 of clubs is always the lowest trump card, no matter which suit is trump.

At the start of the hand, if you're at 5 points or less, you may no longer drop.

I've got a bunch of random ideas for projects in my notes. Here are a few of the more interesting ones.

Cookbookbook would be a "social" website for recipes. I know there are millions of cooking websites out there already (I even briefly had a food blog ~10 years ago).

But searching, collecting, and sharing recipes is an awful user experience today. Really, I want something web-based and text-focused like HN or Tildes but for recipes.

As a new dad, I've gotten to read a lot of really fun kids books. I thought it might be cool to try and make one which takes concepts from coding interview prep like leetcode or cracking the coding interview, and turns it into bite-size books that kids can still enjoy while parents learn something new! Not sure if this is an amazing idea or some sort of SF tech bro dystopian one.

It took my longer than I care to admit to remember what I meant by TypeWriterScript in my notes.

I think that the idea here was writing a language that could compile to TS that looks like this:

# Hello World

This is an example of a hello world in TypeWriterScript.

  console.log("Hello World!");
    

The idea here is that the documentation (either in plaintext or markdown maybe) and the code are one in the same, but written with the docs being any unindented lines. The indented lines would be valid TS.

Essentially, I could write a blog post that is perfectly valid TypeWriterScript.

Credit for the name on this one goes to my amazing wife.

This would be a mobile game that uses your daily step count to solve Sherlock Holmes mysteries. Your step count would be used to power Holmes traveling around London getting clues to solve the crimes (so you can play at home or wherever).

Anyways, that's it for ideas for now.

With the addition of the 🐦‍🔥 emoji in Unicode v15.1, it occurred to me that Unicode has a lot of birds.

This post contains a history of every bird added to the unicode spec. And to answer the age old question, which came first the 🐔 or the 🥚, the chicken was added in Unicode v6.0, but the egg wasn't added until v9.0.

• [1F413] 🐓 - Rooster (v6.0)
• [1F414] 🐔 - Chicken (v6.0)
• [1F423] 🐣 - Hatching chick (v6.0)
• [1F424] 🐤 - Baby chick (v6.0)
• [1F425] 🐥 - Front-facing baby chick (v6.0)
• [1F426] 🐦 - Bird (v6.0)
• [1F427] 🐧 - Penguin (v6.0)
• [1F54A] 🕊️ - Dove of peace (v7.0)
• [1F983] 🦃 - Turkey (v8.0)
• [1F985] 🦅 - Eagle (v9.0)
• [1F986] 🦆 - Duck (v9.0)
• [1F989] 🦉 - Owl (v9.0)
• [1F99A] 🦚 - Peacock (v11.0)
• [1F99C] 🦜 - Parrot (v11.0)
• [1F9A2] 🦢 - Swan (v11.0)
• [1F9A4] 🦤 - Dodo (v13.0)
• [1F9A9] 🦩 - Flamingo (v12.0)
• [1FABF] 🪿 - Goose (v15.0)

This list has customizations of bird emoji that can be done using the 200D Zero Width Joiner.

• [1F426 200D 2B1B] 🐦‍⬛ - Black bird (v15.0)
• [1F426 200D 1F525] 🐦‍🔥 - Phoenix (v15.1)

• [1F357] 🍗 - Poultry leg (v6.0)
• [1F595] 🖕 - Middle finger (v7.0)
• [1F3F8] 🏸 - Badminton (v8.0)
• [1F95A] 🥚 - Egg (v9.0)
• [1FAB6] 🪶 - Feather (v13.0)
• [1FAB9] 🪹 - Empty nest (v14.0)
• [1FABA] 🪺 - Nest with eggs (v14.0)
• [1FABD] 🪽 - Wing (v15.0)

• 🇦🇱 - [AL] Albania (Eagle)
• 🇦🇸 - [AS] American Samoa (Eagle)
• 🇧🇱 - [BL] St. Barthélemy (Pelican)
• 🇨🇽 - [CX] Christmas Island (Tropicbird)
• 🇩🇲 - [DM] Dominica (Parrot)
• 🇪🇨 - [EC] Ecuador (Condor)
• 🇪🇬 - [EG] Egypt (Eagle)
• 🇫🇯 - [FJ] Fiji (Dove)
• 🇬🇸 - [GS] South Georgia & South Sandwich Islands (Penguin)
• 🇬🇹 - [GT] Guatemala (Quetzal)
• 🇰🇮 - [KI] Kiribati (Frigatebird)
• 🇰🇿 - [KZ] Kazakhstan (Eagle)
• 🇲🇩 - [MD] Moldova (Eagle)
• 🇲🇪 - [ME] Montenegro (Eagle)
• 🇲🇽 - [MX] Mexico (Eagle)
• 🇵🇬 - [PG] Papua New Guinea (Bird of paradise)
• 🇷🇸 - [RS] Serbia (Eagle)
• 🇸🇭 - [SH] St. Helena (Plover)
• 🇸🇽 - [SX] Sint Maarten (Pelican)
• 🇹🇦 - [TA] Tristan da Cunha (Albatross)
• 🇺🇬 - [UG] Uganda (Crane)
• 🇻🇮 - [VI] U.S. Virgin Islands (Eagle)
• 🇿🇲 - [ZM] Zambia (Eagle)
• 🇿🇼 - [ZW] Zimbabwe (Bateleur)

I put together a minimal website over at cookbook.lol to start storing my recipes. This is the start of the cookbookbook idea from project ideas i.

It does a couple cool things:

• You can track when recipes are variations of other recipes (either on the site or from a blog or cookbook)
• It has some basic organization (tracking saved/tried recipes)
• It has support for basic comment threads.

It is definitely missing a lot of features and there’s tons more I want to do here, but the core functionality is there.

If you're reading this and you want an invite, just shoot me an email.

There is an existing nginx instance running for this blog which routes traffic to new local frontend and backend servers. In the future, we can split one or both of these out into their own instances.

The frontend serves all requests to cookbook.lol. These requests are served by a nodejs server that renders a bunch of closure (aka soy) templates.

The backend server is a minimal (mostly aip.dev style resource-oriented) API implemented in nodejs with postgresql for storage.

I ran into this by accident when playing around with styles for my site awhile back. Turns out this specific optical illusion is more pronounced for near-sighted individuals when wearing glasses. An oddly specific super-power.

Here's an example:

If you don't see the illusion, the red parts seemingly jump out of the screen, and the blue parts are sunken further back. If you search around on the internet, there are tons of cool examples that people have made running with this technique.

Some thoughts on daylight savings time.

A few weeks ago, my one year old and I dug through a storage bin in the basement and found my old lime green gameboy color, with batteries that were decades old and just a little juice left. Contained within, a cartridge of pokémon gold ready to play.

Starting a new game and listening to some 8-bit music, we pretty quickly got to the question: "Is it Daylight Saving Time now?"

I didn't know the answer back when I first loaded up the game as a kid and nearly 25 years later, honestly? I can still never remember which half of the year is DST.

So fun new thing as a dad of a 1yo: this is the first year where spring forward was the good one.

Usually fall back gives the extra hour of sleep, but the clock isn’t the deciding factor of when I wake up anymore. My kid is. So spring forward gave me the extra hour of sleep.

More or less.

Consider pitching the concept of DST today if it didn't already exist. Everyone once a year decides to move the clock one hour forward, and then six months later, we reverse it and move the clock one hour back. It's unimaginable.

Nobody would agree to it. So how did we get here?

Well in the US, it started in March of 1918 with the Standard Time Act which followed similar changes in Europe during WWI. The library of congress has some fun newspaper articles available online from the time.

In 2018, Prop 7 passed when I was in California, showing a positive sentiment towards getting rid of this twice-a-year nonsense. But, that prop didn't actually implement the change, just the ability for a change to happen in the future.

So is there any hope of change? Right now, with the Uniform Time Act of 1966, states can decide to have permanent standard time (an hour back), but cannot have permanent DST (an hour forward). The only states with year-round standard time today are Hawaii and Arizona.

In 2022, the Sunshine Protection Act proposed permanent DST at a federal level, but after passing in the Senate, it never made it to the House. Newer iterations of this bill have been suggested, but nothing has come of it yet. There is also the open question of whether year-round standard time or DST time is better, for some definition of better.

I don't have a specific story I can think of offhand, but I agree with xkcd/2867 so very much.

Leap second bugs, timezone bugs, dst bugs, browser time vs server time, on and on.

Any part of a software project dealing with datetimes makes me a little extra cautious.

I started coding in Kotlin for a project recently and have mostly enjoyed its philosophy and syntax. But I wanted to try and articulate my thoughts on extension functions because I'm not sure yet if I like them or hate them.

Other languages have supported extension methods for years, but most projects I've worked on have been in languages without support (or at least without idiomatic usage).

The super quick overview is that they let you add methods directly on classes that can be imported independently of the original code. Read the docs linked above for a better description.

I think I instinctively hated them, because coming from something like javascript where this would be done via modifying the prototype, you have no guarantees that you aren't clobbering some other code on the page. Or with something like dependency injection where your framework makes data available for you, but it isn't always easy to know where things are coming from without decent tooling or prior knowledge of the codebase.

But neither of these are concerns in Kotlin. Handling for extension functions occurs at compile time, so you don't have the same concerns around breaking other parts of the codebase. And extensions are imported so its easy to find where they source from.

I guess there is a worry that a poorly organized codebase could become even more cumbersome to work with if past authors scattered different parts of an object around incomprehensibly. But this seems like it would be easy enough to reconcile and reason through.

On the flip side, I like the syntactic sugar it brings and the simplicity of adding utility functions on top of core functionality that I don't own.

So I guess it's just the change aversion that comes with learning something new. They'll probably grow on me as I learn all the shiny new best practices that make use of them.

I started to play around with the Godot game engine a few days ago, since I have some ideas for games I'd like to make. Whether or not any of them come to fruition is still tbd, but I decided to give it a go to see what comes of it.

I searched a little bit about each of the major engines these days (godot, unity, unreal, etc) and for the sort of 2D games I'd be working on (and the positive feedback from developers that use it), Godot 4.2 with GDScript seemed like a pretty good fit to try out.

To learn the basics, I did the intro "dodge the creeps" game in their docs and then tasked myself with implementing solitaire. Not trivial, but not super complex either.

This isn't my first foray into game development. Twice before I've made a few simple games.

As a kid, I learned Visual Basic and used that for assorted projects. One was a game where a box would jump around the screen and you had to click it to progress.

Later in high school, I took an intro class that I vaguely recall had me make some game in Java to learn OOP, but I can't recall offhand anymore what it entailed.

So I got the solitaire game working.

And then I remembered, I don't really like solitaire.

I prefer freecell.

Most of the programming was pretty straightforward and provided a good opportunity to learn the Godot UI, GDScript syntax and all that.

There was an interesting issue I ran into around handling a click event that overlapped with multiple Area2D nodes (e.g. when dragging a card that was on top of another draggable card). I originally used the input event signals, but I had no way of figuring out which one should handle the event (no deterministic order I could figure out as to which node triggered first or to prevent propagation).

I eventually stumbled upon the intersect_point function I could use to get a list of everything that overlapped with the click and filter that by z-index to get the node I wanted.

I've added an rss feed to the site. Well, not actually. Everyone calls it an rss feed, but when digging into it, the recommended choice in 2024 for the fast-moving technical domain of metadata syndication is Atom, defined in rfc 4287.

It's auto-generated using the same scripts I use for generating this hilbert curve journal itself, so it should stay up to date with the latest 20 entries.

And at least at time of writing this entry, it is a valid Atom 1.0 feed.

If you're on desktop, you can now zoom in and out on the grid with the [ and ] keys.

I'd previously toyed with animations that would zoom in and out when navigating far distances on the grid, but never settled on one that I liked. This feature takes inspiration from that, letting you just zoom in and out as you please.

At 3 levels of zoom, the performance of the page is a little janky, but not so bad that I want to do further rendering improvements (above and beyond the switch from position offsets to css transforms in this commit).

The fifth ffxiv expansion, Dawntrail, releases in exactly three weeks on June 28 and I'm excited to continue the adventure. It will be interesting to see how the dynamics between the Scions plays out on opposite sides of the political struggle over in Tural and Wuk Lamat seems like a great addition to the cast.

If there isn't a battle on that big bridge between the northern and southern regions, I will be sorely disappointed.

I have one book left in The Murderbot Diaries series by Martha Wells after which I'll be caught up (until the next three come out). The titular character's relatable perspective on this dystopian universe has been entertaining to read, what with Murderbot's growth as a person, their relationships with ART, Ratthi, Mensah and the rest of their crew, and all the various adventures, missions, and whodunits.

It also looks like there will be a TV adaption of the books expected to release later this year, about which I'm fairly optimistic. Although with Murderbot processing so many things in parallel, it will be interesting to see how it translates to this medium.

I tried out the Ultracube mod a few weeks ago, but the weekly friday facts for Factorio really have me excitedly waiting for the next expansion that might be releasing later this year.

The fff-414 details earlier today about resources spoiling on Gleba have me cautiously optimistic, since it's a mechanic that penalizes my habit of frequently buffering materials on my bases.

The music in fff-406 is amazing. A lot of my spotify history is whatever game soundtrack I'm listening to in the background while coding (FFXIV, Frostpunk, Octopath Traveler, etc) and this looks to be a strong addition. (The rest of my history is my kid's playlists.)

I have such a huge amount of respect for the factorio devs. It is obvious how much love and polish they put into the game and I can't imagine this expansion will be any exception.

After writing the first three sections of this journal entry, I looked up out of curiosity if there had been any news on when Civ 7 was coming out, and lo and behold, it was accidentally announced earlier today! (It has since been formally announced for 2025.)

With previous Civ games releasing in 1991 (I), 1996 (II), 2001 (III), 2005 (IV), 2010 (V), 2016 (VI), this 9 year gap will have been the longest between major releases.

Last year I launched roze, a discord bot for tracking turns for multiplayer Civ 6 games. She is now a verified bot in over 100 servers, which is pretty exciting (and required me to do some extra data retention cleanups and introduce a terms of service). If they don't include built-in support for something similar in Civ 7, I'll try and get roze up and running pretty quickly after launch.

While digging through some files, I found my college transcript from — looks at calendar — 13 years ago and thought it might be fun to jot down some recollections.

I took a few AP tests in high school and got out of intro chemistry and calculus freshman year. It's still interesting that after a decade or so of school before university, a handful of transfer credits from AP tests are the only things that carry over.

I actually started off as a mechanical engineering major, hence taking ME 170. It was cool to play around with CAD for a semester, but ultimately not what I wanted to do.

My AP physics score wasn't high enough and I didn't pass the exam to get credit for multivariable calculus, so even though I took both PHYS 211 and MATH 241 in high school, I had to take them again freshman year.

After quickly deciding to switch to a CS major, I had to take CS 125 and CS 173 before the program would let me officially transfer in. Which is so much easier than today, since you're apparently no longer allowed to move into the CS program from within the university.

Learning data structures in CS 225 was one of the top two most important software dev classes I took. Useful for day to day coding fundamentals (and passing interviews).

CS 460 was pretty neat, we read articles like Smashing The Stack For Fun And Profit and then implemented them in the computer lab.

I remember putting a certain penny arcade comic in a presentation for my CS 210 ethics class. Got a good score on it, but the teacher wasn't fond of the cursing.

Hands down, my favorite class at university was CS 242. This class had a weekly programming assignment that you could solve in any number of ways, but you would have to present your solution to a small group of other students and a TA. This meant you saw the different ways that everyone in your group approached the same problem.

Some weeks, the assignment would build up off of the previous week's work (without you knowing ahead of time), so if you wrote clean code, you'd be better off than if you quickly hacked something together and left yourself with a bunch of techdebt.

Probably the closest class in college to actually having a software job.

Along with data structures, CS 473 is the other of the top two most important software dev classes. I still recommend Erickson's algorithms textbook as a canonical reference for learning or reviewing this topic.

When a student asked if dancing would be on the final exam, the answer was yes! We got a final where every question was about dancing.

During my last year of college, I had to finish up some gen eds and decided to start taking some extra math classes like MATH 417 for fun.

CS 598 with LaValle was a super interesting dive into relevant research papers (and the only grad-level course I've taken).

My senior thesis for CS 499 was A primer on covering spaces for robot exploration, which was essentially a variation of the art gallery problem.

Spacetraders is a "space-themed economic game with HTTP endpoints for automating gameplay and building custom tools." It's had me hooked during the evenings the last couple weeks in a "just one more turn" sort of way.

I kicked off with a minimal web frontend to explore the mechanics of the game. Specifically, I built three pages to get started:

A UI where I could see all the waypoints with markets and what goods they were importing and exporting.

A UI where I could see the status of my ships, have them mine an asteroid, refuel as needed, transfer cargo to other ships, or fly around the system.

An builder UI where I could hit arbitrary endpoints to fill the gaps for things that I hadn't added in my other pages (buying ships, handling contracts, etc).

Once I got the hang of the intro mechanics to the game, I started building some basic automation. This was done with a nodejs server that tells the fleet what to do.

Each ship is given a persona that assigns it an event loop which contains a sequence of actions to repeat indefinitely. I currently have five personas in use:

The persona (for my freighters) which has them fly to an asteroid, load from other ships until their inventory is full, and sell their cargo at a nearby market.

The persona (for my drones) which has them fly to an asteroid and mine whenever they have cargo space.

The persona (for my probes) which has them fly in a loop around the different markets and stores the prices of goods in a postgres db. This uses a naïve nearest neighbor algorithm for generating the route.

The persona (for my frigate) which has it find the most profitable good in the system, pick it up at one waypoint and sell it at another.

The persona for ships I have stationed at a particular waypoint (e.g. so I can buy ships at its shipyard, etc).

The game is an alpha, so the server resets every ~3 weeks. The last reset was on 2024-08-11 (I started playing a little before then) and my handle for this reset is

At time of writing, I have ~1.5 million credits which puts me 14th on the leaderboard for this reset (mind you there are very likely more players that have more credits than me if you include the cost of their ships (and 1st is currently at ~500 million, so there is that)).

My fleet consists of 21 ships: 1 frigate, 4 probes, 1 shuttle, 3 freighters, and 12 drones.

There are several directions I could go from here. Some ideas in no particular order:

• Add ship information to my DB and build a live visualization of where all my ships are at any given time.
• Add a queueing server between my API calls and the game server, to avoid hitting the dreaded "You have reached your API limit" errors.
• Explore more mechanics of the game (visiting other system, market dynamics, etc).
• Port my codebase to another language. I built this out quickly using JS, but it could be fun to try out Rust or Elixir or something well suited to this sort of project.