All of that made me think about ownership, or rather the lack thereof. We rent so many things these days. We rent our homes, our cars, our software, our music, our books and movies, hell, even our relationships. At the same time, physical paper books, indie bookstores, small makers of various shapes and forms, are seeing a resurgence. I don't know if ownership is a fundamentally human desire (I suspect it's not), but going against the current trend, pushing back against prevailing forces, definitely is. A growing desire to re-assert ownership is indicative of that.

Personally, I have always had a soft spot for physical books, and have thought of Kindle books as being rented and temporary. Owning movies (or TV shows) has never appealed to me. When I was a child, we rarely bought tapes or DVDs, and I could only watch things when they aired on TV. Ephemerality in visual media seems natural. I bought CDs of music for a while, still have most of them, I would like to do that again, perhaps starting with the Attack on Titan soundtrack. No, I've never been into vinyl, and can't imagine that I ever will be. I think software subscriptions are fine, especially if you get regular updates. I am conflicted on home ownership.

Over the years, I have come to experience and internalize that all things are temporary (for better and for worse). And yet, I like to have nice things, be grounded in relatively unchanging routines. Maybe this is a chance to embrace what the Buddhists mean by "non-duality"? Making peace with being in a sea of rentals and temporaries is perhaps accepting the world as it is. But perhaps carving out islands of ownership and relative permanence is nudging the world towards the way we would like it to be.

• Transport Fever 2 (Deluxe Edition and the Early Supporter Pack). I'm a sucker for city-builder style games. This seems fun and simpler than, say, Cities Skylines.
• DOOM Anthology. This includes DOOM, DOOM 2, DOOM 3, DOOM 64 as well as DOOM 2016 and DOOM Eternal that I already have. Is that enough DOOM in my life? Maybe, at least until DOOM: The Dark Ages comes out later this year.
• Quake Collection Bundle. This includes Quake 1-4, Quake Live and Quake Champions. We first got a computer at home in 2021, and that Christmas I got Quake 3 Arena. Sadly, I never got to play it because our computer didn't have enough RAM to share with the video card, and my dad wouldn't spend on a graphics card. Better late than never I suppose.
• Frostpunk 2. I played the original Frostpunk during the COVID lockdowns and really enjoyed it. From what I can tell, this is just as good, but different, so I'm looking forward to it.

Talking about better late than never, a few years ago for Christmas I got myself a Christmas tree and a Nintendo Switch. As a child I always wanted a Christmas tree and a Nintendo console, but never got them. There is something very healing and empowering about giving yourself the things you were once dependent on others for.

Blue Eye Samurai (particularly episode 7, Nothing Broken)

Blue Eye Samurai was a hit anime on Netflix last year. The whole show (8 episodes so far) is worth a watch, but one part in episode 7 stood out to me. At this point, the main character, the titular Blue Eye Samurai, is in a rough spot, where it seems like everything has gone wrong. The mentor figure/adoptive parent in the show gives this advice about being an artist:

To be an artist is to do one thing only. Look at me, I cannot fight, or weave, or farm. I make swords. I cook for strength to make good swords. I study the sutras to cleanse my heart to make good swords. Swords, pots, noodles, death. It is all the same to an artist. An artist gives all they have to the art, the whole. Your strengths and deficiencies, your loves and shames. Perhaps the people you collected.

There may be a demon in you, but there is more. If you do not invite the whole, the demon takes two chairs, and your art will suffer.

I only know how to make swords. Each morning I start a fire. And begin again.

Jerry Seinfeld, Ichiro Suzuki and the Pursuit of Mastery via Thorsten Ball's Joy and Curiosity #24

Jerry Seinfeld, without a doubt, is a master comedian. Seinfeld, in his 70s, and fabulously rich, continues to practice his craft, traveling regularly and performing as a live comedian. In this article, Trung Pham tracks down an old Esquire magazine article that Seinfeld has mentioned as influencing his views on mastery. The article pulls out key parts of the article, and places it in context with Seinfeld's career and practices, giving us a very grounded perspective on why mastery is important and a worthwhile pursuit, and how to go about achieving it (as well as how to avoid falling into some common pitfalls).

Exploring Kodawari

I recently learned about the concept of kodawari: the disciplined pursuit of perfection, while being attuned to the fact that perfection is impossible. For me, it is another step on the path of focusing on the process and the daily actions, rather than the destination or goal. Curiously, I learned about this while reading an article about the Sigma BF camera, a camera that seems to be pursuing the platonic ideal of a digital camera.

About a Nightingale, Day 8

About a Nightingale is one of Craig Mod's pop-up newsletters: a daily newsletter over a short period of a time that followed one of Mod's explorations through the less-well-trodden paths of Japan. These newsletters are members-only, so I don't want to point to the whole thing, but I wanted to share this excerpt from the final issue where Mod talks about why he prefers to do his walks alone over a number of days:

The point of creating patterns or rhythms of full days (and doing so in sequence — this is critical, the sequence part, the unbroken part, the sustained part) is to produce an “imprinting” of that fullness on the mind and in the muscles. And then to bring back that fullness into the play of “non-work” time. To bring that register of fullness to the people I love, my step-daughter, my parents, etc. So these are important days because they inform so much else in life!

And you have to fight to create this time. Life (now more than ever given current events and global chaos) conspires to steal from you days and hours and years to “work.” Conspires to steal from you the ability to hone that “fullness meter” in your heart or brain or wherever it lives.

Like artistry, mastery, and kodawari, fullness is another concept and experience that I am working to have more of in my life.

Last year, Maggie Appleton connected this variety of personalized software to the growing power of Large Language Models, suggesting that LLMs would make it easier for non-professional programmers to write small pieces of usable, working software for their friends, families and communities. The Good Place is proof-of-concept for Appleton's idea, since Mod has been leaning on Claude to get things up and working quickly. The software itself seems to be functioning with only minor issues, and the community itself has quickly grown into it. Withig a few days interesting links, personal projects, stories and experiences have been shared, difficulties have been commiserated over, some advice given, and even a couple of physical-world meetings arranged. When contrasted against global social media networks like Twitter, Threads, even Mastodon and Bluesky, the analogy between industrial farmed and processed junk food and a home-cooked meal writes itself (though The Good Place is perhaps more of a local potluck dinner).

For a long time now I have been skeptical of two things: first, the idea that "everyone should learn to code", and second, that LLMs will be widely useful in any significant way. I'm still skeptical of both those ideas in isolation, but The Good Place might have convinced me that together they make for a useful combination. The computing industry as a whole has not made much progress in lowering the bar for non-professional programmers to build usable software for themselves. I think it's still pretty hard for someone with absolutely no programming experience to do useful things, but with LLMs it seems possible for people with some experience to do things that are more advanced, and to do it quickly.

Finally, to stretch the cooking analogy a little bit more, I wonder what will be the equivalent of the local grocery store, or the farmer's market for home-cooked software. A large part of the difficulty in building usable software is that the individual components are quite tricky to plug together and keep running. But if home-cooked software movement does gather enough momentum, it may make sense for people to start offering services around hosting or packaging components for them. I can imagine something like Heroku or Fly.io, but even simpler, and maybe with LLM integrations that make it possible for communities to build and manage independent systems with little administrative overhead. Maybe something like this exists already?

Gina Trapani's Life in Weeks

A lovely little piece of digital art tracking major events in Gina Trapani's life, a week at a time. Personally I feel like I have a pretty bad memory for things that have happened in my life, and having something like this would be a good way to keep track of things that have kept track of things in my life. This version is based on a similar page by Buster Benson, which links to its source code, if you want to make a version for itself.

Why I Choose Lua for this Blog

Since Version 4 I've been using Lume to build this site, and though it's generally pretty great, I am occassionally concerned about the amount of TypeScript dependencies that I pull in every time Lume is updated. Something like Andre Garzia's homegrown, low-dependency blogging software has a certain appeal to it: knowing in more detail what you're depending, and that it won't change very often. In the distant future, if I ever do get around to building my own website generator, these are concerns that will be top of mind. Relatedly, as a programming language nerd, I find the appealing of a slowly developing, mostly language very appealing. I think that some languages (Java comes to mind) should be frozen once they get to some "good enough" point, rather than continually evolving and sprouting new features.

The Partially Dynamic Web

Continuing on the topic of rolling your own website software, Michael Winston Dales describes his motivation and solution for going from a purely-static Hugo-based site to a slightly more dynamic one. Dales' main motivation is to more efficiently handle large amounts of images. As I get more into photography, I'm starting to think about how I can start adding images to my journal entries, or have galleries of images, without too much manual effort, or slowing down the site's build process. This also makes me wonder how static site generators can better support "partially dynamic" features. Yet another thing to think about for my future-website-generator project.

Life in Code by Ellen Ullman

Part memoir, part snapshots of life in the software industry, and part social commentary, this book is a must-read for anyone who wants an inside look at the software industry from a unique perspective. Ullman's personal anecdotes are interesting and often entertaining, but her thoughts on how the internet will lead to the spread of misinformation and decline of communities are prescient by several decades.

In the process of writing this, I realized I don't like how the links and their descriptions look, so there will probably be some CSS hackery in my near future. I suspect that in the process, the limitations of Markdown will cause me more frustration and make me wonder once more if I should just write HTML5 directly. I also really want to have a non-chronological notes section (for documenting things like the aforementioned move to Lightroom), but I will inevitably overthink and over-engineer what that looks like. Webweaving can be such an addictive hobby, one thing always leads to a whole bunch more.

I'm calling the script wyd. Given a directory (or just the current one) it looks into each subdirectory and checks if it is a Git repository. If it is, and there have been commits made today (technically since midnight), wyd reports the commit messages, and if the commits haven't been pushed to a remote. (All my repositories have a remote they backup to.) At the end, wyd also lists the repositories that have uncommitted changes. At the end of the day (or really at any time), I can get an overview of everything I've done so far.

The script is available in my public scripts repo. Feel free to copy, fork or send any suggestions my way. I went to some pains to make it POSIX compliant, so it should run with common shells (namely bash and zsh).

On Shell Scripting

I initially wanted to write this in Rust, because I have been wanting to sit down and learn Rust for a while. But that seemed a bit overkill, and I wanted to see if I could get a shell script to do what I wanted without too much trouble. Though I've been writing and using small shell scripts forever, this is the first time that I really sat down and understood what I was writing, rather than just copying-and-pasting something that worked.

This process made me realize that for a long time, I've been afraid of writing serious shell scripts. The shell script language is definitely quirky. Depending on external programs for most of the heavy lifting means diving into documentation spread across different places. On top of that, being POSIX compliant borders on masochism. But once I did decide to dive in, I was pleasantly surprised by how quickly I could not only get something working, but understand what was happening (leaning heavily on Stack Overflow and the Unix Stack Exchange). Am I going to build large pieces of software in shell script? No, but I am much more comfortable reaching for it to write small utilities and automations.

Over the last few months or so, I have been making it a point to dive into new languages, technologies and problems. As a result, I have felt my abilities and confidence as a programmer grow steadily. Not only do I know a lot more, I also feel more comfortable with just diving into a new problem, domain or codebase and figuring things out. I hope to write more about this later. Happy scripting!

IndieWeb's 3-Body Problem

As far as I can tell, this is the piece that sparked off this round of discourse, written a couple of months ago. The premise is simple:

IndieWeb is a social club for developers, and apparently, not for me.

IndieWeb, a standard for the smart ones. Its member exclusivity works really well, as it prevents people like me from joining. It’s a coding test for the privileged. If you can solve it, like the Three Body Problem, you’re one of the cool kids—just like the Oxford Five. You’re deemed worthy to fight against the Big Tech, for us. It’s up to you, yes, you, to decide how the future of the web can be shaped. It must feel good to be an elite.

Smaller

Simone Silvestroni makes a succinct (and very good) point: the IndieWeb's various new technologies (like webmentions, microformats and even ActivityPub) are mostly unnecessary. Basic websites with RSS feeds and email are all the communication and dissemination technology that we need.

Has the IndieWeb Become Irrelevant?

Matthew Graybosch's post on the topic is probably the most comprehensive, includes links and excerpts to related posts, and Matthew's replies to some of the discussion that has been sparked. Matthew makes arguments that are technological (the complexity of ActivityPub, webmentions, etc. are too much for non-techie web people) and philosophical (it's not the website author's responsibility to support integration with other services). If you're going to read one post on this matter, this is the one to read.

Ownership, Connection and Control and POSSE is still cool

Jason offers a different perspective on the matter, and addresses some of Matthew's points in the previous article. Specifically when it comes to the idea of POSSE (Publish on your Own Site, Syndicate Everywhere), Jason notes that he wants to make it easy for his readers to find and read his material on a variety of different platforms, even if that means more work and technological effort on his part. This is something I agree with (for the time being at least), and is one of the motivations behind Project Bracer Phoenix.

Has the IndieWeb become Discourse Again?

Marty McGuire mentions Micro.blog, a service that lets you have all the goodness of the IndieWeb without having to engage in any technological wizardry (unless you want to). I personally find Micro.blog not quite to my liking (for various, mostly aesthetic reasons), but it is a good place to get started with your own site. I also heartily recommend Omg.lol for something more... quirky? Marty also makes the point that a lot of the technologies on the official IndieWeb wiki are recommendations and experiments, not hard and fast rules. This is a interesting counterpoint post, and if you do read it, I also recommend going back and reading Matthew's response.

There are a couple of other posts on the matter that are worth reading, you'll find them linked from the others above (yet another reason why links are awesome). Personally, I'm conflicted. While I do want to Syndicate Everywhere, I am put off by the technical complexity that engenders. I am also put off by the ad-hoc-ness of the various technologies involved, and that new ideas keep adding. I would love to have a cohesive set of technologies that act together (along with an easy-to-use implementation), but that seems to not exist. But one can dream...

None of this is new. Back in 2019, Anil Dash observed that the popular "link in bio" idiom for Instagram was a slow knife, simultaneously constraining our ability to link and lowering our expectations of web-based platforms. Robb Knight reminded me that services like Linktree only because platforms are link-averse. Commercial social media platforms have a vested interest in not supporting links, especially links to places outside the platform. As Adam Mosseri, head of Instagram puts it, the value of these platforms is "reach and engagement" (not connection, or information, or education), and you can't get your users to "engage" if they are following links to elsewhere. But it is especially annoying that platforms that want to be different and better are also falling into this same trap (looking at you, Mastodon). I personally consider Mastodon's lack of proper HTML support to be one of its original sins, and Bluesky seemingly wants to be different just for the sake of being different.

All of this has been swirling in my head for a while now, espcially after Jim Nielsen reminded me that links are powerful. But what prompted me to actually sit down and write this was discovering Eric Bailey's post "Websites are like bridges". This seems particularly apropos because the primary function of a bridge is to take you somewhere else, and a good website more often than not, takes you somewhere else (via, you guessed it, a link). Remember when we used to talk about "browsing" or "surfing" the web, verbs that imply swiftly and easily moving from one place to the next?

So, gripes and rage baiting aside, where does one go from here? Well, one of the benefits of coming back to my own website is that I can freely have as many links, with whatever text, wherever I like. One of my drivers for building a working writing and publishing system (for both short and medium form) was to be able to write with rich styling and lots of links. Project Bracer Phoenix is an effort to smooth over the deficiencies of the various platforms, and will hopefully be one of a number of tools I build to make writing and disseminating text-centric material easier (i.e., textcasting).

A glimmer of hope in this whole sordid situation is that everyone's free to write websites, and it's never been easier. The beauty of being on the web itself (as opposed to a siloed platform) is that one size doesn't have to fit all. You can be at whatever level of technical detail you choose, with options ranging from rolling your own HTML and running your own server, to using a static site generator, all the way to content-management systems like WordPress and Medium, and including quirkier options in the middle like omg.lol, Pika, or Blot. If you do decide to go down one of these routes, come join us at 32-bit Café, a new community for fellow web-heads.

Happy hacking and good blogging!

Berkeley Mono and Sono

I stumbled across these two lovely monospace typefaces a while ago. Sono is a "soft monospace" with a proportional variant. While it's clearly a monospace, it doesn't look as "techie" as such fonts often do, it works well as a body text typeface. By contrast Berkeley Mono (via BSAG) is a much more industrial-looking typeface that feels like at home in a software manual, or part of futuristic UI. I'm not sure I'd use it for programming, but I would definitely consider for a website, poster or book. Except one or both to show up here in the future.

The Monospace Web

This is a gorgeous, minimalist, design exploration by Oskar Wickström that I'm going to keep close at hand for future reference. If I ever create a purely technical website, or app, this is a style I'm going to come back to. I love how a considered use of monospace (and monochrome) can look both classic and modern at the same time. Not only did I see some fun design in action, I also learned about the existence of box-drawing characters.

Phosphor Icons

Not quite monospace, but keeping with the minimalist theme, Phosphor is set of over 9,000 icons with a consistent, clean design and a variety options for use (including with popular web frameworks). Personally, I've been using Feather icons and while they're lightweight and easy to use, the selection is pretty limited. I'm probably going to swap them out for Phosphor sooner rather than later.