Transforming a Vaio lamp from Craigslist into a modern lava lamp for my desk

Back in September of last year, on one of my irregular jaunts through the local Craigslist, I stumbled upon a listing for a “Sony Vaio Lamp.” As you might imagine given my retro computing proclivities, I was intrigued.

It took me a little while to actually go collect the thing, but bringing it home, it appears it was likely some sort of store display in its former life; it comprises a metal base and a square prism of polycarbonate which allows the light to pass through. Providing the light was a specialised fluorescent bulb, with the appropriate driver/ballast stowed in the base.

Well, this was cool, I thought, but it got me thinking - but what if it were colourful.

I happened to have been given a roll of LED tape by a friend, and that was the spark; I would love to make this an addressable LED lamp.

Cue several rounds of researching options, before ultimately deciding on the WLED project. They provide a firmware for ESP32 microcontroller boards most commonly used to run addressable RGB LED strips, with straightforward integration with Home Assistant.

The only catch is the LED tape I’d been given wasn’t the right sort, so I picked up a roll of WS2812 LED tape, a pre-assembled WLED module, and got plotting.

Photo of the Vaio lamp on a desk with an assortment of bits sat on top; a plastic bag for LED tape, the LED tape, and a controller for it.

With all the electrical bits together, I took a dowel maybe an inch in diameter, then cut, sanded, and painted it white for a uniform background. I gave it a test with just the polycarbonate piece and wrapped in LED tape temporarily with painters tape.

Photograph of the black polycarbonate piece with the Vaio logo in white, in a temporary setup without the base, showing the colourful lights of pink, blue, and aqua shining through the text. A pack of wall hangers is used to block light from escaping the top of the tube.

Thrilled with the result, got to assembly. My goal was to preserve the device’s exterior aesthetic, while adding a little functionality of my own. The device came with a grommet in the rear through which the AC power lead ran, and I planned to reuse it for power too, though with a screw thread type panel-mount USB-C socket instead. I wrapped the dowel permanently in the LED tape, trimmed the excess and put the LED controller in the base.

For interactivity, I could’ve just let it be a Wi-Fi device, but I also wanted to be able to turn it on and off physically, so I picked out the nicest of the buttons my local electrical component shop had to hand; this handsome and slightly industrial-looking brushed metal momentary button. I drilled holes for it and a circular panel-mount USB-C socket, and got it buttoned up.

Photograph of the Vaio lamp in pride of place on a light-coloured wooden desk; lit up in purple, pink and red.

At this point I was ready to let it sit in pride of place on my desk; I even hooked it up to Home Assistant and made it turn on when my computer was active. But I wasn’t quite finished yet; having gotten into woodworking I decided to build myself a lid for it, as it had never come with one. Presumably it might’ve come with some plastic or metal piece, but the seller hadn’t had it to hand, so I improvised; I took some of the plywood I had to hand, trimmed it, added a nice chamfer with my router, and a groove for the polycarbonate to seat in.

Photograph of the Vaio lamp showing it fully assembled with the plain wood lid. It's glowing various shades of blue.

I’m super pleased with how this turned out! I may paint the lid in the future to make it match the base, but for now I’m just happy to have it all buttoned up, and frankly I kind of dig the plain wood look it’s rocking.

Vancouver Game Garden 2026 Picks

I attended Vancouver Game Garden 2026 on Saturday, and I think they’re getting better at this! This year’s event prioritised more space for foot traffic over density of exhibitors and I think everyone sighed in relief at how much more chill and also… cooler it was in the Roundhouse. In any case, there were still plenty of exhibitors with new stuff to show, as well as some familiar faces.

So, in no particular order, here are the games that excited me at this year’s Game Garden!

Abyssal Merge by Time Flies Studios

Slated release: November 2026

The Lovecraftian horrorchild of Suika Game and a vertical shooter, Abyssal Merge is at its heart a fishing game; collect and merge fish in your net while horrors descend from the top of the screen.

Matching some fish in Abyssal Merge

Your boat can shoot upwards to start with, but can be upgraded in a multitude of ways. If the horrors reach your net they get in the way of matching your fish.

Initially it feels a little like rubbing your belly while patting your head but once you get the rhythm it’s hypnotic. A charming spin on the two formulae I can foresee sinking a lot of time into on my Steam Deck!

Caught horrors are being attacked, indicated yellow lines

Open playtest available now on Steam

Desktop Explorer by Recurring Dream

Slated release: 17 July 2026

I’m so glad we have a good name for these now; an Interface Mystery. It’s a computer from the 90s. You are exploring it to find out about what happened. It’s clearly been crafted with a lot of love and care for how computers were in the era. I didn’t get a chance to play the demo at the event, but I did hang about for a good while watching how things played out.

Browsing dw33bnation and exploring the tower in Desktop Explorer's... desktop!

Demo available now on Steam

SporeBloom by Antigen Interactive Inc.

Slated release: ???

SporeBloom is a fascinating little turn-based strategy game wherein you are a fungus trying to be the dominant mycelial network in your area. Think Civ but if you were a cute little mushroom instead of Caesar. You get different traits and abilities depending on what type of fungus you are.

A pair of mycelial networks vie for dominance on an abstract map

Honestly I think I’m doing a poor job of explaining it - suffice it to say I spent a while playing it (and constantly checking that was okay! haha) on the floor. It gripped me. One to watch, I think!

Demo available now on Steam

The Labyrinth At The End Of Time by Debonair B. Bear

Slated release: When it’s Done

Alex describes this game as having a “PS1 plus” aesthetic and I think that’s a good way to put it; not content to simply ape the limitations of graphics hardware of the time, Labyrinth instead feels more like an FMV cutscene of the era come to life. Plenty of polygon and texture detail, but presented in imitation 240p.

An orange-suited figure stands on a platform above electric zaps in a scene whose visuals are somewhat remeniscent of the opening train area of Final Fantasy 7

Fast and fluid, this is a platformer in the vein of Mario, but that wants to challenge players rather than hold their hands.

Wishlist now on Steam

Moon Drop Miners by chipMicro

Slated release: 2026

A large mining operation on a moon in the prime of operation

In Moon Drop Miners, you are charged with retrieving resources from a moon. To do so, you are provided with several classes of structure, each of which will contribute to your mission in different ways; some provide power, others consume it, some will mine a little bit, some a lot, but you have to be conscious of the order of operations, and make sure you don’t get too greedy and have the whole moon implode.

They caught a lot of eyes with their Commodore monitor showing off their mime-esque mask character on the side. Sadly the game won’t ship with CRT support :P maybe in a patch.

The mask character chides you for not meeting your mining quota

Demo available now on Steam

Cabs of Chaos by Macrocosmic Software

With how Sega have dishonoured the legacy of the beloved franchise, Cabs of Chaos may be the take on Crazy Taxi we need right now.

Okay, that’s out of the way, let’s talk about it on its own terms; this one is going more for a couch competitive chaos type vibe; you’re all competing for each fare and to be the fastest cabbie in town.

A split-screen view of two cabbies competing to drop off a fare at The Hardware Hut

The graphics and typography feel a little simplistic for 2026, but I hope that maybe with some more polish (maybe lean into the Simpsons Hit and Run of it all) alongside the gameplay it can stand on its own.

Wishlist now on Steam

Jumper by Extremely OK Games, Ltd.

Slated release: ???

Jumper is the next title from Extremely OK Games. It’s a Maddy Platformer, very clearly building on the foundations she and her team have made over the course of several releases, but the obvious comparison is Celeste. Maddy herself suggests it has more momentum and jump height over Celeste, and I would agree.

It’s familiar but new, at least to me, and I’m excited to see where it goes. Naturally when I played it there was already someone in a Games Done Quick hoodie trying to find the speed tech, and even in my clumsy hands the early levels already feel challenging in a nice way.

Have a feeling I’m going to be sacrificing another set of Joy-Con sticks to Maddy with this one 😅

So yeah, that’s the crop of games I saw that I was particularly excited about! Did you spot anything interesting I missed? Let me know!

New thing! How To Be More Creature Online

I had a wonderful time at Furality Ultra this past weekend - I hope you did too!

Following on from that, and a wonderful time spent chatting with some fellow weird furry and therian VR enthusiasts about our arcane setups, I was inspired to make a wiki for the purposes of such species-affirming tech, and it is now on the web!

Please visit it at howto.bemorecreature.online

SuperStation One is the PSone-shaped, ultra-flexible console I have been after

I’m not gonna blow smoke up your arse and tell you FPGAs are magical and superior, but I have been thoroughly enjoying my time with the MiSTer-based SuperStation One since I got it; having access to consistently-performing, distraction-free, HDMI-optimised (no frame drops between interlaced and progressive modes, with no expensive scaler!), TV cabinet-friendly PS1 and N64 (and way more) is great!

I grew up with the PS1, so NES emulation has never really been my jam1, but having instant access to the best (and, in some cases, worst :P) of the 32-bit era from the couch with comfy modern controllers, and access to a whole library of quirky retro ones? magic.

Back on our honeymoon in 2019 I picked up a NeGcon from Game Tanteidan: Retro TV Game Revival 『ゲーム探偵団 RETRO TV GAME REVIVAL』 in Nippombashi, Osaka. I used it a few times, but only having a PS22 at home my options for exploring it were limited, and sadly in my experience it doesn’t play well on PC emulators unless you have very particular adapters, but with the SuperStation I am finally able to pass it through to use it how it was meant to be enjoyed!

Yes, it’s a quirky, very unusual controller from an era where we didn’t quite know what to do with the concept of analogue input yet, but I think it’s charming and honestly underrated despite largely being the forgotten underpinnings of the PlayStation’s support for racing wheel controllers. And for something so full of plastic mechanisms it’s still working fantastically 30 or so years on! Namco cooked on this one, it’s really quite remarkable.

There’s a core memory of mine of playing Crash Bandicoot on a family friend’s gigantic rear-projection TV, and the sight of the titular bandicoot wiping out on a turtle and spinning in a black void on that big, slightly-fuzzy TV has never left me.

I am delighted to say that the MiSTer core with a nice scanline effect and in HDR (to compensate for the brightness loss from adding scanlines) on our lovely LG OLED TV is about as close as I have gotten without making a regrettable purchase of an3 enormous and difficult to maintain relic of times past lmao.

For a long time I’ve been a staunch adherent to integer-scaled, hard-edges for retro games on modern displays, but this is the first time the softer side has truly appealed to me.

If there’s one major critique I have of the MiSTer experience it’s the user interface. It’s not especially intuitive to navigate for someone more used to PC emulator frontends. Rather than pick a game and launch it, it instead asks you to select a “core” first, and then pick a game within that core. It’s fine, but it leaves a lot to be desired. You can create “shortcuts” which point at a combo of core and ROM, but it’s rather clunky and manual effort. I remain hopeful that the “Console Mode” interface that Retro Remake have been working on will be a big quality of life improvement there; adding box art, metadata and familiar categorisation would be really lovely here.

Got any fun or odd or interesting retro games or MiSTer tips you’d like to share? Hit me up! Or go check out the PAL and Japanese region-exclusive PS1 title “Fluid,” (a.k.a. “Depth” in Japan, (a.k.a. “Sub” in France 😅)) it’s really something!

  1. I actually did, for a while, have an NES at home, but I didn’t know what it was because I had only known the NES to be the big VCR-shaped thing Americans are so familiar with; mine was the later, top-loader revision with the “dog-bone” controllers; despite the “Nintendo Entertainment System” emblazoned on the top case I never clicked at the time that they were the same thing under the hood. This of course also means that unlike most American readers, I got the European-flavoured Super Mario Bros. / Tetris / Nintendo World Cup multi-cart rather than the one with Duck Hunt, so sorry if I never developed a taste for seeing that smug mutt. Anyway, to this day I have little time for anything less sophisticated than a Super Nintendo. What can ya do, eh? ↩

  2. To clarify; the PS2 (SCPH-50002 to be specific, a later “fat” model in the common parlance) can of course play PS1 games, and use the NeGcon. The issue is more that unless you have an original disc or mod chip, even with a soft-modded PS2 you can’t really play PS1 game backups reliably. The software PS1 emulator that Sony shipped back in the day is fine for the games it was intended for, but a bit of a janky mess for anything even remotely obscure. Yes there’s the PS1 hardware and the swap trick for using burned discs, but that’s much easier on the slims, I don’t have ready access to a PAL swap disc, and I don’t really want to break the tray of my childhood PS2, or spend a bunch of CD-Rs on the task! ↩

  3. Well, another; we already have a gigantic CRT television I need to find a good space to use more… lol; p.s. if you know anyone in the Vancouver, BC area who can do a CRT recap and tune-up housecall… hit me up ↩

More links that moved me lately

Hello and welcome to a second experiment in linkblogging; like the first, this one is also a roundup. Life has been a bit difficult of late (some self-inflicted, some not), so I’m working to be kind to myself. I hope you are too.

A Trip to 90s Kansai: Exploring the XD FirstClass Network BBS by Misty De Méo

I think this might be one of Misty’s best blog posts ever, and I’m only slightly biased - this one is a delve into a rare snapshot of a moment in time of BBS culture from the early 90s in Japan. Go find the easter egg I suggested! :)

New Technology: Whose Progress? (1981) by Education Media, with the assistance of British Film Institute

I’ve been reflecting on this one a lot. I happened to catch it on The Old Timey Computer Show a while back. Technology has always been built at the whim of the rich and powerful and the rest of us don’t tend to get a say. We should fix that.

Responsibility is the Human Moat by James Bach

A much better-worded version of a concept that’s been bobbing around in my head for some time. My drafts are full of versions of this that I’ve left on the cutting room floor. Anyway, from James’ post:

AI cannot behave responsibly. Only natural persons can.

In a vital sense, humans have a “moat” about them, with respect to human society, that no AI can cross. AI can pretend to be a part of society, it can interact with human society, but it cannot actually be a part of society.

Sounds oddly familiar! James goes on to list real, tangible problems of this technology, which it seems we are being asked to ignore in favour of the “what if” of it becoming better than it is now.

I strongly believe that LLMs in their current and anticipated forms represent a major workplace and social hazard that we are not treating with anywhere near the gravity it deserves. At this point I can only really hope we’re at a turning point.

i’m not mad, i’m just disappointed by ian walker

An excellent article discussing what rubs the author (and, frankly myself) finds troubling about the framing of indie games journalism darling aftermath’s coverage of the high-profile boycott, divest, and sanction targetted game release from Microsoft’s troubled Xbox division.

Microsoft are actively abetting what experts are calling a genocide, and I think we need to have a real discussion about what that means for consumption of their products, particularly at a time when fascism and its outcomes are so plain to see in the world.

I gave aftermath a yearly subscription after reading Jackson Ryan’s story on the creep of gambling into games media and related websites; some actual, solid journalism. I have withdrawn said subscription because frankly I think they should want to do better in the world than to launder their feelings about a tired game series that doesn’t need any advertising.

The Clowns and Criminals of Titanium Court by Patrick Klepek

A wonderful exploration of one of the most engaging games I’ve played in recent memory; Titanium Court is an absolute masterpiece, and I think everyone should play it. I don’t think I’ve ever felt quite so in communication with a game designer. Go play it, then read this.

also available wherever you get podcasts

Do I belong in tech anymore? On quitting, the spread of AI, and the loss of an ideal. by Ky Decker

It wasn’t a bad job, not by most metrics. It ticked the boxes a job is supposed to tick: good pay. Health insurance. Remote work. Time off. Nice coworkers. […] I was doing good work. And yet, work was rendering me increasingly miserable. I questioned myself. Why am I here? Does any of this work actually matter? And if I stop caring about the quality of my work… will anyone notice? (An uncomfortable thought.)

Honourable Mentions

Cool Tools

  • Hister
    What if your browsing history was more useful? Hister is a self-hosted service and a companion browser extension that indexes and allows full-text search of pages you’ve visited. Like if Windows Recall was built humanely. I’ve had some trouble with the browser extension’s reliability, but it has a lot of potential - I’ve wanted something like this forever.
  • eightyeightthirtyone

    crawls the links between 88x31s on the Internet, which are small badges on websites that link to other websites

This is part of me approaching blogging with an eye to making the ephemeral stories I read feel more a part of an ongoing record. Written, as always, with my own two paws. Please feel free to contact me with any feedback or comments!

A smattering of links

Hello and welcome to my first experiment in linkblogging; I’m still deciding on a format for these, but this first one is going to be a bit of a roundup.

To those of you who are monitoring my feeds or otherwise reading my site, thank you for the feedback on linkblogging! I am pleased that several of you reached out and I’m flattered to even have readers, let alone readers who will respond to questions! 💜

How to Ban Conversion Therapy by Philosophy Tube

Abi delivers a scathing takedown of the trans medical system in the UK in particular, and how it wrests control from patients in the name of continuing to hold power over them. She further provides a pathway to ending these practices decisively. None of us are free until all of us are.

also on YouTube

Why craft-lovers are losing their craft by Hong Minhee

The tension between craft and efficiency doesn’t disappear if you remove capitalism from the picture. LLM coding assistants produce faster results whether anyone is being paid or not, and any community, however it’s organized, will eventually have to reckon with what to do with that speed difference. Capitalism gives the harshest possible answer to that question: the slower worker loses their livelihood. But the question itself would survive capitalism. Other forms of social organization might answer it more gently, but they’d still have to answer it.

I’m not sure I fully agree with the assertion that LLM use is inevitable even free of capitalism’s influence, but the story here resonates with me. The tensions and contradictions of worker power in the tech industry continue to escalate.

I also believe that the speed increase is an illusion; there are many reasons to distrust the output which turn the accelerated speed into a slowdown. I think there’s a blog post in this from me to come; stay tuned.

h/t to James Healy for sharing

Voyager - Across the Unknown Explores Darker, Colder Space by Dia Lacina

[…] on my 9th run of Voyager: Across the Unknown […] it’s finally settling in just how ambitious this game […] is. […] It’s safe to say this narrative survival adaptation of a major television property is quite the scope-up in terms of project size and complexity. Voyager: Across the Unknown is ambitious. It’s not quite the ambition equivalent of launching a new network with a woman at the helm of a non-Enterprise Star Trek series in the mid-late 90s, but it comes close.

Great piece about one of the most interesting games I’ve played recently. If you haven’t tried it, I highly recommend it. It’s both a loving ode to the series and a wonderful game exploring all the negative space around the canonical narraive of Voyager.

Ageless Linux & the Ageless Device

Every device carries a printed label:

AGELESS LINUX
General Purpose Computing Device
OS: Debian Linux (RISC-V) · Provider: FFwF Robotics LLC
AB 1043 STATUS: NONCOMPLIANT
This device does not collect, store, or transmit the age of its user.
This is intentional. · § 1798.500(g) · § 1798.501(a)(1-3)

A thought-provoking form of protest of the wave of age-verification laws, and subsequent debate within the Linux community over compliance. Obviously much of Linux is contributed to by corporate interests, so I cannot truly begrudge the efforts to comply even if I disagree with the laws that are pushing for this. Anyway I want one. Or maybe fifty.

via jwz.org

Age Verification Lobbying: Dark Money, Model Legislation & Institutional Capture

Meta spent a record $26.3 million on federal lobbying in 2025, deployed 86+ lobbyists across 45 states, and covertly funded a group called the Digital Childhood Alliance (DCA) to advocate for the App Store Accountability Act (ASAA). But the operation extends beyond Meta.

Meta, a company who stands poised to have a strategic advantage over competitors due to their platforms already collecting age information, have been lobbying to the tune of tens of millions of US dollars, to make age verification a requirement.

That they are doing this on the sly is really rather telling. Push back against these measures in your jurisdiction.

China to help Cuba with solar energy amid US oil blockade and total power outage

China is offering solar energy to Cuba, an old ally left without power amid the US oil embargo, in a show of strength for Chinese renewable-energy credentials and a sign of Beijing’s economic footprint in Latin America.

An unequivocally good move here. Cuba is being strangled by the cronies in charge of the US. Frankly, Canada should be making similar moves in solidarity with other targeted nations in our region. No more wars in our names. Fossil fuels are an international security problem.

The very serious part-time unofficial hobby of videogame fan-translation by Hilltop

In my recent romhacking video I used a thumbnail with the simplest, most direct translation I could find in the script, specifically so that nobody could complain. The line was:

友枝小学校の４年生

I’m a 4th-grader at Tomoeda Elementary.

[…] someone showed up in the replies and tried to argue that the order of the words should be 100% preserved through translation and that it should be the perfectly SOVLfully accurate:

At Tomoeda Elementary I am a 4th grader.

Amazing. No notes. So yeah, I don’t talk about localization.

A great writeup about the thankless job of video game translation, and the ways in which people have pushed against the very things that make a good translation, and the infection of the community with latter-day LLM-based machine translation.

I’m still a bit heartbroken over Segagaga.

Wikipedia has prohibited use of LLMs to generate or rewrite article content

Text generated by large language models (LLMs) often violates several of Wikipedia’s core content policies. For this reason, the use of LLMs to generate or rewrite article content is prohibited, save for the exceptions given below.

A 44-2 decision is a strong signal. Shoutout to Wikipedia editors for fighting the good fight here!

This is the beginning of me approaching blogging with an eye to making the ephemeral stories I read feel more a part of an ongoing record. Please feel free to contact me with any feedback or comments.


You there! Yes, you with the RSS reader - I need your opinion!

Hello, dear reader. I am writing to you rather directly because I am thinking about how I approach this blog and I would like to include you in the decision.

I have on occasion wondered if I should consider adding “linkblog” style posts to this site. With that, I could publish smaller posts which point you towards (and/or comment on) other interesting things I’ve read online.

You can see this on many other sites around the web. I figure it’s a good opportunity to highlight interesting and important things in a way that’s perhaps more permanent and accessible than a boost on social media.

So, as a reader of my blog, let me know your thoughts! Thank you <3

A machine cannot feel embarrassment, therefore it should not be trusted with a critical decision

I’ve been trying recently to articulate the reasons why I don’t think LLMs are a trustworthy technology. It’s a shame that the machine learning profession has largely had its reputation stolen and twisted by LLMs.

Machine learning is great! Machine learning can help find the good photos in your sprawling photo library. It can be used to find photos of your family. It can react to patterns humans struggle to, and produce useful outputs based on that.

But LLMs are a different beast altogether. As commonly used in chatbot situations they break one of the fundamental rules of our own cognition; for all of human history, the use of language has been enough to differentiate a real person from an automaton emulating one; you have been able to trust that if you can have a conversation you are speaking with an intelligent being with interiority, thoughts, feelings and emotions.

LLMs can express all those things, because those things exist in its training data. Every apologetic character from fiction. Every Reddit thread of someone expressing their frustration. It’s all present, but it’s disconnected. It doesn’t connect to an actual feeling, thought, or intention.

An LLM cannot anticipate being wrong. It can express but not actually feel remorse, embarrassment, hope, or fear. And that is truly the thing that separates us. And that, I feel, is the thing that makes them dangerous for our brains.

In a professional context, say if I am asking someone for code review, there is a degree of confidence that their thoughts must pass before they’ll express them. If they’re unsure, they can ask for someone else’s opinion. An LLM has a confidence score for its impulses, but it cannot feel concern for being wrong, for wasting someone’s time, or seek input from a human to confirm or dispel them.

While it is very impressive that an LLM can take a question and formulate a reasonably coherent, convincing answer, the fact that it cannot want to do a good job, and in fact is trained to be confident above all else, is why I don’t believe this technology is going to stick around for much other than producing enormous quantities of garbage to trawl for ad revenue.

I am open to the idea that machine learning may get us there one day, but to me LLMs are clearly not the technology that could get us there.

I think we're going to need to seize the means of computing, folks

In the 1970s the Personal Computer was heralded as a great revolution, bringing the power of computing to the masses. Democratising it. Taking it from something you had to be affiliated with an institution to enjoy to something many people could afford.

Halfway through the 2020s the Personal Computer appears to be a threatened species.

Decades of complacency have enabled corporate giants to own vast swathes of the computing market, and increasingly enclose their users with more and more restrictive computing platforms, and now with the advent of the AI bubble’s reshaping of the computing market, they appear to be preparing for a killing blow.

Gone are the days where a computer magazine would act as a hub for fellow readers’ shared source code, coach readers on how to approach computing problems and even programming. I fear we are in for a decade of the corporate interests pivoting us into owning nothing and enjoying only slop.

As a long-time Mac user it pains me to say it, but I think we’re gonna have to seize the means of computing. We must be free to use our bicycles for the mind lest it be taken from us. macOS has been slowly degrading for years now. Windows is way ahead of the curve on this. But from this position it feels inevitable.

You should be able to control your devices. You should be able to repair them. You should be able to upgrade them. You should be able to keep them going longer than the manufacturer would like to keep their profit margins up. You shouldn’t have to rent computing power from a datacentre which is poisoning your fellow human and the environment we share. And in the year where solar is starting to beat coal for the first time, we shouldn’t be investing in more dirty power to do it.

Hack your devices. Resist corporate control. Delete corporate bloatware. Run community-maintained software. Host your own services. And teach others to do the same. Keep using that “old” RAM they already sold you. It’s not broken, corporate software is just bloated.

I fear it’s our only path through this bubble.

Breaking up with iTunes Match

I’ve been using iTunes Match almost continuously for my music since September 2012. That’s a scarily long time. I’d been using iTunes even longer.

In 2026, I’m finally ready to move on.

iTunes Match, for those unfamiliar, is effectively a predecessor to Apple Music. Rather than a subscription to an entire catalogue of music, it was a subscription to stream your music collection. Sounds a little silly, but it did offer some tangible benefits;

  1. You don’t need to keep your entire library synced to your iPhone1
  2. Matched tracks could be “upgraded” to iTunes Plus quality if you had, say, an inferior 128kbps MP3 copy
  3. Easy playback across devices, streamed from the internet; your music follows you anywhere
  4. You could always download songs anyway; if you were going on a flight you could grab a few albums you planned to listen to and they’d be there for you

Ultimately, though, iTunes Match was a bit of a dead end, and perhaps Steve Jobs was the last Apple personality to truly care about it. Eventually Apple Music subsumed its features, but the legacy subscription stuck around, largely unmaintained, ever since.2

For a few years now we have kept a NAS at home. It’s great! It stores a bunch of our personal files, backups, and other precious data, like our movie and TV show collection. We have a Plex server running off it. Plex led me to think about my use of iTunes Match more, as it offered Plexamp, a quite decent little music player app for Plex.

Plexamp, however, just left me wanting more. It had issues on certain platforms, some playback and stability issues, and seemed to be written by people with some degree of contempt for anyone who thought any differently to them about how things should behave.

That led me down the path of trying out Navidrome. Navidrome is the Jellyfin of music; it’s open source, pretty quirky, and the default user interface is pretty rubbish. But it opens up one particularly intriguing option; it has an entire ecosystem of apps which are compatible with it.

With a Navidrome server in place, I am able to pick and choose player apps as I please, and the collection, playback stats, and scrobbling are all centrally-managed. Nice!

For Desktop, I settled on Feishin, a stylish web and desktop app (yeah, sorry, it’s Electron) which I find fairly intuitive and usable. I’ve run into a few quirks, particularly due to my music library being largely AAC and ALAC from years of iTunes use, but I have confidence they’ll be sorted out. The developer is quite responsive.3

On my iPhone, I like Arpeggi, which is sadly currently only available as a TestFlight beta, but is one of the nicest music apps I’ve ever used. Simple, modern, and reliable. I will gladly pay for this once the author is ready to publish it for real.

After a few months of running this setup, I was finally convinced; I didn’t need iTunes Match anymore. On the 2nd of February, it expired. And I’m pleased to say that I barely noticed!

Well, today, dear reader, I reached another milestone in this journey. I added my first new music without touching the Apple Music (née iTunes) app. FLACs put through Musicbrainz Picard for tagging and then transferred to the NAS. Navidrome immediately picked up the new albums, and they appeared like clockwork in Feishin and Arpeggi.

And it’s pretty great.

Do you have a favourite Navidrome client? Any fun ways you’re using third-party tools? Drop me a line 😄

  1. Flash storage was a much more precious commodity at the time. Yeah, even compared to the bubble we’re in now 😓 ↩

  2. It even kept the Lucida Grande era branding and an ancient iTunes screenshot on its emails until at least 2024! ↩

  3. Feishin is also pretty great on Steam Deck if you trick it into running full-screen lol ↩

You should listen to the Three Cheers for Sweet Revenge remaster

I usually think of album remasters as cynical cash grabs but I was alerted to the quality of the 2025 remaster of My Chemical Romance’s Three Cheers for Sweet Revenge the other day1 and I’ve been obsessed ever since.

The cover to the 2025 release of Three Cheers for Sweet Revenge, depicting a set of rosary beads on a black background with small red text

Maybe it was what the band were going for at the time but the original mix always felt pretty busy and claustrophobic, hard to distinguish anything within the wall of sound; the 2025 mix is punchier, cleaner, and better distinguished.

I can hear vocals I didn’t even know were present in the original mix, it’s great! Honestly give it a go if you’re a fan. The live set at the BBC included as bonus tracks is also a fun time.

Embracing woodworking to soothe my tech brain

I have worked in tech for, goodness, sixteen years now. And if you do, or have, worked in tech, you probably know I have a problem1.

My brain is absolutely dead at the end of the day.

Now, I consider myself quite good at my job, but something I’ve learned is that being good at your job only makes people ask more of you.

But, that’s part of the problem. Now, I don’t have any clever solutions to that (other than being defensive of your time and kind to yourself) but I have found something that I think is helping me achieve balance, and it is woodworking!

Woodworking feels like the opposite of tech. Solutions are concrete, physical, measured in millimetres and minutes. You’re thinking not in abstract terms, but in real concrete terms. It also helps that I am working for myself.

In working for myself, I am learning to accept mistakes. I am learning to embrace the space between periods of intense focus. And I am learning to use my hands to make something tangible.

Slowly, and with an assortment of other life changes I have been making (not to mention therapy), I feel I am slowly fixing my addled tech worker brain.

I am really only getting started, but I’ve built a couple of little projects out of plywood, some power tools and some gumption and planning out on paper. It’s great. I feel so accomplished every time.

So if you feel similarly, I encourage you to find a way to build something physical. Find a problem in your life or just something that could be a little better and explore a solution. Build something for yourself and your loved ones, which will outlast the tech hype cycle.

And, frankly, I think this is making me better at work too.

  1. I hope it goes without saying that this post is my personal opinion and not that of my employer. ↩

J2ME development on modern Linux

For some reason I have decided I want to build a J2ME app in 2026. Don’t live like me.

Anyway, thankfully other people have already blazed this trail.

Specifically, this post is to provide an addendum to this post by the author of Microgram, a project to run Telegram on your old phone.

tl;dr:

  1. The author was using Ubuntu 24.04 LTS (or a derivative)
  2. To enable i386 packages, run sudo dpkg --add-architecture i386; sudo apt update
  3. If you are not using an Ubuntu desktop installation you MAY need to run the JDK installer (and NetBeans and other installed components) with _JAVA_AWT_WM_NONREPARENTING=1 prepended
  4. Reboot before you run the MIDLet emulator to save yourself some panic
  5. If you are using a different distro and/or distrobox, read on for relevant info

in more detail

So, what I’ve found as someone who is using a Fedora Kinoite installation in 2026, and therefore doing this via distrobox, is that there are some specifics which are difficult to ascertain from the above-linked post alone.

First of all, their distro. In a previous post they mentioned using Linux Mint 20.3. In the latter post, they don’t clarify. My first instinct was to use a similar-vintage Ubuntu distrobox, and 22.04 was a fairly close guess, with the caveat that libxt6t64 was missing; it appears it was named libxt6 in 22.04. That perhaps should’ve been a clue, but I didn’t learn that until later!

Also worthy of note that the incantation to add i386 package support on an x86_64 Ubuntu or derivative is to run sudo dpkg --add-architecture i386 and then run sudo apt update to fetch sources.

Anyway, I proceeded with the instructions under the distrobox on Fedora, and once I got to the point of trying to install the JDK, I ran into a bit of a brick wall. The installer’s GUI was blank. I am going to skip the full rundown but I installed a separate Ubuntu desktop VM to test which side of the Distrobox the issue was on, and it worked fine. Okay, so it’s something to do with my machine. Well!

It turns out that this is a well-known oldschool Java bug and was documented on the Arch wiki; Ubuntu’s default GNOME compositor for Wayland seemingly has a workaround for this implemented by default, but Fedora Kinoite’s KDE compositor for Wayland does not, so I ran the installer with the _JAVA_AWT_WM_NONREPARENTING=1 variable set, and it worked fine. Phew!

Now I know in theory distrobox has tools for exporting programs to the host system but the JDK installer had already created a shortcut to NetBeans on my desktop for me, so I rolled with it, updating it to run distrobox instead. I also added another Java workaround from the Arch wiki, which I don’t know if I specifically needed but it didn’t hurt. What I ended up with is as follows;

Name: NetBeans IDE 8.2 (Distrobox)
Program: /usr/local/bin/distrobox
Arguments: enter j2me --additional-flags '--env _JAVA_AWT_WM_NONREPARENTING=1 --env JAVA_OPTS=-Dsun.awt.disablegrab=true' -- $HOME/netbeans-8.2/bin/netbeans

I updated both the desktop shortcut and the launcher shortcut to match. I then rebooted. I am not sure why I needed to; perhaps my machine or my installation were haunted, but I could not get the emulator to launch a MIDLet without that step.

Then, magically, mercifully, it worked on my host machine running Fedora Kinoite 43. Ultimately having dug further I realised that the microgram author must been using Ubuntu 24.04 or equivalent, and their instructions would’ve worked fine there, but I only had the date and a hunch to go by!

So yes, running NetBeans circa 2016 and an SDK from 2008 on Linux circa 2026. I hope this mess of keywords saves someone some time.

Sorry for the lack of posts lately, things have been… weird.


88x31 button updated!

Well… Aura gifted me an animated 88x31 button right after I published my first. The old link will continue to work, but there’s a new animated version now!

drac.at on the web

Now available from my links page!


I have an 88x31 button now!

I’ve been collecting 88x31 buttons for a while but I never had one of my own for this site.

Which, well, isn’t actually true, I did make one back in September 2024, I just neglected to actually share it. 😅

…until now!

drac.at on the web

I think it’s pretty cute and nice and if you want you can put it on your website to link to me!


Feed Buttons

BTW does your site have RSS and/or Atom feeds? Modern web browsers not having feed buttons getting you down? Need a good way to advertise as such?

I pixelled some basic little buttons (in 1x and 2x density) like how I remember them being back in the day!

RSS

Atom

These can be used, with support for high-DPI displays, like this:

<img src="/img/rss.gif" srcset="/img/rss@2x.png 2x" alt="RSS" width="32" height="11" decoding="async" loading="lazy" />
<img src="/img/atom.gif" srcset="/img/atom@2x.png 2x" alt="Atom" width="38" height="11" decoding="async" loading="lazy" />

The 1x images are gifs, because gifs manage to be smaller than pngs at that size, while at 2x density, png wins. SVGs are provided in case you want to modify these further, but are much larger than the raster images. Text is based on the free Visitor font, a real classic.

Individual files:

You may use these however you wish. Those who care may consider these licensed under the Creative Commons Zero license.


you just have to do the thing

I keep finding myself struggling with the same problem, which is that I find it difficult to start something if I’m not sure I can finish it

maybe it sounds like a #humblebrag but it totally isn’t, it’s a massive impediment to getting better at stuff by starting out not being so good at it, or to accomplishing things that I can’t confidently polish off in an evening

so, yeah, with this refresh of my site, and this blog, I’m trying to just do the next bit, whatever it is, and accept that as progress even if it feels imperfect. I don’t have to have suddenly launched a next-generation, kickass version of this in the handful of weeks since cohost’s demise was telegraphed, I have done enough, and I will continue to do enough

relatedly, this post lives in the cohost post template not because it was posted to cohost, sadly that is no longer an option, but because it’s the only post style the site currently has other than the one for the art gallery

and that’s okay, because I can write here, and you can read it, and that is enough. I just have to keep being okay with taking the next step


cohost weaned me off Twitter

and if nothing else, that is a massive success. Twitter was part of my life for more than a decade, I met my wife on it for goodness’ sake. sadly, by 2021 it felt like the worst place to be on the internet. where you go to get in arguments with strangers for no reason

do I regret leaving Twitter once cohost became home, and once a certain billionaire took it over? in spite of the artists and acquaintances I’ve lost touch with, no, it’s done a lot for my personal health

so, for the short time it has been around, cohost felt like the place for me. assc’s manifesto still resonates in spite of their ultimate lack of success with it

the lack of numbers convinced me to turn off notifications for other platforms, and reduce the noise I have incurred by life with the ambient internet, I have no evidence but I would swear to you that my average resting heart rate is lower as a consequence

the comment structure made it feel like a place you could have conversations in context, with people, instead of shouting them out loud in a room filled with noise

frankly I always had anxiety around whether I was “mutuals” with people on Twitter, and on here? that wasn’t a thing, what a freeing experience

for all this and more I will be forever grateful to @staff for giving this a shot. I hope they can get the rest they deserve after all they’ve dealt with these past two years

so what are we left with, in cohost’s absence? a lot of people are retreating to Bluesky, some even to Twitter. hell, someone asked me for my Instagram the other day. none of those really strike me as home.

I’ve been on the fediverse the whole time. it’s still the fediverse. it makes a lot of the same tradeoffs as Twitter and its clones. it’s open source but in the way where nobody can practically steer it except one guy. don’t get me wrong, it’s a lot better, but this doesn’t feel like home. perhaps the Website League’s vision will bear some fruit, we’ll see

there’s a movement to the archaic internet, of RSS feeds, weblogs, link buttons and webrings, and I am participating in that with some degree of hope. back to the days of going on the internet, rather than the internet always being a waiting queue of content. I hope it’ll stick, and I hope basic HTML can become a form of baseline computer literacy again as it was in the 90s

whatever’s next, I hope it can be ours, and I hope to see you there.

thanks, cohost


hello, drac.at/posts

this is the first post that I am making in the cohost editor and which I will be posting to cohost.org/ticky and to drac.at/posts simultaneously

I plan to make this a more regular blog over time, but for now it’s mainly an archive preserving some of my favourite CSS crimes and cohost posts

I wanted to get this in before cohost goes away so I can cross-link them for when cohost goes read-only

I’ve done my best to achieve parity with the cohost format, which thankfully hasn’t been too difficult due to them using CommonMark (nice one @staff) and me never bothering to use the fancy math syntax

if you’re subscribed to the main drac.at feed, you don’t need to do anything; that’ll still show you everything I put here. for those who are more choosy, there are separate feeds for my art gallery and this posts section, and in both RSS and Atom formats

once again, working to accept progress above perfection. a better web is possible.


◍ + 👨‍🦲 are becoming Plungus™

Learn More

Originally posted to cohost