Latest Posts (20 found)
Jason Scheirer 2 weeks ago

A Revival of Sorts: Getting my iPod Classic 6G Working Again

I’ve been very happy with my Y1 MP3 player over the past 9 or so months. I take it with me everywhere! It’s a companion on my commute, it’s a focus tool in my open office, it’s a way to have a single-purpose device that doesn’t have the distractions of my glass Everything Rectangle and, as the phone ages, a way to mitigate its now-horrible battery life by using a different device with a different battery. As God confounded the language and scattered the people building the tower of Babel, I have confounded the functionality and scattered the responsibilities of the apps on my iPhone. My wife brought up a point that is completely fair: why am I using this $60 piece of crap when she, through great sacrifice, bought me a top-of-the-line iPod Classic 160GB for the same purpose? Sure, that was in 2012, but it was expensive . It’s still worth $350+ today, right? So what the hell, I dug it out of my Closet of Cables and Mystery. Plugged it in. Battery charged. It booted. My music was still in it, last addition to the library wa 2014. Fantastic! I bought a protective case, some new 30-pin USB cables because the ones I had remaining were all frayed and kind of scary, and I got ready to swap the Y1 with the iPod for a while as an experiment. Then my first hurdle: I wanted to add some songs to it. I know Rhythmbox , my player of choice 1 , has an iPod plugin on its list of installed plugins. I plug the iPod in, it shows up! Hooray! I try to drag music onto it: no dice. Checking I see some very threatening notices that HFS+ with journaling is not supported by Linux at all . So I know on Mac it’s a simple command line call to turn journaling off on a volume so it’s probably a trivial process, but I have no working personal Apple desktop machines. Have no fear: I found a chunk of unvetted C that directly alters the raw filesystem to do it for me on Linux! Boom! We’re in business! Back to Rhythmbox. Drag the music I want over to the iPod. It copies! Bingo! Only: no bingo! I disconnect the iPod and it says ’no music.’ The music is on the device, but the iPod’s music database got clobbered. Well crap. So now I know Gtkpod is purpose built for this. Apparently the iPod Rhythmbox plugin isn’t any good on these models, so let’s try that. No dice. It repeatedly hangs, crashes, and when it does work it still fails to correctly update the database. Still ’no music.' Maybe this is all because it’s still HFS+ and not FAT? It seems like most tools assume you’ve liberated your iPod and you’re using it in Windows mode, not Mac mode. So I attempt to wipe the drive, but can’t for the life of me figure out how to do it correctly with Gtkpod or just plain old partitioning tools. Looks like I need to restore the hardware from iTunes for this route. What about Rockbox ? I use it on my Y1. The annoying thing is that I have to manually update the database on the actual device, whereas the typical iTunes stock experience is one that updates the database iteratively as a matter of course of adding music. But the trade-off is no more struggling with Gtkpod and friends, which is higher friction than the drag-and-drop experience of putting music on my Y1 anyway. And I saw this totally cool skin on Reddit I want to try ! I already have the Rockbox utility on my machine from installing it onto my Y1. It sees my iPod but dies on an SSL handshake talking to rockbox.org while downloading resources. I don’t remember this happening last time I ran this. I downloaded and ran the utility on another Linux machine and got the same result. I gave up about 45 minutes into building the tool myself from source. Now I need a Windows machine to use iTunes in Windows to reformat the iPod. I have a debloated Win11 VM in Gnome Boxes, I fire that up and go in to iTunes, I plug in the iPod, then I go to set up USB forwarding so the VM can do its magic and – “USB Forwarding is Not Supported in the Flatpak version of Boxes.” So I uninstall the Flatpak and migrate my disk images from to somewhere less Flatpak-specific and install the dnf version of Gnome Boxes. I migrate the machine over, set up forwarding, everything seems to be working. Only USB forwarding forgets the device when it disconnects and I have to reconnect multiple times. It also doesn’t see the device when it’s in that raw flash mode, so it can’t forward to install the iPod firmware. This is a dead end. Okay, so I have one Windows machine in my house: my kid’s 2013 Intel Macbook with Boot Camp and a debloated copy of Win10 we solely use to play Minecraft Java together with. Only ever since I set up a local server with GeyserMC and Floodgate we’ve been playing mixed me-on-Java/him-on-iPad-or-Switch-Bedrock so the laptop is mostly neglected. So I install iTunes and wipe the iPod. Takes awhile, because I have to install a cascading series of drivers, but it eventually works. The firmware was the latest for the Classic, released 2009. Then I remember that 18 year old bit of early enshittification of iTunes: the iPod can’t simply be its own library you add/remove items from. I was falling out of love with Apple about that long ago , and I had forgotten how low and slow we’ve been dealing with the world of You Will Own Nothing enshittification that’s been inflicted on us. No wonder we’re so complicit, we’re pushing a quarter century of Everything Rental now. So to do iTunes proper I’d need enough storage on this laptop to hold the music in my library on it, be logged in, and sync a selection of it to the iPod. I remember this now: they made life harder and worse on purpose. And now we have Spotify, where we never had freedom or affordances at all. I remember thinking what an incredible act of charity it was that Spotify let your have an offline playlist on your device. I would have expected offline first as a matter of course in prior hardware/software cycles. Rhythmbox and Gtkpod still don’t sync correctly. Same database issues, so nothing I’d done with wiping the iPod had fixed the fundamental first issue. So I install the Rockbox utility on the Windows machine. I have to install some additional Windows components to get it to load, but it works. I flash the iPod. It doesn’t boot. I flash it again. It boots. Hell yes. And I have my cool theme. So I drag music over. 16000 tracks to start, takes 2 hours to copy. HDDs are slow . Afterward I have to manually update my database from Rockbox, which takes hours . I fall asleep as it runs. I can hear the physical spinning platters. It’s a very strange experience having a device with a real life magnetic disc hard drive again. The future we occupy today is strange in the UX of the iPod and its software feels modern enough but small aspects like an HDD feel anachronistic. The Rockbox experience is a lot nicer on the hardware it was designed for than the crappy Rockbox-in-emulation on an Android device that has absolutely no business whatsoever claiming it can run Android. It is responsive, it doesn’t crash, all the plugins work, etc. Next rabbit hole is investigating battery/storage upgrades. There are cheap and expensive options, I need to go through them. As is my wont, I do not need bluetooth on anything I own, but a modern USB-C connector might be nice? Do I want to go the SD card route or a proper SSD? That is for another time. Anyway, no normal person would inflict this experience on themselves willingly, and would likely give up at some point close to the beginning. It is a reminder that much like if you stay very quiet near a playing iPod you can hear the whir and rattle of the HDD. If you stand very quietly near me you can hear the fluttering and tapping of dozens of moths smashing their bodies against the inside of my skull in the space where a brain should be. I am not aware of any other MP3 player that can handle large music libraries this well and still have a presentable UI. TUIs usually suck, “new” apps are all super slow because of Wirth’s Law.  ↩︎ I am not aware of any other MP3 player that can handle large music libraries this well and still have a presentable UI. TUIs usually suck, “new” apps are all super slow because of Wirth’s Law.  ↩︎

java Java hardware Hardware linux Linux music Music
0 views
Jason Scheirer 3 weeks ago

xteSTINK: A command-line utility for doing things to the Xteink X4

Top Matter : Codeberg for the library , doc for the library . I bought an Xteink X4 ! It’s small! It’s cheap! It’s hackable! I have an e-book collection, if you could call it that, but I always purchased PDFs. This thing reads epubs by default, which I don’t have. I want to use this to read my books! So I’ve been hacking on a little go command line utility to get stuff on to the One huge unlock for me was the native XTC/XTCH format the reader can render which gives you the ability to render bitmapped, pre-rendered pages. I’ve abused this to a great extent! I found many of the Epubs I was building were too big to render on the limited resources on the local reader hardware, so having visual artifacts prebaked and ready to go gave me a huge amount of additional control over what I saw on screen. The subcommand I’m happiest with is – I find most of my books are perfectly readable in good light when scaled down to the size of the small screen. was incredibly easy to implement as well, as it uses Playwright on the backend to render a PNG and slices it into visual pages on the device, and even cuts multiple URLs in the same doc into chapters. Current help:

open-source Open Source hardware Hardware programming Programming
0 views
Jason Scheirer 3 months ago

Saddle Creek: Add Out-of-Band Metadata to PE Executables

Top Matter : Codeberg for the library , doc for the library . I took the one part of the shuttered Omaha open source project by Google and I’m making it a library I can consume elsewhere. That is all.

programming Programming
0 views
Jason Scheirer 3 months ago

Golang Webview Installer for Wails 3

Top Matter : Codeberg for the library , doc for the library . I’ve forked Lea Anthony’s library that eventually made its way into core Wails for two reasons: So here we are. I want it in Wails 3 and it’s not there I want to shave a meg off the binary size by not providing the embedded installer exe

web-development Web Development open-source Open Source go Go
0 views
Jason Scheirer 6 months ago

Steam on non-Conventional Desktops (Niri)

I’m trying out Niri ! You know how I encourage getting used to the defaults ? Well I’m not following my own advice ! I’m using it with Dank Shell too, also ignoring my own advice ! Anyhow! One liner! If Steam isn’t working do this: edit and change the line to this:

gaming Gaming linux Linux shell Shell
0 views
Jason Scheirer 6 months ago

Learn To Live With The Defaults

Every deviation from default is slowing you down from getting started and making it harder to help others. I use about 5 distinct laptops/desktops on an average day, not to mention the VMs within them and other machines I shell into. Having a consistent experience is useful, but equally important is my ability to roll with the punches and get productive on a new computer without much ceremony. One thing I do to cope with this is a dotfiles repo with a dead-simple installation method , but also note how conservative it is. No huge vim plugin setup. Very minimal tmux config (which is still bad, and I’ll explain why later). Not a lot going on. Moving from the defaults to a custom setup might make you more effective in the immediate term, but it makes it harder long-term. You have additional complexity in terms of packages installed, keymaps, etc. that you need to reproduce regularly on every system you use. As I complained about in Framework Syndrome , flexible software just moves the problem along, it does not solve the problem. Having a tool that’s flexible enough to get out of the way so that you can solve the problem yourself is double-edged: it does not provide the solution you want, it provides an environment to implement your solution. This seems to mean that everyone new to the software will not see it as useful as it seems to you, right? To them it’s a blank slate, and is only useful with significant customization. This also affects teachability! With your hyper-customized setup you can’t be as effective a mentor or guide. One thing that makes it harder for me to advocate tmux to new devs is that I use one thing sightly idiomatically: coming from the older tool screen means I remap Ctl-B to Ctl-A for consistency. This has bitten me many a time! One example: Once I had set up a shared VM at work and had long-running tasks in tmux that my teammates could check in on. The entire setup was stymied by the fact that nobody but me could use tmux due to that one customization I had set up. Learn to lean in and be as functional as possible with the default setup. A kitted-out vim is great but learn the basics as muscle memory. Prefer tools with good defaults over good enough tools with the flexibility to make them as good as the ones with good defaults.

devops DevOps shell Shell programming Programming
0 views
Jason Scheirer 7 months ago

A Series of Vignettes From My Childhood and Early Career

A short set of anecdotes, apropos of nothing. When I was younger, I really liked programming! I loved the sense of accomplishment, I loved the problem solving, I loved sharing what I made with the people around me to both amuse and assist. One particularly wise adult (somewhere around 1996) took me aside and said, “You know, you’re lucky you enjoy programming, because you won’t be able to make a living on it in the future. Doing it for love over money is a good idea.” “Coding is over, with Object Oriented programming one person who is much smarter than any of us could hope to be will develop the library just once and we will all use it going forward, forever. Once a problem is solved it never needs solving again. “In 5 years there’s going to be a library of objects, like books on a bookshelf, and every software problem will be solved by business people just snapping the object libraries they need together like LEGOs. They won’t need you at all.” I thought about this advice, and how Software Engineering would be ending by the time I entered school. I realized I had not even thought about my education yet. I was in middle school. Programming was not it, though, I knew that. I’m here nearly 30 years later and software continues to pay my bills, despite everything. Open source exists, there are libraries I can use to piece things together to solve all the time. New problem sets not covered by the garden path come up all the time. Clicking the LEGOs together continues to be a hard task. Every time we fix it at one level of abstraction we operate one level higher and the world keeps turning. Whenever I’m threatened with a good time and someone proclaims “this is it for you” all that happens is my job becomes more annoying. Haven’t gotten the sweet release of extinction quite yet. Around 1993 or so was the advent of the “Multimedia Age.” Multimedia was the buzzword. Software has to be multimedia ready . Education had to teach children to be ready for the multimedia age . If your tool, however inappropriate as it was, did not have multimedia features, you were going to be left behind. You needed a video guide. You needed to be on CD-ROM. This is just the new normal. “Multimedia” just means “sound and video.” We had a high concept term for a very direct, low concept concept. And the multimedia boom fizzled out. It became boring. Nobody is impressed by a video on a website and nobody thinks less of a website that doesn’t use sound and video if it’s not appropriate. You pop a tag in your HTML and your job is done. The amazing thing became mundane. The dream of “multimedia” became commonplace and everyone just accepted it as normal. I’m not aware of any industries that collapsed dramatically due to multimedia. Nobody really reskilled. Video editing is still a pretty rare thing to find, and we don’t commonly have sound engineers working on the audio UX of software products. In 2000 a coworker took me aside and showed me his brand-new copy of IntelliJ IDE. “It’s over for us,” he said, “this thing makes it so programmers aren’t strictly necessary, like one person can operate this tool and they can lay the rest of us off.” I was pretty awestruck, he got some amazing autocomplete right in the IDE. Without having to have a separate JavaDocs window open to the side, and without having to manually open the page for the class he needed documentation on, it just was there inline. It gave him feedback before the compile cycle on a bunch of issues that you normally don’t see until build. That was a nice bit of preventative work and seemed to have the potential to keep a developer in flow longer. And then he showed me the killer feature “that’s going to get us all out of a job:” the refactoring tools. He then proceeded to show me the tools, easily moving around code to new files, renaming classes across the codebase, all kinds of manual things that would have taken a person a few days to do on their own. It was magical. After some thought I said, “that’s amazing, but does it write new logic too or does it just move code around?” He didn’t seem fazed by that, and doubled down on the insistence that these powerful tools were our doom. I made a distinction between “useful” code and “filler” code, but apparently what is valued is not the quality and nature of the code but its volume and presence. This tool definitely gave both volume and presence to the tiny human-written nuggets within. At my first job in High School I was working in an office in a suburban office park with programmers from many different local agencies. One guy I chatted up was a contractor: these people were highly regarded, somewhat feared specialists. The guy in question was working on a multi-year migration of some county health computer system from MUMPS to a more modern relational system. He showed me the main family of problems he was solving to show off how smart he was for solving them; they were largely rote problems of migrating table schemas and records in a pretty uniform way. But there were a lot of them, and he was working hard to meet his deadline! I thought about it, and seeking his approval and validation, set out to help him. To show what I could do. I wrote a Python script that could solve the 85% case (it was mostly string manipulation) and even put a little TkInter dialog around it so he could select the files he wanted to migrate visually. It ran great, but he looked a little afraid when I demonstrated it to him: “You didn’t show this to anyone else, did you?” “Nope.” “Oh thank God.” I take it he used my tool because he had a lot more free time to goof off for the remaining six months of his contract. I don’t think he told anyone else what he had either, but I’m guessing that he had a lot more MUMPS migration contracts lined up when he could finish them in a matter of days. At the same job, I was paid to maintain a series of government agency web sites. One of my main tasks was to keep a list of mental health providers up-to-date on an HTML page and upload it to the server. This process was pretty mechanical: take Excel sheet from inbox, open in Excel, copy Excel table to HTML table. Within a month I had a fully automated workflow: I lived in fear of being found out, and told no one that the thing I was getting paid to do was no longer being done by me. About 9 months later the department in question hired a full-time web developer for $45k/yr to bring their website in-house. I was costing them about $25/hr, probably skating under $2000/yr for my outsourced services. This was clearly not about money. And what I feared did not happen. When I no longer had that work to sustain me my managers just put me on something else. There’s always more work. In my last years of undergraduate education and my first couple of years out of college I worked on projects that did some sort of Natural Language Processing tasks. For these we required training data, and the more the better. On that, though, we had responsibilities. We had to make sure the data we had also came with some sort of license or implicit permission. You didn’t just steal a pile of PDFs or scoop up a person’s web site and put it in your training set. There were ethical constrains, and legal consequences. You acted above-board when training your AI models. There were times we’d train models on Wikipedia dumps. They were always comparatively amazing results when we trained on good, large data like that. Cogent. Interesting. Even a simple Markov chain on Wikipedia looked smart. When we wrote web crawlers, we wrote them to respect . We kept them on local domains. The field of the crawlers included our email address, and if an angry webmaster didn’t like the way we were crawling them we’d fix it. Getting crawled aggressively at once taxed servers and spammed logs so we’d space it out to hours or days. If their was missing or malformed and they still didn’t want us there, we’d block the site from crawling. We made sure we had explicit permission to collect data for our training corpora. The dot com boom was a crazy time. The internet has just become mainstream and there was a new gold rush. Money was there just for the taking, so many VC funded business plans were just “ traditional business X, but on the internet! ” and the money flowed . How it flowed. Most of these companies, however, didn’t really have a solid business model other than buying some servers and a domain name and “we’ll put this thing on the internet.” Out of this crash came green shoots: Web 2.0, which used the web natively, organically, gave a good web-native experience. Eventually the dream of the internet, the promise of the hype, was made manifest after a lot of people learned a lot of really unnecessary, really painful lessons. They spent less and put their things on the internet because they made sense on the internet of the present, not because the internet was the next big thing. The dream of the widespread, ubiquitous internet came true, and there were very few fatalities. Some businesses died, but it was more glacial than volcanic in time scale. When ubiquitous online services became commonplace it just felt mundane. It didn’t feel forced. It was the opposite of the dot com boom just five years later: the internet is here and we’re here to build a solid business within it in contrast with we should put this solid business on the internet somehow, because it’s coming . This is indeed a set of passive-aggressive jabs on the continuing assault on our senses by the LLM hype lobby. I used Windows Automation to watch my Outlook inbox When an email came in from the person who sent me the Excels it would download it Open the Excel file in excel using Windows Automation Export it to CSV from Excel (the automation did this, I simply watched a ghost remote control an Excel window that opened and closed itself) Run a Python script that would inject that CSV data as an HTML table into the file Run another Python script that would connect to the FTP server and upload the file. It would randomly pause and issue typos so it looked like the FTP session was being operated by a human at a keyboard so nobody thought anything on my plot.

programming Programming career Career python Python html HTML
0 views
Jason Scheirer 8 months ago

The Innioasis Y1 Music Player

I’ve been enjoying standalong MP3 players! The Innioasis Y1 kept coming across my radar, I like the the form factor, it was $50. What the heck, why not. The community for this thing is insane , it’s just as active as the people doing weird things with my RG35XX . It’s really cool seeing so many people doing neat things with such a simple piece of hardware. And like the RG35XX, part of the value proposition is this is a cheap peice of commodity hardware that would not have been possible in this way even 5 years ago, but is now inexpensive enough and flexible enough to be an incredible product for the money. I saw you could put a flavor of Rockbox on the thing so I did that. The UI out of the box isn’t nearly as polished but there’s a neat community-supported updater that goes so far as to install skins for you. I’m currently using another Adwaita adaptation I found on the y1 subreddit which handles CJK correctly, which turns out to be important to me. Rockbox has the ability to create a play log file, so I can scrobble my commute/work listening again ! I use the LastFMLog plugin to manually create a file, then use rb-scrobbler to upload it. It’s a manual process I only do every week or so, but it’s okay. This is awesome. Three surprises. Not quite complaint territory but worth knowing about: This thing is a lot of fun to use, though! The novelty will eventually wear off but it feels good to have something iPod shaped in my life again. No external storage. My Shanlings had a TF card slot so I could expand and swap the storage easily. This is internal. 128GB so it won’t hold my whole library but it holds everything I care about. No touchscreen. Again, coming from Shanling this took a little bit of getting used to. Pure iPod classic ergonomics, buttons only. Build quality is not super solid. The screen is plastic, not glass, and it scratched almost immediately. You can definitely “feel” a center of gravity while the majority of the device fgeels light. No metal in its construction. It doesn’t feel brittle but by no means is it a luxury experience.

music Music open-source Open Source hardware Hardware
0 views
Jason Scheirer 8 months ago

The Shanling M0s Music Player

My Shanling Q1 died after a couple of years of heavy use and I think it was probably fixable with some soldering but I don’t have time for that. In a rush, I bought an M0s not reading the page and thinking it was the higher-end M0 Pro , but ultimately this was not a big deal! This still works like the Q1: it has a custom proprietary Shanling OS which works pretty well. It does Bluetooth fine, it manages music via TF card the same. The thing I did not realize was how much I’d like how small it is, it’s a tiny little 1.5inc square that’s about half an inch thick and has barely any mass. It just dangles at the end of the end of my headphone cable. It would be very easy to lose if I were more careless. In all, it has served me for a year. No complaints.

music Music
0 views
Jason Scheirer 8 months ago

What You Do and Who You Are

Similar to The Wrong Conclusion is another cognitive anti-pattern: in the pursuit of identity we see ourselves as being something as an inherent quality of ourselves versus doing something as a role, a quality being temporarily practiced. This can lead to bad behavior on our part and it limits our ability to grow and engage in introspection. I have this hard-won lesson over my lifetime: when I stake my identity as being a thing, when I can no longer do the thing I lose my sense of self and spiral into crisis. I am a good programmer, so I am a thing with a clear identity only when I am programming . If I spend a week in the hospital, as I did a few years ago, I am no longer programming, I am infirm . I have lost my anchor. This is upsetting! I need to find a better I am or accept that programming is an I do . Alternatively, I never really thought of myself as a parent or a future parent. Then I had a kid! Whether I cop to it or not, I am a parent – I do parenting things every day . People who take moral stances often consider themselves to be the good guys , a quality inherent in themselves, versus being people who happen to strive to do good things . When you are tautologically good, this is bad! You can do evil but by definition you are doing good because good is what you are . I see it in this quote: Conservatism consists of exactly one proposition, to wit: There must be in-groups whom the law protects but does not bind, alongside out-groups whom the law binds but does not protect. Francis M. Wilhoit Entirely an “we are” versus “you do” mentality: the law is meant to protect citizens from harm . We are citizens, you are harm. We do bad but do not run afoul of the law, you do bad and have committed a crime. That is; if you cannot see what you are doing is counter to your identity , you can be convinced that what you are doing is inconsequential to your identity . Going from an I am to an I do mindset has been a tough lesson for me, but it’s been transformative in my worldview. It is, in many ways, tied to a growth mindset versus a fixed mindset: (I am / I am only) and (I do / I can become) being very closely related. I don’t ‘consider myself’ a Javascript Guy, for instance, but if I make a presentation about some Javascript thing I researched and shared I have done Javascript Guy things , and I am technically a Javascript guy. As far as this goes, back to the growth mindset: you are only alive so long as you are active, and there is plenty of life left in you to be active. You are not a (metaphoric) flag planted in the (metaphoric) ground, you have (metaphoric) legs and can (metaphorically) walk to the next (metaphoric) place. You are capable of cruelty and laziness, and you are capable of kindness and industry. You need to be constantly vigilant to make sure you are currently doing good and do not rest on the laurels of I once did good or I am good so what I do is, by definition, good . Anyway, these were a few short paragraphs summarizing a spiritual crisis I spent 4 years mulling over. Enjoy! Hope you learn the lesson the easy way (from the mistakes of others, namely your supremely handsome narrator) rather than the hard way (having your own mistakes serve as a warning to others).

career Career javascript JavaScript programming Programming
0 views
Jason Scheirer 8 months ago

I Don't Like that I Like Starship

I have my seed Starship config up as a Gist. Starship is a tool that frustrates me because it seems so bikesheddy and unneeded: a custom prompt manager. We already had shell prompt customization! I blindly install on new machines for prompt customization! And Starship is written in Rust. People just use Rust to be cute. Then I realize that it’s okay to have nice things. The command line environment from the 90s can change. doesn’t suck. is right up there with Perl in opaque tools other people do interesting things with and then I steal the interesting things for myself. The TUIs don’t suck. I mock Textual and Charm openly and unrepentantly and still hypocritically use and enjoy the tools built with them. The terminal is changing because the people using it have the agency and hubris to use it differently. I can have a growth mindset and accept that But its killer features: So here it is. All this talk for something that doesn’t look substantively different but compounds into feeling different over the course of the days and weeks I use it: The preferred command line tool to do a thing can change in my lifetime and The preferred workflow to accomplish a task can change in the face of new tools and The people who invented the old tools we’re replacing were not gods, they were just as fallible as us so Writing new tools that learn hard lessons from decades of using the old ones is fine. It’s not sacrilege. Limited/Constrained/Opinionated : Other prompt customization schemes I’ve used have let you do anything , but you had to know how to do anything . I can’t think of cleverness I want in my prompt, I just want to see which Git branch I’m on. We’re all doing that. We all want that. Starship has a way to do that which isn’t brittle bash I have to maintain myself. Multiplatform : I have Windows Bash, Windows PowerShell, Linux Bash/Zsh, and macOS Bash/Zsh all driven by the same seed config. I can have it show the same information everywhere. I can use little emblems to let me know if I’m on my Mac, on a Linux machine, if I’m on Windows and if that’s PowerShell or Bash all right there. The Nerd Font Dark Horse : This system takes advantage of the glyphs in Nerd Fonts and normalizes abusing them. This adds the additional “burden” of installing a Nerd Font enabled typeface on all my apps with terminal editors, but I’ve already normalized that.

Perl open-source Open Source rust Rust Bash Powershell
0 views
Jason Scheirer 11 months ago

There and Back Again: My Journey Into (and out of) Tailwind

I’ve been using the Tailwind CSS Framework for about two and a half years (as of July 2025) for my personal projects, and I used it professionally in my time at unstructured as well. In all, I think it’s a good thing. However, I don’t find myself thinking I’d use it for new projects at this point: I believe I have outgrown it. Here is a short story. Tailwind has a set of that encourages a specific styling worldview that gets you a “modern” looking UI which doesn’t feel offensive versus any other site on the internet. The pre-baked color choices and reasonable framing around layout options like columns give you a framework (ha!) within which to work and make a decent site. It’s less greenfield than a blank page and an open documentation tab . I’ve been doing web development on and off for…well, since my teenage days but I don’t specialize in frontend dev and every few years when I come back to focus on it, things have changed. I don’t know what’s popular, I don’t know what’s possible, I don’t know what’s popular (hint: if it’s easy to do in CSS, it’s popular). Flexbox in particular is something I had a somewhat superstitious, vague understanding of. I could do little bits and pieces and sort of make it work, but being able to set the properties with just a few attributes made it faster. Getting that coding live-reloading in my browser as I made mistakes was the best interactive, REPL-like experience I’ve had to gain an intuitive understanding of the new CSS layout options. For that, Tailwind was invaluable. It certainly is a lot less work to type than . This, I think, along with popular UI tricks like blurred, semitransparent elements being baked in as simple attributes are the killer feature of Tailwind. When I wanted a button-like element, I found myself going through a series of stages: And then custom elements that further refine the style: I think this is where I begin to stray from the way most people use Tailwind: I have a tendency to write lots of very small components and I am a bikeshedder in my visual styles so keeping all the elements in sync requires the type of DRY the on every tag makes difficult. Eventually I find myself wanting a having a class with all the inline constants pulled out. For the sake of readability and some light DRY, I’d rather my buttons be than ; I then find myself doing this: That is; the @apply directive gives me Tailwind-style class macros in my CSS. I can now use a more traditional approach to CSS: rather than a blast of appearance-based attributes on an element I can give it a semantic role which has an associated set of visual properties attributed to it. I also needed this because I wanted to do some things with nested selectors that Tailwind did not entirely make possible (or at least easy). I found in places where it could do selectors the shortened Tailwind attrs beyond and were harder to skim than fully-expressed CSS versions. Anyway, is awesome and I don’t see its use encouraged much in the literature I’ve seen online. There are probably pedantic reasons for it that are all very reasonable but don’t work for me. This worked for me. And, once I am happy with the style and relatively certain they won’t change much going forward, I transform the core CSS with d styles to fully expanded CSS and cut out Tailwind as a build time dependency. I have used Tailwind as a way to bootstrap my design system, and then cut it out of my build when it has overstayed its welcome. So that has been my last two years with Tailwind. I like it, it helped me get back up to speed with modern CSS and get a working visual prototype out the door quickly, I got tired of its class soup usage pattern, and I found that once I returned to old-fashioned CSS classes I could wean myself off it. Style a single JSX component as-is with individual tailwind styles iteratively until I get to something that I like Start on a second component that I want to make look similar to it Copy/paste the style Eventually I want to do a refresh system wide, so I find all the overlapping classes, cut them out and reuse them in backticked strings; I have something like

web-development Web Development css CSS frontend Frontend
0 views
Jason Scheirer 11 months ago

Here's The Interesting Part

For the 500 lines of boilerplate, what are the five truly interesting lines that solve the problem ? In the course of my problem solving, especially when I am solving a problem I think shouldn’t be hard , I think back to the essential core of the problem Typically a fix is a couple of API calls or a single clever data structure. As a consequence of this, I typically annotate my Pull Requests with an annotated “this is the interesting part” section. I recommend this approach: you are asking your reviewers for special attention on the original thoughts, and signaling the other code you had to write is something you do not feel is important, and are far more open to nitpicking on (versus thoughful review). Pair this with This Shouldn’t Be Hard : You just need those five lines of beauty and then you need to do the 500 lines of ceremony to make them real.

programming Programming
0 views
Jason Scheirer 12 months ago

This Shouldn't Be Hard

As you are working your way to the goal, you should be able to be clearminded about how you aim on getting there and when you are slowed down for silly reasons do not accept them as excuses. Say to yourself, “I know what I need to do, it shouldn’t be this hard.” Overall, you should be working toward writing functional code that achieves a goal. Oftentimes we get stuck on the actual specifics of a task: how do we set up the branch, how the dev environment gets up and running, etc. These are obstacles. Keep your eyes on the prize. If you can imagine a problem, if you have a clear vision for how to do it, this is all that matters. Process and tools are a means to the goal, not necessarily a requisite of getting to it. You must achieve your end goals despite everything in front of you. You have a vision and it should be as little friction as possible. This means minimizing unnecessary steps and busywork. Automate it. Don’t do it if you don’t have to. Always stop and think to yourself: I know how to solve this, it shouldn’t be hard to get that solution out into the world . Treat each hindrance with the utmost of contempt. Document and work around pain points, make sure they do not remain painful for long. Can you make them less painful via changing the broken parts? Can you make them less painful via “sharpening your tools;” i.e. getting better at working with the system via practice? Break the rules if the process gets in your way: you have a goal and it is not your goal to get stopped by the journey.

career Career devops DevOps programming Programming
0 views
Jason Scheirer 1 years ago

Chains: My Attempt at an Itertools for Go

Top Matter : Codeberg for the library , doc for the library . It’s been six months since I’ve done this, but I’m finally writing about it! Go recently added proper iterator support to the language, which is something of an improvement over the prior pattern of spinning up a goroutine and communicating via a channel in a to get a stream of values. One of the tools in my toolbox that I use in coding interviews and some light data processing work is Python’s itertools . The nice thing about this library is it gives you a good set of conceptual building blocks to use as a frame around a problem and a fairly clean way to use them. Once you’re familiar with, say, and you can take a harder problem and decompose it into those recognizable parts and then have a stdlib function that’s already bug free and readily available. While I was inspired to author this, I was writing a lot of Ruby and Typescript. Both Ruby and Javascript do processing over lists in a very chainy way; for example Ruby likes to add lots of compact calls etc. as well to handle bad data. Having this syntactic sugar makes it easier to write complex logic, and it also helps conform the logic to one’s brain. Anyway, Go has iterators now, and I like using iterators. The first thing I wanted was my brain poisoning syntactic sugar from Ruby/Typescript; how could I go about doing in Go? Once I had a framework cooking I could start thinking of examples. My test suite took a backseat to being a test of cases I cared about , in cookbook form. Some things I wanted: Not much to write home about here, anyone can write these and I encourage each person to do it themselves using Go iterators. There’s the usual suspects here along with various types of that all boil down to the base . Similar to the above, pretty trivial to write. Can even treat these as specific cases of . is one such case . I wanted and , so those were high on the list. I found myself writing the code more and more generically as I went along, eventually ending with the mess that is . Funnily enough, each combinatorial case was some combination of: Not as high-level as Combinatorics, but I wanted to take a window of N at as time. I like to concatenate iterators sometimes; e.g. to process the results of two tasks in a single queue. works in Python, it was a breeze to write here as well. Got the base case down. works in a pinch for “gluing” together iterators in Python , and that was pretty trivial to write ( I called it “Flatten” because the first iteration took an iterator of iterators, thanks to Ruby/Javascript brain, then made one that took variadic iterator args as too). Some common use cases I find myself writing a lot just aren’t in the stdlib in Python. They generally involve taking many iterators and unifying then in ways dependent on the structure of the iterators themselves; either by length or by value. Another use case is Round-Robining a set of iterators until they are all exhausted. We’ve got a set of inputs and want to consume from all of them until we run out. For that, there’s , which works exactly as expected. Each iterator can have a variable number of entries but all entries are considered by index, so we don’t exhaust one before going to the next but try to consume them all equally. Use case: I have 3000 CSVs, each has rows in order of date. The time ranges may overlap in some cases. I wanted a unified stream of all the rows in order. For that, I wrote , which is at its core a pull-on-demand heap. Once the smallest value has been pulled off the heap, yield it, then grab the next value from the iterator that provided the value to place on the heap. I wanted to be able to chain calls like I can in those other languages, so the first thing I thought to do was design some sort of struct or interface with various / / /etc methods – so like Immediately there’s a problem because Interfaces can’t use generics in Go, so we do a struct instead: This sort of works! You can see in the type we do this. But to do two types (say we’re mapping from to ), we have to have an that does two, and so was born . Now how to get from an to an ? I decided on a top-level function to create a type called a to go from a one-typed chainable to a two-typed one. Now what if we need a third type? This is getting messy 1 . This pattern works for simple cases just fine, but it falls down once we get into the variadic world. Go’s obviously stunted-on-purpose generics are preventing us from doing this syntactic sugar in a clean way, but it is also suggesting a different way to do it. I was in love with my ability to do chained iterators, but they got clunky. Go generics only apply to functions and you can’t template an interface. So while is fun and cute, in Go you’re better off doing something like: …which, quite frankly, feels a lot more Go-like and less foreign than the cute way we do it in other languages. You can see I gave up on chaining in the above examples and just do individual iterators. And so, instead I found myself using single iterators via the adapter function and back to slices with . As I got further into implementing the various functions I wanted, I moved away from the pattern into simple functions. It’s still ugly to do because the pipeline appears in opposite order but doing each as an assignment keeps the order at the expense of slightly more verbosity. It’s not as aesthetic but it works. I think the most practical example I can give is the test in the cookbook that generates a sequence of fights in Street Fighter. If you play as a playable character you can fight all the other playable characters and each of the bosses. You cannot play as the bosses. As such, we have two separate matchup types: And gluing the two together is pretty clean: Once I started doing things the Go way, it really increased the pace of development as well. Being able to implement each iterator as a simple function meant I could focus on implementation and not boilerplate. Constraining myself to and relieved me of the analysis paralysis of variadic iterators: one value, and when it made sense, two. I’ve made this available as an importable library, but many of these patterns are easier to just copy and paste into your code. They should also be inspiration: this is a fun problem to solve! Solve it yourself! After posting this, I discovered another person had done something similar but used runtime reflection to sacrifice compile-time safety for the syntactic cleanliness I was going after. I respect the approach and I like the depth of knowledge of the language needed to do this. I’m going to say which implementation is better is a matter of knowing the tradeoffs: are you willing to sacrifice compile-time safety for convenience? It’s probably an unequivocal yes if you have good test coverage.  ↩︎ Map/Filter/Reduce Cleanups (compacts, nonzeroes, etc) Combinatorics Higher-level stream processing (various merges) There’s an and an for testing for conditions satisfied by the entire sequence; gets just the length. All are useful with ! and do what they say; is the CAR/CDR you didn’t know you needed. You can and to start/end at a particular point. You can an item N times, a slice so the first N items are moved to the back, each element N times, and the iterator which is just infinite repeats. takes an iterable of iterables and turns it into a flat iterable , but it doesn’t do it to arbitrary levels of nesting like Ruby does. There are rules here dude. is largely useless , kind of like a forEach or a visitor that passes the item along. splits an iterator into two based on a partition function, allowing you to e.g. split good/bad inputs into separate pipelines. A simpler function just returns the first value of each key grouping instead. Similarly, takes an ordered set of items and “bins” them based on a key function, allowing you to There’s also a to get a tear-off copy of the iterator. Length (one, fixed, variable) Ordering (in order of occurrence, free variance) Repetition of elements (on/off) Player v Player, unordered Player v Boss, unordered After posting this, I discovered another person had done something similar but used runtime reflection to sacrifice compile-time safety for the syntactic cleanliness I was going after. I respect the approach and I like the depth of knowledge of the language needed to do this. I’m going to say which implementation is better is a matter of knowing the tradeoffs: are you willing to sacrifice compile-time safety for convenience? It’s probably an unequivocal yes if you have good test coverage.  ↩︎

python Python javascript JavaScript programming Programming go Go
0 views
Jason Scheirer 1 years ago

Hate What You Know

You should be familiar enough with your tools to be able list critical things about them, even though they’re what you still choose to use every day. The title Hate What You Know is admittedly vague, I know! There are a handful of directions I’ll go with it, bear with me. And at least to some extent I don’t mean “hate,” I just mean “know well.” I’m an advocate of being a craftsperson out of anger. Do things because they need doing and dammit, you can’t live in a world where this problem isn’t solved anymore. I’ll make a deliberately outrageous claim: doing something out of need is craft, doing something out of compulsion is art. A craftsman makes a work because they see something missing in the world. An artist makes a work because they see something of a possibility in the world. Professionally, at least, you’re hopefully operating in the former camp. In the latter, you’re either outrageously lucky doing what you love in a place of psychological safety or secretly pissing everyone off around you by being a perfectionist primadonna. Hate is a strong feeling that can only really be fostered from a sense of strongly personal experience with a topic, person, or a tool. To truly hate a tool, that means you understand it intimately. To understand it intimately means you can use it well. It’s easy to dismiss a technology based on a pro/con list or based on some surface feature. “I could never use Python,” developers say, “the indentation-based syntax feels too weird.” After a month of actually using Python, a surface judgment like that probably goes away, overshadowed by actual deeper issues with the language. There’s plenty of worse things down the line once you’re familiar with the system. I’ve seen the same said about Go and its (lack of) solid OOP principles. Once you’ve used Go and you’ve gotten things done, there’s a whole new world you discover: how to live without that core thing, and brand new things to dislike but tolerate further down. In this, actually using a tool gives you real things to dislike . An analogy I like to give when discussing this: imagine two people, a person who cut you off in line at the bank and a close relative who happens to be a tremendous fuckup. You’re standing in line at a bank, you’re 45 minutes in, you’re one or two people away from being helped. Someone walks in front of you, and saunters right up to the counter and starts getting helped. You’re seething with rage. All you feel for this person is the singular emotion of anger . They could be someone’s most important person, it may have been a mistake, it may have been for an excellent reason, but that person cut you off and 100% of your interactions with them is that act of being cut off . All you know about this person is this one moment where they wronged you. Then think about a loved one, a close relative. You grew up with them. You watched the process of them developing as a person alongside you. You love this person. Now, this person may be something of a fuckup: they’ve been in and out of jail, they do things you know are bad for them, they continue to hurt themselves, with you as collateral damage. Last time they stayed at your place you came home and they’d sold your TV. More often than not, they hurt you. But you remember the good times. You have a history. This is a fully developed human being and a core character in your life. Their rate of wronging you isn’t close to 100%, but the number of times they’ve wronged you dwarfs that stranger at the bank. You should hate this person with the same intensity as that person at the bank, maybe more. But you don’t, it’s more nuanced because of your shared experience. You should aspire to withhold loud, public judgment of a technology until you can look at it with a “well, sure, but…” and a long wistful sigh. Hate, as opposed to love, comes with a much different place of origin when setting expectations: you presume the worst rather than the best when approaching it. This sets you up for disillusionment when the thing you can see no wrong in eventually fails you in some way. Never meet your heroes. Heck, never watch a hero on a bad day. You should look for ways to hate your tools. By this, I mean the following: Every time you have solved a problem with a tool, try solving it with a different one. The contrast should give you a wider perspective on how your preferred way of getting things done could be better. I’ve gone in a full circle with CSS frameworks recently: I forced myself to use Tailwind until I liked it, then at work I had to use Sass. Sass sucks in a lot of ways that Tailwind doesn’t. The converse is true, too. But knowing how other people solved the same problem, you can recognize the clumsy parts of other approaches and take the good parts of one and try to apply them to the other. And the conclusion to this, of course, was me becoming better at vanilla CSS, because the world should not make sense. Again with the not-quite-hatred talk: I suppose I should write a conclusion. Here it is. No more words. In what ways was it easier? In what ways was it harder? Are you growing in your use of the tool? Is it making you hate it uncomfortable? Is it because you’re in a place where you’re being challenged by it and growing? Are you putting in the appropriate effort to give the tool a fair shake? Is it because you’re outgrowing it? Have you grown out of it for Absolute reasons (it is not a useful tool) or Relative ones (you’ve mastered it or you want to grow in a different direction)? After you’ve outgrown it, can you still advocate it to others who are earlier in their journey?

python Python programming Programming css CSS career Career
0 views
Jason Scheirer 1 years ago

Kicking the Tires on Harper

I’m trying out the Harper tool in VS Code . I’ve been aware of its existence for about 6 months and didn’t think much of it, just another rehash of existing tools. But I’ve come around to the idea that rewriting old tools isn’t a horrible notion: the people who crafted them were no smarter or dumber than ourselves, they were just before us. Rewriting the same thing is a lot like repaving an old road: core infrastructure, but with new materials maybe we can do it better. If anything, we should share in some of the fun the people who came before us got to have. Anyway, so far so good. It’s a little less annoying than piping my posts through , which is about as sophisticated as I’ve managed to get myself in this writing workflow. It’s nice. I thought it was too invasive and preachy on first use long ago, but this time around, at least on first shake, it’s just enough for grammar/spell smell testing. It can run ubiquitously as it builds to WASM and an LSP, so I can learn to lean on it in a wide variety of environments.

programming Programming
0 views
Jason Scheirer 1 years ago

The Wrong Conclusion

A common pattern I have come to recognize everywhere as an anti-pattern: I see this a lot because step 1 seems to short-circuit the thinking behind the rest of the logic. You got a narrative going that you feel compelled to continue to listen to, and assume it’s only going to stay sane. However, just because you have identified a problem does not mean your solution is correct . Stop yourself when a person opens a paragraph and remember that where they are going is not necessarily the only path, or even a sensible one . The example I see most currently is about AI somehow matching human intelligence, usually in the context of LLMs. I’ve decorated these talking points with some straw men that are usually left implicit when usually presented, but you get the idea. Just because you agree with the problem statement does not mean that the conclusion is correct just because it occurs a few sentences later. Here’s an interesting blog post elsewhere on narratives and being misleading as well. Person starts a dialogue establishing a problem, upon which we all agree Person continues on down the same line of conversation, outlining a solution Person comes to a conclusion which is a call to action LLMs exist, and seem to have novel properties that mimic humans producing language Therefore, LLMs are the same as humans producing language Therefore, LLMs have all the other capacities of humans and have been endowed with human nature and will replace us all

writing Writing ai AI
0 views
Jason Scheirer 1 years ago

Throw your Team a Bone

You like little treats , so do other people! Sometimes you gotta give your peers something to make them feel good too. Sometimes this is called the hairy arm technique , sometimes it’s a manifestation of Cunningham’s Law . What you do is this: present something obviously wrong or easily correctable when you make a presentation. This lets the other person feel smart (“you need to fix this broken thing that you made broken”) and changes the focus of the argument so that you get the nitpicking out of the way early and the uneeded but inevitable stage where it happens doesn’t derail the conversation when discussing actual work. Some examples I’ve used over the years: Just a suggestion! Name a new source file slightly wrong in the PR. You’ll be asked to correct the name and then you won’t have to have a long conversation where you get the reviewer up-to-speed about the logic of the method that you spent a week thinking through, they’re satisfied with the feedback they’ve given and do not feel obligated to go further. Just ship a new UI without being provided a design. You’ll get some minor markup corrections but you don’t have to go through a lengthy planning phase.

career Career business Business
0 views
Jason Scheirer 1 years ago

Give Yourself Little Treats Sometimes

Palate cleanser tasks are morale boosts I’ve found that my productivity varies wildly based on the time of year, the phase of the moon, my mental state, the quality of my morning shower thought session, and how enthusiastic I am about what I’m working on . I like to give myself “little treat” issues occasionally to keep my motivation up: issues I pick up outside of the product backlog that are quick to fix, fun to do, and give me little dopamine hits. These keep my average motivation level up, give me a chance to remember why I’m a software engineer, and don’t really eat into the time I already have budgeted to work on other stuff: I’m highly motivated to get it done, so I do it quickly and enthusiastically and it doesn’t make my slog work any slower to deliver. Give yourself treats. Side projects. Little tweaks. Fun stuff.

career Career
0 views