It's December Days time again. This year, I have decided that I'm going to talk about skills and applications thereof, if for no other reason than because I am prone to both the fixed mindset and the downplaying of any skills that I might have obtained as not "real" skills because they do not fit some form of ideal.

12: George

I call it a habit of mine that I can make outdated hardware do things it may or may not have ever intended to do. "I" is not quite right in this statement, because much like how my cooking is following recipe and then being surprised that it turns out delicious, much of my computer touchery is following recipe that others have developed, and occasionally deviating from it if I need to for troubleshooting, or to mess about in the thing that the original creator said could be messed with or customized to meet the needs of the person using the software.

Much of the confidence and practice I have with computer touchery comes from having had a machine to experiment on, one specifically designated as the one that if things explode, I can reset back to a working state and then go forward from there. I don't actually want to have to do that kind of thing, because resetting an exploded machine usually means losing progress or having save files get nuked that I want to preserve, but there is a certain amount of risk affordance you can put on your spare machine that your main machine won't get. Spare machines are the best kinds of machines, usually put together from spare parts, or specific small parts that have been purchased to swap out from one thing to another. They're great for people who want to experiment or to learn how to assemble their own machines, or who want to try some other operating system. Everyone should have a spare machine somewhere along the way, preferably one they've assembled or that they've changed some components on, but single-board machines and spare phones are also ways of doing some amount of experimentation, even if you can't change their components quite so easily.

Spare machines are great for working through problems that arise when you do things. Like when I finally saved up enough money to purchase a 3dfx Voodoo2 3D rendering card. I thought I was going to be blazing hard through various games now, with my relatively unimpressive machine (it barely met the specs for Final Fantasy VIII!), but after I'd dropped it in, and tried to boot up my machine, having hooked it all up, the motherboard beeped at me and refused to boot. After a certain amount of troubleshooting, I finally figured out the thing that hadn't been obvious to me at the start: the 3dfx card was a companion to the video card I already had installed, and that other port on the 3dfx card wasn't for show - I needed a specific cable to take the output from my video card and feed it into the 3dfx card, and then after they'd daisy-chained their way merrily through the requirements, they gave me the output I desired. Which made Final Fantasy VIII playable. (And then I would have a bit of a time with the game wondering why I was seeing things like "B6" during Zell's Limit Break instead of the keyboard controls I wanted. Eventually I figured out that I needed to unplug the gamepad that I had connected to the machine and that it was detecting and assuming that I was playing the game on the gamepad primarily. This was back when discrete sound cards were a part of your rig, and they often also had a port on them for gamepad input.)

So I've done a lot with spare machines, tinkering, experimenting, and trying things with them that I wouldn't do to the "family computer" and that I wouldn't do to my work computer. My "spare" machines have proliferated in my adult life, as I continue to move things around and new machines enter my life. But also, so have my appliance machines. Instead of a full tower desktop running in the bedroom, I have a singe-board machine there. Much quieter and less of a power draw, still does all the desktop environment things I want (as well as some other things, like allowing me to remotely control the TV it's attached to, the one without a working IR receiver.) I definitely had a second machine for much of my time in the bad relationship, and for a time, I used a cell phone dock and some nice cabling to turn a single-board machine in to much more of a laptop. It could at least run XChat at a few other things at the time. A secondhand Surface I'd gotten from someone served as my "work" machine during the shutdown, before receiving an official work laptop. (That Surface eventually suffered from the batteries trying to burst forth from the casing and had to be retired, but we salvaged the SSD from it for purposes.) And I kept two desktops working side-by-side as soon as I reclaimed my house, so that one machine could be used for media purposes and Windows stuff, and the other could be used for Linux purposes and handling all the things I was doing with Android phones and other things where it turns out to be easier to do things from a terminal on a Linux box than it is in Windows. And since nothing "vital" was on the Linux box, I could experiment with it, change distributions, and otherwise use it as the spare that it was. This combined with the experience I had from using Linux as a driver since graduate school to make me comfortable enough to use Linux as the driver on my main machine as well. Something that started because one of my classes meant learning a little Ruby on Rails, and it's way easier to run a local Rails server from Linux than Windows has now come around to being a machine that I can watch streams on, game on (all hail Proton), and otherwise continue to give life to, since I wanted a machine that I could buy and hold as much as possible, instead of thinking I needed to change it from one thing to the next.

After purchasing my first phone with an aftermarket OS on it, I have basically been doing the same thing to every phone I've owned since, especially because those phones would otherwise have reached the limit of their manufacturer OS updates, and instead, I can merrily roll along on old hardware until the things physically give out themselves. They do sometimes complain when I try to do things like play Pokemon Go on them, but it's fine. And by the time I have to be in the market for a new phone again, so many of the flagships of a previous time will have come down in price to the point where I might consider them, or consider asking for them as holiday gifts from people who like to spend money on me, despite my clear failures at capitalism.

So as a cheapskate with regard to technology, it's always nice when I can take the old things and make them run smoothly and swiftly with new software or by respecting their limitations enough to not tax them with software that's not suited to them. (One of my next projects, whenever I have actual need to do so, is to do some exploration of software that can be run from the terminal, so that my spare Model B won't feel left out from the fun and can contribute to some important part of house functions.) That cheapskate nature meant that when I got to examine the original model of Chromebook, and was told that I could do what I wanted with it, since the original model Chromebook stopped receiving updates at Chrome 65, I consulted the Internet, and while there wasn't much information available, there was a website that was dedicated to the prospect of converting such a Chromebook into a fully-fledged Linux machine by replacing the firmware on it with a specific kind of compatible BIOS, and then from there making it possible to put a Linux on it. (It's a very nice machine, actually - 64-bit, a couple gigabytes of RAM, and a 5GHz-compatible network card internally.) Well, I should say the website existed at some point in time, but didn't actually do so at the moment I set my mind to it. Thankfully, the Internet Archive had crawled the entire thing, and I could download it into a zip file, giving me the opportunity to follow the instructions and examine the pictures. I was initially stymied by the first instruction of turning the developer switch on, because I couldn't see a developer switch in the spot where the pictures said it was, but once I discovered that it was behind a small bit of electrical tape, we were ready to go. (That piece of electrical tape would come in handy later, as the thing that was used to disable the write protection on the firmware on the laptop.)

Again, low stakes project, no worries if things didn't go according to plan, because it was otherwise not being used, and great potential for use if it succeeds. Which it did! I followed the recipe exactly as the website archive instructed, got the new BIOS in it, and then put a Chromebook-related Linux on it, boggling the developers of it, because their Linux was not meant for a Chromebook that old. They weren't even sure it would run on it, despite me showing up with such a thing. Eventually, I scrapped that project, since it hadn't updated in a very long time, and instead went with the distribution that was powering one of the "spare" work machines that had been designed with Windows XP in mind and had fallen out of use as a mobile reference tool. I had been using those machines for all kinds of shenanigans and other material that official machines were not being used for, and they have served me well, even if only one of the original pair survives.

That Chromebook still runs BunsenLabs, and does so wonderfully. So long as I don't try to tax it too hard by running too many tabs on it, it rewards me with snappiness and speed, and most importantly, a system that can be updated and kept patched against security vulnerabilities. (When the second of the pair of netbooks finally refuses to boot, this Chromebook will likely take its place as machine-outside-of-boundaries.) And having done it once, when I was alerted to the possibility of getting another Chromebook of a later parlance for a little bit of nothing and doing the same thing to it, I jumped at the chance, and with a similar sort of process, and using some scripts developed by others, I now have a compact and useful Linux laptop that I do a lot of composition on, and that I can take with me to events like the local GNU/Linux conference so I can do interactive bits, or run programs, or just hang out in the chat rooms and post on social media my running commentaries about the sessions that I'm listening to. I've also used it as a presentation machine for such things, when I'm the one doing the presenting instead of listening. After trying to run a form of Arch on this Chromebook, and eventually running into the problem of install creep and strict size limitations (as well as the nasty tendency for it to hard freeze at some point when it ran out of memory and swap), I put BunsenLabs on it during this last update cycle, and it's much happier with me and seems to function better. We'll see what happens when BunsenLabs finally makes the jump to a Trixie base instead of a Bookworm one, but I feel pretty confident I'll be able to get all of that to work, and it'll be nice to have old hardware running modern systems.

I'm doing this because of the work that other people have done to port boot systems to Chromebooks and other machines, and to automate the process of installing things to the right places, and the people who build and maintain the packages and the installers so that all I have to do is download the image, run it, install, and then run the update commands on first boot to get to a system that's ready to work. It doesn't feel like computer touchery to do this, because it's just using other people's stuff, but there's the tale of knowing where to make the chalk mark as one side of it, and the other being whatever arguments you want to bring to bear about how "not invented here" is terrible as a practice, and therefore if someone else has created the thing that you want to use, use the thing they've created and spare yourself the turmoil. (Or, in my case, use the thing because you couldn't create it yourself anyway, and be grateful to the people who are using their time and knowledge to make it so that you can do this thing.) Doing things in userspace is still valid, and as an information professional, a lot of my skills are in finding and surfacing the thing that will be useful for the situation, rather than in trying to create the thing completely from scratch, or in trying to get the person I'm helping to do the same. The world is too large and complex for any one person to understand, or even to necessarily understand the entirety of their discipline, and so it should not be a mark of shame to rely on the work of others and to trust that their work will be accurate and not malicious. (It just makes me feel much more like a script kiddie playing in the kiddie pool instead of a Real True Technologist, even if this is another one of those situations where if you press me on the matter and start making me tell stories and explain myself and solve problems, the claims I'm making look flimsier and flimsier, a fig leaf of modesty because I'm still afraid of the reaper looking for tall flowers.)

There's a lot that I have done, and that I can and should justly consider as achievements and Cool Things. Doing things like December Days and the Snowflake / Sunshine Challenges and other such writing prompts are my way of indirectly getting at those and showing them to others. If I came out and said it directly, I'd be worried about it sounding like boasting or penis size comparison, and someone else would come along to put me in my place. But if I'm talking about how there's a wealth of software and instructions out there to extend the life of old technology, and I'm a cheapskate who's willing to invest the time in following those instructions and prolonging the life of that old technology, it doesn't sound like I'm boasting about anything other than getting some extra cycles out of my machines, and that is something I can safely be proud of. (Why? It's not saying I have any particular skills or capacities, just that I know where to look and how to follow recipes.) Indirectness is one of the best ways to get me to show you my actual potential and abilities, and I can do it to myself just as well as anyone. Full understanding may need a little bit of either reading between the lines or knowing me well enough to see what I'm doing, or to ask the right question that makes me squirm or tell stories. (Please do.)

It's December Days time again. This year, I have decided that I'm going to talk about skills and applications thereof, if for no other reason than because I am prone to both the fixed mindset and the downplaying of any skills that I might have obtained as not "real" skills because they do not fit some form of ideal.

11: Geocities

I made my first website on Geocities, and that probably tells you more than you wanted to know about how old I am.

The concept of Geocities was pretty genius, though. Less so the conceptualization of Geocities as divided into various neighborhoods, loosely based on what the person signing up for Geocities might make their website about, as it turns out that we didn't really need to map physical space onto virtual space. But the idea, basically, of offering someone a few megabytes of space to build whatever they wanted to, so long as they could provide the code (and so long as they ran Geocities' ads on all of their pages, because ad revenue is still the way that a lot of places think is the best way to get money - that, or venture capital.) There was no need for buying your own domain, or for learning how to administer a Linux system, or any of the other highly technical obstacles that would prevent most people from showing their own pages to the world. This was before blog software replaced the idea of having a personal page, and before content management software replaced them both. And so, people went off in every direction they could, bounded only by the restrictions on what the code could do and what things were permitted by the host. Things past what the sandbox of Geocities provided would be the kind of thing that you would get your own domain and hosting for, and therefore you'd learn all those things you weren't learning immediately by using Geocities.

The Web was not quite corporatized, and was not quite in the place where slick Javascript and CSS were considered standard parts of the Web experience. What you received, essentially, was an entire hodgepodge of material, based on how much the person creating the page wanted to learn the coding and how much the person making the page just wanted to get the content out. It was a time of great personality in pages, even if it also sometimes meant choices from the CGA era for text or backgrounds, or that you had to work with someone who didn't believe much in the paragraph tag, or the idea that a web page was designed for a specific resolution and wouldn't look right on any other resolution. Or that it was meant specifically for one browser over another, because it used tags that the one would recognize and others would not. It was a time of guestbooks and webrings and, I strongly suspect, an awful lot of fic archives. If I had been the kind of person who wrote and put their fic online at the time, it might very well have been a windfall to have 100MB of space to put all of my formatted HTML onto so that my epics would be readable, and possibly, I might collect the fic of others, too. It is also the era where search engines actually crawl and search, rather than some other purpose, and they would obey the instructions given to them in files like robots.txt. Discovery was still tough, of course, but people found ways of doing it all the same, through hypertext.

At that time, though, I used the space I had on Geocities as a sandbox to learn all kinds of things about HTML, and how to make links, and show images, and make images into links. I may have picked up a little CSS along the way, so as to make things more easy to control globally, and as well as to do things like use image files as my background for the page. Mostly, it was there as a personal page, constructed haphazardly, with plenty of animated GIFs, pictures from the Internet, and links to other places that I thought were interesting. A professional web designer's nightmare, in a phrase. But mostly it was articulating to myself what I wanted to do, and then looking on the Internet to see if someone else had done it, or if there was a keyword to zero in on, then consulting a reference work to find the appropriate tags and the appropriate place to put them, and then tweaking it until the rendered page actually looked and functioned the way I wanted it to. As I learned more, I put more of that learning into the pages that were there, sometimes adding new things, but often, refining what was there so that it was more specification-compliant and easier to handle later on. Even on the site that I have been neglectful of maintaining that holds my professional CV and as much of the presentation slides and commentary as I have stuffed into it, most of what I'm doing there is following my own template after having figured out the thing I wanted to do. At this point, I believe I've reintroduced frames to the site, because I don't want to have to recode the entire navigation into each page. It's likely the best solution I have for navigation involves Javascript in some way, but I am also the kind of person who wants their site to function properly without Javascript, and therefore I would have to learn how to encode a proper fallback from it.

This approach, "figure out what I want to do, then consult the reference works to figure out how it's done, then see if it actually does what I want, then refine it until it does" is probably much, much close to the actual process of people who code for a profession or a major hobby do, rather than the idea that I might have in my head of someone who, when presented with a programming problem, simply magicks the thing up out of the ether in a flurry of code and it works. (Well, hopefuly there's a test suite in there, too, but…) In the same way that I have a persistent belief that "real cooking" is not "following recipe" but instead "making delicious dishes from a basket of ingredients and your own knowledge", I have bought into some of the belief that "real coding" does not involve following recipe or template, unless you've developed the template yourself, too. That particular belief always gets mugged every time I start trying to get Home Assistant to do something new, or I decide that automation is the best way to do text string manipulation, because I can see how to do it in an automated manner, or when I need to push a change to a great number of records in a work system so that nobody has to do it by hand. (I tested that one on small batches first, because nobody wants to intentionally wreck production.) Or when I'm making changes to my professional website pages. Or the project that I built in one of my graduate school classes to pass a foundations course. The UI was terrible, but UI wasn't something I needed to think too hard about over functionality, and it was something I built for me (as well as an assignment).

For as much as I think of myself as a user, rather than a coder, if you start asking me what I mean by that, or start pushing on my self-imposed boundaries about where "real coding" starts and stops, you'll find all kinds of interesting treasures surface up as I start telling stories or start trying to justify how this thing that I did isn't really the thing it is, because it's someone else's code, tweaked to do the thing that I want it to do. Or because it's not elegant, polished, and efficient code like someone who knows what they're doing would turn out. I have ten thousand excuses to avoid taking credit for anything, or to admit that I might be practiced at or knowledgeable about something. The experiences of my childhood, and the mockery that accompanied when the supposedly perfect child made a mistake, has me perpetually looking out for the scythe and the reaper wielding it, the one ready to cut the tall plant for daring to peek its head above the others. I would say quiet competence is my sweet spot, except I also want to be recognized for the quality work that I do on a regular basis and not have it just be the expectation of me, unworthy of further comment other than "meets standards."

The older I've gotten, the more I realize that an excellent way of getting me to approach a problem or try to figure out how to make something work better is to present it to me as a sandbox, a puzzle, or some other thing where there's no pressure for the thing itself to be perfect or that it needs to be turned around in a short time. Something that is being solved for its own sake, and not because you have to provide the solution to a sudoku puzzle to your past self so that they can get out of the predicament they're in and survive long enough so they can become you and give the solution to themselves and generate a stable time loop. The less stakes there are in the situation, the more I feel like I can bring myself to bear on it, and not to get caught up in the twin weasels of "must be perfect to be seen by others" and "anything that fails will be viciously mocked." I realize this is maladaptive, and most other people do not suffer from these fears in their own lives, but it works, and therefore I do my best to make things as non-important in my head as I can, simply so that I can function in the moment.

I demonstrated that at work today, actually. There was a monitor at one of my locations that was rotating too easily in its housing, and so I tried to figure out what the problem was with it. Checked the screws and the like, and they were holding, and eventually, I concluded that, once I'd gotten the monitor off the clip that was holding it in place, that the bit that attached to the monitor and the clip was too loose, since I could spin it with my handss. There was a pair of pliers in the tool chest at the work site, so I tightened things up, and when we re-clipped the monitor on, it stopped wobbling so easy.

Thanks, Pops. Not just for the whole "can use hand tools" part, but for the bit where you encouraged me to think systematically about problems, to work methodically through possibilities, and to come to conclusions and test them to see if they're correct. You did exactly the thing you were supposed to do to help me achieve not only answers, but processes and analysis. Even though I really just wanted answers at the time, rather than to be led through a process of figuring out where my mistake was, or where I had overlooked something, or whether an assumption I was making was actually correct. It serves me well, just so long as I keep thinking of it as a puzzle rather than something of importance.

But also, if you are interested in the same sort of spirit, try Neocities, and maybe you can start building your own personal page or interest page or another fic archive.