💾 Archived View for rawtext.club › ~winter › gemlog › 2023 › 4-15.gmi captured on 2023-06-14 at 14:13:54. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2023-04-19)
-=-=-=-=-=-=-
You've noticed search getting less useful, but it's hard to tell for how long. Google's top results are full of garbagey sites; you're adding +reddit to every query to get what seems like the last non-astroturfed source of opinions online, and who knows how long it'll be before that's gone as well. Your days in front of the screen seem repetitive, predictable, tightly circumscribed. Maybe these are all related. Maybe not. But surely you've wondered if things have always been quite this bad.
❦
Search has been bad before, but in a different sort of way. Search used to have poor coverage, indexing a very small portion of the websites that were out there. Querying was imprecise, often accepting just a single search string applied against the text. In this void, the link directory proved its worth. Hand curated, made by people, they represented a selection of pages on a topic, or maybe just what a particular person was interested in. And they also suffered from poor coverage. But the quality was there. Sites were selected not for opaque SEO reasons, but because someone liked them and decided to add them. Link directories often caught things search engines missed. For a while, they were absolutely good enough.
❦
I first came online in 1994 or 95. I'm not exactly sure which, but I know it was somewhere in grade eight. A couple of my friends had good computers (at that point, mine was almost ten years old). They had modems. They showed me how we could dial phone numbers and connect to bulletin board systems, or BBSs. This was amazing. In retrospect, you'll never meet a weirder bunch of people than those people who wanted to dial boards in the mid-90s. We made lists of phone numbers. Printed them on dot matrix paper, passed them around, annotated them in pencil and pen. Strike through: dead board. Asterisk: something special.
❦
That summer I went to the nearest computer shop. Put down $30 and bought a used 2400 baud Supra modem. Figured out how to put it in my computer. Fiddled with the on-board jumpers to set COM port, IRQ. Eventually I got the configuration right. My life would never be the same.
❦
BBSs didn't last. They couldn't. At the same time I first came online, the web was just getting going in its nascent form. There were other browsers but the one everyone remembers is Mosaic: if you had a good enough home computer (let's say a 386 running Windows 3.1), you could view web pages. There weren't many of these. In 1995, the number of websites was barely more than 20000. But more and more people were coming online. They were learning how to make their own pages. HTML was relatively simple then. Headers, bold and italic, links, colours if you were fancy. CSS hadn't been invented; relatively few people knew how to write JavaScript. You could make something good enough (or view source, steal, and edit to fit) fairly quickly, hit refresh in your browser and see the results immediately. You could put it on a shell account, or upload via FTP or a web form, and be proud of it. Eventually, someone might find it, and add it to a link directory. People might bookmark it, and return periodically. Regardless, it was a small shoot. Something to tend. Over time it might grow. You might add to it. Hobbies. An About Me page. Maybe a link directory of your own.
❦
When I first tried the web, I was underwhelmed. I was at my mother's office, using her 386 and accompanying high speed university connection. I knew to try a search engine. I went to webcrawler.com. I can't remember what I typed in, but I remember the results being garbage. I didn't think the web was going to be anything special.
❦
High school was not a happy time. Some people had it worse, but I didn't have it good. I'll spare the details but only because I'm never sure who's reading. Needless to say, the modem's shriek was a release for me. BBSs might've been dying, but they were good. My daily escape. Eventually, though, I surrendered to the inevitable. I got online, in the current, internet sense. I found the web. I got an email address. Opera and Pegasus Mail. I started exploring.
❦
It's easy to forget that the web is not the internet, though for many people, it might as well be. A collection of various electronic protocols and services including the web, email, gopher, Usenet, and new protocols such as gemini, the internet allows us to exist in a second space, detached and distinct from the physical sphere in which we live our lives. It lets us be anyone. Meet anyone. For a lonely kid from the prairies, it was a way out. A lifesaver.
❦
My friends were making webpages on GeoCities so I made one too. Green text, black background, hr gifs and skulls. I had a grandiose alias. It was my Domain. It was sick. A couple years later I fumbled my way into a subculture of journallers. We wrote down our lives, linked each other's sites, spent long nights fussing in notepad and Paint Shop Pro, getting our sites just right. Left messages in each other's guestbooks. The effort we put in for our tiny audience, the strange openness that comes from something being technically public, but hard to find. It's not a stretch to say that some of those girls knew more about me than anyone in my offline life.
❦
Over the years, that space we collectively refer to as "the online" became more populated. All the people who bullied you in high school? Yeah, they're on Facebook, getting online only when it became as easy as tap, tap, tap. Right now they're anonymously scrolling Pinterest and Instagram, crafting careful and beautiful lies for the consumption of others. Posts about the vacation; pictures of the kids, of the husband who's been working longer days and who they think is cheating on them; pictures of their tropical paradise, filters applied, care taken to smoothe wrinkles and remove any visible veins. We all know platform curation algorithms will send the posts to others. But what about the details? We don't control who can see it, nor can we see how many people it's reached. Those are carefully-guarded trade secrets. We don't get to think about the how.
❦
Before all of this became the norm, we had to rely on our own wiles to find things. No platforms, only people. And more people meant more pages, back when that's what people did with their time. Even in the era of small disks and space quotas you'd actually hit, there were more pages than you could hope to read. A good page, an interesting page, a regularly updated page - that was worth something. That was saved. We kept lists of links. Carefully-categorized bookmarks. But these required tending. Sometimes the links broke. Any given link directory on a personal website, unless it was meticulously maintained, was probably half dead.
❦
It's hard to overstate just how good Google was when it started. At the time we were shuffling between Lycos and Altavista and other also-rans as we got results that appeared to be ever-so-marginally better. And then Google came in and...it was fast. Results in a second or less. And it was _good_. It seemed to return only relevant results. This was it, we thought. Search's final form. It couldn't get any better. Google was the goose that laid the golden egg. Surely they'd never screw this up.
❦
The need to find things has always been one of the central concerns online. Years ago it was collectively decided that corporations were best-positioned to deliver us these results. This was, in retrospect, an enormous mistake.
❦
Taking Notes for a Bitrotting Internet
A week or so ago I was reading Clarissa's gemlog on notetaking in this crumbling online world. Broadly, fae talks about how we can no longer really trust search: not just because pages can die, or become inaccurate, but because the rise of large language models (LLMs) and their integration into absolutely everything in the last year. These forms of generative text and images, which are taking over the term "AI" in the imagination of the public, add a new kind of incoherence to our lives. LLMs are not search engines. They're essentially sophisticated methods for predictive text and images, and if they think something sounds reasonable, if it's highly weighted, they'll emit it. Truth isn't really a consideration, just the coherence with the data they've been trained on. And while it's bad enough they're being positioned for use in search, they're also now being used to crank out reams of SEO-gaming pages, formerly the realm of the poorly-paid freelance writer. Soon the web will be flooded with noise to an extent we can't even imagine now. The prompts that generate these pages will instruct the LLM to tick particular boxes, to have SEO features allowing them to place high in search results. And these pages, in future, will get fed into the next generation of machine learning data sets.
❦
I realize there's an irony in complaining how bad things are right now. Things can always get worse. They definitely will.
❦
I'm passionate about this weird, stateless space because it's so deeply important to me. I've been online for nearly thirty years. I've downloaded files via X-Modem and written them to 5.25" floppies, had flamewars on Usenet, kept websites and journals, met a lover on a forum for fucked-up teens. And then there was one of the few what-ifs of my life, a friend of another friend who came to visit her the summer after grade twelve. They knew each other through fanfic sites, and began chatting on ICQ; and that summer, my friend introduced me to her, the girl from out east, the girl with the intense stare, the girl with the long red hair. It was immediately electric and after she went back home, we kept talking for years after that. We could never seem to arrange a visit. We were young. We were broke. After that summer, we never saw each other again.
❦
To reject the online would be to reject so much of the strange and serendipitous beauty of my life - as well as the accompanying mess, which is just as important.
❦
It feels like the web is dying. It didn't used to be, but it is now. The rest of the net is (largely) good, but sparse, perhaps spared from ruin because the laser eye of capitalism deems the audience too small, too anti-commercial to throw up thousands of garbage sites about, I don't know, Finding An HVAC Contractor In Des Moines in 2023. But the rest of the internet, the outer wilds, can be difficult to get into. How do you find interesting people? How do you find a trail, or blaze one for others?
❦
I remember hearing about geminispace and being fascinated. When I finally found my way here, I knew I was going to create a journal. And I considered what else I wanted to add to my capsule. The old About Me page, though largely empty due to "winter" being an alias; some writing, and of course a links directory. When I first started exploring using lagrange, it was a little overwhelming. Not because there was a firehose of options, as on the web, but because there wasn't. But even in the general paucity of sites and people, there was still enough that I didn't know where to start. So in the absence of guideposts, I made my own. Found some gemlogs I liked. Noted them on my links page. Read some interesting articles or posts. Added them too. But this had a very expected structure: people; articles; random/misc. And this works. It was organized, and it's good enough. But I started thinking: why am I placing one category after another? One article after the next? You could argue there's an implied connection by having one item follow another. So why was I ordering them this way?
❦
Looking back at the links pages on my old sites in the Wayback: everything uncategorized. One link after another, no real connection between them except that they're sites and people I found interesting. Good writers, fascinating lives, people whose paths I wished I could cross in a meaningful way. Looking at my links page today: good writers, fascinating lives, people whose paths I wish I could cross in a meaningful way.
❦
Clarissa argues that a note-taking system is essential. I agree. Fae also recommends using open-source software, being able to classify information by tags/metadata, that all this should be stored in a format open to text searches like grep/etc. All good. But what I'd like to recommend is that, as the modern web has become more homogenized, more overgrown, there's an opportunity here to get back to what made things good so early on. That we should collect the links we find interesting, and make them available for others; that in offering these, we should be less concerned with classification, as in the traditional link directory, as curation. If you were growing a garden, what would you plant and where? If you were intending something for others to find, what would you leave?
❦
At my old journal site, the last one I kept before I migrated my life to LiveJournal, the links page lists Forum 3000; then a Swedish anarchist hacker, then a journaller, a couple angband variants, and then more journallers. The lack of classification gave these something of an unearned proximity, but it also said that these were all important to me, and perhaps they could be important to you as well. A suggestion on my part to give any or all of them look.
❦
I still think about the people I fell out of contact with from that era, and wonder what they're up to now. There as here, we went by names that weren't always real; we rarely used last names at all. Jess, Mary, Emilie, Véro, I hope you're all well.
❦
Yesterday I started restructuring my links page. I've done away with the idea of a link directory (with its implicit theories of classification, nested headers, maybe even ordering by title, etc), and I'm now tending a link garden. Placing complementary things together. The plan to come back and check on it regularly. Weed the deadheads. Plant new flowers. And hope that perhaps, without my knowing it, other people might find it useful too. If you're not scrolling social media, you're here for a reason. Hello from the outer wilds. Welcome. Maybe we can find our way through together.