The Joy of Discovery
I've been sparked into writing mode by two things:
- The issue of people leaving Mastodon because they find it hard to discover things.
- People on YouTube mentioning that they loved "finding" my channel.
These are sort of connected, because there's something those of us who try to bring back a little of the Old Weird Web also bring back from that era: the Old Web was not discoverable in the same way.
The modern web approaches discoverability in the same way as broadcast TV. You sit there in your chair, prod whichever sequence of inputs activates scrolling on your device of choice, and interesting things are shown to you. There's a lot of fluff about tailored feeds and clever algorithms and probably these days something about AI, but mostly it's doing the same old trick the gogglebox was forced into by the nature of its medium: "a lot of other people like this sort of thing, so we'll show it to you in the hope you also like it".
This is why if you look at monstrously successful YouTube channels, the ones who don't spend 20 minutes talking about bus simulation, you'll see they optimise heavily for that, "this is like the things a lot of people like". As a result, some are increasingly adopting the tactics of a certain rapid-fire, attention-grabbing style of broadcast TV. From thumbnail to script (created in that order, to match the viewer journey) it's optimised for metrics that put it front and centre in those "for you" feeds, making it a "watch next" that everyone watches next rather than surfing for something else.
Passive
The Old Web did not have this kind of passive discovery. If you sat in front of your screen doing nothing but scrolling you'd just move the AltaVista portal page up and down. It wasn't even infinite, we had this thing called the bottom of the page which signalled "no more information", rather than waiting for you to realise you haven't seen anything but AI-generated grey goo in the last few minutes.
Ahem.
We treat this as a problem to be solved, but I think it's the opposite: this need to go out and look for things was crucial to the feeling and enjoyment of the Old Web. Everything required active discovery; coming up with search terms, clicking on links, subscribing to newsletters and even asking your friends, "seen anything good lately?" Often you wouldn't even find what you were looking for on the first link, the search engines being so primitive and the newsletters naturally time-constrained so you'd click whatever looked promising and see if it had links anywhere else.
There's a feeling of reward to this kind of active discovery. I get to feel it from the other side of the screen, when people tell me, "I tried searching for this old game I used to play and I can't believe someone made a video about it". It's the satisfaction of putting in some work and getting something in return which made using the early web such a unique and addictive experience. Realising that if you added "midi" to a search for "star" and "wars", you'd find someone across the globe who'd transcribed the themes from all three films into a 30-40KB file you could listen to in glorious wavetable synthesis, should your computer be so equipped.
Interact
This isn't all, though. The need to actively discover, to search and click and navigate, made people interact differently with the medium. Remember broadcast TV? When it only had five channels it could show only the most popular content. There was some rudimentary segmentation; a "big tent" channel with broad appeal, a slightly fusty and educational one, an unashamedly populist one, one with a more alternative and subversive bent, and one for people who liked the populist stuff but more cheaply made and with a layer of analogue fuzz. But outside of some weirder things in the small hours it largely showed what was guaranteed to be popular with as much of one of those big segments as possible.
You couldn't make a TV programme and hope. You had to get it in one of those five available boxes. And that's the same with the Internet of passive discovery. There's millions of creators, but while there's a little bit more segmentation you're still chasing six boxes for recommended videos, three boxes for recommended images, or one of the 300 allotted Tweets some poor person is desperately trying not to use up on adverts for backpacks.
If I make a video about bus simulators and there are already 6 others, even with the most accurate segmentation going and the most perfectly focused bus simulation audience, I'm not on that page unless mine happens to be deemed more popular and engaging than the others. On some level I'm in competition with all those bus simmists. Every audience member I give away to them is less clout for me, more chance they are the ones who get that coveted Watch Next placing.
It's a very different world to the Old Web. While we had search engines, they were largely OK at best and much of the actual discovery came from clicking links - the art of using the search engine to find something a little interesting, which linked to something more interesting, which linked to something very interesting. There were newsletters, directories and webrings dedicated to showing you things you might find interesting; whole sites dedicated to you clicking on a link and going to another, different site.
(Old users may remember the days when Reddit only gave karma for links taking you outside of the platform)
That's the thing with active discovery. You don't just get the joy yourself, you want to share that joy - the "hey, I found this" moment. It changed the way in which people made things, because if you wanted something to be seen by a lot of people you only had one real option: make something that inspires such devotion other people will link to it on their sites. The Old Web threw up weird and niche content so much more often because that was the stuff people cared about, the discoveries they wanted to share rather than go, "huh, I guess I feel precisely 1 entertained by that, what's next?"
Used
This is a thing we're not used to, because the centralised Internet of the 2010s swept it all away and most of us are trying to rediscover these norms, if we even care about bringing that old frontier era feeling back at all. I feel like too many are trying to bring it back but still treating the tools which could do so like a modern platform. Don't give away your clout. Don't link to things which will send people elsewhere. Don't even give the people who are making good things encouragement, because they might be future competition.
This becomes a problem when the tools themselves are designed around those Old Web assumptions, that discovery happens from people sharing things they like and linking. It's strange, but in some sense I think we're returning to those constraints. The application of passive "things you might like"-ism to search engines has made them arguably worse than those primitive days of Lycos and Webcrawler for finding information on all but the most banal of topics, and recommendation algorithms have become so tail-chasing that relying on YouTube to suggest what to watch next is like picking one of those five channels and accepting whatever comes up. Probably the cheaply-made one with the bad reception.
So if we want the Old, Weird Web back in some sense, we need to bring back the Old, Weird Behaviour; where you really didn't care about being "big" or "influential" (not like there was any money in it!) but making the best possible thing you could in your niche area of interest and sharing everything you liked, because it only worked when everyone shared their personal joys of discovery.
I suddenly feel like I need a "links" page. If only I could find some things to put on it.