OK we need a bit of nuance there: I do reckon there are many niches that don’t need to cater to a general audience to begin with. Specialised conferences on InfoQ are a good example: their audience is very specific, and is best reached through aggregators like HN or /r/programming. Another example is online courses, that are not casually watched.
On the other hand, some important topics are aimed at a general audience: news, politics, scientific popularization, infotainment, entertainment… People don’t actively seek out those things, they stumble upon them and select what they like… with the help of the "algorithm".
Whether that’s a good thing is another matter. The way YouTube works is eerily close to doom-scrolling, and I’ve lost a lot of time there. My point is, to even have a chance of reaching a general audience, right now the algorithm and people’s behaviour is such that the only place is YouTube.
Now I’m not saying that everyone deserves to reach a wide audience. For one that’s flat out impossible (10 minutes watched by 100K people means over 13 years of total watching time), and most content is either niche or crap anyway. High quality content however, that makes a positive impact on the world (for instance by helping, informing, or entertaining people), does deserve a chance.
Does it deserve any particular way to the top? No, of course not. But I do submit they deserve at least a fair chance of being widely watched. And right now, again, that only chance comes from YouTube.
---
As for what we should do about it, I see two routes. One is regulation. We could officially recognise that YouTube basically holds the only meaningful key to an important kind of public discourse, such that any video they refuse to show is effectively censored. This makes them a public utility, and should be treated as such. I’m not sure what that should entail for the search & suggestion algorithms, but it sure means that taking down a video is an infringement on Free Speech, and so should be approved by a judge.
Yeah, that will never work out. Too many videos to process, not enough judges. So I think we’d much better take another route, if at all possible: find a way to severely reduce YouTube’s market share, and have actual competition between many platforms. That way if someone is kicked out of one platform, they can still use another. Or we could expand self hosting, or multiply the PeerTube instances…
What I absolutely do not want (though unfortunately it looks like we’re headed there), is the kind of regulation where YouTube is mandated to filter videos, in such a way that the only solution is an unsustainable level of automation with lots of false positives, no due process, and no way to appeal (like right now in fact, only it’s official).
---
A better regulatory route would be giving platforms a clear choice: either behave as a utility, which means utter neutrality: no integrated search or suggestions unless it’s demonstrably neutral, no filtering, and no arbitrary take down; on the upside, if they happen to host illegal content, they’re not responsible until a judge tells them to shut it down.
Or, retain the biased searches and suggestions (they do have value), all the filtering they want, arbitrary take downs and bans… but then they are treated as editors, and become responsible for everything that happens on their turf. If someone manages to publish some illegal content, they are penally responsible and may in extreme cases go to prison.
Either you’re a carrier, or you’re an editor. Under that rule, YouTube would either become a mere host, at which point we don’t even care about their market dominance (though I suspect their market share would drop as they’re deprived of most of their network effects); or they would become an editor, and that’s so unscaleable they’d need to shrink like hell to be sustainable. And again, they’d lose market share and we’d get the diversity that is needed to make sure that being banned from one platform is in practice very different from actual censorship.