It wont be, the scale of service and ease of revenue sharing will keep it as the king of video distribution untill Google kills it (like they do to all their products). FOSS projects and self hosting can not accomodate a viral hit (the slashdot effect), and also a self-hosted project like that would have to find a way to make money for the host to keep the lights on, and even Youtube fails at that one.
scale
Who does scale really benefit, though? I don’t see how it matters from the audiences’ point of view. Say I watch Youtube for fishing videos - all the competitor needs to do to attract and keep me is offer fishing videos. I don’t really care that I can’t watch music videos on it, or cookery, or make-up tutorials, etc.
The preoccupation we have with scale should be re-examined when it comes to video distribution. A combination of user-friendly banner advertising, modern codecs, and P2P hosting should go an awful long way. If I knew ad placements provided material funding for a video site/community I loved, I’d whitelist the URL.
Video needs fragmentation.
I think scale matters because almost no person is as much of an island as your example fishing video guy. I actually have noticed almost the opposite in most people I know, YouTube is the default place to get entertainment. Across all their interests.
From both sides the network effect might be strongest with YouTube, the creators can’t leave because YouTube has virtually all of the audience, and consumers don’t want to watch singular people on other platforms because on YouTube you can stumble over interesting videos and all the people you like to watch are already there.
The only way I see for other platforms to actually grow is forced interoperability, as in videos of other platforms appearing in the YouTube frontend. Which Google would never do so the government would need to force them.
Yep, my entertainment is 90% YouTube and the rest some show. On YouTube I find everything: from a dude that does reviews of air filter for cars to somebody explaining some obscure Japanese woodworking techniques to the omniscient Indian dude that explains complex programming concepts. If there was fragmentation I wouldn’t be even able to find stuff, like in the early days of the internet that you knew the website existes because somebody shared the URLs in some usenet or some forums, before search engines became a thing.
You make good points, but I still think what I envision would be able to attract enough people interested in specific hobbies, without achieving anywhere near Youtube’s scale. I’m thinking of a scenario where the video platform is more an extension of a web community, such an an old-school forum, rather than a straight video host where the primary aim is to gain any engagement whatsoever, and where (let’s face it) all engagement is generally fungible. It’d be something member-funded and run, like good torrent trackers, and the content is an interest ‘ecosystem’ - so not only fishing content, but fishing gear coverage, and camping and hiking stuff, and meat prep and storage, and boating, etc.
This couldn’t be any worse for either creator or viewer than what YT subjects them to. There would be no having to optimize for an opaque algorithm. The pressure to self-censor would be greatly relieved. Monetization scope and content guidelines would be accountably managed - ie. by the community itself. Creators would still have their Patreon/Liberapay/etc income streams. The platform can place the odd banner ad too, like 4chan.
I wonder how much convenience and (perceived) income security is a passionate creator prepared to sacrifice in order to start exercising power over Youtube by uploading elsewhere? We all know creators hate the place…
The benefit of scale is it attracts the creators. The people making the content we want to watch aren’t all doing it as a hobby, so the chance of attracting a large audience needs to be there. Otherwise they won’t come and the site is populated with really random, low-choice stuff.
All you need is a federated link aggregator like lemmy/mastodon with a UI made for videos.
You post a link to whichever video hosting service and attach a bunch of metadata (thumbnail, description, tags) and the comment section is built in already for each post. Nobody cares where a video is hosted, as long as they can follow creators and topics.