PeerTube is fantastic with its decentralized model that prioritizes user privacy and control. However, it still struggles to gain widespread popularity.
What do you think could be done to enhance PeerTube’s appeal and functionality, possibly even becoming a serious alternative to YouTube?
Personally? Onboarding.
I’ve looked at the instructions on how to install peertube several times now, but its just not worth the hassle at this point. Until I can run it in just a single docker image, without an external database or email service required, then I’m not going to bother.
Its really frustrating, because I really like the project, but I just don’t have the ability to use services like it without docker or podman.
That’s not going to happen. At least as I see it. They pretty much follow best pratices. Lots of webapps use a database, redis, a reverse proxy and sometimes they’re able to send out mails. That’s exactly what’s happening with Peertube, too. And splitting it allows for customizability, different setups, you can maintain one part of it at a time or keep them updated. And not everyone needs to reinvent databases, they regularly better use the official postgres container. Docker is a container platform. If you merge everything together into one large thing, that’d be more a classic install without containers and everything runs on the same OS.
And they let you do that. There are packages for like 3 distros: https://docs.joinpeertube.org/install/unofficial
Or you need docker-compose and it becomes easy to manage the 3 or so containers. But it’s exactly the same for Peertube as it was for all the other services I installed on my server.
Napster only required one app. I don’t see how forcing everyone to use three apps together helps with adoption.
Isn’t Napster a desktop app for streaming? That’s a very different kind of software. Peertube needs to provide content to several people, make it accessible to potentially the whole internet. It connects you with hundreds of other instances and hundreds of thousands of other videos, accounts and comments. All of that needs to be organized, handled and stored somehow. There is just no way around a proper setup. I mean you also can’t build an entire house with just your trusty Honda Accord and a cordless drill at your disposal. You need professional tools for that. And sure, you can build a dog shed with that. But it doesn’t really help discussing “adoptability” of houses if they were to be constructed with just a cordless drill, if physics doesn’t allow for that… The answer is to just use docker-compose in this case. It’s easy to use and does the proper set up.
I’m talking about the original Napster, which would function shockingly close to how PeerTube presents itself. The only thing missing from there original Napster would be a media player.
If that can be done in the 1990’s, there doesn’t seem to be a reason why it can’t be done today.
I think we had several of those. I remember a few other projects that did some peer2peer content delivery and streaming. I’m not sure. Usually with those, people start complaining that it’s “too complicated” because they need to install some software just to watch a video. Or they want to watch it on their phones and that’s kind of impossible with p2p without draining the battery real fast. Or they’re used to videos playing almost instantly (from using YouTube) and now it needs to buffer (more than) a few seconds for less common or trending videos and that’s inconvenient, too.
And the crowd on the internet and usage patterns have really changed over the last decades.
for end users too. the federated model can get a bit confusing at first and can use some onboarding.