What 2023 will bring for PeerTube
joinpeertube.org
joinpeertube.org
May I ask - do the non-peers get any access at all? Even perhaps as a degraded, low resolution service? If so, I'd like to know where to register my disapproval.
Audio could be auto dubbed into received pronunciation and any food onscreen automatically replaced by deepfaked cucumber sandwiches.
Also, remote transcoding of videos seems like a huge win! Makes it so you won't need as beefy as a system to run PeerTube.
I wonder why FOSDEM doesn't use PeerTube to host their videos.
I didn’t watch any of their videos yet this year, but perhaps the thing they are using to host their videos is something they have built on their own, before PeerTube even existed? And so if what they have works then that’s a good reason to stick to using that.
You can technically do it if you're willing to hack around. Since it uses the ffmpeg binary you could technically replace it with rffmpeg[1], the setup isn't pretty but it's doable. I haven't done it myself but have seen reddit posts on the jellyfin community that it works pretty well once setup.
Just goes to show how much potential small teams or even individuals have that gets wasted in big corpos.
Self-hosting is still not quite as easy as it could have been, though.
The biggest issue with Peertube/Youtube is related to creators and how they get paid for content they create.
Youtube has an avenue for that to happen, and Peertube doesn't, or at least in a way that works.
Unless Peertube wants to just be "the place you host your videos from other platforms as a backup" something would have to be done.
The ones that do aren't surviving on YouTube ad revenue alone since the adpocalypse. They're doing so based on ad reads and crowdfunding, both of which work on Peertube just fine.
The only thing keeping them on YouTube is audience reach.
There are certainly plenty of stories of people who produce content funded directly by their audience, but none of them to match the money made by the top Youtubers.
I'm a bit glib here, but the distinction is actually important. The ad industry, in particular, tends to use "educate" when it means "influence."
But this isn't 15 years ago. There wasn't a huge player in this space when Youtube was getting started, but now there is. And now people know it's theoretically possible to make decent money creating content.
I'd like to see an invite-based signup option like Mastodon, but other than that, it's a nice roadmap!
Edit: Created an idea post: https://ideas.joinpeertube.org/posts/139/invite-links-to-cre...
But it is mostly a discovery problem. Once you have found a channel that you like it is easy to follow via RSS or ActivityPub. I would love to see discovery services built on top of PeerTube, it seems natural to separate the viewing infrastructure from the recommendation service.
A content creator producing quality content has spent a lot of money in order to produce the high quality content. How much? Estimates from YouTube creators who are currently in the top creators get us numbers like $25,000-$1,000,000 per video. If they do not reliably make more than that much from trying to produce at that quality, they cannot survive while doing so.
So a larger problem than content discovery is that by default people producing quality content will die.
There is a huge difference between "there is lots of crap content to wade through to get to the good content" and "I can't find content that isn't crap". My foray into Peertube over the years is that the latter is much closer than the former.
Well, most of everything in the Universe is “crap.” But, to each their own, and I find YouTube to be a treasure trove of educational content, among other things.
What's the point of self hosting if you still need Google?
Maybe you don't like that, but that's what software freedom looks like
Similarly, most people on YouTube are not expecting to strike it big with videos of their cat, or their hobby project, or whatever.
Related, that’s probably not the best picture to show if they’re going for mainstream acceptance. I suspect it’s probably a pretty sedate video actually, so the author of the post just didn’t think about how it looked, but just choosing something with less apparent nakedness would probably give a better first impression if PeerTube is going for a widespread audience.
Of course, Torrent has the same problem but I think in that case people are more aware of it.
The problem here was only ever partly technological, and the rest is a business problem. On one end, you have the ability of Youtube to store huge amounts of video, transcode it, and then serve it to basically every corner of the globe, eating up significant portions of global bandwidth doing so, all with incredibly low tail latency (load up random 0-view videos and they still load quickly).
On the other, you have creators actually getting paid. Yes, they're not paid very much, and yes there are endless issues with advertisers getting unhappy or creators complaining about content policy, but by and large it's sustained an incredibly diverse ecosystem of people making videos because enough people watch them to make it a living. That's not to add the embedded advertising that usually pays much more than YT's own ads.
Any platform that wants to compete must solve all these problems simultaneously: can you serve video at global scale, not go bankrupt doing it, and split enough of your money video creators that they actually want to host their videos with you.
The problem is as much with the business model as it is technological, and anybody that tries to build a better youtube just by throwing code at the problem is doomed to fail.
And then after you have all of that you still end up with the network effect: YT has all the users and all the video, and it's a self-reinforcing system where one begets the other.
Though I guess I could see some channels that are too heavily restricted on youtube using peertube as a repository of all videos, and youtube for the videos that youtube currently deems acceptable enough.
https://www.eff.org/deeplinks/2009/01/youtubes-january-fair-...
(To be fair there also was that Viacom et al. lawsuit for $1 billion in damages...)
I'm 100% convinced that NOTHING of significant value would be lost if the revenue that Youtube pays directly were to go away. It would be a rough transition period, but pure-ads just doesn't make anything good. Patreon-style is ideal, but Netflix is fine too.
Use Qualys [1] to test the domain in question to link here or use the testssl.sh [2] code only depends on openssl and bash to test from your machine. If one of the many self hosted nodes, see if you can find a way to reach out to them and kindly suggest they set up certbot or a cron job to renew their certs.
Joinpeertube.org looks good to me [3] so I assume you find a self-hosted node that needs some attention.
If someone here knows of a way to query a list of all the self-hosted domains joined into peertube perhaps we could run testssl.sh against all of them to generate reports. I am not opposed to doing this if someone knows how I can get a list of all the domains using curl.
[1] - https://www.ssllabs.com/ssltest/
[2] - https://github.com/drwetter/testssl.sh
[3] - https://www.ssllabs.com/ssltest/analyze.html?d=joinpeertube....
Clicking on the HN link, as a user how would I know this is to be expected? or that it's one random node?
If you are experiencing issues on joinpeertube.org then it may be worth running an OS update and ensuring you also have the latest version of your browser to rule out CA certificate store issues. CA stores are periodically updated and LetsEncrypt did go through a change that will eventually invalidate its older signing keys and intermediate certificates. This would only impact people that have stagnant operating systems or browsers unless one is being routed through an outdated MitM Man-in-the-Middle HTTPS proxy.
We may be able to rule out MitM with openssl, see if the fingerprint is the same for you.
openssl s_client -servername joinpeertube.com -connect joinpeertube.com:443 < /dev/null 2>/dev/null | openssl x509 -fingerprint -noout -in /dev/stdin
SHA1 Fingerprint=C0:4E:F2:F6:EA:2B:72:C5:84:E0:73:2C:2C:2B:BB:FB:A1:34:C8:20It's showing as some kind of cicso cert valid for a couple of days.
There's no interception on this network. Other let's encrypt certs work just fine.