Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

PeerTube v2

framablog.org

Post reply on HN
Print view View original

131–140 of 159 posts

Re: PeerTube v2

#131

Post by bogwog »

majewsky wrote: post #121
bogwog wrote: post #83

Earlier quoted context omitted.

Anyone can start an instance and post whatever they want. Just because it has "peertube" in the URL doesn't mean it's an "official" instance or anything like that.

Whether a service is centralized or decentralized might matter a lot to us here, but to 99% of people it's utterly irrelevant. An implementation detail. Therefore their mental model will equate Youtube (a service) with Peertube (an application used by multiple services). If illegal content is on some Peertube, it's "on Peertube" and it drags down all the other stuff that is "on Peertube". Just like how big advertiser…

The solution to that then is to not pretend like PeerTube is centralized by hijacking the name for your URL. If you're hosting an instance, give it a unique name. If MyCoolVideos.com is a peertube instance with a bunch of child porn, then MyCoolVideos.com is going to get labeled as a bad site, and no one will know or care if the underlying technology is PeerTube.

This is a problem with federated services in general. People always seem to want to register on the "official" instance, when really there is none. I think there should be a solution to help people make the decision, or better yet not force them to make a decision at all. Maybe some OpenID-type login/account system should be used instead of having to make an account on a single instance. Or simply stop trying to market the underlying tech, like Mastodon or PeerTube, since that's not going to make a difference to the end-user and will just confuse them.

Re: PeerTube v2

#132

Post by dmos62 »

boomboomsubban wrote: post #109
dmos62 wrote: post #20

Earlier quoted context omitted.

Well that's the real problem when you start competing with Google and the like. They're data behemoths. As long as you don't have as much data collected, they'll have a significant advantage. The solution to this is really simple and hardly conceivable at the same time: make competitively important data public. In this case, that's anonymized video browsing/viewing data. Market competition would flourish, the big guy…

>In this case, that's anonymized video browsing/viewing data So the view count publicly available and something like the Google search analytics already available? Personalized data offers a huge competitive advantage.

Personalized (as in per user) and anonymized are not mutually exclusive.

Re: PeerTube v2

#133

Post by dmos62 »

pault wrote: post #64
dmos62 wrote: post #20

Earlier quoted context omitted.

Well that's the real problem when you start competing with Google and the like. They're data behemoths. As long as you don't have as much data collected, they'll have a significant advantage. The solution to this is really simple and hardly conceivable at the same time: make competitively important data public. In this case, that's anonymized video browsing/viewing data. Market competition would flourish, the big guy…

But then who will be incentivized to gather the data?

Anyone who needs more, I guess? Or, is the premise that gathering data is resource intensive (like innovation and patents for example)?

These aren't big problems, I think. The problem is that the people who hold the monopolies on data at the moment are also the people who are extremely powerful lobyists.

Re: PeerTube v2

#134

Post by throwaway9x »

MayeulC wrote: post #26

I really like the platform, but I think it would shine even more with scalable video coding support, to the point that it would be a game-changer. Unfortunately, the issue was closed [1]. Do any of you know open source libraries, or documentation on SVC for mp4, VP9 or AV1? I don't think ffmpeg supports this at all, which is a shame. [1]: https://github.com/Chocobozzz/PeerTube/issues/99

https://en.wikipedia.org/wiki/Scalable_Video_Coding

From what I can tell SVC adds lower resolution streams to the H264 bitstream so that devices with weaker decoders can still decode the video. This would increase, not decrease, the bandwidth.

>Do any of you know open source libraries, or documentation on SVC for mp4, VP9 or AV1

It appears to be defined as annex G in the H.264 (mpeg4 part 10) spec, so it's not available in VP9 or AV1.

The status quo is to use Adaptive Bitrate, where you publish streams at different quality levels and have the client choose which stream to use to adapt to network conditions.

Re: PeerTube v2

#135

Post by timwaagh »

detaro wrote: post #122
timwaagh wrote: post #24

It seems like they did not consider how they want peertubers to be able to monetize. On the contrary, it seems they have something against 'capturing your attention and selling it to Fanta'. I want this piece of work to be exhibit A in how staying inside your ideological bubble will ensure failure. I think it's especially sad since they used crowdfunded money so they aren't just failing on their own dime.

Most of the YouTubers I follow have ads they embed themselves in their videos, sponsored content or patreons. I'm not sure the platform running ads is actually strictly needed. (+ of course it doesn't have replace YouTube to be valuable) It'd be interesting if an instance could add them itself - I think doing it dynamically would mess with Webtorrent support?

Might work for bigger channels with a stable following. Which is great but they need to start somewhere. I have an old friend who owns a history channel on yt. History Hustle. His vids are quite good but he's no pro just kinda doing it on the side. He doesn't have any real kind of sponsorship aside of his one patreon and yt. I think a lot of guys started out this way too. But I remember how other friend bragging when he got like a few euro through it. I think that inspired it. I don't know how that kind of thing would work without the yt model.

Re: PeerTube v2

#136

Post by wbillingsley »

Have you tried talking to universities or university video providers (e.g. Echo360)?

A lot of universities are part of networks (e.g., the one I teach at, UNE in Australia, is part of the Regional Universities Network.) So they seem to fit the pattern of organisations that produce and host a lot of videos (course videos), don't like YouTube (we sometimes get told off for putting course videos there) and are part of networks that might like federated access to each others' video collections (while we mostly produce our own - it's not tv it's a teaching course and students usually want to hear from their teachers - sometimes we'll want to include something like a guest talk from industry).

Universities may be less interested in the bittorrent streaming aspect (they already successfully stream video and don't face the ridiculous peaks in demand because most videos aren't public) but ActivityPub to be able to search selected videos from each others' repositories could be interesting.

Re: PeerTube v2

#137

Post by progval »

craze3 wrote: post #130
lewisflude wrote: post #43

Went on https://peertube.video/ and immediately got a bunch of porn. Not really sure from a consumer aspect this is yet fully there from a UX perspective! But the tech is really interesting and as a federated alternative I could see it doing really well. Best of luck to the team!

I believe the correct URL is https://peertube.social/ ?

That's only an instance. Frontpage would be: https://joinpeertube.org/

Re: PeerTube v2

#138

Post by detaro »

savingGrace wrote: post #77
lewisflude wrote: post #43

Went on https://peertube.video/ and immediately got a bunch of porn. Not really sure from a consumer aspect this is yet fully there from a UX perspective! But the tech is really interesting and as a federated alternative I could see it doing really well. Best of luck to the team!

This was my first thought. With all of these YT alternatives I always end up asking who determines what content is saved/available? What if it doesn't go along with their beliefs or political stance? There's a fine line between going too far in each direction. e.g. China vs Pirate Bay Edit: "a system where videos require manual approval by administrators before posting them" If you're seeing a ton of porn, then clear…

And the great thing is that you then can use an instance that has a sysadmin that filters better for your needs.

Re: PeerTube v2

#139

Post by MayeulC »

throwaway9x wrote: post #134
MayeulC wrote: post #26

I really like the platform, but I think it would shine even more with scalable video coding support, to the point that it would be a game-changer. Unfortunately, the issue was closed [1]. Do any of you know open source libraries, or documentation on SVC for mp4, VP9 or AV1? I don't think ffmpeg supports this at all, which is a shame. [1]: https://github.com/Chocobozzz/PeerTube/issues/99

https://en.wikipedia.org/wiki/Scalable_Video_Coding From what I can tell SVC adds lower resolution streams to the H264 bitstream so that devices with weaker decoders can still decode the video. This would increase, not decrease, the bandwidth. >Do any of you know open source libraries, or documentation on SVC for mp4, VP9 or AV1 It appears to be defined as annex G in the H.264 (mpeg4 part 10) spec, so it's not availa…

> This would increase, not decrease, the bandwidth.

Not if that part of the stream isn't downloaded. It's quite easy to skip over some chunks in either HTTP (range) or bittorent.

> The status quo is to use Adaptive Bitrate, where you publish streams at different quality levels and have the client choose which stream to use to adapt to network conditions.

This works well if you have a good connection (uninterrupted, so that you can switch to adapt to the available bandwidth), however you cannot progressively load better and better quality content, as the low quality one will be wasted. Moreover, if the connection is interrupted, you cannot fall back to the low quality content, as you probably haven't downloaded it.

However, that's just wasteful, as it would "just" be a matter of presenting data in a manner that can be better chuncked. And for adaptative bitrate to work well, you need a reliable bandwidth estimate from the start...

Luckily, network infrastructure is improving everywhere. But that shouldn't be a reason to be that wasteful, especially given that a) most of Internet's bandwidth is dedicated to video b) it would lower transcoding energy costs as well c) if everyone gets the same file, it's much, much more effective for p2p, such as in that case (especially for a moderate number of viewers).

Edit:

> From what I can tell SVC adds lower resolution streams to the H264 bitstream so that devices with weaker decoders can still decode the video.

It's more like organizing the data so that some can be dropped, and it gracefully degrades quality (FPS, resolution...) instead of dropping some images. I tried to explain it with my own terms there: https://news.ycombinator.com/item?id=17307277 before knowing the proper name.

Searching HN, I found https://news.ycombinator.com/item?id=18045494 which kind of answers my question.

Re: PeerTube v2

#140

Post by boomboomsubban »

dmos62 wrote: post #132
boomboomsubban wrote: post #109

Earlier quoted context omitted.

>In this case, that's anonymized video browsing/viewing data So the view count publicly available and something like the Google search analytics already available? Personalized data offers a huge competitive advantage.

Personalized (as in per user) and anonymized are not mutually exclusive.

Sure, but non-anonymized personalized data still provides a competitive advantage.
Post reply on HN