Peertube is software to run a video hosting website. There are many unaffiliated sites that use it. Would you expect nginx.org to drop you "into" nginx and complain that it is not Facebook, or would you expect to find information about nginx?
If you want to visit a particular installation of the software, you go there. e.g. KDE has an instance for their videos[0]. They link to some of their big users, but the site you're commenting on is not itself a video hosting service.
I wouldn't expect that, but I can't tell how much of that is my intuition, and how much of that is my cynicism speaking. He has a point. Nginx is marketed towards people who know what the hell an http server is, merely being one of the best (or one of the only, I suppose) http servers is enough... no more marketing is needed. But with Peertube, it does not matter how good the software is. That sounds insulting to type out, I keep wanting to backspace over it, but it's true.
It matters how many people use it. Basically, every single person who ever steals a glance at it needs to be a convert. People who can't appreciate the beauty of the technical solution.
>If you want to visit a particular installation of the software,
You're reading him wrong. He doesn't want to "visit a particular installation". He wants it to succeed. So do I. I like the idea enough that I want it to succeed for that alone, and not necessarily because if it did it might start murdering Youtube (though that too). But it can't do that when it looks like a turd and behaves like a turd to any Joe Schmoe who happens to stumble across it.
Why does everyone need to be a convert? I'm not even sure that anyone needs to be a "convert" for it to succeed. e.g. MIT OCW uses youtube as its main host right now, which is not a great fit for CC BY-NC-SA material (ads + difficult downloading directly contradict that license). They could run a peertube instance to make it easy for users to exercise their rights while still posting it to youtube for reach. Or similar to how the mathematics community has moved from Twitter to Mathstodon, they could run a peertube instance for lectures, lean conference talks, etc.
Joe Schmoe doesn't need to know what peertube even is. Whoever is going to set it up for an organization/community does. If Joe is in such a community, finding their instance works the same as finding their youtube channel: you link to it. Actually it works better because by nature of being small and focused, the instance lets you easily discover related channels or make your own and have it discoverable by the community. Youtube discoverability is actually kind of trash by comparison, and relies on search (which requires you to know what you want already) or popularity/"the algorithm" for recommendations.
Since Joe doesn't need to know what it is, they don't market to Joe. They talk to community managers and IT people, e.g. on their front page:
> PeerTube is a tool that you install on a web server. It allows you to create a video hosting website, so create your "homemade YouTube".
It's never going to achieve "success" in the sense that the latest Mr Beast or Megan Thee Stallion video is there, but that's okay. Those aren't the target audience.
They could. But then 6 months later, they'll review the logs, see no one's really watching it there, and will get bored with supporting it. Then it's on the chopping block. Which, because of networking effects, subtly hurts other instances at other organizations. As there are fewer things to view with Peertube, fewer people go out and discover any further content with it because they never started watching it in the first place.
Sort of reminded of this one shopping mall I used to go to as a kid in Ohio. Looked it up a few years ago, completely empty except for one store on the bottom flood... do you think that store sees any extra foot traffic?
>Joe Schmoe doesn't need to know what peertube even is.
Someone who wanted to kill Peertube dead with fire couldn't murder it as thoroughly as your attitude does. You get that right?
I mostly agree with the criticisms of design. I'm endlessly frustrated that Mastodon was first on the block, but Blue Sky ate its lunch, for reasons that seem almost entirely to hinge on having been smarter about design. And even Mastodon was a huge leap forward from the characteristic quality of open source social media alternative sites that had existed previously. I still think in its own way Mastodon's design is quite impressive. But Peertube is dropping the ball, I think, compared to some other Fedeiverse projects. And peertube, I would say, is arguably one of the most important projects.
When it comes to the joinpeertube domain, however, that follows a long-standing tradition of fediverse sites to use those as a jumping-off point to discover instances. I don't think that can be roped into the category of transgressions along the lines of the previous ones.
> Can't they afford $5K for websites and UI design? Everything here looks absolutely terrible.
Can you give examples? People in the thread aren't even clear whether you're talking about a Peertube instance or joinpeertube.org.
> but they're asking me to consider this as an alternative to YouTube - how it presents is critically important.
I'm probably in the minority, but I've hated Youtube's design since Youtube first came out. Really cluttered. There's a reason people made several alternative frontends to Youtube, and it's mostly not to avoid ads. Youtube's design just sucks (according to many).
Completely agree. I actually think Sepia search is quite good. It's what peer tube desperately needs and it's so much better than the individual search features on all the different instances which are a complete mess. So I don't understand why it's pulled out into a separate brand and separate domain.
I'm honestly confused. Are you complaining about "joinpeertube.org" or the typical Peertube instance? If the former ... why would you want it to have any search capability? It's not a video host. If the latter (e.g. https://peer.tube/), the search is not at all hard to find.
Think you are confused about what it is, it's a piece of software for hosting videos. Might be worth reading the frontpage https://joinpeertube.org/ .
As an example, blender has https://video.blender.org/ with blender videos. It's kind of like mediawiki is wiki software but it isn't wikipedia.
The part which seems to be confusing you is that these platforms can ( don't have to ) network together to make a larger apparent platform that has some resemblance to youtube.
Having said that, the actual platform is kind of meh in terms of UI design. It does look like a very budget youtube clone as if it was a "learn to make a youtube clone in 1 hour using html css!"
Do you have a more substantive reply to the user you’re replying to? These are genuine observations/experiences. If you disagree or think they are misguided, then point them out.
The issue is that the main comment in the thread complained about design, but gave little concrete information to discuss. If I come and say "Design matters, and HN's design sucks. Go and compare it to Reddit/whatever other site," you likely will have little idea of what my beef with HN is.
Of course it is. It's usually avoided depending on the level of formality/style of the writing, but there is no grammatical rule that says a paragraph cannot be one sentence long.
> .. improves support for Podcast 2.0, allowing users to subscribe to channels and play the video audio stream (if available) using classic podcast applications
A youtube channel url provides a rss link too, but it's not advertised much and you need a player that can extract the audio stream. I often yt-dl the audio from talks/interviews, but it could be easier.
To find the channel ID:
1. Visit the creator's page
2. Click on `more` for the full description
3. Scroll down to `Share Channel`
4. Click and copy Channel ID
The script doesn't work anymore, `externalId` is no longer found in the page source.
That's okay, I've found a regular type="application/rss+xml" <link> is in the page head of YouTube channel pages but only if you load that page directly; if you follow links within the site to the channel page, it's an SPA and you don't get the channel URL's <link> elements.
The channel page also has a <link rel="canonical" …> and that URL ends with the channel ID, even if it has an `@` handle.
I like the idea of a decentralized Google free YouTube. But I find these posts about PeerTube, Frama, Mastodon, protocols, etc just very confusing. And it’s hard to understand how everyday people can use this platform or it is even meant for that. Years after first hearing of this, it has gone nowhere.
I know this is a negative take but does someone have an explanation or an understanding of whether this project means anything from a practical standpoint?
They don't have a plan to conquer the world, a business plan, or anything else. They're just FOSS developers doing stuff that interests them. The majority of framasoft developers are French and European teachers and academics.
Framasoft is the antithesis of y-combinator.
Sometimes they reach a wide audience, like framapad and framadate, which are quite popular in France and Western Europe. Partly because they just work and don't require you to create an account.
Sometimes nobody really knows who it's for.
Most of the time people often mistakenly think it's an isofunctional 1:1 alternative to another product. When it's not.
Regardless of whether the developers of PeerTube intend for their work to be widely used and successful, it doesn’t refute the argument that “if this software/platform/service is to be successful and widely used, it needs more marketing and/or funding and/or clearer UI/UX”. If they’re not doing that, then they are either unable or unwilling to. But that’s a separate issue.
> it’s hard to understand how everyday people can use this platform
Because they are not platforms: they are software, free to use. Their audience is systems administrators, not users.
Either the user is it's own systems administrator, or someone else has to do it... And in any case there is infrastructure to be paid for. Most users currently prefer someone else to do it and the price to be the watching of advertisements.
I believe a sweet halfway point is hosting cooperatives - not free of cost, but delivering user-control at a reasonable cost to users who do not wish to take on the burden of sysadmining.
Some youtubers are posting their videos to Peertube based video sites, like makertube.net , tilvids.com , tinkerbetter.tube , just to name a few. Please share your favorite Peertube instances.
Sure, there are way less videos to choose from compared to youtube, but that leaves more time for productivity or for time away from the screen.
A lot of this is about control over your information.
It isn't about being on YouTube. It's about being able to share your media/information in a way that you feel comfortable.
These projects are for people to explore self hosting or creating a platform based around the ideology the software respects.
I personally self host nearly all my stuff. I'm a photographer/videographer/tech contractor and it allows me full control over my ecosystem without sacrificing my data.
The folks running these aren't really interested in building a big social network or spending hours everyday scrolling what other people are doing.
> Peertube specifically is unusable, has no content and has no incentive for content creators
Peertube users do not need content nor incentives: they just want to publish their videos and post links to them to their audience.
My photo gallery doesn't need content nor incentives for content creators either... I supply the content and my content creator incentive is showing my vacation pictures to my friends !
It's presented as an alternative to youtube for current Framasoft members and users.
It's meant to be understood as “Don't put faculty course videos on youtube, which you can't necessarily rely on if the subject is sensitive, put it on your college or association's PeerTube server.”
Framasoft is originally a non-commercial association of French academics. They have no desire to replace youtube for music videos, unboxing and Minecraft lives.
> PeerTube is specifically advertised as a YouTube replacement.
And about 99% of people who upload to Youtube are not concerned with any kind of incentive other than "Let me share this video". For 99% of people, it is a perfectly capable Youtube replacement (well, beyond being able to search well - a problem the whole Fediverse has...)
> Peertube specifically is unusable, has no content and has no incentive for content creators.
I've been using it since its release and it has been usable since then. It certainly has content, i.e. the content I put on it. I run three sites for different purposes, all from the same server which also hosts dozens of other services. The incentive for the 'content' (a disgusting concept if I can be so bold, as if its role is comparable to that of ballast on a ship instead of being the reason the ship sets sail in the first place) creator - which is me for the first site, others for the other two - is that I get to distribute video and audio without getting censored, without advertisements, without data leeches trying to profile viewers while still being able to reach across the world from that server-under-the-stairs.
Because nobody will see them. Moreover the instance that host them will be blocked by the main nodes as much as on any mainstream social network and managing a server is more work and more risk than buying a pirates account to post
Because it turns out hosting even terrorist videos on existing platforms (and dealing with takedowns, etc) is still a better user experience than using the "fediverse".
I know this will come across as negative but design matters for something like this.
Can't they afford $5K for websites and UI design? Everything here looks absolutely terrible.
Its a negative take but they're asking me to consider this as an alternative to YouTube - how it presents is critically important.
Also, the website root should not talk about peertube, it should drop me into peertube
And a final gripe..... Can't they get a better domain than "joinpeertube.org" - really almost anything would be better.
Sorry this is such a grumble but so many obvious self sabotaging obstacles to success.
Peertube is software to run a video hosting website. There are many unaffiliated sites that use it. Would you expect nginx.org to drop you "into" nginx and complain that it is not Facebook, or would you expect to find information about nginx?
If you want to visit a particular installation of the software, you go there. e.g. KDE has an instance for their videos[0]. They link to some of their big users, but the site you're commenting on is not itself a video hosting service.
[0] https://tube.kockatoo.org/home
>Would you expect nginx.org to drop you
I wouldn't expect that, but I can't tell how much of that is my intuition, and how much of that is my cynicism speaking. He has a point. Nginx is marketed towards people who know what the hell an http server is, merely being one of the best (or one of the only, I suppose) http servers is enough... no more marketing is needed. But with Peertube, it does not matter how good the software is. That sounds insulting to type out, I keep wanting to backspace over it, but it's true.
It matters how many people use it. Basically, every single person who ever steals a glance at it needs to be a convert. People who can't appreciate the beauty of the technical solution.
>If you want to visit a particular installation of the software,
You're reading him wrong. He doesn't want to "visit a particular installation". He wants it to succeed. So do I. I like the idea enough that I want it to succeed for that alone, and not necessarily because if it did it might start murdering Youtube (though that too). But it can't do that when it looks like a turd and behaves like a turd to any Joe Schmoe who happens to stumble across it.
Why does everyone need to be a convert? I'm not even sure that anyone needs to be a "convert" for it to succeed. e.g. MIT OCW uses youtube as its main host right now, which is not a great fit for CC BY-NC-SA material (ads + difficult downloading directly contradict that license). They could run a peertube instance to make it easy for users to exercise their rights while still posting it to youtube for reach. Or similar to how the mathematics community has moved from Twitter to Mathstodon, they could run a peertube instance for lectures, lean conference talks, etc.
Joe Schmoe doesn't need to know what peertube even is. Whoever is going to set it up for an organization/community does. If Joe is in such a community, finding their instance works the same as finding their youtube channel: you link to it. Actually it works better because by nature of being small and focused, the instance lets you easily discover related channels or make your own and have it discoverable by the community. Youtube discoverability is actually kind of trash by comparison, and relies on search (which requires you to know what you want already) or popularity/"the algorithm" for recommendations.
Since Joe doesn't need to know what it is, they don't market to Joe. They talk to community managers and IT people, e.g. on their front page:
> PeerTube is a tool that you install on a web server. It allows you to create a video hosting website, so create your "homemade YouTube".
It's never going to achieve "success" in the sense that the latest Mr Beast or Megan Thee Stallion video is there, but that's okay. Those aren't the target audience.
>They could run a peertube instance to
They could. But then 6 months later, they'll review the logs, see no one's really watching it there, and will get bored with supporting it. Then it's on the chopping block. Which, because of networking effects, subtly hurts other instances at other organizations. As there are fewer things to view with Peertube, fewer people go out and discover any further content with it because they never started watching it in the first place.
Sort of reminded of this one shopping mall I used to go to as a kid in Ohio. Looked it up a few years ago, completely empty except for one store on the bottom flood... do you think that store sees any extra foot traffic?
>Joe Schmoe doesn't need to know what peertube even is.
Someone who wanted to kill Peertube dead with fire couldn't murder it as thoroughly as your attitude does. You get that right?
I mostly agree with the criticisms of design. I'm endlessly frustrated that Mastodon was first on the block, but Blue Sky ate its lunch, for reasons that seem almost entirely to hinge on having been smarter about design. And even Mastodon was a huge leap forward from the characteristic quality of open source social media alternative sites that had existed previously. I still think in its own way Mastodon's design is quite impressive. But Peertube is dropping the ball, I think, compared to some other Fedeiverse projects. And peertube, I would say, is arguably one of the most important projects.
When it comes to the joinpeertube domain, however, that follows a long-standing tradition of fediverse sites to use those as a jumping-off point to discover instances. I don't think that can be roped into the category of transgressions along the lines of the previous ones.
> Can't they afford $5K for websites and UI design? Everything here looks absolutely terrible.
Can you give examples? People in the thread aren't even clear whether you're talking about a Peertube instance or joinpeertube.org.
> but they're asking me to consider this as an alternative to YouTube - how it presents is critically important.
I'm probably in the minority, but I've hated Youtube's design since Youtube first came out. Really cluttered. There's a reason people made several alternative frontends to Youtube, and it's mostly not to avoid ads. Youtube's design just sucks (according to many).
> the website root should not talk about peertube, it should drop me into peertube
> Can't they get a better domain
Maybe the website you were looking for is https://peer.tube
> Can't they afford $5K for websites and UI design? Everything here looks absolutely terrible.
Have you attempted at doing that contribution yourself? Be the change you want to see.
I also don't understand having a search interface not front-and-center and under the weird sub-brand of "SepiaSearch".
Completely agree. I actually think Sepia search is quite good. It's what peer tube desperately needs and it's so much better than the individual search features on all the different instances which are a complete mess. So I don't understand why it's pulled out into a separate brand and separate domain.
I'm honestly confused. Are you complaining about "joinpeertube.org" or the typical Peertube instance? If the former ... why would you want it to have any search capability? It's not a video host. If the latter (e.g. https://peer.tube/), the search is not at all hard to find.
Think you are confused about what it is, it's a piece of software for hosting videos. Might be worth reading the frontpage https://joinpeertube.org/ . As an example, blender has https://video.blender.org/ with blender videos. It's kind of like mediawiki is wiki software but it isn't wikipedia.
The part which seems to be confusing you is that these platforms can ( don't have to ) network together to make a larger apparent platform that has some resemblance to youtube.
Having said that, the actual platform is kind of meh in terms of UI design. It does look like a very budget youtube clone as if it was a "learn to make a youtube clone in 1 hour using html css!"
A sentence is not a paragraph.
Do you have a more substantive reply to the user you’re replying to? These are genuine observations/experiences. If you disagree or think they are misguided, then point them out.
The issue is that the main comment in the thread complained about design, but gave little concrete information to discuss. If I come and say "Design matters, and HN's design sucks. Go and compare it to Reddit/whatever other site," you likely will have little idea of what my beef with HN is.
Of course it is. It's usually avoided depending on the level of formality/style of the writing, but there is no grammatical rule that says a paragraph cannot be one sentence long.
> .. improves support for Podcast 2.0, allowing users to subscribe to channels and play the video audio stream (if available) using classic podcast applications
that refers to https://podcasting2.org/podcast-namespace
A youtube channel url provides a rss link too, but it's not advertised much and you need a player that can extract the audio stream. I often yt-dl the audio from talks/interviews, but it could be easier.
The v7 interface came out nice, for examples see https://instances.joinpeertube.org/instances?sort=-version&p...
Kind of a pain to find but it's essentially `https://www.youtube.com/feeds/videos.xml?channel_id=UCY1kMZp...` and you swap out the channel ID.
To find the channel ID: 1. Visit the creator's page 2. Click on `more` for the full description 3. Scroll down to `Share Channel` 4. Click and copy Channel ID
A bit of a pain but I'm glad I know this now!
a quick shortcut flow in a browser on the youtube channel page (a click below the video title):
ctrl+u ctrl+f "rss+x" -> shows a click+copyable video.xml link
I understand people being pleased to learn this, for me the peertube announcement was learning about p2, as it's been around a few years already.
alternateEnclosure is a real gem in the p2 rss spec, the thing that makes multiple channels and alternative formats/codecs/resolutions work.
This bookmarklet will show the RSS feed URL in an alert window:
Original Source: https://webapps.stackexchange.com/a/116549
The script doesn't work anymore, `externalId` is no longer found in the page source.
That's okay, I've found a regular type="application/rss+xml" <link> is in the page head of YouTube channel pages but only if you load that page directly; if you follow links within the site to the channel page, it's an SPA and you don't get the channel URL's <link> elements.
The channel page also has a <link rel="canonical" …> and that URL ends with the channel ID, even if it has an `@` handle.
The rss feed provides a link that you can open in a web browser to play the video. You don't need a tool to extract it.
You do if you want to use a podcast player application, which is what the comment you're replying to is talking about.
Framasoft is really an amazing team. They are basically the reason I discovered FOSS when I was younger.
The fact they are still here and supporting so many free services is a miracle.
I like the idea of a decentralized Google free YouTube. But I find these posts about PeerTube, Frama, Mastodon, protocols, etc just very confusing. And it’s hard to understand how everyday people can use this platform or it is even meant for that. Years after first hearing of this, it has gone nowhere.
I know this is a negative take but does someone have an explanation or an understanding of whether this project means anything from a practical standpoint?
They don't have a plan to conquer the world, a business plan, or anything else. They're just FOSS developers doing stuff that interests them. The majority of framasoft developers are French and European teachers and academics.
Framasoft is the antithesis of y-combinator.
Sometimes they reach a wide audience, like framapad and framadate, which are quite popular in France and Western Europe. Partly because they just work and don't require you to create an account.
Sometimes nobody really knows who it's for.
Most of the time people often mistakenly think it's an isofunctional 1:1 alternative to another product. When it's not.
Regardless of whether the developers of PeerTube intend for their work to be widely used and successful, it doesn’t refute the argument that “if this software/platform/service is to be successful and widely used, it needs more marketing and/or funding and/or clearer UI/UX”. If they’re not doing that, then they are either unable or unwilling to. But that’s a separate issue.
> it’s hard to understand how everyday people can use this platform
Because they are not platforms: they are software, free to use. Their audience is systems administrators, not users.
Either the user is it's own systems administrator, or someone else has to do it... And in any case there is infrastructure to be paid for. Most users currently prefer someone else to do it and the price to be the watching of advertisements.
I believe a sweet halfway point is hosting cooperatives - not free of cost, but delivering user-control at a reasonable cost to users who do not wish to take on the burden of sysadmining.
Some youtubers are posting their videos to Peertube based video sites, like makertube.net , tilvids.com , tinkerbetter.tube , just to name a few. Please share your favorite Peertube instances. Sure, there are way less videos to choose from compared to youtube, but that leaves more time for productivity or for time away from the screen.
A lot of this is about control over your information.
It isn't about being on YouTube. It's about being able to share your media/information in a way that you feel comfortable.
These projects are for people to explore self hosting or creating a platform based around the ideology the software respects.
I personally self host nearly all my stuff. I'm a photographer/videographer/tech contractor and it allows me full control over my ecosystem without sacrificing my data.
The folks running these aren't really interested in building a big social network or spending hours everyday scrolling what other people are doing.
Why didn't I hear about this earlier? Strange.
The 7.1 version was just released today.
In summary: because it sucks.
All the software from FramaSoft is built from the point of view that "FOSS" is all they need and they can ignore everything else.
This means that the UX for all of their tools is not just subpar compared to alternatives, it is borderline unusable.
Peertube specifically is unusable, has no content and has no incentive for content creators.
> Peertube specifically is unusable, has no content and has no incentive for content creators
Peertube users do not need content nor incentives: they just want to publish their videos and post links to them to their audience.
My photo gallery doesn't need content nor incentives for content creators either... I supply the content and my content creator incentive is showing my vacation pictures to my friends !
PeerTube is specifically advertised as a YouTube replacement.
It's presented as an alternative to youtube for current Framasoft members and users. It's meant to be understood as “Don't put faculty course videos on youtube, which you can't necessarily rely on if the subject is sensitive, put it on your college or association's PeerTube server.”
Framasoft is originally a non-commercial association of French academics. They have no desire to replace youtube for music videos, unboxing and Minecraft lives.
> PeerTube is specifically advertised as a YouTube replacement.
And about 99% of people who upload to Youtube are not concerned with any kind of incentive other than "Let me share this video". For 99% of people, it is a perfectly capable Youtube replacement (well, beyond being able to search well - a problem the whole Fediverse has...)
A veritable wall of ignorance, masquerading as being in-the-know.
For example: the Intergalactic Wasabi Mixes, findable here, are real gems I've been trawling over for weeks now:
https://toobnix.org/c/snowdusk_channel/videos?s=1
You say it is "unusable" - can you describe to me what happens when you (or anyone else) clicks on one of those videos? Do they not work for you?
Worked for me without issue from my phone. Although I suspect that might have been more of a rhetorical question.
> You say it is "unusable" - can you describe to me what happens when you (or anyone else) clicks on one of those videos? Do they not work for you?
I was very clear that the UX is so bad it's unusable.
But if you want something actually unusable, no problem.
1. Open their "Browse content" page: https://joinpeertube.org/browse-content
2. Search the first example: "music"
3. Click the first video that shows up: https://peertube.dsmouse.net/w/bc181d28-3284-4882-b7a3-314e2...
4. "Failed to play video. The video failed to play due to technical issues.", with a month old comment saying the same.
> Peertube specifically is unusable, has no content and has no incentive for content creators.
I've been using it since its release and it has been usable since then. It certainly has content, i.e. the content I put on it. I run three sites for different purposes, all from the same server which also hosts dozens of other services. The incentive for the 'content' (a disgusting concept if I can be so bold, as if its role is comparable to that of ballast on a ship instead of being the reason the ship sets sail in the first place) creator - which is me for the first site, others for the other two - is that I get to distribute video and audio without getting censored, without advertisements, without data leeches trying to profile viewers while still being able to reach across the world from that server-under-the-stairs.
I've always wondered, why doesn't ISIS or other terror groups just use this type of distribution for their videos?
Because nobody will see them. Moreover the instance that host them will be blocked by the main nodes as much as on any mainstream social network and managing a server is more work and more risk than buying a pirates account to post
Because it turns out hosting even terrorist videos on existing platforms (and dealing with takedowns, etc) is still a better user experience than using the "fediverse".
This has not been my experience. Sorry it doesn't work for you, though.