PeerTube v5: the result of 5 years’ handcrafting | Hacker News Reader

PeerTube v5: the result of 5 years’ handcrafting

framablog.org

196 pointsealhad72 comments

This actually makes me sad :(

Hear me out: the Internet was supposed to be about peer-to-peer connected computers, and the privileged roles ISPs and later "cloud" providers assumed changed that for the worse.

It was SUPPOSED to enable me, myself, hosting my videos, on my computer(s) and making them available to whomever I want to, including everyone. This is how early protocols were designed. Everyone was supposed to be a SMTP (e-mail) host. Everyone was supposed to run FTP and HTTP. Everyone got an equally routable (the link quality depends, of course) address, not some 3rd level NAT retail monstrosity. If you needed aggregation, you make search sites like Google (and AltaVista and others before it) and RSS to pull data from multiple sources and CACHE IT LOCALLY.

Of course I welcome projects like PeerTube, but I'd much rather go back to the original idea. No ISPs or Clouds, only Peers.

With Internet like water grid - a utility.

> It was SUPPOSED to enable me, myself, hosting my videos, on my computer(s) and making them available to whomever I want to, including everyone.

I think maybe you misunderstand PeerTube? It is exactly what you say the internet should be.

You can use it to host your videos yourself, on your own computer, and make them available to whomever you want to.

You can use a hosted version too, but it's not required. It's also completely possible to host in your own cupboard, or to use a cloud server you control. On whatever instance you use, you can still talk to your friends directly, through peer-to-peer connections to whichever instance they choose to use. It does even support RSS, and other PeerTube instances do indeed cache the video locally!

It has its own new set of problems of course, but it does seem like it's a strong step in exactly the direction you're interested in.

> Hear me out: the Internet was supposed to be about peer-to-peer connected computers, and the privileged roles ISPs and later "cloud" providers assumed changed that for the worse.

The main problem aren't privileged actors like ISPs (although shit like asymmetric DSL or CGNAT definitely prevents people from self-hosting)... it is abuse and the complete unwillingness of almost everybody from private actors over governments to international organizations to put a fucking stop on it.

You open up a server on the Internet? Not even sixty seconds and the first Shodan or whatever using script-kiddies will attempt to hack you. And god forbid you run some popular software that can be sniffed like Drupal or Wordpress - you end up in Shodan just as well and will be automatedly exploited as soon as the CVE gives enough hints to people to write an exploit. You wish to send your own emails? You find yourself greylisted by almost everyone in their futile attempts to keep their users from spam. You wish to communicate with someone? Better read up on crypto because governments and ISPs just love to mine data. Operate a service that allows user-generated content? Beware for a deluge of everything from warez groups to CSAM spreaders that can and will expose you to serious legal liability.

The old protocols were all designed with implicit trust in mind and the assumption that no actor on the internet would abuse their position. That worked reasonably well as long as it was only universities (but even then, first viruses appeared from enterprising prankster students)... but once the Internet got mainstream, all of that broke down, and it completely collapsed once people started realizing they might make money shilling grey-imported penile enlargement pills. And the more people were on the Internet, the harder the work of "abuse departments" got, which led to most organizations simply dismantling the department or redirecting complaints to /dev/null. The fact that some governments (particularly China and Russia) take a completely blind eye towards hacking originating from their countries as long as they themselves aren't targeted (just look how many malware samples have a dead-man switch when they encounter information that the target might be Russian) just makes the problem worse.

Unfortunately, by that time the old protocols and standards were so widespread in use there was no chance to replace them, and so layers upon layers upon layers of bullshit got placed over the old layers in the end.

You're right that there are a lot of bad actors out there, but in my experience they are pretty easy to deal with if you set things up right. The biggest annoyance for me self-hosting has been ISP refusing to give static IP and decent upstream bandwidth.

I've hosted my own website and email server for decades. It does take a little work to keep up with things like DMARC, reverse DNS etc, but if you get a good score on https://internet.nl/test-mail/ and don't spam anybody, self-hosted email works fine. FYI you are misusing "greylist."

> You're right that there are a lot of bad actors out there, but in my experience they are pretty easy to deal with if you set things up right.

Sure, it's possible to defend against hackers to a degree, but even using a completely static website still leaves you open to attack surfaces in the webserver software or to remotely exploitable vulnerabilities in the Linux network stack.

> FYI you are misusing "greylist."

I assume we share the definition of "greylisting" to be the receiver MTA blocking the first delivery of an incoming email with "try again later", and the sender MTA then retrying after that time frame? If yes then this exactly describes my experience in administrating self-hosted mail servers with popular large mail providers.

Ah, my apologies, I jumped to conclusions that you were misusing "greylisting". I've never actually checked if I'm getting greylisted sending to large providers, but I think initial greylisting of new addressses is pretty reasonable.

My mail server (MailInABox) has postgrey enabled by default, greylisting incoming email each time email is received from a new address. I thought that was a little overzealous so turned it off.

It might be reasonable, but highly annoying if you run a, say, sign-up double opt-in or an email-based second factor/OTP. Customers don't like waiting an hour or whatever greylisting period, and so as a SaaS operator you are all but forced to go to one of the big e-mail senders like AWS SES because you don't stand a chance otherwise.

Ah, darn... I'm starting a web app with email verification now, using my own email server for sending and I didn't consider this. Thanks for the warning. Guess I'm about to find out how bad it is.

Pretty annoying to test for this reliably as well...

Hosting large files from your house is a good way to have a saturated upstream all the time. P2P protocols like gnutella and later bittorrent are a way to distribute files without putting all the load on one host. Not to mention all the liability for pirated content.

Only because the ISP's made it that way. Having 1Gbps down and 10Mbps up. You are forced into consumerism of bandwidth instead of a peer. ISP's don't want competition so you'll never be a peer.

The asymmetry stems from the days where we connected through the phone lines, which was not designed for data. There it was possible to have a higher bandwidth out from the phone switches than in. That is, dial-up modems and DSL modems could have a much higher download speed than upload.

I don't know about cable, but I suspect it's the same there.

OTH, fiber-based internet often has symmetric speeds. Don't know the stats, but I suspect most fiber ISPs gives you that.

The asymmetry stems from ISDN. Bonding pairs, twin lines, etc etc. that bled over into the dialup era and bled over still into the cable internet era and has even bled over into the satellite era.

Fiber internet often gives you 1Gbps up.

Fiber isn’t available in 95% of the US. They give you 1Gbps up because they don’t really have saturation.

> Hear me out: the Internet was supposed to be about peer-to-peer connected computers, and the privileged roles ISPs and later "cloud" providers assumed changed that for the worse.

No it wasn't. The internet was supposed to be a a global network of networks, and it is.

> It was SUPPOSED to enable me, myself, hosting my videos, on my computer(s) and making them available to whomever I want to, including everyone.

No it wasn't. ARPANet was supposed to be a network of computing facilities that could withstand a nuclear attack, and Internet was the effort of opening up such infrastructure to scientific and commercial entities (universities and companies). For quite a while the idea of the general public having internet at home wasn't even an idea.

> Everyone was supposed to be a SMTP (e-mail) host.

No they weren't. Pretty much nobody had their own dedicated computer, and they just had a shell account on a shared computer dedicated to a specific organization. And the administrator of such computer would set up mailing facilities on such host. Hence the local-delivery (/var/spool/mail and stuff).

Wait, what exactly makes you sad about PeerTube ?

Also, how would you connect to the Internet without an ISP ? (Well there are mesh networks I guess, but AFAIK they are much slower, both in throughput and latency ?)

You can perfectly run peertube at home.

What you want is Urbit (though hn hates it for political reasons)

The current state of cryptocurrency should be enough evidence that people want centralization as it affords the most social investment value.

I think it shows that what most people actually want are banks and brokers, and they don't know or care about how things work on the backend, or understand what banking regulations do for them when they use a real bank. The robinhood crowd and early NFT enthusiasts are in some ways two sides of the same coin.

PeerTube is a free, decentralised and federated video platform, and a part of the fediverse, meaning you can interact with it via Mastodon and other tools, thanks to the ActivityPub protocol.

There is only one paid developer, working for the French non-profit Framasoft – and I think it's important to share this in non French-speaking spaces.

If you want to help them "Collectivise / Convivialise the internet": https://soutenir.framasoft.org/en/

> ... and I think it's important to share this in non French-speaking spaces.

Is there a french version of this article?

Yes there is! [0]

[0]: https://framablog.org/2022/12/13/peertube-v5-le-resultat-de-...

Not a single embedded video on the page showing off the platform. Technologists' blindness to product and market building is so painful.

The goal of Peertube and Framasoft, the org behind it, is not product and market building. Product and Market building is exactly what led us to the default video repository being run by a for-profit company with only its own interest at heart, arbitrary censoring based on american feelings for the entire planet, arbitrary demonetizing videos/channels but still getting the revenue stream, mining every bit of data from users for ad placement.

That's not the spirit we want. Peertube is not a product to compete with existing video websites. Peertube is a product you use to self-host your videos, for you, for your band, for your org, for your company, anything. It's about taking control, not selling junk.

EDIT: I fear I might have been unclear, I am in no way affiliated to Peertube

You're proving the point. You see no difference in presenting a product well and "selling junk". The Peertube website's job is to tempt people to try it out. Getting people to use Peertube is the only way Peertube can change anything about the situation where the default video repository is being run by a for-profit company with only its own interest at heart.

That requires what iandanforth refers to as "product and market building".

Making a good, ethical product look attractive isn't evil.

You're right, and Framasoft (the French non-profit behind the project) knows it. If you can help, or know people who can, I think they'd be glad to hear from you!

This is an update announcement on the company's blog. It's not the project's landing page.

If even a household name like Apple puts a so called boilerplate at the bottom of their press releases [1] this reinforces that's a good practice.

[1] https://www.apple.com/newsroom/2022/12/apple-announces-bigge...

Neither the homepage of the company, the linked page to get the update, nor the top level https://joinpeertube.org/ contain any video.

I thought the front page's "platforms" and "videos" section were very good. Got me quickly to the search box and I was watching vids about volcanoes in no time.

then what's this?

https://i.postimg.cc/vTHXhxFC/7089-F268-8-C2-A-4935-A280-841...

A below the fold, secondary tab that contains a link to a separate page with a video?

That's like asking AWS's landing page to have a live database environment embedded into the website. PeerTube explicitely links you to example videos and channels to showcase the platform. If they just embedded a video you wouldn't understand the scope of the project

I came here because the walls of text on this tell me a bunch of very specific features (especially new ones) without telling me what this does. I assume it’s a decentralized video app, that you run yourself, but I can’t find even a little bit of description.

Good.

PeerTube is basically developed by a single guy and has solved the infrastructure problem that many organizations have when wanting to host videos. Any organization can now host VoDs and live events on their own server(s) without being bled to death by the hosting provider since the badwidth will be shared among the concurrent viewers.

This is such an underappreciated piece of technology that I hope many intitutions will adopt, at least as a backup against using the usual Youtube/Facebook/Vimeo/Twitter/Instagram as content hosts

Is it possible to embed PeerTube videos within your own custom frontend or are you forced to use theres? As in, could someone create a clone of the New York Times for example and within each article have a video embedded (from PeerTube tech)?

I've wanted to create a combat themed video site for years, since most sites get censored, and PeerTube always seemed like a perfect solution.

I'm not sure if the following link is exactly what you need (you'll need some more info)...But, i thinki it is in the direction that you need: https://docs.joinpeertube.org/api-embed-player

Basically, what you are asking is possible, and you should navigate around the documentation specific to the Peertube api. (Sorry, trying not to sound like a guy who responds with "RTFM"....but, i guess i sort of am; sorry :-)

But, hey, what you want is possible! :-)

Yup, it's that amazing.

That's why I think said institutions should donate to the project ;)

I truly like PeerTube as a technology, however I still haven't managed to build even a minimal stock of PeerTube based content creators I'd enjoy to follow.

On Youtube it's easy, there's much more high quality content about any topic imaginable than I ever could watch in my limited time on this planet, but on PeerTube instances I haven't found anything that made me come back to that particular creator.

So my question is, can you recommend a good cooking PeerTuber, or a good one discussing the newest single board computers? It's really difficult finding something worth watching.

> I still haven't managed to build even a minimal stock of PeerTube based content creators I'd enjoy to follow.

May I shamelessly suggest that you check out my PeerTube channel that I host on my server? :)

I have only one video so far, but will make more videos in the future.

This channel is about programming, system administration, and computing.

In the first video we (royal we) look at using ChatGTP to solve the first of the Advent of Code problems in Rust.

The video is called “No Brain Required - ChatGPT solves Advent of Code in Rust, episode 1”. So named for reasons that will become apparent when you watch the video :)

https://video.nstr.no/w/6z7PxB4J92H3NHhgMmfYVw

That was a nice first video. Thanks for making it!

Fedi.video has a lot of recommendations, which you may find useful:

=> https://social.growyourown.services/@FediVideo

Why?

From your comment I get that you haven't checked into peertube in a while.

There is tilvids.com that is slowly becoming the premier high quality peertube instance that is taking baby steps into being an educational instance. They will gladly host any creator if they align with their goals.

Besides, if a cooking creator wants to roll their own instance, invite other cooking creators to join in, tilvids.com would gladly peer with them.

The point is, tilvids.com is showing how you can build a good community outside of YouTube. A good step

Thanks for mentioning tilvids.com! You got me to subscribe to like my second Peertube channel: https://tilvids.com/c/veronicaexplains_channel

I'll say that the depth and breadth of topics still seems lacking (lots of meta content about FOSS and the fediverse, which is only interesting to early adopters), but if it becomes easier to start, join, and use Peertube instances, I feel confident this will change over time.

Hey, TILvids here. Veronica Explains is awesome, glad you found their channel!

As far as breadth of topics...you are correct, it's still very FOSS-heavy. That said, we do have some other interesting channels worth following. Here are a few:

Yesterkitchen, dedicated to the history behind recipes! https://tilvids.com/c/yesterkitchen_channel

Let's Talk Philosophy, focused on learning about great philosophers from history! https://tilvids.com/c/letstalkphilosophy_channel

ArtsHole, who covers artists throughout history! https://tilvids.com/c/theartshole_channel

FastPassFacts is dedicated to talking about different historical/social topics from theme parks. https://tilvids.com/c/fastpassfacts_channel

The big challenge is trying to find creators who are willing to use a video site that doesn't have direct monetization. I can talk to them until I'm blue in the face about the importance of controlling your own means of hosting and distribution, etc. but at the end of the day, making content takes a ton of time. They won't do it unless people are watching. So if you want to support alternative platforms, the best thing you can do is USE them, and reach out to your favorite content creators and let them know that you're using them.

sure. this seems to be a one passionate guy operation so its only so much the guy can do.

i literally remember this guy coming on peertube github and reading about him explaining how he was planning to set up a educational instance.

anyway, as i said, this is an experiment supposed to urge more people to take the plunge into becoming hosts and managing their communities. If there is a bad apple down the road, the community can self adjust accordingly.

the point is, people need to be aware about choice. this is doing a good job

TILvids looks quite well curated, seems to have some interesting content.

My main entrypoint is usually sepiasearch.org, where I'm looking for specific topics. And more often than not the results were a bit disappointing.

Question about Peertube: Is it possible to personally view and verify each uploaded video, as the host/admin of the service? (time constraints aside)

My concern is, that if I host a peertube instance and anyone uploads illegal stuff, I am going to hang for it. So perhaps I would like a Peetube instance just for my friends an me, where I can personally review each video, to not get into trouble. Basically doing the job which the big platforms are too high and noble to properly do and use their algorithms for.

Yes, there's a global setting that will mark newly uploaded videos as blocked automatically, they'll show up in a special moderation area where you can decide what to do with the videos. You can also mark users as trusted in which their videos would bypass the automatic flagging.

@zelphirkalt I am not a lawyer, but big U.S. companies like youtube, vimeo, etc. leverage Section 230 of the (U.S.) Communications Decency Act of 1996 in order to mostly protect themselves from scenarios that you referenced (https://en.wikipedia.org/wiki/Section_230). Yes, they still take stuff down, but the intent of this section of the law was to protect YOU and those other companies. Of course, there is big discussions about removing this protection...But for now, this is what those big boiys/girsl use, so why not use it youyrself? Of course, *DO CONTACT* your legal technology representative to be sure.

Governments are trying to make it harder for the average person to host a community, while claiming to hate big tech.....

I thought that platforms were protected as long as they take down illegal content when they are informed about it?

In your country, maybe.

But in most places, that isn't the case. If there is something illegal on your server, you're going to be held responsible for it, especially if you don't have robust evidence of who else put it there (ie. a verified and accurate name and address of the uploader - something that illegal video uploaders rarely leave behind).

Even in the US, I would assume there's some practical threshold where you can't hide behind DMCA and say "I wasn't aware 100% of movies I store are copyrighted and I've acted on all requests so far".

Even if that's the case in your jurisdiction, you can spend a lot of time in jail and a lot of money on lawyers before a judge agrees with you. This is also the reason why it's not recommended to run tor exit nodes at home.

Looking at the article, I can't help but to find the summary of the past releases to be really nice:

  PeerTube v1 (Oct. 2018) allows you to create a video platform with federation, peer-to-peer streaming, redundancy, search tools and multilingual interface.
  PeerTube v2 (Nov 2019) brings notifications, playlists and plugins.
  PeerTube v3 (Jan. 2021) adds federated search, live and peer-to-peer streaming.
  PeerTube v4 (Dec. 2021) allows to customise each platform’s homepage, to sort and filter displayed videos, and to manage them more easily.

Kind of makes me wish most software projects had summaries of changelogs like that. For example, for versions of PostgreSQL/MySQL/React/Vue/Java/.NET or anything else - just to see what the most notable features have been in the releases over the years.

Also, PeerTube itself is pretty nice, I'm still hosting v4 for my own needs and use it as a solution for backing up and encoding stream VODs from Twitch. Might eventually get a YouTube account, but still keep it as a backup just to minimize the risk of losing the videos, though storing hours of them does definitely take up some space on my server's HDDs, backups of those included.

Peertube is amazing. My cheap VPS punches well above its weight thanks to it, and I can host videos that have had (low) hundreds of thousands of views without having to rate limit anybody or force registration.

I'm always quite concerned that it's practically a one man project, though. I hope it can build the community of developers that it deserves.

Has anyone here experimented with IPFS to create a mirrored video archive for things like their YouTube channels?

I don't hear much about IPFS these days.

I have - I've used IPFS to archive the channel of a creator I like (in true data-hoarding fashion).

It is neither convenient nor especially easy. It's mostly managed by a home-grown bash script that uses yt-dlp to archive videos, organized by playlist, and then update a IPNS key.

I plan on writing up a blog post about it at some point, if there's interest. I mentioned the project to the particular creator's fan discord (the creator doesn't participate with the community at all themselves) and there was a surprising amount of backlash from the community leaders. Most of the concerns revolved around the notion that I was somehow stealing ad revenue from the creator (despite my explanations that IPFS is not at all a convenient platform for video consumption). I might take the IPFS node off of the public network to circumvent this worry, and only peer with other parties interested in archival of that specific channel.

IPFS still has poor performance for very little innovation. Bittorrent still works perfectly fine, and in fact peertube uses webtorrent to distribute content from browser to browser to offload the server. It is perfectly possible to mirror peertube channels on bittorrent

IPFS got their fat investment check and got to work on their own crypto-currency (filecoin). I kinda feel like the project lost its way.

The incentives are aligned towards token value and investment growth instead of technology development, inside the sausage factory.

I hope they get back to making useful software soon, but I'm not holding my breath.

At least it's all open source.

I love the idea of PeerTube and often browse videos on tilvids (a fairly curated instance).

However, lately I've been wishing for a tool where lectures could be hosted that has nice searchable transcripts, allows people have have comments and discussion based on the time. Something kinda like Loom but more oriented around making lectures annotateable by students for referring to later.

Someone could write a server plugin that passes the video's audio through the latest Mozilla's STT language model to get the transcription with timestamps. The tools are already there, if you know how to make it happen you shouldn't wait for a single guy in France to reach feature parity with Google

Basically, what I'd like is an open source version of otter.ai. It's doable but I think the key is making something with a nice design :)

What are the privacy implications of using peertube for the viewer? Is his IP address exposed to other viewers? Does peertube offer a way to shield viewers from one another?

It's exposed to anyone who cares to packet sniff. There are features for users to opt out of P2P data sharing, but no way to hide the user IP with it enabled, short of the user connecting with a VPN.

I tried PeerTube but deleted it after it ignored my custom media profiles for AV1 and silently encoded the videos as AVC. No official way of using FOSS formats because the head of PeerTube does not want give people the choice of cutting off Apple users and the official plugin for creating custom encoding profiles just encoded as default AVC when MY AV1 profile was selected.

> because the head of PeerTube does not want give people the choice of cutting off Apple users

Why do you think that ?

AV1 seems to be supported, albeit still with some bugs maybe ?:

https://github.com/Chocobozzz/PeerTube/issues?q=is%3Aissue+a...

Reads like a bug.

Did you file a bug report?

Reply on news.ycombinator.com