If we want good content und independence from YouTube, perhaps it is time to see, that it might cost us a tiny bit of money each month, to uphold our freedom. One rented server can easily support many users, each of those can chip in to finance renting the server. It can be done transparently, so that users always know what they are paying for and how much is covered already for the month. It is time to learn, that we need to support what we like financially, so that people can actually live from providing us with it. As long as we do not do so, people can only provide us with their creations by generating revenue in other ways.
This is the answer to so much of what's wrong with the modern web. We need less business and to get more communities back into the center of the web... it should be even easier today since IaaS is so much cheaper, I think it's just a focus problem, nothing is relevant to the big sites that act as the lens to the web unless it's directly pumping money somewhere.
From a purely-technical lens, that may not seem like a fatal flaw. When viewed through a whole-product lens, it is. PeerTube's anonymous, censor-proof nature is enough to attract a certain class of content that repels advertisers, but then what? Until the dark side of distributed technologies is addressed instead of touted as a feature, this doesn't have a chance of popular success.
that's still "generating revenue". maybe not profit, but servers and bandwidth cost non-trivial amounts of money and somebody has to pay for that. whether it's donations, or user-fees, or ads, or private benefactors, the revenue has to come from somewhere. but if a video service has no answer as to how it can sustain itself, it's questionable whether it's worth the effort of developing it or developing your audience on it.
There are also people who get funded through sites like patreon, who might not care too much for youtube advertisement or YT freemium. Sometimes you even see youtube personalities switching to twitch or other platforms just in order to escape youtube and do legal activity which is just close to impossible on youtube. There is an endless stream of youtube videos complaing about youtube copyright system and how it makes their job close to impossible in some cases. Game reviews and movie reviews are two areas almost dominated by the issue of false copyright claims, which has forced some creators to abandon the platform as a revenue sources.
There is even float plane which was designed from the ground to solve the youtube problem, created by a company that fully relied on youtube for its existence. If your company existence is depended on a free service provided by someone else, you might want to reconsider how safe you should feel. PeerTube might serve a similar purpose.
On top of that we have scientists who might want to know that their research isn't taken down because of false copyright claims or fleeting politics. A company that has linked instruction videos in their books/teaching material might want to be sure that links still work 10 years from now. Activists might want to avoid the risk that their documentation doesn't go dark from a political requests or false claim. Police brutality victims might want to be sure that the copyrighted music playing in the background, placed there intentionally to get copyright claims, does not stop them from uploading videos.
What is the incentive for anyone to run a Tor/i2p/IPFS/Torrent seed node?
This is difficult for normies to understand, but money isn't the only thing that incentivizes people to do things. There are many people for whom "Fuck YouTube" is the only incentive required to participate.
Work. Usually for a three letter agency of some sort -at least for the first two.
For exmaple Wikipedia, thausands of people write articles, for free, they got no money. Huge amount of open source developers publishing their work for everybody, free.
Waste? They'd be using storage that would otherwise most likely be sitting there, unused. Or in other words, wasted.
The incentive is to make content you like accessible to more people. Of course you can do it in your "garage" if you have huge hard drives and good connectivity, but you can also do it as a collective by setting up your own instance and mirroring other instances aligned with your values/interests.
There is exactly zero incentive for an instance to mirror public instances where people can upload terabytes of trash content. But there is strong incentives to replicate well-moderated instances with quality content, because the cost of that is very low (you can rent a VPS with 2TB HDD for ~20€/mo these days).
> Exacerbated by the fact that the entire thing mostly seems to attract people who get banned from anything else
I don't think that's the case. For now i've mostly seen established non-profits (eg. libre software/culture NGOs and workers cooperatives) who were disenchantized with Youtube. I've seen exactly one white supremacist instance, and it got delisted from "official" instances list in under 24h.
This could be an avenue for tech savvy sex workers who need an alternative to OnlyFans.
There needs to be some kind of peer to peer rating system so you can find content suited to you (reinforce your bubble I suppose) because there is going to be no central filtering authority, but that's part of the point.
The trending page on PeerTube shows a list of videos with at most a few dozen views.
Is PeerTube supremely unpopular (seems to have been launched several years ago: en.wikipedia.org/wiki/PeerTube), or am I missing something? Does it really have an order of magnitude more Github stars (https://github.com/Chocobozzz/PeerTube) than daily viewers?
Another major problem that prevents creators from hosting on PeerTube is that there are no ads - many see direct-from-YouTube revenue driven by Premium views/Ad views generate a significant portion of their income. Even Linus Tech Tips, with their audience of gamers who mostly running an ad blocker, sees YouTube generate 26% of their profit[0], almost exactly as much as they get from sponsors spots in their videos. If everyone were to move to PT, we'd either need an intermediary for ads, or only creators with audiences that could afford to personally finance them (via Patreon or similar) would be able to survive and make a living off of running their channel.
If you move your content to PeerTube and want both an audience and to be paid, you'll need a combination of third party services and other social media, and will likely end up with similar problems re: advertisers cutting out "unacceptable content".
I do think it'd probably be better if the content provider was divested from this layer and focused on only providing media, but Peertube doesn't really solve these social/economic problems people complain about with Youtube.
Is it so hard to be on both? Something I like about YouTube compared to some other platforms is they never (to the best of my knowledge) engage in exclusives or impose exclusivity conditions. And even when they develop new features, I've rarely seen Google try to restrict those.
You can consume peertube content with any activitypub client that supports its types. In short, peertube is an accessible way for users to publish their videos to activitypub.
[1] https://github.com/owncast/owncast/tree/gek/activity-pub-1
One reason of seeming more inpopular than it actually is, lies in that while there are many PeerTube instances, they are not all federating together. They are selective and some are not federating at all.
Also I am not sure if counting of views works all that well yet, and Likes require either signing up to individual instances, or spreading of video link via Mastodon to get liked (via ActivityPub integration).
Both the federation and the level of integration with other Fediverse apps is a work in progress and steadily improving.
...that's quite a lot even on YouTube, no? I always assumed the videos recommended for their sheer popularity were a "top 1%" sort of situation.
So I'm guessing as long as you post regularly you can get thousands of subscribers without much effort.
That's because their “landing pages” (project advertisement websites) aren't instances. The home page of any instance (e.g. https://diode.zone/) will show you the content visible from that instance.
But, if you’re going to be called “join peertube” you need to have the sort of content that makes people want to join: copy about what peertube is + a list of instances isn’t that.
That is a technically correct answer, but it misses the point: they're asking users to sign up before they even see the product. They're expecting that the premise of the site alone is enough to entice users to join, which is very optimistic. It would be better to immediately direct the user to content, and then provide reasons for why they should get an account. At a high level, that's what YouTube does.
Try PeerTube. Here's a cat video. Posted 1 week ago, 38 views, duration 10 seconds.
https://peertube.tv/w/tsvCimhEaLSTZD16B3gqBQ
Even as the only watcher, and with gigabit Internet both ways at my end, it stuttered, then stalled completely.
For short videos, you'd be better off putting them on a shared hosting site as .mp4 files.
Peer to peer video hosting is just not a good idea for bandwidth reasons.
Peertube, on the other hand, encourages distribution of content in a peer-to-peer manner. As a person, you can use your favorite Webtorrent [0] client to help seed content from far away servers to your neighbors. As an instance operator, you can opt-in to seeding for another instance (with a disk quota limit) from your own network, helping videos published on the other side of the world reach your local audience.
Example: take three hypothetical FLOSS-oriented instances floss.fr (France) floss.cn (China) and floss.ml (Mali). If those instances all "follow" one another, with sufficient disk quotas, all videos will be replicated across the 3 servers, and video streaming will be smooth from those 3 locations (and neighboring regions).
In all cases, even when initial streaming is slow, Webtorrent scales well because most clients (unless they opt-out) will seed videos via Webtorrent protocol (WebRTC + STUN) so as a video becomes more popular it becomes easier/faster to access without placing the infrastructure/economic burden on the Peertube instance. Clever stuff.
[0] There is a reference Webtorrent client, but libtorrent recently started implementing Webtorrent support so that should become more broadly available "soon".
Instances like https://tilvids.com/ or https://video.ploud.fr/ have better bandwidth.
It's both. PeerTube instances federate with each other, but the video player uses P2P to spread the load between viewers.
It's just like you create a blog: you need to find a host for it and most websites don't allow you to just start uploading content.
A growing number of non-profits are starting to embrace it and for that usecase Peertube is perfect. If there's no peertube instance for your community/usecase, you could start one with a few friends/neighbors.
How did this even make it to the front page?
How do people believe this is even remotely true?
I have so many options to host videos, least of which is just housing itself on my own damn server!
I don't know if the complaint here is that no one else will do it for me for free, or that no one else will just hand me viewers for free.
I don’t think Facebook owns YouTube :)
Sadly peertube seems to have to buffer from the beginning to play from this position.
That's weird. What browser are you running? Is it downloading video directly from the server or is it doing WebRTC P2P download?
The former (direct download) is intended as fallback mode and i believe it uses HLS chunking which may or may not be well supported by your browser.
Why would any creator spend all that time and money making content to give away for free? Do you give away your daily work for free? They can easily use something like Patreon or one of the many pay-only tube sites if they're giving up on ads and asking for donations/subscriptions/etc.
I mean, I'm not against the idea, it just seems to fundamentally misunderstand why creators choose youtube. If Youtube paid $0/1000 views... the majority of the top 1000 creators would be gone overnight.
I understand the point of big budget video creation, but that's not most creators, just like most companies aren't Google and shouldn't be trying to solve their problems that are a 1000th the size with Google-scale solutions.
Simply, there is a place for small budget videos. I'd even argue that the majority should be small budget, and that advertising and budget inflation have on average been more bad for YouTube than good.
Most creators/uploaders do not monetize their works. This is just the tiny minority that Youtube highlights to keep their business flowing. A lot of content is produced by amateurs and non-profits, and Youtube does them a huge disservice by displaying ads (which they don't benefit from), tracking their viewers, recommending bad-quality/outrageous content (eg. conspiracy videos), and copyright-striking their videos even when it is clear no law was violated (eg. fair use).
> Why would any creator spend all that time and money making content to give away for free? Do you give away your daily work for free?
I definitely would not give my work for free for companies to profit from, but i'm happy to engage in free labor for non-profits and for my peers. This is a well-known paradox in the libre culture/software, especially with non-copyleft licenses (eg. MIT) which enable a for-profit to siphon of volunteer work without ever contributing something back in return. See also: OpenSSL maintainership "scandal" back in the day.
> If Youtube paid $0/1000 views... the majority of the top 1000 creators would be gone overnight.
Would that be such a loss? Looks like these channels are just outrage/advertisement machine with very little content creation going on... at least form an outsider's perspective.
Well, creators can still point to their Patreon (or Liberapay) without problems. I see many creators relying more on Sponsorships and other income streams as significant (apparently YT doesn't pay that much unless you're incredibly popular).
Dave from EEVBlog is extremely transparent on YT income, worth checking out:
https://www.youtube.com/watch?v=wU3FAIaAF0I
It seems LBRY gets around this with micropayments. Could be an interesting avenue for PeerTube. Maybe you could "charge micropayment credits" and those would be redistributed to creators according to your likes/dislikes/subscriptions.
Have a contact form where people can request an account. Create accounts for trusted peers. I'm unaware if Peertube has a "moderate first" setting but it should not be hard to implement... it's just not the assumed usecase of Peertube so far.
However, this concern is mitigated because Peertube uses open standards, so it's very easy to use various proxying services for privacy concerns. It's unlikely that any such 3rd party clients will ever get blocked.
Comparing to Youtube: a single malevolent company tracks everyone's actions for nefarious purposes, purposefully changes internal "protocol" regularly to break 3rd party clients (youtube-dl, Newpipe) and blocks scrapers and other archival systems.
Overall, Peertube has much more privacy-friendly capabilities than Youtube ever will have.
> Friendly Reminder: the sharing system used for this video implies that some technical information about your system (such as a public IP address) can be sent to other peers.
GitHub issue upon which this banner was decided upon[1]
PeerTube uses BitTorrent under the hood.
With libtorrent starting to support Webtorrent, we should "soon" see a new generation of torrent clients who can bridge the gap between those two protocols.
Peertube reuses an established social networking protocol, adding Bittorrent-like P2P video seeding on top (using Webtorrent protocol, not Bittorrent), taking the best of both worlds in a "low(er)-effort" manner.
twitter . or fb or mastodon or whatever. social and video are orthogonal problems to solve (utube didnt start as a social network)
To break it down further - it's not much of a deal breaker to go to "coolvideoguy.com" instead of searching "cool video guy" in the youtube search bar.
A big part of what I consume on YouTube is on my TV, thanks to subscriptions to channels. Since there is no Android TV app for Peertube at the moment, the only way I would watch content from a Peertube channel would be on my computer. Which is not how I like consuming this content.
Slightly off-topic, but do you think having a computer you don't control running in your TV is such a great idea? It sounds like an IoT/security nightmare, unless you can flash a system you control on there.
This video is not very good however, there's something not very stimulating to the imagination about moving from youtube yet still using shallow buzzwords like "content creator", "monetise", "open source".
- every node needs to keep a complete history of the ledger so as the network grows only the biggest players can stay up... what does that say of Sybil/51% attacks?
- every video published has a public record, and i assume everyone paying to watch a video is also recorded publicly; no private communities
- their blockchain can only (according to odysee website) process 10K events per second... which they're proud of for some reason? that's a very low number compared to eg. Bittorrent DHT or Fediverse activity
- LBRY appears to favorize traditional "intellectual property" and pay-per-view model, not a donation-based economy of libre content
- LBRY relies on a "cryptocoin" to achieve their economic model, and their tech was not conceived to prevent speculation; i would even argue it's architectured for speculation, as it's based on Bitcoin without learning from any of Bitcoin's weaknesses/mistakes (from what i could read) ; Tribler/Taler sound like much more reliable solutions for such microtransactions
- LBRY is pushed by a for-profit company, whereas Peertube is developed by a bunch of volunteers, coordinated by the french libre culture/software non-profit Framasoft as part of their "Degooglize Internet" campaign [0]
Peertube should fare just fine with that. It's just important that instance operators consider the broader implications of running an instance:
- you should find an organizational/financial model to still be around in a decade
- you should find like-minded instance operators to seed one another's content across continents, to ease reachability
If those two aspects are taken care of, "long tail" delivery should really not be a problem for Peertube.
On mobile I'm not the kind "I need an app to access that service", but on TV, where I consume most of my YouTube content, I DO need an app. Because otherwise the UX will be an absolute nightmare.
https://diode.zone/download/streaming-playlists/hls/videos/b...
How do you get rid of the giant left column, other than zooming way in?
- clients viewing a video will seed (share) it along, so as a video becomes more popular it becomes more widely available
- other instances can seed one another's video to help them reach a broader audience (geographically speaking)
I have no clue about how this instance is operated but it's a fair bet that it's not replicated by other instances (yet), and that it's operating from a network location that doesn't have very good connectivity with you.
How so? Just because something is not popular (yet?) doesn't mean it's lame. Also, view counts on a single instance are not representative of the entire federation.
Although i agree removing view counts would be good, because it produces bad incentives to artificially inflate your view count (which i've never witnessed on Peertube instances so far, but is possible).
So yes, exploited by ads.
Basically, it's user experience degradation until you're forced to sign up.
How about trying to just reach out and offer a service without killing my current experience?
I pay for other subscriptions so this isn't a money thing, but I won't for youtube on principle. They ruined that platform.
