this post was submitted on 16 Jun 2024
65 points (95.8% liked)

Tech

461 readers
1 users here now

A community for high quality news and discussion around technological advancements and changes

Things that fit:

Things that don't fit

Community Wiki

founded 9 months ago
MODERATORS
 

With YouTube leveraging its dominance to make the service shittier and shittier, we're forced to consider our future. Yeah we have Peertube, but Peertube is shitty. I consider myself techy and I can't find a peertube instance that's not just one single users' "boring" videos.

So in order to move away from YouTube, we're facing two major issues. No three!

  1. Service: Even in its state of enshittification, the YouTube app is still a million times better than Vimeo, DailyMotion, etc. Introduce ReVanced into the equation and YouTube has a lock-in.
  2. Hosting: Hosting video is expensive as hell and that's a major hurdle to toppling Big Tech.
  3. Audience: People stay where the people are, because that's how they generate money. Peertube sucks because I can't just put in a URL and find random content. Without audience you don't have discoverability, without discoverability, you don't have monetization, without monetization, you only get "boring" videos.

Okay, so the third point is a bigger one and I actually think we need to adopt the Blendle-esque model, until we overthrow capitalism and live our post monetary wealth utopia.

What's this Blendle-esque model you speak of? Blendle was a great app idea that was blocked by corporate greed. The idea was that if you wanted to read an article from a newspaper, rather than pay a subscription, you could just pay 10 pence for the pleasure of reading the article. Win-win? Wrong! Most newspapers wanted a subscription or nothing.

Okay, so how does that work with videos? The idea is that users would put money into a pot. So let's say I have £10 in my pot, at the end of the month, the app would divide that £10 across all the videos I watched in the month and send it to all the videographers. If my pot was £1 the share would be smaller and if it was £100 it would be larger.

Okay, so the service issue. When are they going to finally make Peertube user friendly and discoverable? Wouldn't they be forced to if content creators were attracted? Because it can't just continue to suck right? Anakin? Seriously, search for a video on the Peertube main site and someone in their infinite wisdom thought it would be great to give you a wall of text! Mate!

So now that we got all that out of the way, think of it like salad, this is the real meal now. Let's talk about hosting. Hosting video is expensive and its the barrier to toppling Big Tech. Though middle-size tech should've been trying to do it. If Vimeo added Peertube support, it would be a hegemon, but I digress… Pick the pitchforks back up and re-light the torches! Hosting videos is a huge resource expense. It's why we don't see a crazy number of videos posted to Lemmy, Mastodon and even PixelFed. But what if we could solve that? Not the Fediverse video bit (yes, Peertube, you are a joke to me, kidding!), that's just a byproduct, but what if we could all chip in and distribute the cost? Well, I recently, literally just before I started waffling in your eye. But I present the Interplanetary File System! IPFS for short. Think of it like torrenting, but more user friendly and more seamless. Anyway, I'm thinking this could be the missing piece and it could be the building block that allows video to return to the embrace of the open web? What do you think? Why aren't we leveraging this?

More info on IPFS here: https://ipfs.tech/developers/

For the record, I'm not affiliated with any project, protocol, entity or anything. Peertube didn't kill my puppy and I don't even think my mum even subscribes to my YouTube, so I'm totally looking from the outside in.

all 34 comments
sorted by: hot top controversial new old
[–] [email protected] 24 points 5 months ago (3 children)

It's unfortunate how blockchain-adjacent IPFS is, but aside from that, from what little I know of it, I think it's a fine technology.

But, it doesn't really have any search capabilities, right? Or am I mistaken. If not, it seems odd to criticize PeerTube's search if content discoverability on IPFS is close to nonexistent. (And a cursory Google search on the topic seems to indicate that's the case.)

[–] [email protected] 11 points 5 months ago* (last edited 5 months ago)

Agreed. Ipfs is interesting to build something on top of.

https://github.com/ipfs/awesome-ipfs

[–] [email protected] 7 points 5 months ago* (last edited 5 months ago)

IPFS is doomed because the design does not allow it to grow and stay fast, it's already too slow with barely any users. Trying to find a single file can often take minutes, which is simply unacceptable.

I think torrent 2.0 is a much better technology, one of the very few services where you can actually look up files by their hash, but almost nobody uses it.

[–] [email protected] 21 points 5 months ago (2 children)

because cryptocurrency is a fucking punchline and ipfs is poisoned by it

[–] [email protected] 7 points 4 months ago

I mean there is file coin but that's just to entice people to host files on IPFS. You don't have to touch it at all, I personally don't

[–] [email protected] 2 points 5 months ago (2 children)

What does crypto have to do with IPFS tech? I installed IPFS and nowehere does it mention crypto.

Anti Commercial-AI license

[–] [email protected] 6 points 5 months ago* (last edited 5 months ago) (1 children)

Putting a link to a license at the end of your comment is going to have a severe negative effect on how seriously people will take you.

Edited to remove my disingenuous apology.

[–] [email protected] -2 points 5 months ago (1 children)
[–] [email protected] 7 points 5 months ago (1 children)

You're right. It felt like it was the right thing to say, but it was disingenuous.

[–] [email protected] -2 points 4 months ago

I think explicitly licensing your content is a great idea.

LicenseCC By ND 4.0

[–] [email protected] 5 points 4 months ago (1 children)

NFTs, when implemented properly, would be an item tracked on a blockchain with a link to an asset backed by IPFS, plus some kind of copyright notice granting whoever owned the NFT a copyright licence to do stuff with the asset.

People see IPFS mentioned in the same context as NFTs, and assume it's a scam. People have seen memes about NFTs just being an expensive hyperlink that can go down at any time, so think IPFS can go down at any time. People have seen twitter meltdowns from people who've bought NFTs and then lost access when the previous owner stopped adding redundancy to the IPFS file because the new owner was a moron who didn't know they were responsible for adding redundancy to the files they cared about and had spent their life savings on a esoteric way of getting a commercial copyright licence to something they didn't need a commercial copyright licence to without knowing that's what they were buying.

It's basically just down to NFTs being the thing that made most people who've heard of IPFS hear of it.

[–] [email protected] 3 points 4 months ago (1 children)

So inherently IPFS has nothing to do with crypto, but people read about some external connection and now thing it's crypto or something... Sounds typical.

Anti Commercial-AI license

[–] [email protected] 4 points 4 months ago (1 children)

Yep. You can scare people even more by mentioning that it's also used by Windows Update.

[–] [email protected] 1 points 4 months ago
[–] [email protected] 17 points 5 months ago* (last edited 5 months ago) (1 children)

Why Aren't We Embracing IPFS?

Because it's an overhyped joke successfully utilized by crypto scammers.

Neither content addressing, nor distributed hash tables (or key-value stores, or whatever) were novel ideas.

The combination of the two is not a novel idea.

For p2p, torrents, work as another user already pointed out (initial realease 2001).

For a distributed filesystem, look at Tahoe-LAFS (initial release 2007).

For a full anonymous p2p distributed filesystem, check out (real) Freenet, called Hyphanet now (initial release 2000).

And no, if you need anonymity, an anonymous transport (e.g. using libp2p) is not enough. You need to consider anonymity at each step like Freenet does.

These are three real non-overhyped products one can draw inspiration from. IPFS? not so much.

You can look around for more examples. I always found this Wikipedia page about file sharing in Japan interesting, since it mentions networks not well known to the rest of the world: https://en.wikipedia.org/wiki/File_sharing_in_Japan

[–] [email protected] 6 points 5 months ago

You can look around for more examples. I always found this Wikipedia page about file sharing in Japan interesting, since it mentions networks not well known to the rest of the world: https://en.wikipedia.org/wiki/File_sharing_in_Japan

This is super interesting, I had no idea these networks existed, thanks for sharing

[–] [email protected] 15 points 5 months ago

I mean, classic torrenting has no issue with user-friendliness, look at Popcorn-Time. It's all about the app front-end, proper API to update trackers, and UX. Hell, you could enforce some balance or seed/leech by having a reward system for heavy seeders.

Note that I never used peertube, so it might just be exactly that...

[–] [email protected] 15 points 5 months ago (1 children)

The problem is, you would need a lot of users that are willing to pay for the service to be able to pay enough large creators for more users to jump on board. Considering how many people would rather endure YouTube‘s obnoxious ads than paying their subscription, that wouldn’t be viable. You need both a free tier and the possibility to make money off those free accounts. Because if it’s just free with no ads and voluntary donations, why would the average person pay? And if it’s just paid, why would most people come to the platform if they can get the content free somewhere else? As soon as you introduce the need for monetisation, which a YouTube competitor would need, you’re going to struggle. It’s the same reason paid browsers stopped being a thing as soon as a free version existed. It might have been better but it wasn’t free.

[–] [email protected] 7 points 4 months ago (1 children)

I'm not saying that you're wrong, but I think that people are more likely to put up money if they believe in the model instead of because they are being nagged into it. For example, I have a nebula subscription that I happily pay while I refuse to pay for a yt subscription despite the fact that I watch youtube a lot more. This is more out of spite towards youtube than it not being worth the money (it probably is to be honest). I also donate money to wikipedia while I haven't ever considered shelling out for encyclopedia britannica for example.

Video hosting is of course very expensive so I understand that it's harder to fund wikipedia-style than wikipedia. People are probably happy paying creators they like but less so spending a ton of money on infrastructure.

[–] [email protected] 3 points 4 months ago

That is true for a lot of people but probably not for enough people. I agree, that a platform people believe in is more likely to receive donations but that’s not gonna be enough. Platforms like Nebula and Floatplane only work because there are tons of big YouTubers advertising for the platform on YouTube because they benefit from that. And tons of people primarily subscribe for the extra content provided by the creators they already like, not necessary because they believe in the platform.

[–] [email protected] 7 points 5 months ago (1 children)

I'm not sure what your problem with peertube is.

There a cross instance search. Here are results for "Anakin". Actually every search on the instance searches for videos on itself and connected instances. Here's an example from libre video.

Monetisation can be implemented. Beetoons did it with crypto, so instead of reinventing the wheel with IPFS, it's possible to write a plugin that does exactly what you suggested (Blendle-esque).

IPFS could have a place in peertube too for storage distribution (see issues on the topic on github).

Anti Commercial-AI license

[–] [email protected] 2 points 4 months ago (1 children)

Can you please ELI5 peer tube and how it works?

[–] [email protected] 3 points 4 months ago

Here's a 2 minute video with animations that hopefully explains it better than text.

For a simple text version (not the transcript of the video):

You know about youtube. It's a website that hosts videos, allows you discover videos, search them, comment on them, upload them, and lots more. Maybe you also know about similar sites like vimeo and dailymotion. Those are separate sites. The comments you write on youtube, the videos you upload, the accounts, and so on, aren't visible on vimeo and dailymotion, and vice versa. They don't speak the same language.

Just like lemmy is a federated reddit, peertube is like an federated/interconnected network of youtubes, vimeos, and dailymotions. They can all talk to each other. If youtube, vimeo, dailymotion, and other websites used peertube, you could upload a video to vimeo and watch it from youtube, comment on it from dailymotion using our dailymotion account, and (here's the kicker) like it from your mastodon account!

The language peertube speaks is activitypub. The language mastodon, lemmy, kbin, mbin, misskey, and so on speak is activitypub. They aren't all fluent and don't speak with the same accent or dialect, but most things can be understood. For example lemmy cannot see the comments on a peertube video.

Anti Commercial-AI license

[–] [email protected] 4 points 5 months ago
[–] [email protected] -4 points 5 months ago (1 children)

I didn't read that long post, but creators cannot move to a platform that does not offer some payment for their work.

[–] [email protected] 18 points 5 months ago (1 children)

Not being funny, but why post if you can't be arsed enough to read? What's even worse is that you raise a point that's literally covered. Genuinely the worst type of person on the Internet is the person that wants to enter a discussion without bothering to put in any due diligence.

[–] [email protected] 7 points 5 months ago (1 children)

Some people just want to talk and not listen (or read, in this case)

[–] [email protected] 2 points 4 months ago

90% of Reddit threads are just people talking past one another. Very common use case for social media!