ITT we discuss possible ways to replace YouTube

Why don't you show us your number of YouTube visits? no cheating

Torrents are certainly the way to make this work, maybe with IPFS as an underlying technology. Normies used Popcorn Time, I don't see why, with good content, we couldn't push something like that for user-created videos.

On the other hand, the other big issue is mobility. The vast majority of normies (which is where the revenue, and therefore creators come from) browse exclusively on smartphones now. And I'm pretty sure that an IPFS/bittorrent client would be either disallowed by the major app stores, or such a data hog that nobody would want to use it (though unlimited data is becoming popular again).

I don't use a web browser to watch or search for videos on youtube, fag.

You forgot to shill ethereum and bitcoin too.

Naw.

Unless it runs as fast and as easy as youtube, normies will never abandon YT. They have no reason to.

Worse still, anything decentralized simply can't enforce intellectual property - you'd have to make all monetization voluntaristic. I imagine content creators will not be exactly thrilled with the idea at this time, but perhaps in the future when most of their income comes from patreon....

I don't think the using the youtube app on your apple tv is making you look any better champ...

perhaps he is referring to mpsyt

alright kiddo, go be a gnewb somewhere else.

There is a big difference between total decentralization and partial decentralization. When I download starcraft or wow it goes through a downloaded that tries to spool up a little torrent downloaded to save blizzards server upload sheckles.

A partially decentralized youtube could do somethings similar. The servers could send the first part of the video while some form of p2p network spools up under the covers.

You could pay people in meme coins to host content for you, clearly you would have to send the end user hashes or whatnot to ensure authenticity and report back if some third party send them shit files but that's trivial.

Should you be required to do video take downs you simply stop linking people to those videos and remove it from your servers, you could also send your third party hosts notices of the take downs so they could also remove them.

You only send the first half meg or what ever for each video and random joe with his google fiber sends the rest for a cut in meme coins. Advertisers buy advertisments in meme coins and content creators get paid in meme coins. Users can also buy meme coins to donate or content creators can sell access to videos in meme coins.

make it less racist

Advertisers are the problem to begin with. If monetization is the goal then a model will have to be developed to pay for the infrastructure and the content creators without advertising. Some sort of meme coin is a great starting point.

You can't. Once something gets popular it will stay that way (unless they do something retarded - but even one retarded decision won't be enough to kill themselves). I mean, can you imagine all those people with a million subscribers just moving somewhere else? Impossible.

Media goblin is written in PHP, I don't really like it. Hosting videos is already solved anyway.

The only way to replace youtube is to have massive amounts of hosting and bandwidth. Software is not a problem.

I just want a service that lets you upload whatever the fuck you want without re-encoding. The only restriction really needs to be size and bit rate.

Or maybe we can go back to a time where people made interesting content for the sake of discussion. Pisses me off that once something is "monetized" suddenly no one wants to do it for free anymore. I ain't paying some faggot to rant on camera about their political views or what they thought about the latest cartoon. Most of YT's big channels aren't even fucking content "creators", they're just reaction videos where some nigger makes a dumb face at a camera for 5 minutes while another video plays in the background. They use bots and click-bait to up their views to try to squeeze more money out of the automated system. I'd rather have dedicated users that put out semi to good content from the bottom of their autistic hearts than try to create another "clickbait is my career" website.

The problem you can't simply replace it is the same why you can't simply replace all the other established platforms: you can't copy the brand. Even if you leak the youtube-software and make everything the same, people will go to youtube.com instead.

People didn't suddenly reduce their messages to a few words because finally twitter offers them a platform to do so but because people of status and influence go there. There's a reason why cereal-brands pay known athletes a fortune for pretending to like it. That's how marketing works.

Let's face it: what differs Youtube from other, similar sites is that they have a ridiculous amount of MONEY. People go where the money is.

And that's also the reason why all those FOSS/privacy-attempts are bound to fail: there's no MONEY in it. Social Media / selling user-data is a multi-billion-dollar-business. It's all about the MONEY, folks.

An Interesting Solution
Using Ethereum Contracts, people are rewarded to "seed" the video that are on their computers.
The faster they deliver the "blocks", the more Ether they get.
An average bittorrent block is 256KB and YouTube bitrate is 16Mb (or 2MB) per second
Which means it needs 8 "blocks" per second for video to flow smoothly
i.e. each block must deliver at 125 milliseconds per block speeds.
All we need is a calculation on how much people make by delivering "blocks" on time,
and how much extra they can earn by delivering earlier than expected.

50USD 25Mbps
60USD 75Mbps
70USD 150Mbps
How do we calculate then?

Talking into a camera about immigration or feminism or whatever and getting paid for it by advertisers is going to be a historical anomaly.

Big money is already pulling out of Jewtube because Google allowed this to happen.

The future is in shilling yourself to get Patreon bux.

Dailymotion and Vidme have already gong to shit, they purged all "offensive" content to appease the almighty advertiser.

Ever heard of Myspace? MSN messenger? and soon facebook. Well they had millions of users and they went the way of the dinosaurs real quick and got replaced by something better because they didn't want to evolve and they kept on doubling down on their autism.

What's great about the internet is that nobody is too big too fail. Displease your userbase for too long and you will get fucked sooner or later by a better alternative.

There was a time when people used to create content on YouTube for free without expectation of getting ad sheckels in return.

We got to also remember that it was also the time of low quality video. Bad editing if any, and generally low energy video.

All you're doing here, repeatedly, is proving you're too much of a drooling Arch-ricing freetard to know how to count youtube URLs passed to command line tools.

Shouldn't HTML5 make youtube deprecated? Now anyone can throw up videos on their website using tags. Combine wordpress and tags you're already 75% of the way to making your own tube site.

Hard part of creating video sharing website is paying for data transfer.

Underrated post
Huge companies, bigger than Youtube or even Google itself have failed before, are failing now, and will continue to fail until the end of time. It's even easier for companies to fail when they rely on something as volatile as the Internet.
What we really need is a lot of people doing a lot of projects, and just generally throwing shit at the wall until some of it sticks. If enough people start making decent video sites, something's bound to give.

True, but I was thinking less of a video sharing website, and more of people making their own website to host their own videos. It would make personal webpages more popular again. Instead of sharing on youtube, people could just share with any of the bookmarking and social sharing sites that already exist.

an alternative would need something like easier access to ebegging/donations through 'streamlined integration' with Stripe/Gratipay/PayPal/whatever email-acct. webapp

the more steps there is, the more unwilling the common end user is to skip supporting a site and/or creator (hence why MediaGoblin never took off)

This is different though, you're basically going up against a mix of a CIA front and a multinational company that spans the whole globe, it's really a big malevolent beast that has the backing of the most evil and wealthy people on the planet, and they are deadset on controlling the flow of information

So did the EIC, and they're gone now.
They're a company just like any other. And even huge companies, believe it or not, are much like people.
They live, they grow, they stagnate, they shrink, they die. This has always been true and it always will be, no matter how large or intimidating the company may be, or what government backs it. Google will fall. Maybe not now or even in fifty years, but their time will come.
Given the volatile nature of computer technology and the Internet though, I'd wager they won't last as long as some of the other giants of the world, past and present.
Like I said before, all we need is to have anyone with the brains and the will throwing shit at the wall to see what sticks, because something's going to give.

The problem with flat microtransactions is that bandwidth and storage of popular data is essentially free - nobody wants to pay for free shit.

What you want to do is pay for unpopular, *rare* data. That is, nobody thought about duplicating that data before, you suddenly want it, you have to pay archival node to get it. Price is trivially determined by dutch reverse auction market (normally conducted by escrows).

Next, you dont't need ethereum for this, as theres no way to verify data delivery in-contract - you need an external escrow data will be relayed through. And anything that supports double-escrow will do (this includes bitcoin).

Finally, there's no need for transactions to end up in blockchain for each such exchange - instead, the transactions are probabilistic. Say, theres a 1/1000 chance the transaction will become valid, decided by random number elected by the network (ie next block hash). That way you vastly reduce the microtransaction spam.

I love these RP threads

The moment you centralize it like this there's no incentive to go with all the offloading to p2p network trouble.

There's a reason why pandora, blizzard and steam stopped using P2P delivery, and use CDNs instead. It's better user experience, bandwidth is dirt cheap and they don't need that _much_ storage.

That said, the scheme you described is actually deployed, including memecoins - [email protected]/* */ (ehentai) uses this to host manga galleries. It was mostly relevant 10 years ago when bandwidth was still pretty expensive, but nowadays the user run caches stopped being relevant.

The only reason to go this route is to scale massive storage requirements and crowd-sourcing it improves the overall cost. However for that, your scheme is way too naive, because you also need to ensure retention - an elaborate nash game with replica counts vs popularity.

That's very true nobody's too big to fail on the internet.

And if 2016 taught us something, it's that even $1.2B of backing you can still get fucked by the public.

You would need literally petabytes of storage space to archive all of the current videos. Archival of every video under 6-7 minutes would be almost mandatory in order to get enough users to switch since so many people use it for music.

The other stupid thing about peer-to-peer is that older content gets harder and harder to get. Ever want that really old copy of OS X Tiger for an Intel Mac? Well fuck you (every public and most private trackers). Want that song made by some Russian guy in September of 2007? Well here it is and you can load it just as fast as everything else (YouTube).

But what if, this system allows you to earn free Ether?
Also, good idea on fighting microtransaction spam.

github.com/ethersphere/go-ethereum/wiki/IPFS-&-SWARM

You're a fucking idiot.

When (if?) IPFS becomes native to browsers I think this will solve many of our woes. Basically, the more people that watch a video, the more hosts that video has. Supply/demand (download/upload) should be able to balance itself automatically.

I agree but faggots will always try to monetize the internet rather than get a real job.

This, be it ipfs.js or native integration. If the user has to install something, it's not happening, period.

You want to solve a problem yet are unaware solutions already exist.

Bitchute uses torrent tech to power a youtube clone and it just werks because its built ontop of webrtc.

Why does bitchute serve everything through http, just like youtube does?

Until browsers can seed out permanently your watch history, webtorrent is a cool tech demo, but useless in practice - because browsers simply can't accomodate it.

As a high level guess: Some sort of peercast system and web UI over top SWARM. Whereby the pooling system simply groups together users that are of mutually low latency. Web UI for normies, reeeeeeeee.

SWARM can hook you both up.
Basis of the SWEAR subsystem which deals with receipts for data. SWARM is file-agnostic, so if one block can be used to meet the demands of two different files, that's a bonus. It simply allows the collateralisation, and rent payment for the storage of files.
Basis of the SWINDLE subsystem, which allows for both masked and recursive proofs. Masked proofs if you want to prevent the file custodian from simply caching proofs, and with recursion one can hold the receipt to the digest of the video, and go on from there.
Which falls under SWAP, which handles accounting for data transfer, peer tables, and requests. Last I recall, by default it clears minutely, or when someone accrues 1Eth credit. (Bad behavior gets the boot)
It can be extended to allow for more general payment plans, like having the overarching network subsidise the cost. (Or using state channels with collateral)

And to be clear the blockchain ends up being the worst case scenario whenever it comes to a failure in expectation or validation.

I'm aware of swarm. It's a double escrow, as I mentioned earlier.

Just an outline:
* Litigation creates unpredictable incentives, "ddos everyone else until everybody uses my servers". Just burning the bond is not enough of a countermeasure when you're set to destroy competition.
* The emergent vote weight is sybilable. You position yourself on as much diverse spectrum on the ring as possible. Sure you have to pay the entry fee, but it will be fairly modest.
* Unfortunately SWARM can't make strong preference for nodes with larger bond, because it would create a PoS system - rich get richer (by ddosing weaker nodes with smaller deposit), and the incentives collapse.
* Chunk and ticket relaying has huge scaling implications - it creates incentives for nodes to become high-degree (by sybiling), so as to reduce relaying.
* Finally, relaying ignores geography, which is huge internet peering bonerkiller. Can be fixed by S-DHT, but the spec doesn't mention it.
* Another problem is there is almost no viable altruistic incentive to store & retrieve in presence of nasty, profit-motivated players. No "youtube replacement" for you.

I've sperged about it at length on reddit and jewhub. Still, despite all these shortcomings, it's still probably the best escrow scheme out there. Storj and Maid is complete bollocks.

Last I recall on the swap papers, it was going to be some reconciliation of a few kinds of kademlias. The point to make here is that the system presumes that nodes are transient to begin with, and that they have some semblance of self-interest. That is to say they aren't going to want to their deposits to get slashed. (There are weaker arguments for credit score on swap, and the revenue they'd get from continuing services however)
As far as DDOS goes, this is most applicable over SWAP and SWINDLE. Heretofore, swap, is based on bitswap which isn't unlike the fake economy used to optimise node routing in torrent, set aside two things. One, it being part of a real economy, requiring at least a personally tracked account book, and being pluggable.
With the former there's an incentive to track losses from greedy, or unreliable nodes, as well as nodes that fail to clear their debts. This is a mutual occurrence, and can result in embargoes. A node has full control over whomever it wants to route to or from.
The latter is better described as being able to tune or extend the functionality of the node. Tune as in change the slashing or clearing conditions, or something as basic as how much one charges per Gb. The extensibility is there in case one wants to do something fancy like subsidise traffic, or form dedicated routing agreements.
The only part that needs the Ethereum to intermediate is whenever it comes to payments, and a base contract that points payments to your wallet. This can also be used for peer identification purposes.
In the general sense, there is quite a bit of exposure (~sum of debt is 1Eth, or 1 minute passes), but it shouldn't be too too much.
(Cont.)

(You)
Now where does this exist or appear int he specifications or papers? One can certainly short-change swap, but this comes at the cost of needing to reinitialise nodes and base contracts, as well as likely making node paranoid about nodes that they don't have a history with. It's a collateralised reputational system
As far as incentives go, one thing to not ignore is that SWARM opportunistically caches egress. Chunks are file agnostic, so it stands to reason that the local chunk dictionary will be filled with currently or commonly requested chunks, even if they are being used for separate files. This ends up being one of the natural ways that nodes try to minimise debt, but maximise credit, and leads towards natural caching/duplication of files.
As such, I can consider scaling and relaying for swap becomes a non-issue over time. Especially with the likely existence of big/specialised actors that bring in big collateral.
With the existence of SWORD (an on-demand chain/proof storage system for light clients) there should be some amount of activity going on with the network at all times. It's slated to become one of the solutions to dealing with the overhead that light clients usually put on the Ethereum proper, let alone other blockchain propers.
This can further incentivise strong connectedness for the Ethereum network without consideration for actual file transfer.

(You are doing it for nary a nickel)
But you are ignoring that swap relaying considers segment stability, latency, and most obviously, cost and opportunistic caching.
It's actually economically retarded to over-collateralised a stored file because of fat-tailed risks like outages or partitions. And for someone to be willing to endure the exposure, there needs to be some about of benefit. What the rich people have is economy of scale, but I'm not inclined to believe they'll be the majority of the deposits, let alone storage or routing on the system.
As a side note, it's hard to compare Casper with the run of the mill PoS. The run of the mill is more or less PoW, but with the rewards, difficulty, or expected blockk maker being biased by coin age or something. In either case, the rich folk surely win out in either PoS and PoW because of economy of scale, and the only recourse being that they don't get expected revenue.
Casper is a bit different in that anyone who maintains the consensus of the system is also someone that can manipulate the consensus of the system, and are treated as such. Casper is collateralised, meaning to say that one has to voluntarily create a contract and deposit with the network to actually mine, or validate. Similar situation with exposure, it should be with its benefits, esp. when it comes to validation, in large part because the penalties of failure or damages are actually proportional to the size of the deposit.
SWARM follows this maxim as much as it can in its three subsystems, swap/swindle/swear.

(You are back on track)
Regardless to say, but there are not a lot of people that have the disposable income, capital, or pathology/idealism of altruism. It's more common to expect run of the mill people to want to make an honest buck.
So with that in mind it should be expected that the best system is one where profitability, honesty and altruism line up with the alternative being slashed. Constructing such an economic system is more or less the point to cryptoeconomics.
If the network dies under Casper, then the deposit dies with it. There's no recapitalisation of mining capital like one can do with PoW, Fiat Justia Ruat Caelum.
So if there is a problem, the custodians of files get shafted, and lose their deposits.
The escrow here, iirc, is more or less the custodian's deposit versus the owner's rent. As time goes on it can be topped off, but it's a bad idea to drop your deposit on the floor over the remaining rent on the file. Deposits are static sized.
Swindle is the validation or integrity checking system. Receipts return either the file, challenge response, masked challenge response, or another receipt that is monotonically closer to the content address/hash of the file.
Swindle is also recursive in the sense that one can have a receipt point to multiple receipts that each themselves point to files.
A receipt itself is enough information to request a file, or challenge a file in such a way that a file challenge in a cheap but still slightly profitable operation if it ends up being properly responded to. DDOS eat your heart out.
Because receipts/validation can be recursive, part of the specification is that the receipt holder is often given a continuation of the proof as a sort of heartbeat. The continuation can be used to cache the progress of the validation/request, or simply reduce file indirection.
DDOSing would be a more apt term for it because it abuses the overhead of the system. But because o kademlia... it's logarithmic.
Because of the underlying bitswap system, you are unlikely at being able to sybil swap nodes, they are expectant of pay and reasonable requests.
Because of the the receipt system, someone is going to have to pay if the receipt doesn't clear. And any interim custodian of the receipt or file (because custodians can do that) has to also have an outstanding deposite on the receipt.

Regardless, swap/swindle/swear all have their place, and interfaces. For something like an end user, they just need to set their routing bid per Gb, save their receipts, and pay their rent once a year or something.

As far as being a good escrow... Somewhere down the line there needs to be something leveraged for security. Authority, hash rate, ... Why not just use the money directly, and cut out this song and dance?

The most anyone would be able to do with a DMCA or something is de-link the video from the account, people would still search videos outside of the server and you'd be able to upload videos without an account anyways since the website is pretty much just a search engine for IPFS videos.

Aside from that people would need to be able to profit off of videos so instead of ads there could just be a cryptocurrency patreon alternative integrated into it.

no, it is not. it's written in Python

I agree but sometimes I want my favorite YouTubers to get the money they deserve. So I propose a system where you can choose whether you want to watch ads before a video (By default off)

I like this. If you could use the site and post videos anonymously (without creating an account), I'd be all over it.

so, everyone using the service could get sued / three-striked / van'd / whatever at any time because they're potentially uploading copyrighted material / cheese pizza / other kinds of wrong-think?

Yea, that sounds like the kind of service people would want to use

How about a combination of that and peer to peer. The website itself can be a tracker or a collection of magnet links that videos can be viewed from. That way if a video maker wants to use another website, s/he doesn't have to "reupload" the video.

I use a patreon payment service. Most of my favorite video bloggers have one.

Initially YouTube was releant only because you couldn't just post a video on your favourite forum. Now it's money.
People already tried to replace YouTube with shit like Vid.me. Yet nobody wants to use it because you get even less money than on YouTube (at this point most of the people who want to earn good money off YouTube are forced to use Patreon).
I bet there will be replies about shekels and all that but this is the world we live in: people want to get money for things they invest time and effort into (surprising, right?).
If you'll solve this problem -- boom, YouTube is dead.

Wrong.

To me it kind of sounds like the upcoming LBRY project:
lbry.io
youtube.com/watch?v=DjouYBEkQPY

I get the distinct feeling this will only be used by ISIS, pedos, movie pirates and perhaps a porn studio or two.

Actual movie studios don't need this, and most likely don't want this due to the custom currency, and that there is literally no DRM on the material other than the price, so if they put the movie up on LBRY, it'll sit there for about 5 minutes for $20 or whatever is the LBC equivalent, before being swarmed with FREE! copies, or even worse, people MAKING money off pirated movies by putting them up for 1LBC or something. Which will have the movie industry up in arms worse than torrenting.

Letsplayers, which are a huge part of why youtube got popular to begin with, won't touch this because there's no passive revenue. You cannot put a video up for free and expect to earn anything. Instead you're forced to choose between earning anything, and forcing people to pay which will decrease your fanbase. It just doesn't work at all unless you're already super popular and have enough brand recognition to convince people to pay for your videos.

Reviewers have the same issue. No passive revenue, so they're forced to choose between a paywall or earning nothing.

Streamers can't even use this platform because no inherent support for streaming, so they're out.

Your average normie won't be using this either because it's too complicated. They can either choose to press one button on their phone and upload their newly recorded cat video to youtube and a potential audience of millions, or they can choose to wait until they get home in order to painstakingly upload the video to this complicated protocol for maybe a couple of people to see. It's obvious which they would choose.

And that's already like... 98% of youtube's current content removed. So you're left with the stuff that is not allowed on youtube;

- Various types of extremism. ISIS, Anarchy related stuff, etc.
- Violent and gruesome shit, which you can already find on liveleak
- Pirated movies and porn. Lots and lots of pirated movies and porn.
- Various types of illegal porn not allowed on pornhub/redtube/etc
- Leaked government material, stolen army recordings, etc

Which mind you is great for the people who wants those things without the risk of being caught, but still lightyears off being something that can replace youtube.

archive.is/0Dvcm

Shit like this is why I'm excited for IPFS. People often forget that others didn't just make and distribute content for free, they PAID to host it too, back when it was way more complicated and expensive, and they didn't expect anything in return, nobody to pay for their hobbies. They did it just because they wanted to share the content they made. Something like IPFS reduces the cost and complexity of hosting dramatically, now not only can you make content for free but distribute it for free too. The only thing you need is electricity and bandwidth which everyone pays for already.

With the webui I think it's easy enough already but I'm sure there will be tons of user friendly additions when it's actually finalized that will make it blow up. With the current webui it's as easy as dropping files onto a web page and linking them out, you're done. Want to host a website with videos you made? Make the pages and drop the directory then link it out. No need to buy/rent a whole machine, setup the OS, server stack(s), register a domain, get ssl, etc.
You just add the file, and share it with people and it goes from you to them.

Whether it gets popular with the mainstream or not I couldn't care less, a tool like this is too convenient to not be picked up by people I'm interested in already. Look at the attached webm, someone just added an episode of a show and I played it in my media player, I could easily just have that happen in browser too, no need for YouTube or anything, although you could build around it if you really wanted to. Here's a dumb thing I did to show off webm alpha transparency that used to be hosted on pomf, no need to even have ipfs installed to see it.
ipfs.io/ipfs/QmbWbiZjm3TGBfLsNnQ5bRuN8VjXdMwpYyg8R69hEV9Cmb

I forgot, that transparent webm will only work in webkit browsers or Firefox 53+.
bugzilla.mozilla.org/show_bug.cgi?id=944117

Nevermind I guess the index had to be rebuilt.

Who is this semen demon I keep seeing reposted?

While losing all wireless traffic, dooming the project to the same fate as Tox.

We're not going back to that time because there are no jobs left for us. You own nothing, and the guy with the land doesn't need you anymore. We're facing a great filter event, and only the biggest cocksuckers and backstabbers will survive it. Welcome to hell.

I looked all over jewtube and didn't see this, where do you find this crap?
Closest thing I found was under analytics and I think 4kdownloader fucks with it because it says I've got no activity at all since joining.

I leave the 24/7 stream on all the time for BGM if I'm too lazy for winamp or doing something long-term and don't want to swap playlists.
We're trying to work something out where they can accept "DJ submissions" but it's a bit complicated working with so many people not physically at the office. It's basically a bunch of people yelling into webcams telling a mestizo how to do things.
(Not even the CEO lives in the same state! Internet!)

Share .torrent files and there you go. Popcorn time can be the new "youtube".

I wonder who could be behind this post.

There's nothing worthwhile on youtube to begin with, except for uploads of lectures and talks, and some music. Everything else is just really garbage "entertainment" that will literally damage your brain after one hour of exposure.

cancer