There's something really sad when these sites die, a part of my teenage life is gone for good. I felt the same way when Demonoid shut down.
Can these sites skate the legality line by offering only magnet links? Why are their costs so high if they can just host magnet links and cache the site with Varnish or something similar?
>Why are their costs so high if they can just host magnet links and cache the site with Varnish or something similar?
Just from personal experience, but here is my view of it.
A good search engine for millions of torrents costs several hundred a month. Scrapers to keep torrent seeders / leechers up to date cost another fifty to hundred (depending on how up to date you want your torrents). Now add in a few website servers and a load balancer for another hundred or two. You are looking at three to five hundred a month. Doesn't seem like a lot, but this is just a smallish torrent site. Isn't large enough to generate ad revenue or have enough donators to keep running. Maintaining that by yourself soon get's expensive unless you have a good job with plenty of disposable income.
Trackers already do a lot of work in keeping peers up to date. Adding a subscription system would probably overload many trackers and make running one of those extremely expensive. Granted, I would have loved a system like that to subscribe to new torrents that get announced to a tracker with peer updates every so often, but that sort of addon would be incredibly high resource intensive.
A better way IMO is to allow the DHT network to query a little bit more information and have arbitrary data attached to a torrent. BEP 44 [1] should really help with that and actually might be a start with truly decentralizing the whole system. With a large enough DHT node base, you don't need to run a tracker to get peer information. On top of that, if you created an immutable record for each torrent with the torrent information (or ran a lightweight torrent system to pull torrent information from a single peer) you can easily get the torrent information to store in the search engine, and then query the DHT for unique peers to get peer information. You can then create comments or ratings with mutable data signed by your server to keep track of good / bad torrents.
I built the first half of that (DHT Crawler essentially) but it was always the peer information that was a pain. It was easier to grab the information from a tracker
DHT crawler written in erlang. It can join a DHT network and crawl many P2P torrents. The program save all torrent info into database and provide an http interface to search a torrent by a keyword
I've been imagining a grooveshark like site, run on magnet links, using browser based torrenting, and with all URLs and playlists user contributed. What's the legal precedent for suing that out of existence?
Pretty much all torrent index that have closed ? They were in a very similar position (no content is hosted, but they help piracy) and we can see where that lead them. Your idea is probably even harder to defend, because you'd be sending the webtorrent code directly along the list of torrents, instead of having to copy-paste a link in another application. As a user, it feels like I'm just clicking on a button to download (or play) music, so it feels like your site is directly distributing content (we can talk all day long about the technicalities, but we know they don't matter as much as we'd like)
At some point I hope the media conglomerates sue Google and lose. Long has it been the case where I can search "watch x [for free|online]" and Google will helpfully point me to a site where I can do just that, no need to install torrent software or anything. The difference from a torrent/magnet link hosting site may just be one of broad vs specific searching purpose but is that enough to continuously keep Google clean and torrent sites dirty?
That's because the facts have shown us that when it's about IP, it doesn't matter whether you're right or wrong:
- either you have no money at all, and you'll probably get a pass
- or you have some money, and you'll get sued until all that money goes into the rights-holders' pockets or your lawyers' pockets, whichever comes first
- or you have a fricking huge amount of money, so much that you can withstand any amount of trial thrown at you and probably come out victorious in the long run
It's an attrition fight, and Google is the clear dominant fighter in the arena
Yes, and they don't install malware. (At least, they've never infected me. But it's not like the media center PC is an unpatched Windows using IE and no adblocker. That'd be the honeypot (if I had one).)
What if the website itself doesn't do web-torrenting—in fact, it doesn't even make the magnet links links, instead just showing raw magnet hashes as text—but there also happens to be a well-known "unaffiliated" browser extension that does "gradual enhancement" of any site using the magnet-hash text "microformat", into a web-torrent download button?
The site doesn't link to copyrighted content. It doesn't link to anything, in fact.
And the extension doesn't contain any references to copyrighted content. It contains no content, and is completely unaware of any sources of content. (No explicit whitelist patterns for sites it should run on, etc.)
And—at least as far as anyone can tell—the two things have different authors.
But put them together, and oops! You've got a piracy program! User's fault for putting them together, of course.
Hashes whether linked or not are still gateways to copyright content.
What you're describing is not that different Google in some ways. ISOHunt tried to make the Google argument about just being a search engine... but the distinction is that your visitors come specifically for torrents (mostly of a copyright nature) and your index provides copyright content. Google on the other hand is a general purpose search engine where copyright infringement make up a small percentage of their search. Your site would have to be a general search engine site to have any compelling legal argument.
You define technical subtleties, but again the intent remains the same: sharing content without asking the relevant rights holders for authorization. You will still provide a search box to the user, where they will enter the title of the latest episode of the series they want to watch, and you're going to give them the relevant infohashes. You're trying to obfuscate but we all know what your end goal is.
You could alternatively try to argue that it’s not a “website,” but an app. Google calls certain websites “progressive apps,” for example, so it’s very plausible. That would be interesting.
Yes actually: redacted.ch is the continuation of it.
Quite a large community at the moment. I would however say that the original torrents that were uploaded when they first re-opened it (it was called Pass The Headphones) a lot of the torrents were junk, but the moderators have done a wonderful job allowing torrents to be "trumped".
There's a couple of invite requests in this thread, so I thought I should point out that one of the rules at Redacted is:
> Even if you offer the invite in a personal message it is still considered a giveaway and against the rules. Responding to public invite requests is prohibited.
Redacted has an interview process if anyone's interested in joining the site, but don't know any current members personally: https://interviewfor.red/en/index.html
Sorry no invites to give, unfortunately. I used what.cd briefly, but I received an invite from an admin who is a good friend of mine growing up, and he's fairly high up the chain. There's definitely the same level of love and care put into the site as before.
Of course, like what, it's really hard to get your ratio up without uploading your own albums, or trumping others. They do however give freeleech tokens every so often.
Well there is the public channel but there was initially a private channel recruited from the early public channel that was much better and more useful but we all got into redacted and things died down.
What was one of the first ones I'll never remember the name of? I swear it was part-run by the guy that actually invented Bit Torrent, but it also might've been a spin-off. I can visualise the website in my head but I can't place the name!
If you have time would you mind sharing the top 3 unique/favorite features? I'm not an expert in this space but I'm curious what it was that made BTJunkie so special.
The search results felt a lot like Google's. That is, the end user puts in minimal effort and automagically their torrents showed up with high ranking. The site itself was fast and minimal. The torrents had comments where an active community left notes (e.g. media quality, warnings, etc)
Perhaps your project can include a "memorial" in the readme or some other wiki page as sites shutdown, including the site url, any claim to fame, and the date it died. This page may eventually serve this purpose: https://en.wikipedia.org/wiki/Comparison_of_BitTorrent_sites
As it is, if you were to [edit:actually link] the list of sites you currently support I'm certain your project would draw some unwanted attention!
I'm continually surprised that Bittorrent is still relying on content indexing websites. The edonkey community solved this problem years ago with fully distributed search that only relies on the peers themselves[0].
Can anyone with insight into the BT comunity/technology offer any ideas why they never moved in this direction?
Check out Tribler, they have the torrent search thing working in a similar way to Kad from eDonkey land.
Their anonymous communication is pretty sketchy security wise, I wouldn't rely on it, but their DHT search is pretty good and their client has some cool features like streaming video too.
The other route I've seen people go is that of ZeroNet where you can have a central authority or a service-based approach like traditional torrent sites, but P2P-based distribution of the site.
Overall, Bittorent sort of has moved in this direction, they've added support for trackerless peering, 3rd parties have created search tools, but I think many people still rely on the websites because they offer additional functionality like malware removal and commenting and popularity in general.
You can see the huge benefit of this pretty easily if you compare bittorrent to other things. Almost every time I download a game or app from usenet, it's got some script kiddie's trojan bound to it. Even from the private ones. On the other hand, with the big public bittorrent indexes that is extremely rare because they tend to get flagged pretty fast due to sheer popularity more than anything else. Only the ZeroNet approach would offer a way to solve this that would be comparable to existing torrent sites today.
Distributed indexing is very hard, slow and easily gamed. If you remember Kazaa, for example, it had an enormous problem with low quality content and malware.
The best solution is not fully distributed indexing, rather decentralized indexing, such as IPFS.
Its because bit torrent trackers are much more than just an index of torrents. They need a huge support system with forums, membership systems, point systems, funding drives, etc. If you look at the software that runs these sites you will see that moving to a distributed platform would not be so easy.
> none of that is relevant if you're not using index sites in the first place
One of the reasons that bittorrent has been successful and all the other protocols have failed is precisely because it allows for the use of private index sites.
This allows for the creation of custom software that can provide all those extras (forums, comments, ratings, descriptions, etc.), but most importantly, user tracking.
If you track the user, you can solve the sharing problem. For someone to download, someone else has to upload. Using index sites allows you to solve the hit-and-run problem, where someone just downloads, but then doesn't upload. You just ban those people. You then end up with a community where everyone plays nice, which leads to fast speeds and an eagerness to share content.
The "hit-and-run" problem turns out not to be that big of an issue really - there's so many public torrents out there and so many people that want the files that things mostly seem to work out just fine.
Sure, you won't max gigabit like you will on some private stuff, but unless that's an absolute need, issues with public trackers are fairly rare. In fact I've occasionally found files with only a single peer or two on a private tracker which would barely download which were much more readily available on a public tracker.
> Sure, you won't max gigabit like you will on some private stuff, but unless that's an absolute need, issues with public trackers are fairly rare.
Well, that's kind of the point. If you can download something that's 3GB in two minutes, it fundamentally changes how you deal with media.
There's also the issue of security that I didn't mention. Public torrents are visible to everyone, so your download or upload is an action viewable to everyone. With private trackers, your actions are only viewable to the members.
> Well, that's kind of the point. If you can download something that's 3GB in two minutes, it fundamentally changes how you deal with media.
Not really in my experience. I'm a big usenet guy, but I still horde a local collection - if you're using a private tracker you don't want to be re-downloading things all the time because that'd cost you ratio and you still need to hold them long enough to upload them a fair bit. Usenet is really the closest you get to not needing to hoard, but there's still a risk with that, so you often keep it around.
I wouldn't say having something in 2 minutes vs an hour, but still needing to keep it around "fundamentally changes how you deal with media". Especially when you can freely stream a torrent using peerflix off a public tracker, but can't off a private one due to ratio concerns.
It's 2017 and I'm so surprised that BT is still the dominant way to share copyrighted works illegally. Why hasn't a fully decentralized and E2E encrypted alternative replaced it? Who wants to actually run trackers these days?
People flock to huge public trackers and reputable private ones for the accuracy of finding what you are looking for. On fully distributed systems, anyone can call anything whatever they want, and you need an authorship or rating system for users to give support to reputable sources.
I don't believe any decentralized network, including ipfs, has come up with a way for consensus accreditation of a source as being legitimate. If you look for a movie on piratebay, you have the ratings users give the torrents, the comments they write on it, and the number of active swarmers on that torrent all as credence to the legitimacy of it. If you lose any of that information you lose accuracy in finding what you actually want.
Bittorrent doesn't need trackers anymore. Use DHT to find a node with part of the file you want, and peer exchange to query that node for the other nodes that have the rest of the file.
Because it has the lowest friction ? There are dozens of software speaking bittorrent already (instead of only retroshare speaking retroshare), there is a lot of resource on the web explaining how to start download, there is no need to gather friends as a F2F desires.
If I work for Fox (or whomever), I share a file called Logan.2017.1080p.WEB-DL.H264.AC3.mkv but it's actually random bytes. In fact, I have a bunch of virtual hosts all around the globe sharing the same random bytes under that filename. How do you tell the difference between what I'm sharing and the file you probably want that isn't garbage?
The easiest way to date as been to have a website that indexes torrent files. As that grows increasingly unfeasible, how does a fully decentralized system fulfill the same function?
I learned about IPFS the other day and it seems like an interesting alternative. Although I don't believe that bandwidth can be shared among nodes (seeding).
I believe you can still get caught with such systems, because if somebody can download from your machine, that somebody could just as well be a government agent.
Unless it uses something like Tor, but in that case it could become illegal to run exit nodes. The person running the exit node could legally be considered an accomplice.
It's hard to imagine how you could operate a successful torrenting site in 2017; they represent such a loss of income to so many people with expensive lawyers.
sleepychu's point stands either way. It's not a question of whether there's actually a loss of income. If the people with the expensive lawyers think there's a loss of income then you're gonna get dragged into an expensive lawsuit.
Obviously it's not a 1:1 ratio of lost revenue to pirated material. But I bet many pirates would spend some money / watch commericals if they were forced to for entertainment.
I've seen some suggestion that they spend more on average than the average consumer. But I suspect they'd spend even more. And getting some content free and paying for others has discretionary effects.
There is more free content made every day online than one person could consume in their entire life. Even when 99% of it is garbage, that 1% is still more than you can fill your time with.
I read a few dozen webcomics, most of which have patreons, but I don't get ads on my rss feeds for them, and there are thousands of webcomics. And artists. And musicians. There are dozens of websites for hobbyist, amateur, and even freelance professional games, animations, etc.
There is a fraction of pirates that will pay for content they would otherwise torrent because they want that content. But that doesn't imply they are more likely to buy entertainment. A portion probably just turn to free legal sources.
Heh, easily. If you are technically inclined enough to operate a successful torrent site, it shouldn't be very challenging to operate such a site anonymously.
Buying bitcoins, and then "laundering" the coins is relatively trivial, you just run them through a few of the more anonymity-inclined altcoins.
Using those bitcoins to get torrent-friendly hosting? Hit up Infium in Ukraine, they won't mind a few threatening abuse emails. https://infiumhost.com/
Domains you can get from Njalla, Namecheap and a plethora of Chinese registrars who won't even read English abuse emails.
Want to earn money? It gets slightly trickier. There's a few solutions here. If you get lucky, you might find a well paying VPN affiliate program that'll pay you in bitcoin. But the more common solution is setting up a shell company and bank accounts in a friendly jurisdiction to accept payments from different advertising companies until you find one that won't kick you off.
I suppose it might even be possible (probably not legal however) to do that in an anonymous manner, you can register businesses online with bitcoin (or if you need a mastercard, you can get an anonymous one from https://advcash.com right now!) and opening bank accounts over the internet isn't exactly impossible either. Later you'd just wire all that money to a bitcoin exchange.
Define "successful". Many of them are only interested in floating as long as possible. User donations pay for the server and users have motivation to help keep the site running to keep the ability to pirate. Development and moderation is volunteer work.
For many of these sites just their continued existence is a success - however many years it ends up being.
My memory might serve me wrong, but I recall very interesting article years ago about playing monetary-gain games in Japan. Hazard where you win money is banned, so as a loophole what they did instead is that you were still playing but instead of wining cash, you were winning tiny white plastic balls (with serial numbers). Those balls were technically worthless and just made of cheap plastic. But once you won some, on the other side of the street where you played, was unrelated business owner, who all business was buying sad white plastic balls for hard cash from you.
No law is broken here. You can run a casino as long as people don't win money or anything worth a lot. You can run a store where you buy out white plastic balls at a premium value. And life goes on :)
I think Japanese government had hardship proving that one business is somewhat owned or operated by another and eventually AFAIK they gave up, giving birth to Japanese biggest mafia.
So.. long story short... run a torrent site where all you have is a database of copyrighted content (only names such as IMDB) and provide half of a magnet link :) Quietly find someone on the other side of the world who will host website that does the same thing just provide second part of magnet link. Don't know each other, don't talk with each other; wait until third person builds an app to glue both link and viola! Long live ExtraTorrent/PirateBay/Whatever about to die next.
https://github.com/sergiotapia/magnetissimo
There's something really sad when these sites die, a part of my teenage life is gone for good. I felt the same way when Demonoid shut down.
Can these sites skate the legality line by offering only magnet links? Why are their costs so high if they can just host magnet links and cache the site with Varnish or something similar?