p2pnet - rss feed: URLs, URIs and LokiTorrent
p2pnet.net News Feature:- Done! LokiTorrent’s defense fund went over the required $30,000 on Monday, around 1pm EST.
It won’t be enough, though.
“Recent estimates by attorneys are looking at two or three times this amount in a full-on battle”, the website says.
And that’s not a one-time bill. It would be a monthly requirement.
That’s a lot of money.
As noted in this letter, dated December 14, 2004, the MPAA (Motion Picture Association of America), “have recently filed suit in the United States District court of the Northern District of Texas against the individuals responsible for operating the website lokitorrent.com (…) for copyright infringement.
“We demand that you immediately cease and desist from using any website or BitTorrent tracker to directly or indirectly cause, contribute to, enable, facilitate, induce, encourage, and/or participate in the infringement of MPAA member studio’s copyrighted works.
“The attached list is a reprensentative sample of copyrighted works owned by the MPAA members which you are infringing.
“We are authorized to act on behalf of the owners of the exclusive rights that are being infringed.”
Infringed, infringement, infringing… the word is written six times.
There must be an infringement, wouldn’t you say? But Lokitorrent’s operator doesn’t think so.
“There are many reasons why I am [standing up against the MPAA], but the primary reason is that I run a completely legal website that the MPAA or anyone else has no right to force me to close”, Lowkee told Zeropaid “We run a completely legal site for use by those who wish to share their own creative works, and as much as I realize the MPAA is a huge organization with deep pockets, I have the law, common sense and millions of people behind me.”
There’s no question about the millions of people, but how true will that be if it comes down to to legal arguments?
Identifier vs locator
LokiTorrent currently features 38,109 torrents - 38,109 reasons why the MPAA is suing it for “copyright infringement”.
When p2p hardly existed, the US 2nd Circuit Court of Appeals confirmed a test set up by Judge Kaplan to decide whether or not linking to the DMCA infringing DeCSS source code was itself infringing the law.
He wanted clear and convincing evidence “that those responsible for the link (a) know at the relevant time that the offending material is on the linked-to site, (b) know that it is circumvention technology that may not lawfully be offered, and (c) create or maintain the link for the purpose of disseminating that technology.”
The same tests apply to copyrighted content. Webmasters are aware that knowingly linking to copyrighted material makes them liable, even when the content is hosted on a server which isn’t theirs.
How different can it be with torrents websites?
Most people would believe there is no difference, and that LokiTorrent has no chance to win its battle. But others aren’t quite so sure.
“BitTorrent is an interesting situation”, says Alex Cameron, an Internet Law professor at the University of Ottawa. “I don’t think the legal answers are clear at all”.
Traditionally, linking has been understood as using a URL. The acronym, Uniform Resource Locator, indicates how such a link is used to *locate* content on the Net that will be accessed and downloaded as soon as the URL is clicked.
A typical URL is :
http://p2pnet.net/images/p2plogotemp.gif
Here, the link locates the logo of p2pnet, telling the browser it’s located on the p2pnet.net server, inside a directory called “images”.
Put aside torrents and take a look at what is, for instance, an eDonkey link (no worry, this is free content ; )
ed2k://|file|novishan%20sweet%20november%20[par%20Ratiatum.com].mp3| [cut]
8173815|738EC141A6FCA8F965E367AFD06C5D47| [cut]
h=45KWDXO3TNDO6NRLLEXPFYVDWWV5CQSK|/
The link doesn’t tell you where the content is, but *what* it is.
It comprises a prefix (eD2k://|file|) which enables the machine to recognize it is an eDonkey link, then it’s followed by the name of the file and by its unique digital signature, the “hash code”. Because it doesn’t actually locate the content, but merely idenfies it, such link is called a URI: Uniform Resource Identifier.
Although eDonkey, as with Gnutella and FastTrack, provide the URI, they’ve no way of knowing if the content identified is actually distributed. Links can be created out of any distribution circuit; they’re just the result of an algorithm.
Let’s say I create a URL on my website, linking to: http://www.p2pnet.net/mp3/best-of-madonna.rar
Since the content has never been located on p2pnet’s server and never will be (too bad ; ) I fail the first condition of the test: “know at the relevant time that the offending material is on the linked-to site”. The same applies to p2p links: I can create a link, but if the content isn’t actually distributed, I’m not infringing.
Because a URI is an identifier and not a locator, one can never be found to have, “known at the relevant time that the offending material is on the linked-to” a network. For this reason, I tend to believe websites that merely list eDonkey or Magnet links, even if they they identify copyrighted works, aren’t infringing per se.
When it comes to BitTorrent, however, the 38,109 torrents listed on LokiTorrents are specific kinds of URI that also lead to trackers which are resource locators. Trackers ‘track’ what people transfer and create links among them so they can acurately share the content.
LokiTorrent both hosts the torrents and operates the trackers, pretending it didn’t know the offending material was available on BitTorrent, thus declaring it’s 100% legal, in the process making it a hard game to call.
Before asking for three times $30.000 per month, it seems to me Lokitorrent should explain clearly why it has good reason to believe it can win the case.
I’d be more than happy to join the effort, but so far my analysis of the legal perspectives don’t encourage me to do so.
Guillaume Champeau - Raitiatum
January 5th, 2005 at 7:44 pm
Actually an ed2k link (technically) can’t even identify a file. There is a rare (I mean *very* rare) chance, that there is an other file avaible with the same checksum and filesize.
It’s a common belief that an ed2k (or magnet) link starts a download. Technically it only starts a search, so it can’t be an URI.
January 5th, 2005 at 10:30 pm
Hi Tothbenedek,
As far as I know, it seems to me eDonkey and eMule use an MD5 algorithm and chances of a collision are virtually zero (there had been rumors of the algorithm being broken but I didn’t hear of any confirmation).
> “it only starts a search so it can’t be an URI”
I’m not sure I understand what you mean. To me, it *is* because it is a URI that the search can be started (thanks to the hash code). Am I wrong?
Cheers,
Guillaume Champeau
January 6th, 2005 at 1:25 am
LokiTorrent is both a tracker and an index site. As an index site, why should it be responsible for pointing to possible copyrighted material hosted on another server? Search engines such as google also index and directly point to sites that have copyrighted material but no one ever tries to sue Google.
LokiTorrent has become a very popular site and it would be a full-time job for someone to effectively monitor every torrent submitted to them. They have a link on their site that allows rights holders to notify them if they find their software listed on the site. LokiTorrent claims they will remove it once they are notified.
Although I’m completely behind torrent and verified links sites in general, I realize that it will be difficult for LokiTorrent to defeat the MPAA in court, especially considering the fact that the MPAA has countless politicians in their pocket.
However, I don’t think it’s an open and shut case.
Drake
January 6th, 2005 at 2:49 am
I have a solution so that a site can’t be sued for hosting torrents, since it doesn’t host a torrent in entirety, but stores it with a “twin torrent site”.
Step 0. The torrent clients should be able to retreive “twin torrent” site pairs from SITE0 like this:
SITE1 TWIN SITE2
SITE3 TWIN SITE4
Step 1. you select a twin torrent pair from the list in your client (e.g. SITE1 TWIN SITE2)
Step 2. your client uploads the torrent to SITE1
Step 3. SITE1 generates a hash from the torrent
Step 4. SITE1 stores only the first half of the torrent and displays the download URL for this first half like this
http://www.site1.com?id=hash
Step5. your client uploads the torrent to SITE2
Step6. SITE2 generates a hash from the torrent based on the same algorithm as SITE1
Step7. SITE2 stores only the second half of the torrent and displays the download URL for this second half like this
http://www.site2.com?id=hash
Step 8. when a user clicks a SITE1 link the client realizes that SITE1 is a twin of SITE2 so it downloads the one half first from SITE1, then it merges the other half from SITE2, so the full torrent is assembled.
The conclusion: neither SITE1 nor SITE2 can be sued because they don’t have the full information to commit the “infringment”. (Neither one has the full torrent).
Furthermore SITE1 knows nothing about SITE2 and vice versa.
It’s SITE0 that knows about SITE1 being the twin of SITE2, but SITE0 can’t be sued since it doesn’t have a torrent on it at all.
There can be variations of how sites store the torrent (e.g. SITE1 stores the odd index bytes, and SITE2 stores the even index bytes etc.)
Cheers,
Nani
January 6th, 2005 at 4:46 am
Fighting a legal fight against a bunch of lawyers is a mug’s game. Only the lawyers win. Fight the battle on your own ground. This is one area where we should take a lesson from spammers (yes, spammers - hold your nose and listen for a minute). Don’t fight an expensive legal battle to keep a site running. Shut it down and open a different site, preferably hosted by a different ISP in a different country. Keep changing the p2p network, the algorithms, the search and distribution methods. It takes time to identify each new variation, time to launch a legal offensive, time to get results - and then it just starts over again because the laywers are chasing a moving target. And lawyers cost *a lot* of money. Eventually even idiots like the RIAA and the MPAA will get the message that they’re making a lot of lawyers rich at their expense and accomplishing nothing. Then it will be time to have a productive discussion about copyright, business models, fair compensation and fair use. And hopefully the discussion won’t be conducted by lawyers with a vested interest in promoting disputes that never get settled.
January 6th, 2005 at 10:46 am
actually they use a modified MD4 algorithm (hashing in 9 Mb blocks) but chance for two file to have the same checksum is still very low (it’s ~ 1:2^128), but it is enough for you to say that it can’t “identify” a file.
It can’t be an URI, becouse of this (^^^) and to the definicion or “URI”. Since it’s not an URI it’s just some technical stuff, which looks like an URI.
Why this is important?
Becouse in most country hosting copyrighted material without permission is a crime. But a direct link to a file, hosted elsewhere, can also be a crime.
This URI and search thing can protect any ed2k publisher at any court.
January 7th, 2005 at 10:41 am
A good analogy to show that posting identifiers of copyrighted works on a website is perfectly legal (with or without knowledge of it) are barcodes (or: ASIN/ISBN codes).
Lets say you buy a DVD movie. It has a title, length and barcode on it.
Ask a judge whether posting these three pieces of information on a website (or catalogue) is legal. There is no doubt in my mind that it is because the only thing you do is to identify a (copyrighted) work. If this were illegal then companies like amazon would be violating copyright law.
An ed2k-link contains the same kind of information: title (filename), length (filesize), barcode (hash). These three kinds of information have exactly the same function: identification. If you were to put a couple of these ed2k-links and their corresponding title/length/barcodes on one piece of paper it would become perfectly clear there isn’t any real difference.
When however somebody says to you: “you can get a pirated DVD at that address” he is (most likely) breaking the law. So would you (most likely) be if you told somebody where to get a copyrighted file. But this a matter of localization not identification. Ed2k-links have nothing to do with localization.
While “normal” links point to (or resolve into) a single address (analogy: it gives you the streetname and number of a pirate) ed2k-links only give a description of a (copyrighted) work (analogy: the movie contains 321 scenes, 34 shots are fired and the total age in days of all the actors is 532489: id=32134532489).
Hope this helps.
Just my 2 cents…
PS. What would happen if Amazon would use hash-codes for their “Catalogue Numbers” of DVDs? In other words: is a hash copyrighted itself?
January 20th, 2005 at 5:59 pm
(a point of view on Guillaume Champeau’s article “URLs, URIs and LokiTorrent”)
Hi,
First of all I’d like to congratulate you on your story. I found it very interesting and persistent. I’m also assuming that you are French (bon j’avoues c’est ton nom qui m’as mis la puce a l’oreille mais bon ca fais plaisir d’avoir le point de vue d’un fr
).
I may be not be looking into the technical side of “URI” and trackers as you’ve already mentioned earlier, but i do believe laws on p2p are not making it very clear to users. While you clearly condemn Lokitorrent’s effort to fight back, i think it’s important to take a look at the “bigger picture” of Infringement in today’s Internet society
When you say “infringement” what do you mean in general? Stealing or Sharing illegally someone’s copyrighted work right? While the big guns are being pointed to movie filesharing websites (lately torrent sites like lokitorrent and supernova), what’s happening with other copyrighted material?
Take a look at the number of webmasters who’ve stolen stuff (images, files, material in general) from other websites. Are they infringing? Yes. Do people really care? I guess the people involved do, but what can they really do about it when millions of webmasters out there are doing the same thing. It’s attained such gigantic proportions it’s literally impossible to monitor or control these infringements.
Now let’s take a look at the music industry. The law is changing worldwide as we speak to impose “fines” on people who have downloaded free mp3’s. So while the music industry is loosing loads of money on sales, who do you think is making all the big bucks? Do you think the people who have downloaded alot of mp3’s illegally are the beneficiary’s of all this. Well I’m not saying they aren’t but i still think you should take a look at Apple’s net sales in IPod last trimester before you answer.
I’m just here to argue that companies like Apple who work around the problem of “infringement”, “copyright”, etc… are still making business while other organizations like the MPAA are getting involved in endless lawsuits on Internet filesharing.
One day the law will be clear and straightforward enough so that everyone abides to it. Right now we’re living in an era of online copyright chaos where closing down websites like “Lokitorrent” & “Supernova” only brings the MPAA one step closer to the real problem of p2p in the world today.