IFPI Turning To Lawsuits

Sherman's doppleganger writes "The IFPI (the "European RIAA") has made a lot of noise about filtering this year, but it looks as though 2008 is instead becoming the year of the lawsuit. The IFPI has now sued an Irish ISP in an attempt to keep copyrighted content off of its network. 'The lawsuit accuses Eircom of abetting illegal downloading by allowing copyrighted material to traverse its network unimpeded. The IFPI... wants the ISP to start filtering traffic to scrub all illicitly uploaded and downloaded copyrighted material on its network.' The lawsuit comes less than a week after an Israeli court forced the nation's three biggest ISPs to block access to HttpShare.com."

Comments

"Turning" to Lawsuits? Come Now

[eldavojohn]

IFPI Turning To Lawsuits To say the IFPI is turning to lawsuits is like saying Bob Dylan is turning to drugs. It's an organization of lawyers! What else do they do?!

I recall them dishing out 2100 lawsuits at once in 2005 and 8000 lawsuits at once in 2006! And evidence that it's been going on since 2004.

You might be able to convince me that the IFPI is getting smarter (or stupider, depending on your views) at stopping file sharing by targeting ISPs with lawsuits but to say they're only now with litigating to stop these losses is ignorant.

Re:Good

[Bryansix]

That's not what this is about. It's not about ISPs hosting copyrighted works that the person hosting doesn't own. It's about the ISP's customers downloading copyrighted works that they may or might not be authorized to.

Re:So all traffic should be banned

[Kjella]

Unfortunately the distinction between permitted and not permitted is meaningless, as is the distinction between copyrighted and public domain. The ISPs see bits and bytes, but these are not properties of bits and bytes. The exact same transfer that's illegal today will be legal in life+70 (barring more Mickey Mouse acts), bit by bit. That means the only possible way for ISPs to tell an illegal download from a legal download is to keep a database over all possible illegal downloads, which works for a plain unencrypted transfer. However, as anyone that's worked with SSL knows it negotiates a random session key so there's an arbitrarily large number of streams of bits and bytes that transfer the same data. Once we arrive at this stage the ISP is basicly checkmated, there's nothing it can do.

What they are trying to do is to use the non-authenticated, plaintext nature of the negotiation phase as it is today to determine whether it's illegal or not. Creating an HTTPS version of torrents/trackers that doesn't leak anything to the ISP would be fairly trivial, so would adding authentication if the ISP tried its own SSL connection. At that point, the ISP is quite frankly guessing. They know you connected to TPB, but not what you searched for, what torrent you're getting and if it happens to be a legal download (many torrent aggregators just pick up everything) and you talk SSL to all your peers. There's no possible theoretical or practical way they can tell the difference between you downloading Ubuntu 7.10 (700MB) or a illegal DVD rip (700MB) over a torrent, the traffic patterns would be exactly the same.

To take a practical example where this is already all encrypted, I can connect via NNTPS to my news server. How the hell is my ISP supposed to know what I'm doing? They haven't got the faintest possibility to know anything at all. Of course in this case there's a server at the other end they could go after instead, but in a P2P network it's simply impossible. P.S. For anyone trying to make the lame pun about "The first rule about Usenet..." it's near 30 years old, and everyone that cares to know already knows about it. The only possible way an ISP could prevent copyrighted works from going over their networks is to turn off the lights.