You note that "A localization service doesn't have to discriminate
[between legit and illegit P2P]..."

I don't understand why it wouldn't.  

What's the point of facilitating the illegal distribution of copyrighted
content?  

And how would one justify that?

-----Original Message-----
From: Nicholas Weaver [mailto:[EMAIL PROTECTED] 
Sent: Wednesday, December 03, 2008 3:02 PM
To: DePriest, Greg (NBC Universal)
Cc: Nicholas Weaver; Y. R. Yang; Le Blond, Stevens ; Arnaud Legout;
[email protected]; Craig Seidel
Subject: Re: [alto] Paper on "Pushing BitTorrent Locality to the Limit"


On Dec 3, 2008, at 11:33 AM, DePriest, Greg (NBC Universal) wrote:

> I'm struck by this sentence:  "And for legitimate content, you don't
> have all the legal headaches you do on BitTorrent."
>
> I had thought this entire conversation was centered on distribution of
> legitimate content.

A localization service doesn't have to discriminate, and there some  
who want suitable privacy protections which would allow localization  
to work on BitTorrent regardless of the file being distributed.

> Am I wrong?  If so, why?

And its not just that.

Its that it actually is straightforward to make a BitTorrent cache  
today for an ISP:

You need to just do the following:

Have an IDS get a mirrored copy of the traffic, and look for tracker  
requests.

When an internal node contacts a tracker, have your cache contact the  
tracker to find local peers, and contact JUST those peers, exchange  
blocks with JUST those internal peers (so the cache is ONLY containing  
content that the internal nodes have fetched), and play some lying  
games (eg, tell one peer you have blocks belonging to the other peer,  
and vice-versa, esp using a large block of non-externally-routed  
addresses so you can seem like a whole BUNCH of peers to the internal  
BitTorrent clients).

This would work, would be straightforward to build (probably 2-3 man- 
weeks for a ugly prototype, a man-year or few for a good deployable  
product), would be out of path (so failures are non-damaging), but  
would be an invitation to a pile-of-DMCA-notices which, unless you are  
unable to automate, would cost so much more in person-time to handle  
than the cache would save in traffic.

_______________________________________________
alto mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/alto

Reply via email to