> a) The easiest way to determine how much network traffic is illegal P2P is to ask a large network operator. You'll need to sign an NDA and I encourage you to do so.
It may not be as simple as this. >From page 7 of http://downloads.comcast.net/docs/Attachment_A_Current_Practices.pdf: "The Sandvine PTSes, as deployed on Comcast's network, do not inspect the content. These devices only examine the relevant header information in the packet that indicates what type of protocol is being used (i.e., P2P, VoIP, e-mail, etc.). The equipment used does not read the contents of the message in order to determine whether the P2P packet is text, music, or video ... or try to discern whether packets are personal or business, legal or illicit, etc." Additionally, the network operator may be able to detect that traffic appears to be encrypted P2P by the pattern of traffic, without being able to tell what particular content is being transferred. We have an idea of the amount of P2P traffic, eg see the Sandvine paper on traffic demographics from http://trac.tools.ietf.org/area/rai/trac/wiki/PeerToPeerInfrastructure. We also know about the quantity of DMCA notifications we send to customers for alleged copyright infringement. http://www.comcastvoices.com/2009/03/a-few-words-about-copyright.html -- Rich -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of DePriest, Greg (NBC Universal) Sent: Friday, April 17, 2009 9:21 AM To: Lars Eggert Cc: alto Subject: Re: [alto] Adopting two I-Ds as WG documents You stated: "I would like to point out though that none of the three studies say that illicit content is the root cause of network congestion today (which was your claim that I responded to). Or, to turn this around, that network congestion would disappear if the transmission of illicit content somehow suddenly stopped. "It's clear that some volume of traffic would be removed from the network if that happened, but whether that volume is sufficient to alleviate network congestion to some degree depends *a lot* on many other factors." a) The easiest way to determine how much network traffic is illegal P2P is to ask a large network operator. You'll need to sign an NDA and I encourage you to do so. b) You might also think about ipoque's numbers. The company notes that 66% of network traffic in Southern Africa is P2P, 65% in South America, 44% in the Middle East. What percentage of those numbers do you think is scientific data sharing, Linux downloads, or WoW updates? c) You might also go to any index site and check out the Top 100 downloads in any category and in the aggregate. Perhaps I missed it but I could not find one piece of content in either group that was not copyrighted. It's easiest to turn a blind eye and I hope you won't. -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Lars Eggert Sent: Friday, April 17, 2009 4:46 AM To: DePriest, Greg (NBC Universal) Cc: alto Subject: Re: [alto] Adopting two I-Ds as WG documents Hi, > a) ipoque GmbH's Internet Study 2008/2009 notes that P2P generates > the most traffic in all regions. > > b) Per an AP story about Sweden's new ipred law dated April 1, 2009: > http://news.yahoo.com/s/ap/20090403/ap_on_re_eu/eu_sweden_online_piracy_ 3 > > c) A recent study at Illinois State University in the US found 119k > "likely infringements" in a one-month period (April 2007). thanks for the pointers! They do illustrate that P2P traffic makes up a sizeable fraction of Internet traffic in some regions (a), and that at least some fraction of P2P traffic in some regions is transferring copyrighted content without permission (b + c). > To me, the conclusions are inescapable: P2P constitutes the > majority of network traffic and almost all P2P is "sharing" of > copyrighted content without the copyright holders' permission. I wouldn't feel comfortable making this broad generalization, based on those three studies alone. (My interpretation is above.) But without additional data, we could argue about what is accurate forever. I would like to point out though that none of the three studies say that illicit content is the root cause of network congestion today (which was your claim that I responded to). Or, to turn this around, that network congestion would disappear if the transmission of illicit content somehow suddenly stopped. It's clear that some volume of traffic would be removed from the network if that happened, but whether that volume is sufficient to alleviate network congestion to some degree depends *a lot* on many other factors. Lars _______________________________________________ alto mailing list [email protected] https://www.ietf.org/mailman/listinfo/alto _______________________________________________ alto mailing list [email protected] https://www.ietf.org/mailman/listinfo/alto
