On Wed, 3 Jun 2009, Enrico Marocco wrote:
The WG has been chartered to define a protocol to provide applications with information to perform better than random peer selection; honestly, I don't see how information about its own link capacity -- assuming it is feasible to determine it in a meaningful way without recurring to realtime measurements -- could help a peer in finding good peers to connect to. So I'd say it is out of scope, but I'd be glad to be proven wrong.
Examples of applications that can make use of provisioned link capacity: -voice relaying -p2p video and audio conferencing
OTOH, but as I understand it this is not what you're talking about, information about other peers' link capacity might fall within the scope of ALTO, as it could be useful in peer selection optimization. However, as Rich mentioned, there are many reasons why an ALTO server can't or may not want to pass it.
It is a protocol design issue whether/how provisioned link capacity should be advertised to other peers. I can design a p2p audio conferencing protocol where a peer, based on its provisioned link capacity or its fraction, only advertises the number of audio streams that it can mix.
I do not think that such an advertisement is any worse in the privacy sense than an advertisement which uses measured link capacity to determine the number of audio streams.
-s _______________________________________________ alto mailing list alto@ietf.org https://www.ietf.org/mailman/listinfo/alto
