Comcast's ISP Experiences in a Proactive Network Provider Participation for P2P (P4P) Technical Trial
draft-livingood-woundy-p4p-experiences-10

[Ballot comment]
This is great document and excellent input for the WG efforts.

I do have one question though, and that relates to the complex
selection process and "economics" of p2p peer selection. Basically,
there is a large number of variables that one has to take into account
when selecting a peer or evaluating the effectiveness of a peer
selection algorithm. In particular, download speed or localization is
not the only metric to follow. For instance, the ability to spread
the content in the Internet, grow the swarm, and retain the most rare
segments of data are also important. I would be curious if the experiment
provided any information with regards to possible negative effects of
focusing on localization as the key startup connection criteria? Or
lack thereof.

(FWIW, I do not agree with the requirement raised in IESG discussion about
more extensive security considerations section.)

[Ballot comment]
I support Tim's DISCUSS. I think that RFC 2223 requires all RFCs to have a Security Considerations sections. If security content is null for this document it should explain why.

[Ballot discuss]
The security considerations section of this document reads as follows:

"7.  Security Considerations

  There are no security considerations in this document.

  NOTE TO RFC EDITOR: PLEASE REMOVE THIS NULL SECTION PRIOR TO
  PUBLICATION."

As noted in Scott Kelly's security directorate review, this is inconsistent with current
guidelines and is probably inappropriate for the content.  I would like the authors to
review Scott's comments and the guidelines in 3552, and either develop a security considerations section covering their proposed approach or offer a convincing argument
why such a section is not needed.