Atom Threading Extensions
draft-snell-atompub-feed-thread-12

[Ballot comment]
The 3rd paragraph of section 7 says:
  >
  > Such attacks can be prevented or mitigated by allowing users to
  > explicitly configure the sources from which responses may be
  > retrieved, or by applying heuristics to determine the legitimacy of
  > a given response source.
  >
  s/users/consumers/

  The last paragraph of section 7 says:
  >
  > Digital Signatures as specified in Section 5.1 of [RFC4287] present
  > one possible avenue for mitigating such concerns, although the
  > presence of a valid XML Digital Signature within an entry is not,
  > by itself, a reliable defense against repudiation issues.
  >
  I agree.  Can you provide a pointer to a description of the other
  aspects of a total solution?

[Ballot comment]
I noticed that the namespace http://purl.org/syndication/thread/1.0 is
not an SDO.  I want to confirm that we hae sufficient control over
this specification that we have change control for the future.

IANA Last Call Comments:

Upon approval of this document, the IANA will make the following addition to the
Link Relations [RFC 4287] Registry located at:

http://www.iana.org/assignments/link-relations.html

The value:

replies

will be added to this Registry.

We understand the above to be the only IANA Action for this document.

[Ballot comment]
Explanation of a puzzling choice by the author, following
Gen-ART review comment by Joel Halpern:

Several of the attributes have a type of "atomURI".  However, reading the text carefully, it is clear that they are actually IRIs.  I suspect that those have the same syntax.  But since they have different semantics, it seems odd for the syntax definition to refer to atomURI rather than atomIRI.

At 12:15 AM 6/16/2006, James M Snell wrote:

> The atomURI label is adopted from RFC4287, which also uses the labels to
> accept IRI's.  Yes, it is confusing, but given that this is an extension
> to RFC4287, I felt it was better to reuse the concepts than to redefine
> them.