My recollection of discussions during 5005 was that we had to come up
with "current" precisely because "self" was ambiguous. People may
have a reason for advertising the archive document's URI in "self",
and I don't see what the utility of repeating it is.
YMMV.
Cheers,
On 14/11/2007, at 5:38 PM, Franklin Tse wrote:
Since RFC 5005 does not clearly define if atom:feed/atom:id should
change and which feed should atom:feed/atom:[EMAIL PROTECTED]"self"] point
to, feed publishers may not know what they should do, and
implementations may vary. I think the situation can be improved if
the Feed Validator adds some recommendations.
Based on the discussions before, the following rules should be added:
1. If first, last, previous, next, prev-archive, next-archive or
current is found, issue a notice reminding the publisher that all
of the feed documents of a paged feed or an archived feed should
share the same feed ID
2. If the href of a "current" link is not equal to the href of the
"self" link, issue a warning or a notice
However, Sam's archived feed is not following the recommendations
above:
http://intertwingly.net/blog/archives/index.atom
http://intertwingly.net/blog/archives/2007/10/index.atom
-Franklin
-------------------------------------------------
From: "Eric Scheid" <[EMAIL PROTECTED]>
Date: Wednesday, 14 November, 2007 04:56
To: "Atom Syntax" <[email protected]>
Subject: Re: Paged Feeds Question
On 14/11/07 12:05 AM, "Franklin Tse" <[EMAIL PROTECTED]>
wrote:
o atom:feed elements SHOULD contain one atom:link element with
a rel
attribute value of "self". This is the preferred URI for
retrieving Atom Feed Documents representing this Atom feed.
So, does this mean that the atom:link with rel="self" should have
a href to
the first page of a paged feed or the subscription document of an
archived
feed?
yes.
e.
--
Mark Nottingham http://www.mnot.net/
