___/ On Sat 03 Sep 2005 02:09:42 BST, [Firas D.] wrote : \___
Mathias Bynens wrote:
What might be a good solution, is the use of a new IRI design for
feeds and trackbacks. I'm currently using this on my site:
http://domain.ext/archive/yyyy/mm/post-slug
-> feed: http://domain.ext/syndicate/yyyy/mm/post-slug
-> trackback: http://domain.ext/trackback/yyyy/mm/post-slug
I don't need to tell you my robots.txt contains the following:
User-agent: *
Disallow: /syndicate/
Disallow: /trackback/
How cool is that?
That's an excellent idea. A shame that many blogs out there still have their
PageRank diluted due to (?trackback and) feed URL's...
From a web architecture standpoint, that's a step backwards, if WP
were to adopt it... ideally there should only be one uri per item and
what's returned (atom, rss, xhtml, pingback uri, etc) would be
determined by the request HTTP headers.
<snip />
That's also a valid point. It need to be done carefully though. I once
appended
?from_rss to all URL's contained within feeds in order to perform some
temporary tracking. Little did I know that my feeds were getting
indexed and so
did the URL?from_rss variants, which meant that I had duplicate (up to
3) pages
indexed for each entry (%#@!). After some discussion, I immediately
reverted to
the older version and flushed the duplicates.
I suspect that if HTTP headers were used, the case would be different,
but I am
not too familiar with that functionality... not enough to comment on it
anyway.
Roy
--
Roy S. Schestowitz | "Have you compiled your kernel today?"
http://Schestowitz.com | SuSE Linux | PGP-Key: 74572E8E
2:45am up 9 days 15:53, 3 users, load average: 0.31, 0.47, 0.45
|
|