Re: [syndication] Re: robots.txt and rss

["Phil Ringnalda" <phil@philringnalda.com>]

Bill Kearney wrote:
>> Feed ID 1866 ( http://linuxtoday.com/backend/my-netscape.rdf ) -
identified
>> as RSS 1.0 on the feed info page, but with a default namespace of
>> http://my.netscape.com/rdf/simple/0.9/ that's 0.90 (I found it by
googling
>> for the 0.90 namespace). You aren't trying to identify 0.90 by
version="",
>> are you? I see that's listed as a tag with zero feeds using it, but AFAIK
>> 0.90 didn't have a version attribute, just the namespace.
>
> The feed wraps itself in rdf:RDF tags.   That's not an RSS-0.9x format
feed.
> The feed, however, doesn't contain anything 1.0-like.  So thus it's indeed
> more appropriately listed as an 0.9 feed.

I wouldn't say that 0.9 versus 1.0 was a matter of containing 1.0-like
things, or open to interpretation at all: if a feed contains a
rdf:RDF/http://my.netscape.com/rdf/simple/0.9/channel then it's 0.9, if it
contains a rdf:RDF/http://purl.org/rss/1.0/channel then it's 1.0. If it
contains an rss element, then it's the version in the version attribute
(which has been required since 0.91). If it doesn't have any of those, then
whatever it is, it's not RSS.

> If this is a pissing match about version numbers then the 1.0 count may
well
> be off.

Why does everything have to be personal and nasty with you? If Syndic8 tells
me that there are no RSS 0.90 feeds around, then I'll design my parser to
start looking for http://purl.org/rss/1.0/channel as soon as I see an
rdf:RDF tag, because that's the only valid channel tag I might find in an
RSS file that has an rdf:RDF tag. Whether you report 0.90 files as 0.90 or
1.0 shouldn't be a political question, it should be a question of whether
you want to report useful, helpful statistics or deceptive statistics that
are likely to cause bugs down the line.

Phil Ringnalda