Dare Obasanjo wrote:
>Tim Bray wrote:
>> 3. You get help for aggregate feeds using atom:origin
> Sounds like the <source> element in RSS 2.0
No. The two are in no way similar. atom:origin has no relationship
to <source/>, at least as <source/> is defined in the RSS V2.0
specification. Atom:origin can only be used to "identify" a source while an
RSS V2.0 <source/> can be used to locate a source. These are very different.
atom:origin is a character-by-character copy of an atom:id which is
a URI that can't be reliably dereferenced. An atom:origin might look like:
<origin>tag:blogger.com,1999:blog-555</origin>
RSS's <source/> element[1] provides a copy of the title of the
source feed and has a required attribute which is the URL of the source.
(i.e. a URI which can be dereferenced.) An RSS <source/> might look like:
<source url="http://static.userland.com/tomalak/links2.xml";>Tomalak's
Realm</source>
As I've explained before, atom:origin is largely useless in
aggregated feeds.[2] On the other hand, RSS's <source/> element is useful
since it communicates the most important information that an aggregator
needs to know about an entry's source feed when it displays that entry. The
PaceHeadInEntry defines a mechanism which should be combined with an
atom:source, similar to that defined by RSS V2.0, in order to properly
support the embedding of entries in aggregated feeds.
bob wyman
[1] http://blogs.law.harvard.edu/tech/rss#ltsourcegtSubelementOfLtitemgt
[2] http://www.imc.org/atom-syntax/mail-archive/msg11090.html
