Re: PRISM and RSS

[Aaron Swartz <aswartz@swartzfam.com>]

dave.cantrell@gunter.af.mil <dave.cantrell@gunter.af.mil> wrote:

> My question is, am I "breaking" RDF if I create my own arbitrary elements?
> [1]

Of course not! Creating arbitrary elements (in a proper namespace) is the
whole point of RDF. However you are breaking it if you put arbitrary stuff
in them and don't let RDF parsers know that you're doing that. I comment
more on your example, below.

>> Thanks for the faith. I hope that RDF comes in handy for you someday soon!
> 
> Actuall, that's why I have faith, because I believe it will. I do believe it
> can and will be useful, and I see the promise, I'm just stuck in the middle

Cool.

>   <ilnews:category name="Top Stories">

You can't really stick an attribute right there. And even if you did, you'd
have to give it a namespace too (funky thing in the namespaces spec).

>     <rdf:description
> about="http://www.moreover.com/cgi-local/page?o=portal&amp;c=Top%20stories";>
>       <dc:title>Top Stories</dc:title>
>       <dc:identifier>topstories</dc:identifier>
>       <dc:creator>Moreover</dc:creator>
>       <rdf:description
> about="http://p.moreover.com/cgi-local/page?c=Top%20stories&amp;o=rss";>
>         <ocs:format>http://backend.userland.com/rss091</ocs:format>
>       </rdf:description>
>     </rdf:description>
> 
>   </ilnews:category>
> </rdf:description>

This version of OCS isn't technically valid RDF but forgetting that for a
moment, this would be just fine. If you were to stick in something that
wasn't valid RDF, you'd need to add this attribute:

rdf:parseType="Literal"

that says to stop parsing the element it's on as RDF and treat it as plain
old XML.
-- 
[ Aaron Swartz | me@aaronsw.com | http://www.aaronsw.com ]