[rdfweb-dev] Is it OK to produce valid-but-strange-looking RDF?

[Danny Ayers]

>Yep, non RDF tools that deal with that wrongly are going to get an awful
>lot wrong, so aren't likely to be around too long, all the genuine RDF
>tools will deal with it fine, so there should be no problem.

I've just had the same problem with RSS - few (if any) of the readers seem
to go beyond a vanilla XML interpretation, e.g. so the second item here is
invisible:

<item rdf:about="http://blah/blah.html">
<title>One Page</title>
...
<rdfs:seeAlso>
<item rdf:about="http://blah/blah2.html">
<title>Another Page</title>
...
</item>
</rdfs:seeAlso>
</item>

It's a royal pain in the derriere, but I've worked around it by simply
pulling out the other properties when plain RSS is required.

I've also wasted a lot of time thanks to the treatment of namespaces - I
wanted to produce stuff the Amphetadesk reader could display, but it seems
it only understands the elements when RSS is the default namespace (no
rss:item possible). Not yet managed to set the default namespace (with
Jena), so am now looking at a more general purpose XSLT alternative to
communicate with RDF-shallow apps.

Long term - I think the the weight of benefits from using the RDF model will
outweigh the perceived extra work, but short term...grrrr...

Cheers,
Danny.