I've been a slow convert to the church of XML. And I'm happy to admit that I'm still not a fanatic. I won't be saying just give it to me if it's XML
or But it's XML, you should be able to read it!
any time soon. But, I have finally started to find ways to put XML to work for me. I am now starting to look at how to put it to use for my next phase of plans, especially in regards to euc.cx and/or my online label Rive.
The current Rive catalog is all managed as HTML. This has been fine for individual release pages, whose design is important to me. The rest of the site, however, is missing out on features that are increasingly important to me. I want aggressive cross linking. I want to offer an RSS 1.0 feed for news and releases. I want to take part in this grand old "Semantic Web" idea. And I still want to do this without using a database management system or any server software beyond Apache.
Now that I've had some experience with XSLT from working with an old version of Ftrain Sitekit for euc.cx/ddec, I'm fairly confidant I could handle the new Rive catalog plan from scratch. The main question now is - what format to use for the Catalog? It seems like RDF might be decent source, as I am defining resources, but I come across few people who claim to like RDF/XML. I've found it acceptable (but not perfect) for RSS 1.0, but RDF/XML is a scary beast beyond that, and it still feels heavy to use as a source document. Later versions of Ftrain Sitekit define something Arbs that I like, for some perverse reason:
An Arb is an arbitrary block of content that can contain other Arbs."Arbitrary" is a term I usually wish to avoid - at least, when other people are telling me what to do (ie
We should make it so it can take in any arbitrary object and return an arbitrary result...) - but this might be an exception. For some reason, this feels better than defining my own tagset for the Rive catalog, and far better than dealing with anyone else's set which will be overly complex for my simple needs, or annoyingly simple and limited.