Ok, even nicer now ... after allowing to introduce custom namespace prefixes also with '?' and '=' as separators (Don't know if that is valid for QNames though? Anyone knows?) than before (blog post 1, and post 2):
Code (see code repo) is still quite ugly though. Just typed together to get things working. Will look at structuring things more as soon as I have slept a little better :)
I also got to think of these random issues, which is foremost questions to myself, that might not be fully thought-through yet, but jotting them down here for possible feedback (hoping it's readable ^^):
I now managed, (using ARC2 and SMWWriter, in a MediaWiki extension) to populate Semantic MediaWiki pages with triples, from a snippet of RDF/XML (Thanks to Egon Willighagen for the RDF-ized NMRShiftDB data, submitted further below), yay! But ... as you can see, using the full URI:s as wiki names is not a good idea. URI:s as wiki titles is ugly ... and the predicates in this case even got truncated and all treated as the same one, since they had hashes in it, which MediaWiki obviously doesn't allow in titles), so that's the whole background for our talking so much about the "equivalent URI handler" (mentioned here for example), which is thought to be a configurable handler of mappings from URI patterns to (sensible) wiki titles. Also, optimally the same pattern can then be used both on import and export so that the format is kept, allowing to use SMW as a (collaborative) RDF editor! (which is one of the main motivations for my GSoC project).
Well, the hard bits (URI -> Wiki title mapping) remains. Diving in to it now ...
I finally found a summary that contrasts RDF, RDF Schema, OWL, and rule languages (SWRL in this case) in a few sentences: