Granted, we still need to find the successful patterns of when to use RDF-star vs. when to use quads. But I think there is room for both.
(in the end, all you need are zeros and ones, and still we invent more complex languages 😉)
+1 to what @dret said. In some context, 'Tungsten' would be a perfectly valid identifier for this (and conversely, in some context 'Element 74' might refer to something completely different...). IRIs, of course, make most of the context explicit in their structure.
The various RDF test suites are managed by the RDF-Tests community group w3.org/community/rdf-tests/ . @drobilla what exactly made your previous attempt unsuccessful?
Same as @olafhartig. No ready-to-use tutorial, and no time at the moment to actively work on one. I'm also happy to share my slides, or to review a proposal.