So hey, I'm nobody wanted to see this thread revived, but I'm hoping
you info uri folks can clear something up for me.
So I'm trying to gather together a vocabulary of identifiers to
unambiguously describe the format of the data you would be getting in
a Jangle feed or an UnAPI response (or any other variation on this
theme). "I have a MODS document and I want *you* to have it too!".
Jakob Voss made the (reasonable) suggestion that rather than create
yet another identifier or registry to describe these formats, instead
it would make sense to use the work that the SRU:
communities have already done. Which makes a lot of sense. It would
be nice to use the same identifier in Jangle, SRU and OpenURL to say
that this is a MARCXML or ONIX record.
Except that OpenURL and SRU /already use different info URIs to
describe the same things/.
What is the rationale for this? How do we keep up? Are they
reusable? Which one should be used? Doesn't this pretty horribly
undermine the purpose of using info URIs in the first place?
Is anybody else interested in working on a way to unambiguously say
"here is a Dublin Core resource as XML, but it is not OAI DC" or "this
is text/x-vcard, it conforms to vCard 3.0" in a way that we can reuse
among all of our various ways of sharing data?