> As I previously suggested it, I prefer the broader history!
> But I think I may prefer calling it 'provenance', though not
> particularly strongly.

OK... though I think that implies a narrowing of the semantics too? i.e.
we suggested "history" might be broad enough to encompass "change
history" etc, but "provenance" is typically closer to custodial history,
with Sarah's qualification about scope.

> Using dspace:provenance is attractive because it saves us
> from proposing a new term. However I'm not sure what the
> status is of terms in the 'dspace' namespace. I doubt if they
> are standards in any formal way - so wouldn't gain us much
> over using rslpcd namespace terms. (Assuming the object of
> this exercise is to get a more formal standards status for
> the collection description terms.)

Agreed. Let's forget about using dspace:provenance then.

> If we decided to go with custodialHistory, then I think it
> should be in a more restricted namespace 'cld'. It seems to
> me to be too domain-specific for a 'gen' namespace. At least
> 'custodial history' isn't expression I use on most days of the week :)

OK. (I have too many friends who are cataloguing archivists).

> I think, whatever it is called, it should be an element
> refinement - presumably of description. I doubt if a new
> element would be acepted by the UB.

The perfectionist pedant in me shrieks: "But that would sacrifice
semantic specificity in the interests of political expediency!" The
reluctant realist in me quietly agrees with you.

> (Remembering the infamous
> abortive 'over my dead body' attempt at a 'version' element!)

Oh!

Pete