[OAI-implementers] Reconsidering mandatory DC in OAI-PMH

[Jeremy Keeler]

Are there arguments for or against defining repository types each with 
their own rules for minimal compliance? Metadata requirements would be out 
of the core oai-pmh, but would be defined for communities. So through OAI I 
can say I'm a "stm repository" and harvesters would know if I support a 
certain level of functionality. This would take more thought up front as 
communities form, but only the communities who had the need for defined 
base operability would have to carry that burden. And these communities 
could still be contributing to the whole and not off in their own worlds.

Jeremy

At 07:50 AM 08/05/03, you wrote:
>Zitat von Matthew Cockerill <matt@biomedcentral.com>:
> > I would argue that enforcing a baseline set of cross-domain metadata
> > semantics is *as* important an aspect of the OAI-PMH project as is the
> > protocol itself.
> > (Although certainly, the two could be thought of as logically distinct
> > sub-projects]
>
>I cannot talk for the OAI project, all I say is that from my perpective, the
>two *are* logically distinct projects from a technical point of view: There's
>an OIA-PMH and OAI-DC project. OAI-PMH could recommend using DC, but I would
>not make it a requirement of OAI-PMH. This would be like requiring HTML for
>HTTP.
>
>After all, I am not a member of the community of digital libraries, instead I
>try to use OAI-PMH in a rather unorthodoxical way as I understand now :) My
>comments intend to foster a more generic and widespread use of OAI-PMH, 
>that's
>why I'd like to see it decoupled from the DC requirement.
>
>Sandro
>_______________________________________________
>OAI-implementers mailing list
>List information, archives, preferences and to unsubscribe:
>http://oaisrv.nsdl.cornell.edu/mailman/listinfo/oai-implementers