more changes

[Ray Plante]

Hey Kevin,

On Tue, 27 Jun 2006, KevinBenson wrote:
> I see what you mean, but it does seem strange considering this was the 
> whole reason of RofR to be created.   How about we simply move the 2 
> sentences "Note that this document does not specify how a registry 
> obtains....."  Into the Note section? after all the sentence starts with 
> the word "Note" :)

Well, I can live this this; however, despite the wording, I intended this 
sentence to be considered part of the specification--that is, it says 
explicitly that this standard does not specify how to solve this problem, 
and thus, readers should not infer one.  The RI spec should not be forced 
to solve the whole problem of registry interoperability, especially the 
bits that are less exercised or implemented.  If we decide RofR isn't 
working, we can change it without changing the RI spec.  

> > So where are we on the optional/required harvesting interfaces?  

> Yeah I am in the same area, I am less convinced as well.  I tend to like 
> the REST interface.  

Okay--I think we are resolved, then; the SOAP version is optional.  

> Question/Thoughts: In OAI the child of "metadata" is an "any" element 
> with "any" namespace curious if that will cause any client problems?

Apparently not, since we've been using it fine.  That is, apart from this 
issue:

> Question:  I remember the child of "metadata" being different between 
> some registries and this caused a few problems.  

This is one of the things that I want the RofR to check.  Before it adds 
the registry record to its store, it will actually call the OAI interface 
to make sure it complies with the letter of the RI.  

Okay, great--I think we're in really good shape!

cheers,
Ray