RE: [saml-dev] RE: [security-services] Proposed erratum resolutions

["Scott Cantor" <cantor.2@osu.edu>]

> I can understand wanting to keep format specific processing rules out of
> the specification but, at this point, I'm not sure it's appropriate.  I
> completely agree with Scott that this needs some further discussion and
> treatment in errata.  I've focused primarily on the AllowCreate
> attribute but I think the issue is bigger and involves just about
> anything that can potentially have different interpretation that are
> dependent on name id format.  

This is really where we differ...I don't think there *are* any other major
cases. AllowCreate is messy and use case (I wouldn't say so much format)
specific. I don't see any other major ambiguities.

NIM, for example, is not at all hard for me to see as format-agnostic. The
only place where format specificity arises there is transient, which is
simply ruled out.

-- Scott