OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

security-services message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: PE 49 suggestion


Long-standing AI to clean up text around Attribute NameFormat/Name,
profiles, and how implementations should treat them.

Greg originally suggested text for core at line 1217 in the section defining
the <Attribute> element. It's currently in the errata document. A different
spin follows, but it's essentially the same in spirit.

"Attributes are identified/named by the combination of the NameFormat and
Name XML attributes described above. Neither one in isolation can be assumed
to be unique, but taken together, they ought to be unambiguous within a
given deployment.

The SAML profiles specification [SAMLProf] includes a number of attribute
profiles designed to improve the interoperability of attribute usage in some
identified scenarios. Such profiles typically include constraints on
attribute naming and value syntax. There is no explicit indicator when an
attribute profile is in use, and it is assumed that deployments can
establish this out of band, based on the combination of NameFormat and
Name."

If others have suggestions, please make them. The goal here is to
communicate to implementers that their logic must of necessity be based on
Name + NameFormat, and any profile processing essentially has to be
configured up front based on that combination. Which, BTW, is exactly how
SAML 1.1 worked.

-- Scott



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]