[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Erratum: Descriptions of QNames
I'm discovering that the descriptions of the operation of AuthorityKind and RespondWith are a bit misleading when it comes to QNames. I'd like to suggest a change that I'm happy to consider editorial, but I will wait for the group's consensus on this before making the change. Near lines 716 (all line references in this message are to core 1.0) for AuthorityKind, and 968 for RespondWith, the text gives an example of a QName in use and unfortunately implies (rather more strongly in the latter case) that the prefix must read "saml" when a natively defined construct is being referenced. But the prefix of a namespaced value is never fixed, and we don't clarify that the appropriate namespace must have been defined in the scope of the relevant element where the QName appears. It would be better to say something like this (underscores around new or changed material): For AuthorityKind: "For example, an attribute authority would be identified by AuthorityKind="samlp:AttributeQuery", _where there is a namespace declaration in the scope of this attribute that binds the saml: prefix to the SAML protocol namespace_." For RespondWith: "For example, a requestor that wishes to receive assertions containing only attribute statements _would_ [this was a lowercase "must"] specify <RespondWith>saml:AttributeStatement</RespondWith>, _where the prefix is bound to the SAML assertion namespace in a namespace declaration that is in the scope of this element_." The QName description starting at line 1255 regarding StatusCode seems okay. Eve -- Eve Maler +1 781 442 3190 Sun Microsystems cell +1 781 354 9441 Web Technologies and Standards eve.maler @ sun.com
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]