- From: Joseph M. Reagle Jr. <reagle@w3.org>
- Date: Thu, 21 Dec 2000 18:00:30 -0500
- To: Philip Hallam-Baker <pbaker@verisign.com>
- Cc: "Public Archival Service" <www-archive@w3.org>
http://www.verisign.com/rsc/wp/xml/xkms_spec/xkms_spec_wp.pdf Some of these comments might be the result of my own careless reading, but this is what I scribbled in the margin for what it's worth: >Tier 1 Processing of the <ds: KeyInfo> element by the application is >delegated to a service. This is later called the Locate service, might as well include definition here. >Tier 2 Validation Service As in tier 1, but in addition, the service >reports further information "Validation" might be confused with "Signature Validation" in that a reader might think the service actually validates the signature. (Maybe "Binding Service"?) Either case, it'd probably be good to call it Binding Validation and distinguish between the formal definition of Signature Validation. >Tier 3 Assertion Service Establishment and management of long term trust >relationships. Is this a generalizable assertion service? >of the URI and Type attributes is mandatory Type is now optional (prose and schema disagreed (in your spec too), but now harmonized: http://www.w3.org/TR/2000/CR-xmldsig-core-20001031/#sec-RetrievalMethod ><ds: KeyInfo> > <RetrievalMethod .../> ></ ds: KeyInfo> You should qualify the RetreivalMethod with ds: too . ><Locate> <Query> I wonder if the query/response could be generalized using XPath (perhaps less appropriate) or XML Query (they have a data model and algebra, haven't designated an XML syntax yet though). ><Validate> and <ValidateResult> Is there a data model behind the request, the input, and output? Should the Query have an ID to match one then provided in the ValidateResult? >Authenticating the response messages using the XML Signature Specification. Where in the content model do these Signatures go, and which elements must be signed? >Authentication >An XML document node that provides information that authenticates the >request. Probably better to call this an XML element type (instead of document node). __ Joseph Reagle Jr. W3C Policy Analyst mailto:reagle@w3.org IETF/W3C XML-Signature Co-Chair http://www.w3.org/People/Reagle/
Received on Thursday, 21 December 2000 18:00:45 UTC