- From: Patrick Stickler <patrick.stickler@nokia.com>
- Date: Thu, 18 Mar 2004 13:13:18 +0200
- To: "ext Chris Bizer" <chris@bizer.de>
- Cc: "ext Jeremy Carroll" <jjc@hpl.hp.com>, <phayes@ihmc.us>, <www-archive@w3.org>, <jjc@hplb.hpl.hp.com>
On Mar 18, 2004, at 11:43, ext Chris Bizer wrote: >> I much prefer the swp:certificate property rather than forcing >> the URI denoting the authority to resolve to the certificate. >> > > Absolutely. I think it should be general practice that the URI > identifing an > authority refers to a graph with a self description of the authority > (name, > address, affiliation, roles in different application domains, > certificates > .... using the vocabularies of choice). Another solution would be using > rdf:seeAlso in a graph to link to this self description. > > > > G2 (G1 swp:assertedBy ex:chris. > > ex:chris rdf:seeAlso <http:www.bizer.de/selfdescription.trix>) > The practical difficulty with this is that it's not clear that folks will indicate the certificate of the authority in any consistent manner via an rdf:seeAlso reference. Better to have it explicit, e.g. :G2 (:G1 swp:assertedBy ex:chris . ex:chris swp:certificate <http://...> ) > > > ... "Distrust everything a vendor says about its competitor." ;-) > > > > I think we should derive our vocabulary for expression signatures and > certificates from the W3C "XML Encryption Syntax and Processing" and > "XML-Signature Syntax and Processing" specifications > http://www.w3c.org/Signature/ and the W3C key management work > http://www.w3c.org/2001/XKMS/, rather than inventing our own > vocabulary. A > guy called Garret Wilson started this, but didn't get too far. See: > > http://lists.w3.org/Archives/Public/www-rdf-interest/2003Dec/att-0140/ > crypto > -draft-20031224.html > > > > If we want to go this way, I could derive the necessary vocabulary > from the > specs next week. This sounds like a reasonable way to go, but I think for this particular paper, all we need to define is swp:signature and swp:Signature and simply cover their essential function, leaving it to future work to define the rest. I.e., go ahead and put together a proposal vocabulary, which may in any case be useful in clarifying things for the present paper, but let's keep the two activities distinct. OK? >>> 5) define swp:signedAndAssertedBy as a subproperty of (3) and (4) >> >> Would we not rather be able to infer >> >> ?graph swp:signedBy ?authority . >> >> from >> >> ?graph swp:authority ?authority . >> ?authority rdf:type swp:PGPCertifiedAuthority . >> >> ??? >> >> I.e., if someone is a PGP CA they they are always the CA >> for graphs which they verify/sign. >> >> This allows us to avoid the extra property. If the graph is >> signed with an X509 signature, then you have to specify the >> CA explicitly, otherwise it's the same as the value of >> either swp:authority or swp:assertedBy. >> >> Yes? >> > > No. I think a link to the CA is encoded in the X509 certificate. Thus > it is > enough that an authority signing a graph has a certificate property in > its > self description. An agent who wants to verify the signature can > decode the > certification chain from the certificates. Or we should have a look how > certification chains are handled in the "XML Key Management > Specification" > and use their vocabulary. OK, so either way, all we need is the swp:authority and swp:signature, and then via the URI of the authority we obtain a certificate for the authority, where the CA of that certificate is either the authority itself (PGP) or is specified in the certificate (X509). Right? Patrick -- Patrick Stickler Nokia, Finland patrick.stickler@nokia.com
Received on Thursday, 18 March 2004 06:25:47 UTC