I think the line is determined by balancing factors such as tone, scope, complexity, detail, etc. Since this is the first step, I think that section 3.3.1 strikes a proper balance on introducing this concept, at this point in the project. I don't think we need to go into more detail about a particular infrastructure unless a certain use case is prevalent and relevant. We could, however, add a little more description about who is the subscriber, why the subscriber might be relevant to the trust model, to cover the gap you mentioned between RP and subscriber, etc.
-----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of David Chadwick Sent: Tuesday, October 15, 2013 11:26 AM To: [email protected]; 'Bruce Morton'; [email protected] Subject: Re: [wpkops] FW: New Version Notification for draft-barreira-trustmodel-00.txt In this case where do you draw the line of who to include and not to include. Supply chains are massively long and complex these days. So if the mainframe running the OCSP server crashes due to a fault of the manufacturer, so that no-one is able to check the revocation status of certs, is the computer manufacturer responsible rather than the CA? Should we mention this in the spec? Where do you draw the line? regards David On 15/10/2013 18:06, Ben Wilson wrote: > Concerning " 3.3.1. Subscriber uses agent", David Chadwick wrote, "5. > What is the relevance of section 3.3.1? If a third party is > subcontracted to a party to do work on its behalf, then the party is > ultimately responsible for this and there is no need to mention it." > > David, > > I think it is helpful to mention or flag where certain relationships > might exist that are not apparent by looking at just the > technical/operational aspects to provide context to model. Since we > are trying to explain how things operate, I think we need to go > slightly beyond just the traditional three-party model. > > I don't think that the distinguishing feature here is legal. For > instance, can we say with certainty that one party or another is > "ultimately responsible"? That might involve legal wrangling and it > might ultimately take a judge to make the determinations of who was responsible for what. > The devil is in the details of the subcontract. I'm not saying we > need to get into all of that legal stuff - quite the contrary. > > Ben > > -----Original Message----- > From: [email protected] [mailto:[email protected]] On > Behalf Of David Chadwick > Sent: Monday, October 14, 2013 5:52 AM > To: Bruce Morton; wpkops WG ([email protected]) ([email protected]) > Subject: Re: [wpkops] FW: New Version Notification for > draft-barreira-trustmodel-00.txt > > Hi Bruce > > here are my comments on this version > > 1. There is a potential problem with the scope/Introduction of the > document, since it only covers trust between the browser and the > subscriber, when what really matters is trust between the RP and the > subscriber. How is this gap to be covered? > > 2. Section 2.1. 3rd para insert may -> The root store provide "may" > require the root CA.... > Rationale. If the root store provider can verify a CA simply because > it has been accepted by another root store provider, as per the second > paragraph, then conversely, it may not require it to be annually > audited but may remove it only when the other root store provider removes it. > > 3. Section 2.3 insert may -> The subscriber may identify... > Rationale. This more accurately reflects the current situation today, > doesn't it? > > 4. Section 3.2.3. A third party RA is not identified in a CA > certificate as anything, is it?. Remove "as an issuing CA" as this > implies it is identified as something else. > > 5. What is the relevance of section 3.3.1? If a third party is > subcontracted to a party to do work on its behalf, then the party is > ultimately responsible for this and there is no need to mention it. > > 6. Section 5.2. Non-unique names. It is unclear whether non-unique > names refers to Internet wide unique names, or only to CA wide unique names. > Be explicit. > > regards > > David > > On 11/10/2013 13:02, Bruce Morton wrote: >> The Trust Model draft has been updated. >> >> Bruce. >> >> -----Original Message----- >> From: [email protected] [mailto:[email protected]] >> Sent: Wednesday, October 09, 2013 8:47 AM >> To: Inigo Barreira; Bruce Morton >> Subject: New Version Notification for >> draft-barreira-trustmodel-00.txt >> >> >> A new version of I-D, draft-barreira-trustmodel-00.txt has been > successfully submitted by Inigo Barreira and posted to the IETF repository. >> >> Filename: draft-barreira-trustmodel >> Revision: 00 >> Title: Trust models of the Web PKI >> Creation date: 2013-10-09 >> Group: Individual Submission >> Number of pages: 9 >> URL: > http://www.ietf.org/internet-drafts/draft-barreira-trustmodel-00.txt >> Status: http://datatracker.ietf.org/doc/draft-barreira-trustmodel >> Htmlized: http://tools.ietf.org/html/draft-barreira-trustmodel-00 >> >> >> Abstract: >> This is one of a set of documents to define the operation of the Web >> PKI. It describes the currently deployed Web PKI trust. >> >> >> >> >> Please note that it may take a couple of minutes from the time of > submission until the htmlized version and diff are available at > tools.ietf.org. >> >> The IETF Secretariat >> >> _______________________________________________ >> wpkops mailing list >> [email protected] >> https://www.ietf.org/mailman/listinfo/wpkops >> > _______________________________________________ > wpkops mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/wpkops > > > > _______________________________________________ > wpkops mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/wpkops > _______________________________________________ wpkops mailing list [email protected] https://www.ietf.org/mailman/listinfo/wpkops
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ wpkops mailing list [email protected] https://www.ietf.org/mailman/listinfo/wpkops
