This is excellent feedback, thank you! I'll take a look at the wordings and you're right, I need to circle back to the "and" claim. I'll get that cleaned up.
On RFC 9711... I'm not sure. The problem I'm running into is that $$Claims-Set-Claims is defined in an informative appendix of that document, so it's weird to normatively reference the informative appendix. But it's a real issue, so I'll see if I can find a solution. On Thu, Jul 30, 2026 at 5:15 AM Emil Lundberg <[email protected]> wrote: > > Hi Chris and COSE WG, > > Please find my review comments on draft-lemmons-cose-composite-claims-03: > "Composite Token Claims" below. Overall I think the draft looks fine, but I > myself have no particular use case for it at the moment. > > > Section 3.1.2. nor (Nor) Claim: > 'If any claim set in a "nor" claim is acceptable, the claim set containing > the "nor" claim is also unacceptable.' > It seems to me like this would read better as '[...] the claim set containing > the "nor" claim is also unacceptable' (without the "also"). > > > Section 3.1.3. and (And) Claim: > 'If any claim set in an "and" claim is not acceptable, the claim set > containing the "and" claim is also unacceptable.' > This too might read better as 'If any claim set in an "and" claim is > unacceptable, [...]' > > 'The "and" claim is often unnecessary' > Perhaps include an example of the "and" claim being useful even though it's > often unnecessary? There is one example in the next section, but it rather > undercuts itself by admitting that "and" is not actually necessary for the > "aud" criterion. > > > References: > [RFC9711] should be normative? (Since this document extends > $$Claims-Set-Claims defined by it) > > > Cheers, > > Emil Lundberg > > Staff Engineer | Yubico > > > _______________________________________________ > COSE mailing list -- [email protected] > To unsubscribe send an email to [email protected] _______________________________________________ COSE mailing list -- [email protected] To unsubscribe send an email to [email protected]
