This is enough for me to clear my discuss (as I see v7 has been published
with all the changes).

Deb

p.s. I was also happy to see that the typos in the message (than/then,
TCC/TTC) were not in the draft.  shew....

On Wed, Aug 13, 2025 at 8:01 AM Henk Birkholz <
[email protected]> wrote:

> Hi Deb,
>
> In your DISCUSS, you have highlighted four issues. At least three of
> these have been addressed via
>
> https://github.com/ietf-scitt/draft-birkholz-cose-tsa-tst-header-parameter/pull/61
> .
>
> Correspondingly, I-D.ietf-cose-tsa-tst-header-parameter-07 reflects
> these changes. We hope that issues 2, 3 and 4 in your DISCUSS
> are addressed.
>
> This leaves issue 1: "Names of modes".
>
> The authors discussed this feedback extensively — and I mean
> _extensively_.  Unfortunately, the author's opinion holds (where
> "signing" means "COSE"), as follows:
>
> "COSE Than Timestamp" (CTT) and "Timestamp Than COSE" (TCC) are as
> simple and intuitive as they get. To further address the concerns
> raised, we have created a new subsection of the Security Considerations:
> §5.1 "Avoiding Semantic Confusion", which is referenced from the very
> first paragraph of §1.1 "Use Cases". This presents text suggested by
> Stefan for further clarification. For convenience, we quote this
> text here:
>
> > Implementers MUST clearly differentiate between RFC 3161 TSA:
> > timestamps proving the existence of payload data at an earlier point
> > in time (TTC), and timestamps that explicitly provide evidence of the
> > existence of the cryptographic signature (CTT).  Failure to clearly
> > Failure to distinguish between these timestamp semantics can result in
> > vulnerabilities, such as incorrectly accepting signatures created
> > after key revocation based on older payload-only timestamps.
> > Validators must not interpret protected-header payload timestamps as
> > proof of signature creation time, and should rely exclusively on RFC
> > 3161 TSA timestamps explicitly cover signature data.
> > determining signature validity timing
>
>
> We expect readers to read the specification, and we believe that we have
> introduced a clear distinction between modes and the sequence of
> content. We have also introduced a prioritization of modes of use. Any
> further attempt to address your comment resulted in less readable text
> or redundant reminders in the English text about which mode applies in
> which context. This does not seem to be a naming problem, but a
> presentation problem. We sincerely hope that the compromise in version
> -07 is sufficient to address your concerns by improving the
> presentation.
>
>
> For the I-D.ietf-cose-tsa-tst-header-parameter editors,
>
> Henk
>
> On 05.08.25 21:19, Deb Cooley via Datatracker wrote:
> > Deb Cooley has entered the following ballot position for
> > draft-ietf-cose-tsa-tst-header-parameter-06: Discuss
> >
> > When responding, please keep the subject line intact and reply to all
> > email addresses included in the To and CC lines. (Feel free to cut this
> > introductory paragraph, however.)
> >
> >
> > Please refer to
> https://www.ietf.org/about/groups/iesg/statements/handling-ballot-positions/
> > for more information about how to handle DISCUSS and COMMENT positions.
> >
> >
> > The document, along with other ballot positions, can be found here:
> >
> https://datatracker.ietf.org/doc/draft-ietf-cose-tsa-tst-header-parameter/
> >
> >
> >
> > ----------------------------------------------------------------------
> > DISCUSS:
> > ----------------------------------------------------------------------
> >
> > Thanks to Stefan Santesson for the repeated secdir reviews.  These
> reviews have
> > driven change to create a better specification.  The authors are lucky
> to have
> > drawn a reviewer that is well versed in the subject of timestamps.  I'm
> not
> > sure it is possible to completely eliminate the issue of a choosing the
> wrong
> > method, but I hope recent changes have made it harder for the developer
> to make
> > that mistake.  I will make a few comments which I hope will make it a
> tiny bit
> > harder.
> >
> > Names of modes:  Stefan points out that the two mode names are similar,
> perhaps
> > too similar.  Instead of choosing the mode names by the order of
> operation,
> > maybe mode names that describe the operation.  Perhaps 'certificate
> timestamps'
> > and 'unsigned statement timestamps', recognizing that these don't create
> nice
> > acronyms, and that it is a pretty pervasive change to the draft.
> >
> > Section 1.1:  Add a sentence to the first para that this specification
> outlines
> > the original mode and a new mode, where the security characteristics are
> > different so care needs to be taken to choose the appropriate mode.
> >
> > Section 1.1, para 2&3:  Please consider stating that this use case is the
> > primary, or original use case that most implementations will use.
> Something
> > like 'The original use case....'  and then 'This primary usage scenario
> > motivates....'
> >
> > Section 1.1, para 4&5:  Please consider stating that this is a new use
> case for
> > very specific purposes.  Something like 'The new use case...' and then
> 'This
> > new usage scenario....'
> >
> >
> > ----------------------------------------------------------------------
> > COMMENT:
> > ----------------------------------------------------------------------
> >
> >
> > Section 1.1, para 4:  A small clarification, if the TST is acquired
> before the
> > statement is signed, then the relying party knows that the statement was
> signed
> > by the issuer 'not before' the times specified by the TSA.  This is an
> early
> > bound (vice a no later bound in the first case).
> >
> >
> >
>
_______________________________________________
COSE mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to