On Tue, Oct 8, 2024 at 11:10 AM Brian Campbell <[email protected]>
wrote:

> The OAuth URI registry
> <https://www.iana.org/assignments/oauth-parameters/oauth-parameters.xml#uri>
> on the OAuth Parameters
> <https://www.iana.org/assignments/oauth-parameters/oauth-parameters.xml>
> registry page was established by RFC6755 An IETF URN Sub-Namespace for
> OAuth <https://www.rfc-editor.org/rfc/rfc6755.html>, which says it
> "establishes an IETF URN Sub-namespace for use with OAuth-related
> specifications." The subnamespace of "urn:ietf:params:oauth:" (which I got
> wrong by omitting the "ietf:" part in my prior message, apologies) was
> created stating in the intro
> <https://www.rfc-editor.org/rfc/rfc6755.html#section-1> that "OAuth
> relevant parameters will be established underneath it." The presence of the
> word "oauth" in the URN Sub-namespace itself also implies some relationship
> to OAuth.
>
> COSE and COSE keys and their thumbprints are not OAuth-related
> specifications nor OAuth relevant parameters.
>
> That's why I feel this is incorrect.
>
> JWK Thumbprints URIs <https://www.rfc-editor.org/rfc/rfc9278.html> are
> arguably also not OAuth-related. I'd be pretty sympathetic to that
> argument. I'd also be sympathetic to an argument that that document
> shouldn't even exist. But it's much too late to do anything about that now.
> And JWKs are used in some OAuth-related specifications and the document
> came up through the OAUTH WG so there is some relationship, if a rather
> tenuous one.
>

I agree with your comments regarding the relationship of JWK Thumbprints to
OAuth.
The same public key expressed as a JWK or COSE Key will have different
thumbprints.
For this reason, I think it will be least astonishing for developers to
discover the URI expressions of thumbprints in the same registry.

BTW, this topic was discussed previously here, and I agree with Mike's
comment generally:

https://mailarchive.ietf.org/arch/msg/cose/wGUObrMhC1QmERJ5_l47gK8unF4/

Especially this part: Unnecessary differences when doing the same thing
should be avoided.

While digging this up I realized the document does not explicitly state
that these thumbprint URIs can be used by applications other than OAuth.

I see several paths forward:

- Remove the COSE Key Thumbprint URI from the document.
- Keep it and use a different sub-namespace (not oauth).
- Keep it and use oauth

I prefer the last option, but I believe we should apply Mike Jones'
suggestion to be explicit that these URIs are expected to be used by
applications other than OAuth.


>
>
>
> On Mon, Oct 7, 2024 at 5:36 PM Orie Steele <[email protected]>
> wrote:
>
>> Brian, indeed this was done to align with the existing JWK Thumbprint URI.
>>
>> Why do you feel this is incorrect?
>>
>> Here is the JWK example:
>>
>> urn:ietf:params:oauth:jwk-thumbprint:NzbLsX...
>>
>> In an ideal world, I think names for keys should be shorter... and not
>> protocol specific.
>>
>> Something like:
>>
>> urn:jkt:Nzb....
>>
>> urn:ckt:Nzb....
>>
>> But that's not the way these parameters have been registered historically.
>>
>> OS
>>
>> On Mon, Oct 7, 2024, 5:34 PM Brian Campbell <bcampbell=
>> [email protected]> wrote:
>>
>>> I realize this comes quite late, sorry, probably too late for any
>>> action. I find myself here due mostly to a tangentially related discussion
>>> in a different standards-related organization.
>>>
>>> But can anyone explain the justification for the use of the OAuth URI
>>> registry here? I realize the registry exists so it was probably a
>>> convenient thing to do to carve out a URN sub-namespace. And I know that,
>>> for better or worse, the JWK Thumbprint URI uses "urn:oauth:params:" so
>>> this was likely just following what was done there. But the use of
>>> "urn:oauth:params:" for a COSE Key Thumbprint URI really doesn't seem quite
>>> right.
>>>
>>> On Thu, Apr 4, 2024 at 9:04 AM Tschofenig, Hannes <hannes.tschofenig=
>>> [email protected]> wrote:
>>>
>>>> Hi David,
>>>>
>>>> I hope it is OK for me to do the expert review given that I am also an
>>>> author of the specification.
>>>>
>>>> I checked the text in the IANA registry of the draft against the body
>>>> of the document and the request to add a new entry to the OAuth URI
>>>> registry for the URN: urn:ietf:params:oauth:ckt is correct.
>>>>
>>>> Ciao
>>>> Hannes
>>>>
>>>> PS: Could you remove the Arm email address from the IANA system?
>>>>
>>>> -----Original Message-----
>>>> From: COSE <[email protected]> On Behalf Of David Dong via RT
>>>> Sent: Wednesday, 13 March 2024 20:38
>>>> Cc: [email protected]; [email protected]
>>>> Subject: [COSE] [IANA #1361034] expert review for
>>>> draft-ietf-cose-key-thumbprint (oauth-parameters)
>>>>
>>>> Dear Hannes Tschofenig (cc: cose WG),
>>>>
>>>> As the designated expert for the OAuth URI registry, can you review the
>>>> proposed registration in draft-ietf-cose-key-thumbprint-04 for us? Please
>>>> see
>>>>
>>>> https://datatracker.ietf.org/doc/draft-ietf-cose-key-thumbprint/
>>>>
>>>> The due date is March 27th, 2024.
>>>>
>>>> If this is OK, when the IESG approves the document for publication,
>>>> we'll make the registration at:
>>>>
>>>> https://www.iana.org/assignments/oauth-parameters/
>>>>
>>>> With thanks,
>>>>
>>>> David Dong
>>>> IANA Services Sr. Specialist
>>>>
>>>> _______________________________________________
>>>> COSE mailing list
>>>> [email protected]
>>>> https://www.ietf.org/mailman/listinfo/cose
>>>> _______________________________________________
>>>> COSE mailing list
>>>> [email protected]
>>>> https://www.ietf.org/mailman/listinfo/cose
>>>>
>>>
>>> *CONFIDENTIALITY NOTICE: This email may contain confidential and
>>> privileged material for the sole use of the intended recipient(s). Any
>>> review, use, distribution or disclosure by others is strictly prohibited.
>>> If you have received this communication in error, please notify the sender
>>> immediately by e-mail and delete the message and any file attachments from
>>> your computer. Thank you.*
>>> _______________________________________________
>>> COSE mailing list -- [email protected]
>>> To unsubscribe send an email to [email protected]
>>>
>>
> *CONFIDENTIALITY NOTICE: This email may contain confidential and
> privileged material for the sole use of the intended recipient(s). Any
> review, use, distribution or disclosure by others is strictly prohibited.
> If you have received this communication in error, please notify the sender
> immediately by e-mail and delete the message and any file attachments from
> your computer. Thank you.*



-- 


ORIE STEELE
Chief Technology Officer
www.transmute.industries

<https://transmute.industries>
_______________________________________________
COSE mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to