This feels like a good moment for me to mention that Anne van Kesteren and
I have been writing a new iteration of the Cookies spec
<https://johannhof.github.io/draft-annevk-johannhof-httpbis-cookies/draft-annevk-johannhof-httpbis-cookies.html>
that aims to fix these layering violations with WHATWG and other browser
specs (e.g. Cookie Store API). I presented it to this group a few months
ago and we're hoping to submit it officially very soon (we were mostly
waiting on 6265bis to move out of HTTP WG). We also discussed it at TPAC
with the other browser folks and I think we have pretty good alignment on
this as the way forward.
Independent of what we come up with to solve this particular issue - it
does feel like something that should be addressed or at the very least
considered again in that next spec.
Johann
On Tue, Dec 3, 2024 at 3:09 PM Greg Wilkins <gregw@webtide.com> wrote:
>
>
> On Tue, 3 Dec 2024 at 17:30, Daniel Stenberg <daniel@haxx.se> wrote:
>
>> On Tue, 3 Dec 2024, Greg Wilkins wrote:
>>
>> > make sure that the next RFC attempts to solve those issues so that the
>> > browsers can simply adopt the RFC rather than their own specification.
>>
>> I don't think they have or use any other cookie RFC.
>>
>
> My bad, I was thinking of URLs <https://url.spec.whatwg.org/> . But my
> point stands in general, as I think the issues with various interpretations
> of cookies are very similar to the URL issues that drove the creation of a
> whatwg spec. So before any respecification effort on Cookies in the IETF,
> that there is a explicit effort to engage the whatwg to ensure that they
> will follow such a RFC and not go their own way as they have with URLs.
>
> --
> Greg Wilkins <gregw@webtide.com> CTO http://webtide.com
>