[Rats] Re: Progressing CoSERV and CoTS

[Thomas Fossati <[e-mail]>]

Hi Paul,

[top-posting]

Thanks a lot for the write-up.

To provide some additional information on the status of CoTS: it was
parked while waiting for CoRIM to converge.
CoRIM is in much better shape now than it was three years ago, so it
makes sense to take CoTS out of the refrigerator.

As a co-author of CoTS, my (very fresh) assessment is that there are a
few things that need to be sorted out, but nothing major.
As you said, the authors are having conversations on GitHub about a
few open topics.
The aim is making progress towards bringing CoTS into WGLC-ready shape
in the near future.
Both CoTS and CoSERV are queuing behind CoRIM, though, so we still
have some time to make sure the two efforts align.

cheers!

On Wed, Aug 12, 2026 at 10:48 AM Paul Howard <[e-mail]> wrote:
>
> Dear RATizens,
>
> I’m (belatedly) following up on an action from IETF-126 related to the progress update on CoSERV.
>
> As per my presentation in the meeting, CoSERV [0] currently has a normative dependency on CoTS (Concise TA Stores) [1]. This dependency is at the CDDL data model level, as explained in slide 5 of [2]. This means that CoSERV would be unable to progress towards publication until CoTS was published.
>
> At the time of presenting at IETF-126, the CoTS draft was in an expired state. It was not clear - at least, not to me - what the future of this draft was looking like, despite its being adopted. Hence the dependency caused me some concern.
>
> In my presentation, I mentioned possibilities such as:
>
> Completely removing the CoTS dependency from CoSERV, and relying solely on CoRIM AK triples.
> Creating some sort of dynamic abstraction such that CoSERV would be able to support CoTS objects in the future, but without directly inheriting the CoTS data model.
>
>
> I took an action to start an on-list discussion around these options, and potentially others.
>
> In the meantime, though, the CoTS draft has been de-expired, with version 03 appearing a few weeks ago.
>
> Furthermore, Thomas Fossati has examined the draft with the CoSERV integration in mind, and has raised some issues in the CoTS GitHub repo [3], which have resulted in strong engagement from the editors of CoTS, meaning that there might be a good path forward for progressing both CoTS and CoSERV together.
>
> There’s more thinking to be done in terms of the integration, particularly with respect to whether part of CoSERV’s query model might need to change to better accommodate CoTS. But, for now, the front-running suggestion is to keep the normative dependency from CoSERV to CoTS, rather than try to remove it or abstract it.
>
> This post serves to fulfil my pending action from IETF-126, and provide a status update, and encourage any discussion around the topic.
>
> Thanks and all the best,
> Paul on behalf of CoSERV editors.
>
>
> [0] - https://datatracker.ietf.org/doc/draft-ietf-rats-coserv/07/
> [1] - https://datatracker.ietf.org/doc/draft-ietf-rats-concise-ta-stores/03/
> [2] - https://datatracker.ietf.org/meeting/126/materials/slides-126-rats-coserv-progress-update-00
> [3] - https://github.com/ietf-rats-wg/draft-wallace-rats-concise-ta-stores/issues
>
> Paul Howard
> Senior Principal Software Architect | Arm
> ………………………………………………………………..
> m. [phone]
> arm.com



-- 
Thomas