[Dawn] 回复: A reworked draft charter for DAWN

[Lianhua Zhang <[e-mail]>]

Hi Adrian and Wes,





A minor editorial comment on the current DAWN draft charter: in the “Coordination with Other Working Groups and Organisations” section, WISME should be changed to WIMSE (Workload Identity in Multi System Environments).





Best regards,


Lianhua

-----原始邮件-----
发件人: "Adrian Farrel" <[e-mail]>(由 "forwardingalgorithm" <[e-mail]> 代发)
发送时间: 2026-06-20 18:09:32
收件人: "dawn" <[e-mail]>
抄送: "dawn-chairs" <[e-mail]>,"'Eric Vyncke (evyncke)'" <[e-mail]>
主题: [Dawn] A reworked draft charter for DAWN




Hi,

Now that the BoF is approved and we have some I-Ds to fill in the
background, Wes and I have been working on the draft charter text for a
potential DAWN working group.

You can find the working copy at
https://github.com/danielkinguk/discovery/blob/main/charter/dawn-charter-02.
md 

First step was to halve the number of words (sorry, the initial verbosity
was mainly my fault).

Then we also worked to focus in on the distinctions between the WG's initial
focus and what might come later. This was addressing two key questions:

1. Should the WG work only on AI agent discovery, or should it be open to
more general "entity" discovery?
Here, we noted that most of the drafts that have been posted are dealing
with AI agent discovery, and the need for work in that area could be
described as "pressing".
On the other hand, it seems highly likely that in solving AI agent discovery
we could also solve discovery for a range of similar entities.
That leads us to suggest the WG should initially be focused on discovery of
AI agents, but that the WG is expected to produce results that are general
and reusable within other discovery contexts whenever possible.

2. Should the WG select between a number of possible solution technologies,
or should it build immediately on DNS?
A number of factors influenced us here:
- There is often a preference to use pre-existing IETF technology where
possible.
- A quick start will re-use or modify existing code, but can then be
developed along different lines.
- There are probably two stages to "discovery", the first being relatively
simple discovery, the second being the exchange and negotiation of
capabilities and other information.
- We are not certain at this stage that DNS will do the job.
This lead us to suggest that, "Where possible, any solutions work will be
built in a modular way using existing IETF protocols that provide support
for any needed communication, authentication and privacy. The WG will
consider the DNS as a likely initial protocol upon which to build a
discovery protocol."

We would very much like to hear discussion (or just words of
approval/opprobrium) on the charter text and these two questions.

Thanks,
Adrian (per pro Wes)

_______________________________________________
Dawn mailing list -- [e-mail]
To unsubscribe send an email to [e-mail]