[Atp] Re: Video Meeting on Charter Scope Next Week

[Daniel Holmgren <[e-mail]>]

Thanks to everyone that joined the video call today! We had around 20
people on the call with a good mix of folks from both the IETF and AT
communities. As expected, we spent most of the call circling around this
question of how to handle the identity system in the charter text. I’ll
give a quick summary of the call as well as my read on the temperature of
the room. If you feel like I’m misrepresenting anything please hop in and
clarify.

I would say the stronger opinion in the room was that in the absence of a
specified identity system, the protocol doesn’t allow for interoperation,
and therefore specifying the identity system is necessary. To illustrate:
in the network as deployed today, you would be unable to interact with
other accounts or services if you do not use the associated identity system.

The other opinion is that, while that is true, specifying an interface for
an identity system without specifying the system itself is still a coherent
and useful piece of work. In other words, the working group should keep it
simple to start with.

Everyone was in agreement that we need to at least define a profile of the
properties of supported identity mechanisms. This is the case whether the
working group defines an abstract interface, references DID as part of the
interface, or defines its own identity system.

The general sentiment toward DID seemed to be “not great, but not the end
of the world”. We shouldn’t try to adopt the full, messy, constraining
world of DID. Maybe we can define a subset of the DID spec that is
supported. At the very least we should not support all DID methods and we
should instead define a small set of supported DIDs. The group seemed
supportive that this set of blessed methods could be defined in an IANA
registry and that that set of methods could evolve over time.

There was discussion of test networks, experimental implementations,
isolated communities, etc that basically came down to: don’t be dogmatic
about disallowing usage of other methods. Pave the main happy path for the
blessed methods that you need for interoperation, but leave the interface
open for off-spec experimentation. We discussed the example of Tor and the
.onion special-use TLD in the broader context of DNS as an analogy to this
flexibility.

Bryan and I are taking a stab at another draft of the charter text. We
should have that out on the mailing list by tomorrow. Looking forward to
hearing feedback.


- Daniel


On Wed, Dec 10, 2025 at 12:59 PM Bryan Newbold <bryan=
[e-mail]> wrote:

> Just a quick final reminder, this call is getting started in a minute.
> Video link in the thread, see you there!
>
> On Mon, Dec 8, 2025 at 6:36 PM Bryan Newbold <[e-mail]> wrote:
>
>> Had a few folks respond and the proposed time worked for them, so
>> confirming that we will meet this Wednesday to discuss charter scope:
>>
>> Wednesday, December 10
>>   11:00am – 12:00pm US/Pacific
>>   7:00pm - 8:00pm UTC
>>   Timezone conversion:
>> https://www.timeanddate.com/worldclock/fixedtime.html?iso=20251210T11&p1=234&ah=1
>>
>> Google Meet: https://meet.google.com/hdo-bjdr-pdg
>> Calendar link:
>> https://calendar.google.com/calendar/event?action=TEMPLATE&tmeid=N2JtNWJkNmtwYXVsZTE4aWx2YW9vMXNyazggYnJ5YW5AYmx1ZXNreXdlYi54eXo&tmsrc=bryan%40blueskyweb.xyz
>>
>> The core of the agenda will be discussing which identity components
>> should be included in the charter, though we will also leave time to
>> discuss other aspects of the charter. My goal is to send out an iteration
>> of the charter text to the mailing list after the call, by the end of this
>> week.
>>
>> --bryan
>>
>> On Thu, Dec 4, 2025 at 4:01 PM Bryan Newbold <[e-mail]>
>> wrote:
>>
>>> Hi folks,
>>>
>>> I'd like to schedule a video call to talk through charter scope, and in
>>> particular the boundaries around identity and DIDs. I'll send an open
>>> invitation and Google Meet link to this list ahead of time with an agenda,
>>> and we will take notes and post a summary afterwards. We don't have a WG
>>> yet and this won't be recorded or use Meetcho, but we'll keep it under IETF
>>> norms (Note Well).
>>>
>>> The easiest times for me next week are after Wednesday after 11am
>>> Pacific, or Thursday after 10am Pacific. To propose a specific time, let's
>>> say 11am Pacific next Wednesday (Dec 10th), for one hour.
>>>
>>> Please reply to me off list if you'd like to join and that time doesn't
>>> work for you. If there is a tangle of conflicts i'll send out a scheduling
>>> poll.
>>>
>>> --bryan
>>>
>> _______________________________________________
> Atp mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]
>