Re: [Tools-discuss] "Distantly related Internet-Drafts"?

[Spencer Dawkins at IETF <[e-mail]>]

Dear All, but especially Lars,

On Fri, Oct 1, 2021 at 7:44 AM Lucas Pardue <[e-mail]>
wrote:

>
>
> On Fri, 1 Oct 2021, 07:25 Lars Eggert, <[e-mail]> wrote:
>
>> Hi,
>>
>> On 2021-10-1, at 1:59, Spencer Dawkins at IETF <
>> [e-mail]> wrote:
>> > The first document filename starts with "draft-dawkins-quic". I expect
>> this one to be listed as a related Internet-Draft. The filename is
>> "draft-author-wgname-*", as suggested for Internet-Drafts in
>> https://www.ietf.org/standards/ids/guidelines/#7, and it is related to
>> the QUIC working group.
>> >
>> > The second and third document filenames start with
>> "draft-dawkins-sdp-rtp-quic". I wasn't expecting to see these listed as
>> related to the QUIC working group. These drafts are intended for discussion
>> on the MOQ ("Media Over QUIC'') non-WG mailing list, and (if I understand
>> the guidance I've gotten from the TSV and ART ADs correctly) are actually
>> out of scope for the QUIC working group.
>> >
>> > Is it possible that the datatracker is scanning the filename looking
>> for any occurrence of a working group name, not just in the third position?
>>
>> IIRC, when I was a QUIC co-chair, I manually tweaked the pattern for the
>> QUIC WG in the datatracker settings, because people were submitting
>> QUIC-related I-Ds that were not showing up on our datatracker page and we
>> hence didn't learn about until (sometimes much) later.
>>
>> Since I'm not a co-chair anymore, I don't have access to the "manage
>> document list" button anymore, so I can't verify this.
>>
>
> I've become accustomed to the behaviour now. To my mind out of scope for a
> charter doesn't mean unrelated. I think the datatracker does a decent job
> with tags that it can let us clearly mark drafts that are serious
> contenders for adoption vs other more tangential things.
>
> If there's a problem that the current pattern causes then we can have a
> discussion but otherwise I'd leave it as is.
>

Thanks for helping me understand - and I feel much better about asking this
question in my original post:

>From the "Spencer is not the center of the universe" department, is it
reasonable to list individual drafts as "related", if the filename includes
the working group name in any position? If so, maybe that just needs to be
documented someplace (it doesn't look like
https://www.ietf.org/standards/ids/guidelines has been updated since Russ
Housley was IETF Chair), and I should be aware of that when choosing
filenames.


It looks like the current behavior in some working groups is actually MORE
helpful than what I thought the datatracker-wide behavior was - and I
didn't know that the behavior wasn't datatracker-wide, either.(*)

So, no, please don't change the behavior just to match
https://www.ietf.org/standards/ids/guidelines! You guys are Doing The Right
Thing.

(this is the "Lars" part) But it might be a good thing to update
https://www.ietf.org/standards/ids/guidelines to match current practice -
the only problem I'm seeing is that the current behavior violates the
Principle of Least Astonishment. I thought I knew where my drafts would
appear, but I was wrong.

Best,

Spencer

 (*) I've been co-chair for something like 5 working groups, but 4 of them
were before the datatracker had significant support for working group
chairs, and (as Lars noted) ADs don't see the "manage document list"
button, so I COMPLETELY missed that I have access to
https://datatracker.ietf.org/wg/cellar/documents/manage/. Seriously, thanks
for catching me up. And everyone who knows me, knows that "catching Spencer
up" is a full-time job ...