--On Tuesday, October 13, 2020 13:00 +1300 Jay Daley
<jay@xxxxxxxx> wrote:
> Thank you all for the feedback so far. In order to understand
> how to draft revisions to address the issues raised [1], I
> have some questions. In asking these I want to reiterate that
> Conflict of Interest is, in my experience, often determined by
> individual perceptions rather than an entirely objective
> framework.
>
> 1. The emerging view of the community is that an RPC employee
> *contributing* to a draft is sufficiently distant from the RPC
> process to avoid any conflict of interest. It appears to me
> that *authoring* a draft however does introduce a conflict of
> interest (COI) as part of the RPC role is to enforce specific
> standards that authors must comply with, which is a standard
> COI that is controlled by separation of powers. There is an
> existing situation where RPC staff author RFCs on behalf of
> the RPC but that is different from a personal role.
Jay, this would be fine, except for two things. The first is
that the distinction between "author" and "contributor" (whether
listed as such in the body of the I-D or RFC or not) has been
much less clear in the history of the RFC Series than your
description above seems to imply. Putting the RPC aside for a
moment, I can't count the number of times (including when I was
an AD or WG chair) in which I've helped someone sketch out a
document and contributed strawman text and then dropped back to
Contributor or, more often, a simple acknowledgement. Sometimes
those have even been documents whose positions I don't like but
felt it important that the community have a chance to consider
the ideas. My participation in those ways has never been a
secret and I think there is an ethical (whether contractual or
not) obligation to disclose such involvement.
But I think the general principle is close to the one I
identified about the Nomcom restrictions: The IETF simply cannot
afford to discard (or be deprived of) knowledgeable and
competent input and opinions just because someone who has them
happens to have taken a job with a contractor organization (or
because they work for a company that has become unpopular). We
(and especially WGs and the IESG) made evaluations of the
quality and content of work independent of its authorship and
their connections every day: even if we don't already get it
right, that is the business we are in, just as the Nomcom is in
the business of soliciting and evaluating comments about
candidates. It would be perfectly reasonable for AMS to say to
one of its staff "we have a contract here and you can't work on
that document to which you are contributing on company time".
It is not reasonable, or in the IETF's best interest, to prevent
input, contributions, or participation from anyone. For many
I-Ds and RFCs, for the LLC to do that would constitute
interference in the standards process which I believe the LLC is
expressly forbidden to do (see Brian's note about the Nomcom
part of that).
> I would be interested therefore to hear which of the following
> ways forward you think is best (for now let's leave aside
> whether policy or contract is the best way to capture this):
>
> a. Prohibit RPC staff from *authoring* non-RPC drafts (not
> contributing to, just authoring). b. Prohibit any RPC staff
> that author a non-RPC draft from any processing or discussion
> of that draft in their RPC role. c. No restrictions at all.
Close to "c": this may be an AMS (or whomever the RPC contractor
is) issue, but it is not one in which the LLC should be meddling
wrt the RPC or any other AMS employee.
> 2. RFC 8711 is clear about the role of the LLC that "It has
> no authority over the standards development activities of the
> IETF", which is taken to mean that there should be a bright
> line between the work of the LLC and the standards process.
> The concern is that the LLC and the IETF Executive Director
> (and the Secretariat because they act on behalf of the LLC)
> have significant power in the the IETF, even if it is not
> directly about the standards process, and if allowed to
> participate in the standards process could utilise that power
> to "put their finger on the scales".
Again, if we eliminated participation from everyone whose
affiliations could create the appearance of putting fingers on
scales, we'd have very few people left, including members of the
LLC Board, even those who are not Nomcom appointees... whom I
note this policy does not prevent from contributing to the
standards process.
> With that explanation would anyone object to maintaining the
> restriction on staff and Secretariat being involved in the
> standards process or do you think there is not a COI that
> needs to be guarded against?
I think the only reasonable remedy to actual or perceived COIs
is full disclosure, not prohibitions.
> 3. The proposed policy includes the following statement with
> regard to development of RFCs that nobody has yet suggested is
> inappropriate:
>> Contractors are also expected to ensure that it is always
>> clear to those being engaged with, if the contractor is
>> engaging as a contractor or as a volunteer, as assumptions
>> may vary between people and situations.
>
> Does anyone object to that same clause applying to any
> interaction with the NomCom?
I would be happier if such disclosure requirements applied to
everyone involved, including LLC staff and LLC Board members wrt
any decisions they make.
And that may expose a small pachyderm in this particular room:
In this situation and a number of others, the boundary seems a
little thin between your explaining a proposal from yourself
and/or the Board, asking questions to clarify community
perspectives (as above), and so on on one hand and either
advocating for particular approaches or offering perspective or
advice in your individual capacity. Consistent with what I'd
said elsewhere, I think more transparency would be helpful while
more rules would probably not be. Do you agree and, if so,
could you try to be a little bit more clear and when you are
speaking for the Board versus offering your own perspective and
when you are speaking as an individual rather than Exec Dir?
Finally, coming back to your response to one of Mike's
questions, is it reasonable to assume that, if the LLC makes new
polities that it intends to apply to future contracts and
retroactively to existing ones, those policies will at least be
clearly identified and shared with the community? That would
seem to be consistent with general requirements about openness
and transparency without infringing or interacting with
confidential provisions of any particular contract?
thanks,
john
