Re: Draft IAB conflict of interest policy

[Michael StJohns <[e-mail]>]

On 1/13/2020 2:43 PM, Joel M. Halpern wrote:
> There seems to be an implication in Mike's description that I find 
> troublesome.  It may well not be intended by Mike.  I apologize if I 
> am over-reacting.
>
> If we were to insist that WG chairs or ADs recuse themselves from 
> taking positions on technology that matters to their employer, we 
> would quickly find ourselves in a position where we could not get any 
> work done. Whether vendors or operators, our leaders care about the 
> technology we work on.  They (more accurately, we) work on the stuff 
> daily, and work on the stuff at the IETF.  All of the routing ADs 
> would probably have to recuse themselves from all of the routing 
> work.  That sounds backwards.
>
> Yes, We do ask that folks disclose their primary affiliation.  I 
> believe we ask that for WG chairs, for ADs, and for IAB members. That 
> gives the community the information about the situation. That is VERY 
> different from asking folks not to participate in leadership decisions 
> about work that may affect their employer.
>
> Yours,
> Joel 


Hi Joel -

No apologies necessary  - happy to clarify as needed.

Here's what I thought I wrote (between the lines!):

1) If you've publicly disclosed a conflict, feel free to participate in 
the process of building a standard.   If the standard comes before you 
on the IESG (e.g. standards vote) or IAB (appeal), best to recuse.  I 
mostly don't have a problem with being an active advocate for something 
- but being in the selection process - especially if competing standards 
- is probably reaching over the line.   Mostly I see this behavior from 
the various ADs and IAB members already.

2) If you've disclosed the conflict, and aren't participating in a 
specific standards building process, if the standards comes before you, 
has competitive benefits for your employer (e.g. mainly contributed by 
your employer or co-employees or standardizes your employers 
implementation choices over others), consider whether recusal is 
necessary and whether you need to put your two cents worth in given a 
non-conflicted co-AD?   A "maybe recuse" tick box here is one or more of 
Author is an employee, Editor is an employee, WG chair is an employee.

3) If you done a redacted disclosure, avoid even participating in the 
process of building a standard as its impossible for the rest of us to 
understand that there may be more than just your best technical 
judgement at work.   Having your relationship popup publicly later on 
would tend to taint the standards process and could give a competitor 
some basis for various legal actions both against you and against the 
IETF leadership.

4) If you really must participate, paper the hell out of it and make 
sure you have your leadership peers are on board - e.g. they have to be 
able to read past the redactions and be of the opinion that its good 
technical judgement playing here.

As someone else mentioned - this is one of the reasons why most of the 
AD positions are two deep.

I don't think I'm asking for "folks not to participate in leadership 
decisions about work that may affect their employer", what I am 
suggesting is that when in the "decider" role, that the person takes a 
very close look at the optics, and avoids blatant conflicts of 
interest.   90% of the time these will be "don't care" decisions for 
both the employer and competitors.  It's that other 10% where we might 
end up in trouble.

Later, Mike