Re: [arch-d] Draft IAB conflict of interest policy

[Ben Campbell <[e-mail]>]

> On Jan 9, 2020, at 2:07 PM, Livingood, Jason <[e-mail]> wrote:
> 
> From: ietf <[e-mail]> on behalf of Ben Campbell <[e-mail]>
>> I agree with that for the IESG, but the IAB is not usually in a position to approve or block specific technical decisions. Yes, it sets architectural direction, but see Richard’s comment down-thread.
> 
> [JL] Perhaps we are thinking too narrowly about what the IAB does? Here are some areas for potential conflict that are unrelated to architectural direction:
> 1 - Confirming the IETF Chair and Area Directors
> 2 - Standards appeals
> 3 - RFC Series
> 4 - Liaison roles
> 5 - Advice to ISOC

I think a good part of those categories are covered by the sections on personnel and contract CoIs. I agree with those sections in principle.


> 
> [JL] Even in architectural oversight, what if there was a strong push for a particular approach against a particular protocol or codec and someone on the IAB had an undisclosed financial stake in a patent pool that would directly benefit from a certain decision? Or if they "worked" for a university but were 100% funded by a grant from a government that was underwriting their time specifically to work against a specific upgrade to encryption? Or whatever other cases you might imagine.

I think the IPR pool example is already covered by the IETF IPR policies. I assume those will continue to apply. (And I suppose a reminder of them in any such CoI policy would not be a bad thing.)

In the govt grant example: Is the issue not also true for someone pushing similar views in, say, the TLS wg? It seems to me that an effort to undermine a specific upgrade would be done more effectively at the wg level than in the IAB.

That all being said, I’m almost okay with the proposed text that effectively says “IAB members should use their judgement to decide if they have a problem”, but that doesn’t seem to need to be said in a written policy (unless it’s purpose is to explain why we don’t say more.)

> 
> [JL] So IMO it seems like some CoI policy for the IAB is better than no CoI policy.
> 

I am not arguing for no policy. I agree in principle with the other two bullets.

Ben.