Re: Comments on Composition and Comportment of the IETF Nominating Committee draft

There are a couple of assertions in your email that I question.

II have no disagreement with your observation that diverse teams produce 
(statistically) better results.   However, nothing in your proposal 
seems likely to increase diversity in either the nomcom or the IETF 
leadership.  (I have not commented on your prooposal to require 
diversity in the nomcom, as it appears to have no statistical effect as 
written.)

You then assert that somehow tying the hands of the nomcom in terms of 
the questions they ask of candidates will help avoid discrimination in 
the selection they perform.  That does not follow.  Hel;ing to ensure 
that the base answers are comparable is a good thing.  And is already 
common practice.  Requiring that only the comparable questions be asked 
is a bad thing.

Yours,

Joel

On 7/6/2026 11:10 PM, Mallory Knodel wrote:
> Hi Joel,
>
> On Mon, Jul 6, 2026 at 10:42 PM Joel Halpern <[email protected]> wrote:
>
>     I think without violating the confidentiality requirements, I can
>     make a couple of observations. Other past nomcom members / chairs
>     / liaisons can provide comments as they feel appropriate.
>
>     First, one of the difficulties is that the nomcom must treat the
>     feedback as confidential.   Great care is taken to not violate
>     that.  Nonetheless, committees do learn about issues with specific
>     candidates from the feedback, and it is important to identify
>     those issues. Whether they decide to ask the candidate about
>     aspects of that ought, as far as I can tell, be left to the
>     discretion of the committee.
>
>     In my experience committee members are generally good at
>     separating substantive feedback issues from feedback driven by
>     attributes that shouldn't matter.  There are many dimensions
>     besides the ones you are focusing on that the committee needs to
>     watch for and avoid.   I don't think trying to prevent bias in the
>     feedback is desirable or achievable.
>
> It is the goal of this section of this document to intervene against 
> discrimination in leadership appointments by the nomcom.
>
> You are validating that, indeed, current processes will change as a 
> result of non-discrimination efforts. This is inherently good. Also, 
> diverse teams perform better, which means these process changes will 
> result in better decisions and better outcomes for the IETF.
>
> Win-win.
>
> Best,
> -Mallory
>
>     It is also worth noting that we are interviewing people for many
>     different jobs.  At the very least, even the structured part of
>     the interview is likely to be different for the different
>     positions.  This then gets even more complicated when the
>     committee may be evaluating person A based on strength X and
>     weakness Y, and person B on strength Q and weakness R.  The
>     interview may well want to explore those specific qualities for
>     that specific candidate.  Even when we know the job, we know that
>     people will do the job differently.  And the selection process is
>     even more complicated by the fact that the job itself changes over
>     time.   In fact, one of the tasks the nomcom undertakes is
>     deciding what properties they think are important for which positions.
>
>     I suspect that further analysis would reveal even more nuances in
>     the interview and evaluation process.  Yes, it is important to be
>     clear that we avoid bias in our selection.  I don't think tying
>     the hands of the committee in terms of what they ask candidates is
>     helpful.
>
>     In some sense, the structured part of the interview presumably
>     starts with the common questionnaire.  I am not asking or
>     suggesting that the questionnaire be tuned to the individual
>     candidate.   But that is only the starting point.
>
>     Yours,
>
>     Joel
>
>     On 7/6/2026 10:25 PM, Mallory Knodel wrote:
>>     Hi Joel,
>>
>>     I think specific candidate questions, if not related follow ups
>>     to the structured interview questions, can invite discrimination.
>>
>>     OTOH, the US OPM goes to the other logical end: Employers should
>>     use structured interviews in which all candidates are asked the
>>     same core, job-related questions and are evaluated using the same
>>     criteria. Any follow-up questions should be limited to clarifying
>>     a candidate’s responses and should be used consistently to
>>     maintain fairness and comparability across candidates.
>>
>>     Probably there’s a middle ground.
>>
>>     It would be worth interrogating the role of community feedback.
>>     Obviously this is a unique element and guidance for most hiring
>>     processes won’t have much to say about it. If we worry about
>>     discrimination in nomcom decision making we should probably worry
>>     about it in candidate feedback too. So what would be the fair
>>     mitigation? Giving a candidate the chance to rebut a potentially
>>     discriminatory feedback comment? Or not mentioning the comment
>>     during the interview at all, and giving the nomcom discretion to
>>     consider feedback in its own right to supplement its decision?
>>
>>     Sorry I pointed the question back. But I’m genuinely not sure
>>     beyond my own experiences and intuition, which I’ll keep to
>>     myself as an author of this draft, which can hopefully encompass
>>     community consensus on the topic.
>>
>>     -Mallory
>>
>>     On Tue, Jul 7, 2026 at 11:11 Joel Halpern <[email protected]>
>>     wrote:
>>
>>         With regard to the structured interview item in your draft, I
>>         was unclear whether your documents intends to say that each
>>         interview should include the structured questions or were you
>>         aiming to say that the interviews MUST ONLY consist of the
>>         structured questions.
>>
>>         As far as I know, the former is common practice, and a good
>>         idea.  The latter would unduly constrain the committee from
>>         asking about things they think are important to the specific
>>         candidate, for example based on the feedback they have received.
>>
>>         Yours,
>>
>>         Joel
>>
>>         On 7/6/2026 10:03 PM, Mallory Knodel wrote:
>>>         Hi,
>>>
>>>         On Tue, Jul 7, 2026 at 01:11 S Moonesamy
>>>         <[email protected] <mailto:sm%[email protected]>> wrote:
>>>
>>>             Hi Mallory,
>>>             At 05:13 AM 06-07-2026, Mallory Knodel wrote:
>>>
>>>
>>>             >I'm not sure I remember. Could you elaborate this
>>>             comment please?
>>>
>>>             It's in RFC 7704 (Page 4).
>>>
>>>
>>>         Incredible. Thanks a lot for sharing this. I’ll cite it in
>>>         the next version.
>>>
>>>
>>>             >Disclosure of interest to the chair, who then decides
>>>             if there's a
>>>             >conflict. Simple recusal.
>>>
>>>             It could be a bit more complicated than that.
>>>
>>>
>>>         Certainly but I don’t know if the document can or should try
>>>         to encompass the complexity or simply describe a process for
>>>         addressing it.
>>>
>>>         I am open to text suggestions that better elaborate this and
>>>         any of the comportment sections.
>>>
>>>
>>>             >It's here. I've edited the main copy.
>>>             ><https://urldefense.proofpoint.com/v2/url?u=https-3A__hbr.org_2016_04_how-2Dto-2Dtake-2Dthe-2Dbias-2Dout-2Dof-2Dinterviews&d=DwIBAg&c=slrrB7dE8n7gBJbeO0g-IQ&r=krANNudPSfUTEf2kXiduBUqRjXhDsKNCASr1kibHLfs&m=YIrtQwrRtb_e31ldkUQjV6tcWTEcpf5fCSgJN5EFLRkjqZ3pHWlqlpLKxKQuPKb-&s=MBhR2Q0q-JMD3j71Sm58-ZD61u61e_r2AQAlRgDPkEI&e=
>>>             <https://urldefense.proofpoint.com/v2/url?u=https-3A__hbr.org_2016_04_how-2Dto-2Dtake-2Dthe-2Dbias-2Dout-2Dof-2Dinterviews&d=DwIBAg&c=slrrB7dE8n7gBJbeO0g-IQ&r=krANNudPSfUTEf2kXiduBUqRjXhDsKNCASr1kibHLfs&m=YIrtQwrRtb_e31ldkUQjV6tcWTEcpf5fCSgJN5EFLRkjqZ3pHWlqlpLKxKQuPKb-&s=MBhR2Q0q-JMD3j71Sm58-ZD61u61e_r2AQAlRgDPkEI&e=>
>>>             >https://urldefense.proofpoint.com/v2/url?u=https-3A__hbr.org_2016_04_how-2Dto-2Dtake-2Dthe-2Dbias-2Dout-2Dof-2Dinterviews&d=DwIBAg&c=slrrB7dE8n7gBJbeO0g-IQ&r=krANNudPSfUTEf2kXiduBUqRjXhDsKNCASr1kibHLfs&m=YIrtQwrRtb_e31ldkUQjV6tcWTEcpf5fCSgJN5EFLRkjqZ3pHWlqlpLKxKQuPKb-&s=MBhR2Q0q-JMD3j71Sm58-ZD61u61e_r2AQAlRgDPkEI&e=
>>>             <https://urldefense.proofpoint.com/v2/url?u=https-3A__hbr.org_2016_04_how-2Dto-2Dtake-2Dthe-2Dbias-2Dout-2Dof-2Dinterviews&d=DwIBAg&c=slrrB7dE8n7gBJbeO0g-IQ&r=krANNudPSfUTEf2kXiduBUqRjXhDsKNCASr1kibHLfs&m=YIrtQwrRtb_e31ldkUQjV6tcWTEcpf5fCSgJN5EFLRkjqZ3pHWlqlpLKxKQuPKb-&s=MBhR2Q0q-JMD3j71Sm58-ZD61u61e_r2AQAlRgDPkEI&e=>
>>>
>>>
>>>             Thanks.  The answers to the questions would not be
>>>             relevant in my opinion.
>>>
>>>
>>>         Fair. My thinking was to provide some authoritative example
>>>         of how industry thinks about this. I’m open to citing a
>>>         better and more relevant publication.
>>>
>>>         Thanks,
>>>         -Mallory
>>