Re: Tolerance

[Keith Moore <[e-mail]>]

On 7/16/19 5:55 AM, Dave Cridland wrote:

> On Mon, 15 Jul 2019 at 23:04, Keith Moore <[e-mail] 
> <mailto:[e-mail]>> wrote:
>
>
>     On Jul 15, 2019, at 4:51 PM, Dave Cridland <[e-mail]
>     <mailto:[e-mail]>> wrote:
>
>>     On Mon, 15 Jul 2019 at 18:50, Keith Moore
>>     <[e-mail] <mailto:[e-mail]>>
>>     wrote:
>>
>>         On 7/15/19 12:32 PM, Jacob Hoffman-Andrews wrote:
>>
>>         > To reinforce what Melinda's saying: I dedicate less time to
>>         IETF work
>>         > than I otherwise would, specifically because of the hostile
>>         and alien
>>         > nature of debate here. I have colleagues who feel the same
>>         way, and
>>         > other colleagues who refrain entirely from participating at
>>         the IETF
>>         > because of it.
>>
>>         I believe you.   But could you drill down a bit into (i.e.
>>         define more
>>         precisely) "hostile" and/or "alien"?   Because I suspect
>>         these words
>>         mean different things to different people.
>>
>>
>>     Almost certainly the precise meaning taken differs, but the
>>     essential meaning here is that some IETFers choose to be blunt,
>>     and this drives others away.
>
>     Could you be more precise about what it means to be “blunt”? 
>     Because I’m not sure that I know what you mean. I can think of
>     several possible alternatives:
>
>
> Would "unnecessarily aggressive" be clearer? Perhaps simply 
> "unwelcoming" is enough to cover it.

"unnecessarily aggressive" is still kind of vague, but at least I can 
agree that being unnecessarily aggressive is probably counterproductive.

"unwelcoming" is trickier.   I can certainly understand that a newcomer 
to IETF might not understand the conventions for discussion that are 
appropriate to and supportive of IETF's work.   But it has generally 
seemed to me that newcomers to IETF are treated with a bit more 
deference.   The reason a newcomer feels unwelcome might be more due to 
the newcomer's observation of more experienced persons' behavior toward 
one another, than to the way newcomers are treated.   (I will admit I 
have no supporting data for this.)

>     - let’s say (for the sake of example) that my informed technical
>     opinion is that NATs pose a grave threat to the security and
>     stability of the internet and to the internet’s ability to support
>     diverse applications. [*]  Is it too “blunt” of me to say so? Or
>     am I obligated to pretend, for the sake of “politeness” that NATs
>     are somehow more virtuous than that, because others believe that
>     they are?  In other words, do I have an obligation to be dishonest
>     for the sake of protecting others’ egos or their employers’ products?
>
>     [*] note: this is just an example; my current position on NATs is
>     more layered and nuanced than that.
>
> [...]
>
> Politeness surely isn't a bad thing that needs careful protection with 
> scare-quotes. Surely if you're not polite, you're being impolite? I 
> don't see how that is a useful strategy for either persuasion or 
> encouraging debate.

I put "politeness" in quotes because it's dangerously ambiguous.   Not 
only can it mean a lot of different things, many of those things are 
actively harmful to IETF's work.   Politeness is not about intent, it's 
about adhering to social conventions. Many social conventions are 
actually harmful - sometimes in the sense of being detrimental to 
technical work, sometimes in the sense of perpetuating harmful 
prejudices, sometimes even in the sense of supporting abuse and 
protecting abusive people.

I doubt that many IETFers would support forms of "politeness" that they 
know have such negative effects, but often people have become so 
conditioned to harmful social conventions that they don't question them 
or even realize that they exist.   Politeness also varies from one 
culture to another, and is often dependent on things like status and 
gender which, if followed, would deter us from behaving as peers.

It's for these reasons that I don't think IETF should define appropriate 
behavior in terms of "politeness".

There are alternatives that are more defensible.  As examples, it's 
always appropriate to have genuine respect for others, and being 
considerate is always an appropriate motivation (though it can backfire).

There's another problem with the notion of "politeness" that I should 
mention - it's the convention that when someone else violates one's 
notion of politeness, one has license to disregard that person, dismiss 
their input out-of-hand, and/or be critical of them personally (either 
publicly or privately).


>
> But look at what you're doing here - isn't the premise of your 
> paragraph here is that I am suggesting, in a debate over how to 
> encourage people to express opinion, that they should not express 
> their opinions?

I suspect you and I are approximately on the same page with this, but I 
don't have the sense that everyone who has been participating in this 
discussion is on the same page.

There's a figure named Cassandra from ancient Greek mythology. She had 
the gift of prophecy but it came with a curse:  The more accurate her 
prophecies, the more fervently they would be disbelieved.   I think the 
ancient Greeks were wry observers of human nature, that they 
incorporated such observations into their myths, and Cassandra's curse 
is a good example.

I've often seen a form of this curse operating in contemporary 
experience.   It looks something like this: The more that a particular 
observation or suggestion or bit of foresight challenges people's 
presumptions, the more they disbelieve it and/or resist it.   Quite 
often, they see that observation or suggestion or foresight as impolite 
or even offensive.   And the truer the observation, and the more clearly 
it is stated, the more fervently people are inclined to resist it.   
(Though I should caution, the fact that people see something as 
offensive doesn't mean it's true.)

I will assert that engineers and scientists and people who make laws and 
policies, especially, need to cultivate the discipline of looking past 
their presumptions, and to make sincere efforts to ask themselves 
whether a challenging observation or suggestion or potential bit of 
foresight might actually be true.

I will further assert that expecting or conditioning people to be 
"polite" is counterproductive to this, and that the likely principal 
effect of expecting people to be "polite" (without being more specific) 
is to encourage mediocrity in IETF, and to discourage participation of 
people with keen powers of observation or unconventional points-of-view.


>     - or am I supposed to treat every technical proposal or decision
>     as if it’s a good one, and refrain from pointing out flaws, or
>     refrain from attempting to express just how harmful a proposal or
>     decision appears to me to be?  If I think that launching the
>     spacecraft in below-freezing weather will significantly risk
>     killing the crew, or that a proposed bridge design is likely to
>     fail in a heavy wind, with potential loss of life, am I expected
>     to say that in veiled terms like “I recommend against that”?
>
>
> You should read up on Crew Resource Management, which discusses this 
> kind of behaviour (often with the same examples, so I suspect you know 
> of it) in intricate detail, but the summary is "no". Pointing out 
> serious issues, assertively, and explaining why you feel these are 
> serious, should be welcomed. Treating every technical proposal or 
> decision as if it's made in good faith, though, is good - and if 
> everyone does that, your warnings of doom should be more heeded, not less.

I hadn't thought of it in the context of this discussion, but CRM is an 
excellent example of adopting different standards for behavior in 
situations for which social conventions were actually harmful.  Thanks 
for mentioning it.

(especially since it has fair amount of research behind it, maybe we 
could actually use CRM as a starting point for a better notion of IETF 
etiquette)

>
> But most of what I'm concerned about isn't shielding the leadership 
> likely to make decisions, but encouraging newcomers to engage - 
> inevitably, as newcomers, the stakes are much much lower, and the need 
> for feedback less urgent, so you can deliver the message needed in a 
> calmer way.
>
>     - or am I supposed to avoid technical discussion of details of a
>     proposal or decision that might embarrass the person making the
>     proposal or decision?
>
> You're supposed to discuss technical details and decisions in a way 
> that avoids the embarrassment of the person making the proposal or 
> decision.

One should not deliberately try to embarrass.   But neither, I submit, 
should IETF participants feel compelled to describe problems with an 
idea or decision vaguely in order to avoid bruising the egos of the 
people who proposed them.    (Hyperbole doesn't help either, of course.)


>>
>>     And make no mistake, this *is* a choice, and I've only ever seen
>>     long-time, "senior", IETFers defend that choice publicly.
>
>     It may be that very experienced and competent engineers are more
>     likely to understand the value in it.  IETF was blessed to attract
>     a lot of extremely competent people in its early days.
>
>
> I think it's more likely that most people who have spent a significant 
> period of time on these mailing lists are the ones who have grown used 
> to it and consider it normal, or at least can be comfortable with it.

Perhaps they also understand why it's useful or even necessary sometimes.


> When I joined my first IETF list nearly 25 years ago, I was one of the 
> younger ones. I think I still am. While I appreciate the way the IETF 
> keeps me feeling young, I'm not entirely convinced this is an overall 
> positive outcome for the IETF.
>
>>     This means avoiding being blunt when we offer our opinions, as
>>     well as trying to tease out the useful opinion from the bluntness
>>     of others.
>
>     I disagree with that generality.  I don’t think it’s welcoming to
>     people to not take them seriously.  One way to not take someone
>     seriously is to humor them when they make bad suggestions and act
>     as if they were good ones. That’s not to say that I think they
>     should have to endure scathing criticism of their ideas (and
>     certainly not of themselves).  But sometimes it is necessary to be
>     clear.
>
>
> Yes, you're right. The point I was trying to make was that an opinion 
> delivered aggressively, or with sarcasm, or whatever remains an 
> opinion. We are of course free to disagree with it, but as readers, we 
> should try to understand the actual opinion and disagree with that. I 
> didn't intend to suggest that all opinions are equally correct, but it 
> is often useful to understand others' viewpoints and why they hold 
> them. After all, it allows one to construct more persuasive arguments 
> against them. It is that utility that I meant by "useful".

Certainly agree with that.

>>
>>     To put it another way, we should be conservative in what we send,
>>     and liberal in what we accept from others. Just like that
>>     principle, it should not bar us from noting and rejecting bad
>>     input - but we shouldn't let that be the cause of a fatal error.
>
>     I had started to cite that principle myself but found too many
>     cases for which it produced bad results.  At least we agree that
>     we should be able to reject “bad input”.
>
>
> Sorry, this was not as clear as I'd have liked.
>
> I actually intended the "bad input" to mean stripping out the 
> aggressiveness or whatever and rejecting that, so as to concentrate on 
> the substance. But again that's unclear, since I don't mean to suggest 
> we cannot reject poor arguments, invalid statements, and so on. We not 
> only can, we should.

Agree with that also.


>>     b) Rather than call out that choice of words but answer the
>>     essential points, the organisation chose instead to chastise the
>>     sender publicly and leave the points unanswered - unwittingly
>>     shutting down the discussion.
>
>     Different people will have different opinions, and I don’t want to
>     add fuel to that particular fire, but my analysis of that
>     conversation was somewhat different.
>
>
> Sorry.  "Shutting down" was an unwarranted exaggeration. But I think 
> it did indeed reduce constructive participation.

I certainly agree that constructive participation was thwarted, though 
you and I might disagree as to why.  (or not)

Keith