Steering Council nomination: Raymond Hettinger (2021 term)

Core Development Steering Council Nominations
21

So that others won’t be thrown-off, can you please add to note the part where you said I was “being ridiculous” :slight_smile:

And since we’re on the topic of whether the actions are being applied to “egregious and on-going conduct” as specified in PEP 13, consider that an SC member who is also a CoC WG member has been actively soliciting new reports of micro-aggressions.

IMO that is beyond the scope authorized by PEP 13 and not what we intended the CoC WG to do. If elected, I aim to dial that back to a reasonable level.

22

Raymond, my questions may have been lost in the ensuing discussion, and you’ve touched on some of them in other responses, but I would appreciated your input on the rest if possible. Steering Council nomination: Raymond Hettinger (2021 term) Reiterating that these are important and difficult issues, so I think as much clarity as possible is important so others understand what they’re voting for.

1 Like
23

But… it wasn’t a home invasion. It was an email pointing out that they’d – in Raymond’s words – “said something highly inappropriate on the internet and no one liked it”, and asking them to make amends. Objectively, this is completely different than a home invasion. This comparion seems wildly hyperbolic, especially when in the other thread you were talking about how old-school Usenet flamewars full of actual vicious personal attacks were sub-optimal but like, something you could cope with, no big deal.

Being held accountable can be terrifying. Again, I’ve been there myself, and I totally get it! It’s an honest emotional reaction. But that doesn’t mean that it’s accurate, or that we can skip taking responsibility for our actions just because it’s scary.

I don’t think that’s what PEP 13 says. The bit you’re quoting comes from the section on ejecting core team members:

Ejecting core team members

In exceptional circumstances, it may be necessary to remove someone from the core team against their will. (For example: egregious and ongoing code of conduct violations.) This can be accomplished by a steering council vote, […]

But first, this is only talking about the specific extreme situation where someone needs to be ejected, not all possible responses to all possible conduct issues. And second, it’s saying that “egregious and ongoing CoC violations” are one example of a case the steering council might need to respond to, and there might be others, while you seem to be reading it the opposite way, like it’s saying that’s the only case the steering council should respond to.

OTOH, PEP 13 also says:

Mandate

The steering council shall work to:

  • Maintain the quality and stability of the Python language and CPython interpreter,
  • Make contributing as accessible, inclusive, and sustainable as possible,
  • […]

Tracking and potentially responding to microaggressions seems entirely within the scope of the part I bolded.

In general, PEP 13 leaves a lot of room for the SC to interpret things, on the theory that we’ll elect thoughtful, principled folks who we can trust to make good decisions. So if you think that the best way to accomplish the SC’s mandate is to take a hands-off approach to CoC issues, and the voters agree and vote for you, then I don’t think anything in PEP 13 would stop that. But I don’t think you can argue that the current SC has violated PEP 13’s text or spirit, either.

3 Likes
24

Didn’t say it was. Did say “deep fear” is a far more natural reaction to a home invasion than to your theory that the mystery person suddenly realized they had hurt someone. Fantastic as it is, “home invasion” is more plausible than that :wink:.

Didn’t it bar them from participating in community life for 3 months? “Asking” is a request; banning is not.

Yes, I would have been better off likening it to the police breaking down your door at 3am with a no-knock warrant and shooting your dog :wink:.

“Deep fear”, trauma, and - as Raymond later quoted - saying the experience was “devastating” are obviously natural reactions to a surprise attack. You’re welcome to continue selling the idea those were actually reactions to a sudden realization of guilt, but that’s just not plausible to me.

Power, Nathaniel. On Usenet, people in flame wars usually had no relationship in real life, and nobody had the power to do anything to the other beyond the power of hateful words to hurt (which is certainly real, and which I certainly support having a CoC to oppose). “You’re barred from posting to comp.lang.python for 3 months” wasn’t even a theoretical possibility. Although “you’re in my killfile for life” was.

In the case at hand, one side had real power to directly affect the other’s real life, and the other had none. Fear, trauma, and feelings of devastation are natural when being on the losing end of an unexpected exercise of real power.

[Tim edited this to remove possibly identifying speculation - it’s not helpful to anyone]

If so, it seems to me to be a case of “viewpoint discrimination”, that the action was in part - and possibly mostly - due to what they believed, in addition to how it was expressed. The US government generally can’t do that, but a private party (like the PSF) can. So I don’t claim we “can’t” do that. I nevertheless oppose viewpoint discrimination.

Which is just me. I don’t know what Raymond thinks about that.

1 Like
25

Nathaniel, thanks for chiming in. It is great to hear from the author of PEP 13 :slight_smile:

Like you, I completely support both PEP 13 and the Code of Conduct.

We do disagree on implementation. It seems that the emphasis has shifted away from being welcoming and more towards kicking people out. The frequency of events is higher than expected. Instead of protecting the powerless, the reports so far have mostly come from current or aspiring SC members. And soliciting microaggression reports seems like a recipe for exclusion – we should be more about tolerance of each other’s foibles, temperaments, and communication styles. Already, I’ve had a report from a senior developer that they are contributing less because of worries about CoC reports.

Frankly, I’ve been concerned for a long time. The first CoC action I observed was where a sitting SC member and CoC WG member ejected a participant from our forums for being “distracting”. The person habitually posted lists of bullet points rather than prose. They were “warned” about this, but it was an intrinsic trait they couldn’t change. Research into the person’s internet presence revealed that this was their only communication style. While this person was annoying, the original intent of the diversity statement was to protect people who are neurodiverse. Instead, it was used as weapon against them by a powerful person who was simply irritated. This was pointed out on the list, but no retraction was forthcoming. The participant has never been heard from again.

When you published an essay taking down the creator of the Requests module, I didn’t think it lived up to the spirit of the diversity statement. Historically, we never treated each other this way. Deserved or not, it must have been devastating for the target of that post. You were fully within your rights to make the post (protected political speech), but I became concerned when you told me that elected Python officials had green-lighted the post.

I have huge respect for you and it is okay that we disagree about whether creating a welcoming environment should revolve around non-permissive and aggressive CoC enforcement or whether we want to aspire towards de-escalation and mutable tolerance, saving the big hammer only for the egregious and on-going cases.

7 Likes
26

Ah, I didn’t know it was about that commit message thread. I followed it for a while and didn’t see anything obscene.

It’s possible the posts were removed too quickly for me to read them, however I would be extremely surprised if they contained actual hateful language.

1 Like
27

Doing a Google search on their posting name does show posts on python.org mailing lists as recently as this year. For example, here. I think, from the posting style, we can be pretty sure it’s the same person.

For the record, I was shocked when they were first banned. They never tried to hurt anyone in the slightest. At worst the endless lists were mildly annoying, and at best they sometimes contained an interesting link that was new to me :smiley:.

28

Note that I don’t know either. I’m just pasting clues together and making a semi-informed guess.

[Tim edited this to remove unhelpful speculation]

I didn’t see them post any “racist and demeaning theories”, but it’s possible I missed such - and, of course, that the actual person is someone else entirely.

In any case, I doubt more speculation about this will help anyone with anything, so I’ll stop with this :smile:.

29

Thank you for expressing yourself with empathy.

To be completely clear, I fully support PEP 13, the CoC, and the diversity statement. I also think the way they are administered matters. Official actions should not be frequent. They should be used to protect the powerless rather than powerful. Following our own rules, they must not be punitive or excessive. Fairness and community health are paramount.

7 Likes
30

I adore the phrase “mutable tolerance”! Please don’t repair it. Perhaps it was a Freudian slip :wink:.

4 Likes
31

I’m one - and perhaps the one you’re referring to. But it’s not just disinclination to get bogged down in defending myself against CoC accusations, but that the CoC also gets dragged into the decreasing number of threads I find myself involved in. I don’t want to see other people hassling with that either.

A while back, someone new popped up on python-ideas suggesting that the first line of the so-called “Zen of Python” (PEP 20; import this) should be changed, for “perpetuating beauty bias and containing lookist slur” (it contains the words “beautiful” and “ugly”, arrogantly insisting that beautiful is “better”, but in a context having nothing whatsoever to do with human appearance).

I stayed out of it, until someone made the reasonable observation that, since I wrote it, my opinion of the suggestion should carry some weight. So I popped in a few times overall, briefly.

Of course the thread went on & on without me. At least two CoC bans stemmed from it. They were both against people with names that “sounded Dutch”, one of which posted from an “.nl” domain. To my eyes, yes, their expression was out of bounds by majority American standards. But those aren’t Dutch standards - Nederlanders can be, famously, blunt and direct and “insensitive”.

That’s a tough call. I probably would have let it slide with an off-list warning about cultural differences, but am not really upset by the call that was made. But somehow, and irrationally, I feel bad that something I wrote many years ago, for fun & entertainment & food for thought, played a role in getting those posters shut down. Better to just stay quiet :frowning_face:.

Speaking of Nederlanders, Guido popped into that thread once briefly, to lay out two reasons for suspecting the original poster might be a Russian troll. How welcoming is that?! :wink:

That certainly could have been sold as a CoC violation too. I’m glad it wasn’t, but while I, in general, just can’t guess what will & won’t lead to CoC enforcement, Guido is off limits, so I didn’t expect he’d “get in trouble”. He’s as safe as a rant, however intemperate, from a Huffington Post editorial :wink:.

Which is something I’d like to see discussed more. Preserving the community won’t really do much good in the end if its core CPython deliverable goes off the rails.

3 Likes
32

That happens on the issue tracker as well. When an easily broken test was described as “fragile”, a CoC accusation ensued. When a user’s code didn’t optimize due to a “sanity check”, there was a CoC accusation that the phrase in the code comment implied that the actual user was not sane. As you can imagine, that brought an other productive thread to an end.

3 Likes
33

Well, there’s a problem: as I said before, I can’t guess what will or won’t be considered to be a CoC violation by the CoC WG. So a mere threat of invoking the CoC by a participant can shut down an otherwise productive group discussion. Few people want to spend their limited volunteer time on that, or risk even being banned.

The only thing our enforcement page says will be dismissed:

Reports that are not made in good faith (such as “reverse sexism” or “reverse racism”) may receive no response.

The idea that those two specific examples are “bad faith” relies on arguments that few in the general (as opposed to activist and/or educated young) community are likely to be aware of. The relevant meanings of “racism” and “sexism” don’t match most dictionary definitions, so it’s common as mud in wildly contentious mailing list threads I’ve seen for people to be using these words with quite different meanings in mind, but neither side having the slightest idea that the other side isn’t talking about the same things. For example, absolutely nobody, not even the most fevered academic, claims that members of some race are immune from being prejudiced. That’s a human thing. But only one side believes exhibiting prejudice is the same as being racist. The other side believes that, at least in the US, only whites can - by definition - be racist (it has to do with power structures much more than with “prejudice”).

The CoC page could/should be clearer about this. Not all community members are activists, or even young. But I would also add that applying prejudiced generalizations to another is also a potential CoC violation, regardless of the poster’s race, despite that it’s not “reverse racism”. Treating others with respect has nothing to do with demographic characteristics of the parties involved. If the CoC isn’t about treating all others with respect, I’d rather be rid of it.

So what about describing code as being “fragile” (I recall a case too where calling an algorithm “robust” was characterized as “ableist”)? Is that kind of accusation made in “good faith”? Many “I’m invoking the CoC!” threats center on claiming ordinary English words are offensive, even when they have a long established, universal, technical meaning, connected to human characteristics only if you work to contrive such.

I’d leave “good faith” out of this one, and would add to the CoC page a list of such words for which complaints will not be entertained. It can grow over time :wink:.

Should be needless to say, I’m speaking for myself here. If someone doesn’t like it, don’t blame Raymond.

1 Like
34

Sorry to be the bearer of bad news Tim, but maintaining such a list is untenable.

Senate intelligence reports on foreign interference have shown that social media is being used to rapidly spread such ideas.

The ideas mostly come from postmodern critical theory, which aligns well with the usual tactics of active measures:

In the view of the Russians, words, and the exact meanings they convey, matter more than in Western countries, where the importance of this notion still is taken lightly or even dismissed, whence the success of the Russians in inhibiting the Western society with the concept of “political correctness,” essentially based on new meanings given to words and on the power that can be derived from those alterations. To put the latter explanation otherwise, how to cripple a nation by altering the meanings of its own language and by closely associating violence to as many of its words as possible, in order to ʻpoisonʼ them. The process is invisible because it does not consist in creating new words carrying in themselves influence, but in altering instead the meaning of words that exist already […]

I don’t have a solution, but I’m fairly certain the list would grow too fast to be useful.

2 Likes
35

Doubt it needs to be - any start would be a useful signal. The current page contains only two (and even then very much related) examples of what it considers to be frivolous (well, worse, “bad faith”, in opposition to the “good faith” it’s giving them as examples of being counter to).

I don’t view inherently political positions as being of “good” or “bad” faith. One side only appears “bad” to the other side :wink:.

Without moral judgment, pointing out the long-term neutrality of words commonly used in technical contexts is pragmatic, avoiding needless contention. For example,

Specific words widely used in technical contexts with no history of intent to refer to human characteristics or relationships should be viewed in those contexts. Objections to them can be divisive to no overall good end, and may be dismissed. For example, saying that code is “fragile” is not inherently ableist, and a code comment saying “sanity check” is in no way intended to be commentary on imposed societal norms for human behavior. Such phrases have technical meanings referring to characteristics and behaviors of computer algorithms.

I do not want to accuse people who do object to such phrases of acting in “bad faith”. I do want to say that not all objections are worth the costs of addressing in the ways an objector wants. “Master/slave” may have been - those words have a long history of being highly charged. Objecting to “fragile”, though, is pulling offense out of a vacuum.

5 Likes
36

This is one dimension that’s very frequently forgotten in highly-charged issues; often times, we (as a species) have a tendency to consider the “bad” side the one that happens to go against our own personal biases. So when addressing these types of problems, it’s critical to take a step back and consider the perspective from both sides of the report.

To me at least, “bad faith” constitutes things like the usage of reporting to intentionally knock someone down a peg, pursuing a personal agenda, etc. rather than trying to address a real problem or improve the community (intention matters). That’s almost completely independent of the topic, even if some might be reoccurring themes, which is what I think the current CoC enforcement guide was likely implying (despite the scope of the examples being a bit limited).

1 Like
37

That’s what “bad faith” means to me. I find that sentence in the enforcement page to be jarring, though, because the examples it gave are inherently political judgments. Someone claiming, e.g., “reverse racism”, may be thoroughly sincere, and feel genuinely aggrieved, but not using the phrase with the same meaning the CoC page author had in mind. That meaning is a relatively new construction of what “racism” means, and is simply unknown to many (especially of my advanced age), while others may sincerely disagree with the strains of sociopolitical analysis it arose from.

I personally find a lot of merit in it, but it wouldn’t occur to me to accuse someone who doesn’t of acting in bad faith - unless there was more than just that to justify it (e.g., they’re trying to game the system for personal gain or revenge, just trolling the CoC WG, etc).

2 Likes
38

So we all agree that whoever makes up the next SC, discussions should be conducted in Russian?

2 Likes
39

Haha no, I was just remarking on one of the documented causes of the trend.

1 Like