On kernel mailing list behavior [LWN.net]

On kernel mailing list behavior

Posted Jul 17, 2013 23:36 UTC (Wed) by bojan (subscriber, #14302)
Parent article: On kernel mailing list behavior

First off, the e-mail Sarah reacted to wasn't all that bad, but I understand where she's coming from. Generally speaking, there is quite a bit of abuse on open source mailing lists and other discussion forums (and I have been guilty of engaging in that from time to time - apologies to anyone that was on the receiving end of that stupidity).

Second, there is no doubt that many of high level Linux contributors practice the cult of Linus. It doesn't matter what he says or does, they will always try to defend it - no matter how silly. Of course, Linus is an absolutely brilliant guy. Nevertheless...

The abusive behaviour that happens in all those forums would normally get people fired in any half decent western corporation. And for a good reason. There is _always_ a way to say want you want without abuse.

I find Linus' comments about non-belief in politeness really hilarious. Being polite is not a religion - you don't have to believe in it. We do it in civilised societies because it works. Not having manners is not any particular culture - it's a lack of one. We even have laws that enforce "manners" (e.g. noise). These are purely practical - nobody's freedom is more important. So, we "turn down" a notch what everyone is allowed to do, so that we can all live in peace.

Ingo seems to be confusing conflict (actually, standing up for you own beliefs, opinions, solutions etc.) and abuse. Totally unrelated. In fact, offending people as a way of getting what you want can (and many times does) result in getting the opposite.

Most of the abuse that happens on the Internet is due to the fact that the person you are about to abuse is not standing in front of you. Somehow, many people turn into high school kids when they are about to post online. It is pathetic. Yours truly included at times.

We should just stop with that nonsense.

PS. I wonder if Linus is telling his kids to behave in life (e.g. towards the teachers) like he sometimes does on the mailing list. I would think not, but then again - who knows.

to post comments

On kernel mailing list behavior

Posted Jul 18, 2013 0:03 UTC (Thu) by hummassa (guest, #307) [Link] (13 responses)

> Being polite is not a religion - you don't have to believe in it. We do it in civilised societies because it works.

Do you have any idea how much does your "it works" contradict "is not a religion - you don't have to believe in it"?

The point -- oh, yeah, subscribe me to the cult of Linus (NOT) -- is that, at least for some people, it does NOT work. As Linus himself explained, being polite, at least for him, undermines the way his words convey his emotions relative to some technical topic, and he thinks that said emotions are important in the decision-making process for his project.

On kernel mailing list behavior

Posted Jul 18, 2013 0:16 UTC (Thu) by bojan (subscriber, #14302) [Link] (12 responses)

> Do you have any idea how much does your "it works" contradict "is not a religion - you don't have to believe in it"?

I do. There is no contradiction. It is the same as saying that one does not believe in science - it is not a matter of belief. The proof is in the pudding, as they say.

> As Linus himself explained, being polite, at least for him, undermines the way his words

That is the kind of excuse that many abusers make. For instance, we heard many times: "I have to hit my wife - she provokes me". And some such crap.

PS. Yes, I used a physical abuse example.

On kernel mailing list behavior

Posted Jul 18, 2013 0:21 UTC (Thu) by bojan (subscriber, #14302) [Link] (1 responses)

> For instance, we heard many times: "I have to hit my wife - she provokes me".

OOPS - part of it fell out. Should be:

"I have to hit my wife - she provokes me - if I don't hit her, she won't get the message not to do it next time."

On kernel mailing list behavior

Posted Jul 18, 2013 11:14 UTC (Thu) by simosx (guest, #24338) [Link]

Or "The child is screaming. A slap on the face is what works. Debating with the child does not work."

I think these are the analogies that need to be put forward to show how moronic are the attitudes of Linus, Greg and others.

On kernel mailing list behavior

Posted Jul 18, 2013 2:30 UTC (Thu) by hummassa (guest, #307) [Link] (9 responses)

> It is the same as saying that one does not believe in science - it is not a matter of belief. The proof is in the pudding, as they say.

Nope. I'm sorry, but your "it works" affirmative is not akin to saying "science" but "the effectiveness and safety in the use of anlodipin as a medication for arterial hypertension". Yes, some social constructs are based on politeness. Some professional social constructs are based on politeness. And they work. But you are making the statement that *only* they work, and that they work *because* of the politeness.

Guess what? Being impolite also works. There are lots of social constructs that are based in being impolite as the de facto etiquette. And they don't fall apart. Do you know why? Because it's even more efficient in avoiding social mistakes than being polite all the time. So, if my best friend says to me "you are an imbecile" and I reply "at least I am not as ugly as you", when one of us is *really* upset with the other and calls him an idiot, it may even pass as the "normal banter". And gives time for both to calm down before someone takes offense. But if we refer to each other as "best buddy" and "smart man", the stress situation could escalate more easily.

I suspect the politeness-based social constructs work in spite of the politeness, but I can't prove *that*.

On kernel mailing list behavior

Posted Jul 18, 2013 3:05 UTC (Thu) by bojan (subscriber, #14302) [Link] (8 responses)

First, your example of an exchange between best friends is hardly an example of civility in the workplace or generally in society. The underlying relationship between best friends allows for forms of communication that are simply not possible or acceptable in the workplace, around your neighbours or other citizens you may bump into on the street. In other words, when you say to your best friend "you're an idiot", it doesn't have quite the same meaning that it would if you said it to a person that is building a piece of software with you, works with you or just lives next door. In other words - that exchange is not abusive - because both parties accept (through the nature of the friendship) this kind of communication. So, if anything, this just proves my point.

Second, the human history directly contradicts your proposition that being abusive (impolite) works. It doesn't and as a race humans are rejecting it more every day. You don't have to go back much to find proof. Be it racial, sexual, religious, child or any other kind of abuse is not accepted any more in modern, civilised societies.

So, absolutely, "politeness" (i.e. being considerate and not abusive) is absolutely the underpinning of every civilised society.

On kernel mailing list behavior

Posted Jul 18, 2013 14:31 UTC (Thu) by nevets (subscriber, #11875) [Link] (5 responses)

You start off being polite, and if that doesn't work than you get nasty. Otherwise you just waste everyone's time and people end up even *more* pissed off.

That is basically how Linus does it.

On kernel mailing list behavior

Posted Jul 18, 2013 23:55 UTC (Thu) by bojan (subscriber, #14302) [Link] (4 responses)

This is exactly how it should _not_ work. I still remember telling the general manager (and owner) of the company I worked for (in front of other people that were present in the meeting) to stop yelling at me, because he is not my father. He did stop and never tried that again.

Abuse is never OK. There is a difference between aggressive and assertive. Google it if you don't believe me.

On kernel mailing list behavior

Posted Jul 19, 2013 1:25 UTC (Fri) by nevets (subscriber, #11875) [Link] (2 responses)

IF I WRITE MY REPLY IN ALL CAPS, AM I ABUSING YOU?

Yes? No?

There's a huge difference between yelling face to face and on email. The problem with text is that it's much harder to tell the degree the person is upset with you, or disagrees with you.

When Linus disagrees with me, I take his level of emotion as a gauge of how much he disagrees with me. If he says "you shouldn't do it that way", it sounds like he doesn't like it, but he doesn't really care that much about it, and may still let me do it if I really want to. If he says "Only a f*cking idiot would do it that way", it's clear that I screwed up and need to change the way it was done. It's about communication, that's all.

On kernel mailing list behavior

Posted Jul 19, 2013 2:01 UTC (Fri) by bojan (subscriber, #14302) [Link] (1 responses)

> IF I WRITE MY REPLY IN ALL CAPS, AM I ABUSING YOU?

> Yes? No?

All I can tell is that you are shouting. Now, depending you what you are saying, it may be yes or it may be know. Compare:

WE WON THE GAME! YAY!

with

YOU STUPID MORON!

Pretty easy to figure out, really.

> If he says "Only a f*cking idiot would do it that way", it's clear that I screwed up and need to change the way it was done. It's about communication, that's all.

Well, maybe you are are a buddy of Linus and you guys have exchanges like this in private. If that is the case and you are both cool with that, that's fine.

However, if you are not buddies, this kind of exchange may be considered abuse by you. Also, there is the problem of a the list being _public_. Other people may take offence to that kind of exchange, as it may sound intimidating to them. For instance, Sarah did - which then makes it inappropriate.

On kernel mailing list behavior

Posted Jul 19, 2013 2:02 UTC (Fri) by bojan (subscriber, #14302) [Link]

> may be know.

Er, sorry. Should be "may be no", of course.

On kernel mailing list behavior

Posted Jul 19, 2013 12:23 UTC (Fri) by shmget (guest, #58347) [Link]

"This is exactly how it should _not_ work."

Nonetheless, it does.

On kernel mailing list behavior

Posted Jul 19, 2013 3:10 UTC (Fri) by Tara_Li (guest, #26706) [Link] (1 responses)

And it's not like you can't be quite abusive being extremely polite. The British Upper Crust, and the American Gentile South proved that a long time ago. Why, one might argue that bringing Gentility and Civility to the unwashed masses is the burden of some specific group.

On kernel mailing list behavior

Posted Jul 19, 2013 15:52 UTC (Fri) by smurf (subscriber, #17840) [Link]

Sure you can, but if the recipient is not part of the same society your abuse will flow through their brain without affecting any of their neurons the way the sender intended.

On kernel mailing list behavior

Posted Jul 18, 2013 6:27 UTC (Thu) by b7j0c (guest, #27559) [Link] (5 responses)

while i agree with everything else you say...your relation of noise laws being somehow extended to common civility (speech, general offensiveness etc) carries no water whatsoever.

On kernel mailing list behavior

Posted Jul 19, 2013 0:10 UTC (Fri) by bojan (subscriber, #14302) [Link] (4 responses)

Really curious - why do you think so?

Here is a direct quote for NSW POEO Act (which is the state law where I live):

‘Offensive noise’ is defined in the Protection of the Environment Operations Act 1997 as being noise:
(a) that, by reason of its level, nature, character or quality, or the time at which it is made or any other circumstances:
(i) is harmful to (or is likely to be harmful to) a person who is outside the premises from which it is emitted, or
(ii) interferes unreasonably with (or is likely to interfere unreasonably with) the comfort or repose of a person who is outside the premises from which it is emitted, or
(b) that is of a level, nature, character or quality prescribed by the regulations or that is made at a time, or in any other circumstances prescribed by the regulations.

Obviously, there is the harm/regulations bit, but the rest is really about being considerate.

On kernel mailing list behavior

Posted Jul 19, 2013 18:19 UTC (Fri) by b7j0c (guest, #27559) [Link] (3 responses)

we do not need to consider by metaphor, extension, or suggestion as to how something like a noise law impacts speech. specific limits of speech and free expression have been tested repeatedly in detail in the supreme court, for exactly this reason - so someone can't come and tell me that a noise ordinance somehow implies a speech limitation.

i can't blare my music at 1am but i can most certainly (for example):

(1) create a private club that bans women

(2) start a camping troop for children that bans gays

(3) refuse to offer even trivial assistance to someone in life-threatening danger near me

all of these examples have been validated by the supreme court, and oddly enough no one suggested that noise ordinances were relevant

On kernel mailing list behavior

Posted Jul 21, 2013 12:45 UTC (Sun) by renox (guest, #23785) [Link] (2 responses)

I think that 1 and 2 are also legal in France and I see no real reason why this should be illegal (you are free to make a club with whatever stupid rule you want and I am free not going into your stupid club), but 3 isn't legal here, we have "non assistance a personne en danger": we can be charged for not helping someone in danger, this is rarely enforced but the law is there.

discriminations

Posted Jul 21, 2013 14:18 UTC (Sun) by Tobu (subscriber, #24111) [Link] (1 responses)

(2) isn't legal in France either, and gets taken seriously. You'll probably be able to find examples of (1); but the “private club” defence wouldn't hold for commercial operations and is just there to discourage complaints. As far as most scouting activities splitting children by gender, local organisations do accept both genders and give them equal treatment. This is unlike the US situation where there are gender-specific institutions. </sidebar>

discriminations

Posted Jul 23, 2013 14:52 UTC (Tue) by shmget (guest, #58347) [Link]

"you'll probably be able to find examples of (1); but the “private club” defence wouldn't hold for commercial operations and is just there to discourage complaints"

Really ? there is no women-only gyms ? or do you mean they are 'illegal' ?

" As far as most scouting activities splitting children by gender, local organisations do accept both genders and give them equal treatment."

Separate but equals... hum.. where did I heard that before ... ? ah! yes.. "Plessy v. Ferguson"

On kernel mailing list behavior

Posted Jul 18, 2013 9:07 UTC (Thu) by niner (guest, #26151) [Link] (5 responses)

If with "any half decent western corporation" you mean "any half decent US corporation" then please say so. Because believe it or not, there are parts of the ("western") world that view the typical US politeness as being over polite and dishonest.

This would have been less imposing on the rest of the world:
"Being polite is not a religion - you don't have to believe in it. We do it in the US because it works for us. Not having our manners is not any particular culture - it's just not the US one."

On kernel mailing list behavior

Posted Jul 18, 2013 10:29 UTC (Thu) by Cyberax (✭ supporter ✭, #52523) [Link]

Yup. For example, in Russian there's a phrase "American smile" - it means a smile without any actual feelings behind it.

On kernel mailing list behavior

Posted Jul 18, 2013 23:52 UTC (Thu) by bojan (subscriber, #14302) [Link] (3 responses)

I have worked for Australian, US and Indian companies. All of them had explicit policies that ensured the communication between employees is of a very high standard. No abuse was permitted anywhere.

On kernel mailing list behavior

Posted Jul 19, 2013 1:43 UTC (Fri) by bloopletech (guest, #71203) [Link] (2 responses)

I have worked for several companies, in Australia and the US, and these policies are generally disregarded by the majority of people.

I'm not saying that's good or bad, but the presence of policies does not imply that (a) the policies are good or (b) that people respect the policies.

On kernel mailing list behavior

Posted Jul 19, 2013 2:10 UTC (Fri) by bojan (subscriber, #14302) [Link] (1 responses)

Well, then these companies you worked for are obviously not "half decent", are they? Abuse should never be condoned.

For instance, examples of hitting children and women were mentioned in the comments. So, I guess it would be OK for me to then smack women and younger people in the office around to get what I want, right? Absolutely not.

Nobody ever claimed that no work needs to be done to reduce the amount of abuse in general society and workplace. In fact, this whole discussion started because there is more work to be done.

Policies must be voluntarily obeyed to be effective

Posted Jul 19, 2013 7:51 UTC (Fri) by tialaramex (subscriber, #21167) [Link]

When a policy doesn't work you have to fix the policy. Some of us care about the real world, not your internal emotional state except in so far as that has consequences in the real world.

The pretence that "We have a policy forbidding X, so X doesn't happen" is a terrible sort of delusion because those too far from the action to see X continue anyway might believe you and act as though it's true.

For example, the current _policy_ here says that track workers will all interrupt what they're doing, stand up and walk off the tracks into a place of safety before any train passes. No doubt in the imagination of whoever wrote this rule there was only one "road" and it was mere "common sense" that track workers would want to get out of the way. But in reality a lot of work happens at complex multi-road sites. The workers can see (or think they can see, or even just guess from experience) that a particular train won't pass near to them, and they wave it through but remain on the railway and thus in danger. Approaching trains are loud and (at night) heavily lit, masking any danger from other trains nearby so that even when they're correct about the train they can see, they may be wrong about whether they're actually safe.

People who don't understand why policies don't work usually insist that workers who disobey the policy should be disciplined. That ought to work, right? But it won't, it just creates the impression that management are capricious, arbitrarily punishing some workers whose behaviour was fine, for a merely technical violation of a rule that makes no sense. Even showing the workers why their current practices are dangerous through the video reconstruction of accidents (including fatal accidents) turns out not to make a dent. Humans are largely oblivious to their mortality and will take small temporary gains for a significant risk of death without qualms. The only way to fix the problem is to come up with a policy that the workers are _willing_ to obey but which is also safe.

On kernel mailing list behavior

Posted Jul 21, 2013 0:40 UTC (Sun) by krakensden (guest, #72039) [Link] (3 responses)

> Ingo seems to be confusing conflict (actually, standing up for you own beliefs, opinions, solutions etc.) and abuse. Totally unrelated. In fact, offending people as a way of getting what you want can (and many times does) result in getting the opposite.

It's not a useful distinction. All conflict is abuse when it's somebody saying no to you. There's no endgame where people won't take it personally.

On kernel mailing list behavior

Posted Jul 21, 2013 3:57 UTC (Sun) by raven667 (subscriber, #5198) [Link] (2 responses)

> All conflict is abuse when it's somebody saying no to you.

No. Do you feel that way? If someone contradicts you in any way do you feel personally slighted and abused? I don't feel that way, I'm happy to take constructive criticism, unless the other party goes out of their way to be personal and insulting. I would think it the same in most professional environments, do you have a different experience?

On kernel mailing list behavior

Posted Jul 21, 2013 15:36 UTC (Sun) by khim (subscriber, #9252) [Link] (1 responses)

If someone contradicts you in any way do you feel personally slighted and abused?

Ah, so now my POV suddenly matters? But when supposed "victims" Sarah raises on her flag come and say that they don't feel insulted and abused it somehow does not matter?

And yes, I've seen people who've felt slighted and insulted when people pointed out that they may be wrong (if you want an example the look for the yet-another-tempest-in-a-teapot right here).

I don't feel that way, I'm happy to take constructive criticism, unless the other party goes out of their way to be personal and insulting.

And some people are not happy to take "constructive criticism" at all. There are cultures where you are not supposed to criticize anything publicly - at least not directly. Should we all stop using word "no" on mailing lists to not offend such people?

I would think it the same in most professional environments, do you have a different experience?

First we need to define what "professional environments" even means. Because there different viewpoints. For example there are this video which explains that professionals are guys who never ship shit, who are always ready do deliver, who don't slow down when project ages, who can easily adapt to changes in requirement, etc. Like all the real professionals: good carpenters, good janitors, good cooks. And not like all these amateurs who develop spaceships and cars, mobile phones and large bridges, skyscrapers and yes, software, too.

By this measure I never worked in real professional environment thus I wouldn't know how such work is done. But I've participated in different kinds of projects and I'm ready to admit: as you go from amateurs to professionals (from more challenging work to less demanding and stable one) the more environment represents Sarah's ideal.

On kernel mailing list behavior

Posted Jul 22, 2013 14:46 UTC (Mon) by bfields (subscriber, #19510) [Link]

There is a lot of middle ground between "people should shout as loud as they like and nobody should complain about it", and "nobody should say anything that might offend anyone".

There are cultures where you are not supposed to criticize anything publicly - at least not directly. Should we all stop using word "no" on mailing lists to not offend such people?

I can say "no" in this case while also believing Sarah's request is reasonable. I don't see any contradiction.