On kernel mailing list behavior [LWN.net]

On kernel mailing list behavior

Posted Jul 17, 2013 19:53 UTC (Wed) by rahvin (guest, #16953)
In reply to: On kernel mailing list behavior by marduk
Parent article: On kernel mailing list behavior

I believe that's the most important item glossed over in this whole discussion. Linus sets the tone of LKML and the whole kernel development process. Without his buy in, little will change. As Rusty said, you can be firm without being rude.

On kernel mailing list behavior

Posted Jul 18, 2013 14:08 UTC (Thu) by dgm (subscriber, #49227) [Link] (10 responses)

Linus most important trait is being pragmatic, so let's analyse this from a pragmatical point of view:

Being occasionally rude:
- is easy and natural
- relieves stress
- keeps crappy code out of the kernel (sometimes)
- keeps crappy people out of LKML (sometimes)
- _may_ be keeping good people out of LKLM (unverifiable)

Being always polite:
- is unnatural and sometimes hard
- increases stress
- _may_ allow some shy people to contribute

And a few facts:

- The LKML is one of the busiest in the world.
- Linux is the software project with the greatest number of contributors in the world.
- The number of contributors to Linux has been steadily growing over the years.
- The LKLM has a reputation of abusive language.
- There is really no measure as to how much abusive the LKML is.
- "Abusive" is a subjective term.
- It's not clear that the LKML reputation is actually a bad thing.

Feel free to add your own opinions.

On kernel mailing list behavior

Posted Jul 18, 2013 21:43 UTC (Thu) by diederich (subscriber, #26007) [Link] (2 responses)

'Being always polite'? I think a lot of benefit could come from Linus in specific, and LKML in general, agreeing to the the simple paradigm that, all other things being equal, being nice is better than being nasty.

'Being always polite' is a worse state, by far, than where LKML is right now.

It goes a long ways when people say, "I will make some effort to make my tone more pleasant."

On kernel mailing list behavior

Posted Jul 18, 2013 21:59 UTC (Thu) by dgm (subscriber, #49227) [Link] (1 responses)

> all other things being equal, being nice is better than being nasty.

I think that's exactly how the people on the LKML behave today, unless you can bring proof that people are nasty all (or most of) the time.

The LKML is archived. Just start at any random point and count messages until you find some cursing or nasty behavior. Repeat a few times. You can call it Guerrero's metric if you want. Is the average is less than 2?

On kernel mailing list behavior

Posted Jul 19, 2013 3:35 UTC (Fri) by diederich (subscriber, #26007) [Link]

I don't think LKML is rife with nastiness. But there's definitely some, and what I'm opposed to is the idea that it's ok for some people to be unpleasant, even rarely, if they're very good at what they do.

Linus is a fantastic leader in many ways. By his own admission, he's a bit of a technical hot-head, more so in the past than now. In my mind, it's such a tiny delta to say, "I am a hot-head, and I'll make some effort to tone things down." Rather than justifying (very rare) unpleasant behaviour.

On kernel mailing list behavior

Posted Jul 18, 2013 23:28 UTC (Thu) by brianomahoney (guest, #6206) [Link] (5 responses)

Concur, see above, this silly woman is either motivated by power and status or just cant conceive that not all of us agree with her world view, if she has something USEFUL to say, say it.

On kernel mailing list behavior

Posted Jul 19, 2013 13:36 UTC (Fri) by emunson (subscriber, #44357) [Link] (3 responses)

This is a perfect example of lots of negativity with no addition to the discussion. We would have been better off without the addition of the patronizing sexism and people would be more willing to listen to you. Instead, the community now gets to write you off as a jerk who relies on being bombastic rather than right.

I have contributed, I have been flamed by both Peter Z and Linus, neither really bothered me because I felt that (in my case) they were attacking the admittedly poor code quality. But your statement is far different and is of the type being discussed. You seem incapable of engaging with the idea so you attack the person. Sad, really.

On kernel mailing list behavior

Posted Jul 21, 2013 18:55 UTC (Sun) by hirnbrot (guest, #89469) [Link] (2 responses)

>We would have been better off without the addition of the patronizing sexism

Where was the actual _sexism_ in that? They just mentioned her gender - replace "silly woman" with "silly person" and the message is the same.

On kernel mailing list behavior

Posted Jul 21, 2013 18:59 UTC (Sun) by rahulsundaram (subscriber, #21946) [Link] (1 responses)

If that was the case, why bother introducing the gender? When was the last time you heard "silly man" in similar arguments?

On kernel mailing list behavior

Posted Jul 21, 2013 19:42 UTC (Sun) by hirnbrot (guest, #89469) [Link]

I agree that when it's not important you should leave it out - however, languages being what they are, it sometimes flows better _with_ gender than without (e.g. singular "they" being somewhat uncommon).

However, I was just talking about that one particular comment, and having read the other comments by that particular person in this thread, it's very clearly a case of sex- and/or troll-ism.

On kernel mailing list behavior

Posted Jul 25, 2013 22:52 UTC (Thu) by stevenb (guest, #11536) [Link]

Does "wrote most of the USB3.0 code" count as "something useful to say" to you? Sarah is not just a random person sprouting a discussion on lkml, she's been around for quite some time.

As for your comment: Silly is as silly does...

On kernel mailing list behavior

Posted Jul 19, 2013 11:43 UTC (Fri) by shmget (guest, #58347) [Link]

I'll add

Being always 'polite':
- _may_ be keeping good people out of LKLM (just as unquantifiable)