Not logged in
Log in now
Create an account
Subscribe to LWN
LWN.net Weekly Edition for May 23, 2013
An "enum" for Python 3
An unexpected perf feature
LWN.net Weekly Edition for May 16, 2013
A look at the PyPy 2.0 release
Indeed. There's no pride in contributing to a project that would accept the uttermost trash. Or as Groucho put it: "I don't care to belong to any club that will have me as a member".
LinuxCon Japan: Making kernel developers less grumpy
Posted Jun 8, 2012 1:10 UTC (Fri) by gdt (subscriber, #6284)
And sometime public mocking is counter-productive. Here's an example from yesterday where a developer was mocked for no good purpose (the API documentation was out of date, and they were asking how it now worked). I wonder if the developer feels as motivated about contributing to the kernel as they did before that thread. The level of blame towards the developer rather than towards the patch which didn't update the documentation is interesting.
Posted Jun 8, 2012 1:25 UTC (Fri) by gregkh (subscriber, #8)
We are people, and people are messy, kernel developers have no requirement to always be civil, especially to people to purposefully do not participate in our development process, and then blame us for problems.
What would you do in that situation?
Posted Jun 8, 2012 2:38 UTC (Fri) by rahulsundaram (subscriber, #21946)
Posted Jun 8, 2012 4:50 UTC (Fri) by gregkh (subscriber, #8)
And read that whole thread, this person is obviously trying to work outside of the kernel developer community, they don't want to be involved, they only want validation that they were using some kernel code based on a documentation line, instead of reading the code itself.
So tell me, honestly, how would you respond in that kind of situation.
Now multiply that interaction by 400 every week.
Tell me how you would react then.
I think we all are doing a very good job given the load we all work under. If you think you can do better, please, help out, we can use it, and we welcome help. But criticizing without understanding the issues, environment, or situation surrounding a specific interaction, does nothing.
Posted Jun 8, 2012 19:41 UTC (Fri) by daglwn (subscriber, #65432)
Sure, but so what? That kind of behavior would not be tolerated in any workplace and it shouldn't be tolerated in any project that considers itself professional.
Posted Jun 8, 2012 19:58 UTC (Fri) by dlang (✭ supporter ✭, #313)
or complaining that the add-on component that the outsider designed to plugin to the companies devices didn't work and demanded time from your senior engineers to help troubleshoot the outside product?
I guarantee you that if you did manage to corner senior people like this you wouldn't get thanks and help. At best you would get a polite brush-off (and instructions to pay for support)
so if you're going to start the "wouldn't be tolerated in the workplace" approach, realize that the initial request wouldn't be tolerated either
Posted Jun 9, 2012 3:51 UTC (Sat) by daglwn (subscriber, #65432)
Which is a perfectly fine response, and is what SHOULD happen on LKML.
There is a difference between saying "no" and publicly humiliating.
Posted Jun 11, 2012 5:48 UTC (Mon) by broonie (subscriber, #7078)
Posted Jun 11, 2012 16:44 UTC (Mon) by daglwn (subscriber, #65432)
Posted Jun 13, 2012 13:36 UTC (Wed) by broonie (subscriber, #7078)
It's just a generally bad way of correcting people's behavior.
Posted Jun 13, 2012 22:32 UTC (Wed) by nevets (subscriber, #11875)
I've been complimented on being one of the politest developers on LKML. I jump into heated flames all the time without resorting to insults, even when insults are directed at me.
But as I started working on more core kernel code, and becoming more responsible for code that was used by a wider audience, I started getting some of these crazy requests. I started with the polite responses, but that just seemed to bring more craziness my direction. Then the polite brush offs, which also attracted this crap. Ignoring it made me look unprofessional (not responding to email). Finally I resorted to some insulting feedback and that seemed to work.
I wasted my own time trying to be nice. One thing that you can get from Greg's report is that kernel developers have very limited time. Time is our most precious resource. If being polite wastes time, and a quick jab is more effective, then those quick jabs become the efficient process.
Sorry, but unfortunately it's just part of the job. Now, as we are still humans, some developers can go overboard with their insults. But just imagine getting email every day from people that don't make the effort to understand the process. When being a bit of a jerk back either makes them understand or just go away, and you get more work done, one tends on being that jerk.
If you're creative with your insult you might even get the added benefit of appearing on LWN's "Quote of the week" :-)
Posted Jun 18, 2012 12:36 UTC (Mon) by marcH (subscriber, #57642)
Answering harassment or trolls is anything but professional. Ignoring harassment is definitely professional.
Whom do you trust to define what is professional versus not, the harassers?
Posted Jun 18, 2012 15:05 UTC (Mon) by nevets (subscriber, #11875)
But it's usually people that expect something for nothing, or just do not listen to the responses you give. After a bit of explaining the same thing multiple times, and getting the same answer back without it going any further, is where one starts to get frustrated, and a quick jab can sometimes wake them up.
Posted Jun 18, 2012 16:36 UTC (Mon) by marcH (subscriber, #57642)
This what I called "harassment". Please search/replace with a better term, thanks.
Posted Jun 8, 2012 19:59 UTC (Fri) by branden (subscriber, #7029)
You should watch the movie _Glengarry Glen Ross_ sometime.
Posted Jun 9, 2012 3:49 UTC (Sat) by daglwn (subscriber, #65432)
Posted Jun 12, 2012 12:38 UTC (Tue) by mpr22 (subscriber, #60784)
A number of factors.
Posted Jun 9, 2012 1:52 UTC (Sat) by daniel (subscriber, #3181)
Posted Jun 9, 2012 3:04 UTC (Sat) by gregkh (subscriber, #8)
Posted Jun 9, 2012 3:52 UTC (Sat) by daglwn (subscriber, #65432)
Posted Jun 9, 2012 4:57 UTC (Sat) by rahulsundaram (subscriber, #21946)
Posted Jun 13, 2012 22:47 UTC (Wed) by nevets (subscriber, #11875)
But really, I've seen LKML actually become much more professional and less insulting. 8 years ago it was more vicious. I've seen people get insulted for no good reason, and the insults were quite nasty. Today, and even the example someone posted here, the insults come after a lot of back and forth where no progress is made, and the developer finally gets frustrated. You can even see the frustration in the insults.
Sometimes its just that the person on the other end isn't getting the point of the developer. Lots of miscommunication. That's natural. My point is that on the whole, kernel developers are pretty good to you if you just try to make an effort, and hide your ego. The last thing a developer wants is code coming from someone that says they know better than the developer. Sure, they may know more, but it works much better if you show them benchmarks and numbers to back your work, than just saying "I'm an expert in this".
Posted Jun 9, 2012 19:13 UTC (Sat) by marcH (subscriber, #57642)
The only sane way: like a robot.
Please just stop caring and send 400 pre-canned, polite, terse and relevant answers per week. I am sure you are well able of automating your email software so the time required to select and send such a pre-canned answer becomes negligible compared to the time actually spent reading the corresponding submission. Bonus: you save time for the interesting submissions & discussions.
Real people have feelings. Silencing them when needed is part of being a professional.
Posted Jun 10, 2012 2:08 UTC (Sun) by gdt (subscriber, #6284)
Exactly this. I was reading the list with a view to cleaning and submitting my years-old patches to add NAT for Cisco Sippy phones and for per-route TCP algorithm selection. What that exchange says to me is "it isn't worth the angst and that I'd be happier doing one of the 101 other things competing for time in my life".
Posted Jun 8, 2012 18:37 UTC (Fri) by daglwn (subscriber, #65432)
I would even throw in a "thank you" for finding a documentation bug.
Posted Jul 13, 2012 10:20 UTC (Fri) by dmk (subscriber, #50141)
I guess, that message would have been better with the hint that if the documentation is wrong, the correct thing is to sent a patch to fix it, not to bitch about it.
Copyright © 2013, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds