Code of poor tone setting [LWN.net]

Code of poor tone setting

Posted Mar 9, 2015 18:56 UTC (Mon) by paroneayea (guest, #96661)
Parent article: The kernel's code of conflict

Seems like a pretty weaksauce document. Nearly every other community seems to have a much stronger, more explicit document.

Both the super bad name and the weak wording are disappointing. Okay, I get it's a pun, but really, setting up expectations for new contributors that "there *will* be conflict, get ready for it!" is setting a poor direction for your community.

It's good to know that hopefully there's a group (the TAG) that maybe will listen to your concerns, but it's better to have a document that takes the approach that the community is already taking the collective route of trying to provide a safe and welcoming environment.

to post comments

Code of poor tone setting

Posted Mar 9, 2015 19:51 UTC (Mon) by branden (guest, #7029) [Link] (2 responses)

"but really, setting up expectations for new contributors that "there *will* be conflict, get ready for it!" is setting a poor direction for your community."

It sounds like mature realism to me.

Code of poor tone setting

Posted Mar 10, 2015 6:00 UTC (Tue) by alison (subscriber, #63752) [Link]

branden, my sentiments exactly.

So far everyone in the kernel community has been exceedingly helpful and friendly to me. I must be doing something wrong.

Code of poor tone setting

Posted Mar 10, 2015 16:49 UTC (Tue) by vonbrand (guest, #4458) [Link]

You'll find no conflict whatsoever in either a brutal dictatorship or in a graveyard.

Code of poor tone setting

Posted Mar 9, 2015 20:19 UTC (Mon) by proski (guest, #104) [Link]

Actually, I like the name of the document. It doesn't set rules unless there is a conflict.

Seems reasonable

Posted Mar 9, 2015 20:21 UTC (Mon) by david.a.wheeler (subscriber, #72896) [Link]

Conflict *is* inevitable, and it is realistic to expect it. The goal should be conflict resolution, not the elimination of conflict. I'm sure the document can be improved, but it doesn't strike me as unreasonable.

I tell people to be harsh on the code, and kind to the people. It's important to separate code from people. Yet it's really hard to do that separation.

People's lives literally depend on the Linux kernel in some cases (e.g., a 911 call using Android). I suspect some kernel developers are concerned that a community focus on "nice" could lead to a failure to rigorously question the code (after all, many communities DO overlook performance failures in order to be "nice"). Does one kind of niceness necessarily lead to the other? I don't think it needs to, but I suspect that is at least in part one of the underlying concerns. The Linux kernel developers are clearly focused on quality of results (as illustrated by the first part of this document).

Code of poor tone setting

Posted Mar 9, 2015 21:42 UTC (Mon) by BenHutchings (subscriber, #37955) [Link] (3 responses)

Seems like a pretty weaksauce document. Nearly every other community seems to have a much stronger, more explicit document.

Indeed, this reads as very defensive:

This development process [is wonderful and doesn't need to change]. If however, anyone feels personally abused, threatened, or otherwise uncomfortable due to this process, that is not acceptable.

It puts the onus on those who are not treated fairly, whereas it should begin by stating the responsibility of reviewers and maintainers to treat others fairly. It also doesn't address the possibility of harassment of other kernel community members outside of a review process.

I was asked to ack this before it was submitted and I noticed that although the wording had been discussed by the TAB, not all members had acked it. After talking to one of the exceptions, I didn't feel any more inclined to ack it myself.

Code of poor tone setting

Posted Mar 9, 2015 23:59 UTC (Mon) by josh (subscriber, #17465) [Link]

> This development process [is wonderful and doesn't need to change]. If however, anyone feels personally abused, threatened, or otherwise uncomfortable due to this process, that is not acceptable.

> It puts the onus on those who are not treated fairly, whereas it should begin by stating the responsibility of reviewers and maintainers to treat others fairly. It also doesn't address the possibility of harassment of other kernel community members outside of a review process.

I agree completely; that's a bug. Put that together with the very heavyweight process for actually raising a complaint, and I doubt that any of the direct targets of abusive mails will end up gaining much value from this document.

On the other hand, nothing in this document prevents others involved in the community from raising such complaints in response to such behavior, even if they're not the target. In particular, the next time (and sadly there probably *will* be a next time) that some high-profile kernel maintainer decides to be a terrible person on LKML, anyone on LKML can and should actually exercise this process. Preferably making it clear afterwards what result they obtained from the process.

The first time someone does so, we'll find out if this code actually works.

That's the one reason I felt like this was a sufficiently non-zero improvement to ack it. It's by no means a sufficient code of conduct; it fails several of the standard tests for effective codes of conduct. But at best it might actually provide a channel for curtailing some of the abusive mails in the kernel community, and at worst it's no more ineffectual than not having a code at all.

Code of poor tone setting

Posted Mar 10, 2015 6:04 UTC (Tue) by alison (subscriber, #63752) [Link] (1 responses)

BenHutchings, submit a patch? But maybe we should wait for some test results first.

Code of poor tone setting

Posted Mar 10, 2015 8:22 UTC (Tue) by error27 (subscriber, #8346) [Link]

Patches to the CodeOfConflict aren't going to be accepted for a while. The original text went through a lot of revisions. It's the kind of thing where everyone has an opinion but no benchmark numbers.

Code of poor tone setting

Posted Mar 10, 2015 2:15 UTC (Tue) by charris (guest, #13263) [Link]

I've been missing Al Viro's incisive commentary for years. He had a way with words.