[openstack-dev] [all][stackalytics] Gaming the Stackalyticsstats

Ihar Hrachyshka ihrachys at redhat.com
Mon Apr 11 10:45:09 UTC 2016

Markus Zoeller <mzoeller at de.ibm.com> wrote:

>> From: Masayuki Igawa <masayuki.igawa at gmail.com>
>> To: "OpenStack Development Mailing List (not for usage questions)"
>> <openstack-dev at lists.openstack.org>
>> Date: 04/11/2016 03:20 AM
>> Subject: Re: [openstack-dev] [all][stackalytics] Gaming the Stackalytics
> stats
>> 2016-04-11 9:46 GMT+09:00 Matt Riedemann <mriedem at linux.vnet.ibm.com>:
>>> On 4/10/2016 6:37 PM, Clint Byrum wrote:
>>>> Excerpts from Matt Riedemann's message of 2016-04-09 06:42:54 -0700:
>>>>> There is also disincentive in +1ing a change that you don't
> understand
>>>>> and is wrong and then a core comes along and -1s it (you get dinged
> for
>>>>> the disagreement). And there is disincentive in -1ing a change for
> the
>>>>> wrong reasons (silly nits or asking questions for understanding). I
> ask
>>>>> a lot of questions in a lot of changes and I don't vote on those
> because
>>>>> it would be inappropriate.
>>>> Why is disagreement a negative thing? IMO, reviewers who agree too
> much
>>>> are just part of the echo chamber.
> __________________________________________________________________________
>>>> OpenStack Development Mailing List (not for usage questions)
>>>> Unsubscribe:
> OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
>>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>>> I'm not saying disagreement is a negative thing, I was saying there
> are
>>> times when I've seen people -1 for crazy nits, e.g. there should be a
> blank
>>> line between the bug ref and change-id in the commit message, or for
> asking
>>> questions for understanding (which, btw, I'm fine with -1 for 'add a
> comment
>>> because this is complicated and I didn't get it at first'). And I'm
> also not
>>> crazy about piling on or agreeing with everything either. My point is
> I
>>> think it's appropriate in a lot of cases to just not vote but still
> comment.
>> I think we have some/many implicit rules for our review. There's a
>> document[1] for review
>> but it doesn't mention crazy nits. So should we add what we don't want
>> to see people -1 for?
>> [1] http://docs.openstack.org/infra/manual/developers.html#peer-review
> My basic rule of thumb for voting is:
>     vote | translates to
>     ---------------------
>       -1 |  "I understand the code and your change and I'd rather not
>          |  maintain it. My reasons are [...] and suggestions are [...]."
>       +1 |  "I understand the code and your change. It improves the
>             project and I'd maintain it."
>        0 |  "I don't get the code or your change. My questions are [...]."
> If it already has at least one +2, I (usually) ignore it. The change
> already has attention from the cores, it is unlikely that I can add
> more valuable feedback to that change.

Oh. I think that’s not an ideal approach. As a core, I often times see  
valuable and reasonable comments that were not pointed out by existing  
cores who already voted with +2, from folks with no core status.

As a core for one of projects, I appreciate *a lot* when people do deep  
reviews for patches, or just ask questions, even if not leaving a vote.  
Cores are often dragged into multiple directions, and so can miss a legit  
bug, while some other contributors can pull their attention to a problem.

So please, please, please comment on patches you review. If you have a  
question, please ask it. You will learn something, and maybe the code will  
become cleaner or at least easier to read [f.e. because authors will add  
some comments to their patches to avoid later confusion and questions.]  
Don’t assume cores are gods and spot every bug. Cores also do silly things.


More information about the OpenStack-dev mailing list