Re: [MirageOS-devel] [PATCH v4 6/7] Add guide on Communication Best Practice

[George Dunlap <george.dunlap@xxxxxxxxxx>]

On 1/13/20 9:21 PM, Lars Kurth wrote:
> 
> 
> On 13/01/2020, 19:54, "George Dunlap" <George.Dunlap@xxxxxxxxxx> wrote:
> 
>     
>     > On Dec 30, 2019, at 7:32 PM, Lars Kurth <lars.kurth@xxxxxxxxxxxxxx> 
> wrote:
>     > 
>     > From: Lars Kurth <lars.kurth@xxxxxxxxxx>
>     > 
>     > This guide covers the bulk on Best Practice related to code review
>     > It primarily focusses on code review interactions
>     > It also covers how to deal with Misunderstandings and Cultural
>     > Differences
>     > 
>     > +### Avoid opinion: stick to the facts
>     
>     In my talk on this subject I said “Avoid *inflammatory language*”.  At 
> some level it’s good to have strong opinions on what code should look like.  
> It’s not opinions that are a problem, or even expressing opinions, but 
> expressing them in a provocative or inflammatory way.
> 
> Let me look at this again: I don't feel strongly about it
> 
> I changed the title because I felt that the bulk of the 
> example is actually about sticking to the facts an opinion 
> and the inflammatory element was secondary. So it felt more
> natural to me to change the title.

Right; the point though specifically is that people's natural, and
probably healthy response to poorly-written code, or to
inconsiderately-written patch series in any way, is to use charged
language.  I wouldn't call any code "garbage", but code submitted is
sometimes actually terrible, fragile, spaghetti, inefficient, racy,
messy -- whatever bad things you can say about it -- and any
well-trained developer will have the same opinion.

It's not a problem at all to have opinions on code; I think that's a
prerequisite for being a good developer.  It's also not a problem at all
to say, "This code is great" or something positive about the submitter;
nor is it a problem to talk *together* about something not written by
the submitter ("Wow, this code you're trying to fix is a mess.")  The
point specifically is to avoid things which are likely to provoke a
negative emotional response in the submitter.

> But then looking at the definition of inflammatory language,
> aka  "an inflammatory question or an inflammatory statement
> would be one which would somehow predispose the listeners
> towards a subject in an unreasonable, prejudiced way."
> It is clearly also true that the example is inflammatory.
> 
> I think I may have tripped over an area where there is no good
> language match: the German translations of inflammatory
> aufrührerisch & aufwieglerisch have an element of rebellion
> and mischief to them (at least when I grew up).

"Provocative"? "charged"? "loaded"?  "derogatory"? "contemptuous"?

> I am wondering though, whether it is necessary to include 
> a definition of an inflammatory question or an inflammatory
> statement if we stick with it in the title

I think people should be able to pick up what we mean from the reasoning
and from the examples.

 -George


_______________________________________________
MirageOS-devel mailing list
MirageOS-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/mirageos-devel