Again it is brought to my attention that my gerrit review causes annoyance.
There is a recurring theme here, so I'd like to address all patch submitters.
I feel like it's a Good Thing to mark *everything* I notice: "I've read your
patch, and these are all the details I found, in case you'd like to address
them." I see it as a courtesy towards the patch submitter.
For some reason though I turn out to upset rather than help?? My (wrong?)
impression is that these are normal reviews as I've given them as well as
received them hundreds of times over the last decade+. In my world, adjusting
my patches N times is normal, I make mistakes all the time. But I certainly
don't want to upset: I'd really like to find out what I can change to make
everyone's hacking smoother.
In all my previous projects I got used to very high code scrutiny, and when
starting on Osmocom, Holger's review of my patches has continued on the same
high level ... often I had to revisit large amounts of my code.
Thus primed, I find it hard to ignore irregularities: most of them mean the
code becomes less stable or less maintainable/readable.
Usually I'm trying to understand what the patch does or intends, and want to
make sure future readers of the code can also understand easily. The idea is to
save time in the long run.
If you disagree with me or see inconsistencies, please let me know, whether a
patch is merged or not. Maybe I oversaw something or maybe I'm just plain
wrong.
The mood I'd like to convey is: "yes, I trust you that all these patches are an
improvement. I invested some of my time in your work with the goal to merge it
soon. And here's everything that caught my attention; do you agree?"
Ideally we can reach high code standards and at the same time collaborate
productively.
None of my gerrit comments or -1 votes are intended to convey emotion...
Feel free to mail or jabber or talk to me, also privately, on these
issues anytime! And feel free to ignore nitpicks if you don't care enough.
~N
Show replies by date