On Tue, 14 Aug 2007, Abdelrazak Younes wrote:
> > > > Remember that this discussion started with "if you have 10
> > > > minutes, please merge my branch".
> > And you still think code review is bullshit.
>
> Never said that, I am still waiting the review...
FYI: Regardless of your exact words[1], my impression of your attitude
towards review of your code is similar to what Jean-Marc wrote.
I stated maybe 10 times that I'd like comments on the incremental diffs
in my branch. I find it weird that I receive comments on the small
glitches before receiving any comments on the design.
In my case, I can get distracted if I early on encounter discrepancies or
other unexplained issues. One reason is that without knowing the overall
structure/purpose/design in advance, how can I know what's important and
what is not? So I'd ask questions about it.
I am not interested in the small details at this point.
You are not interested because you know the overall design etc. For me,
the "details" stand in the way of understanding the design by e.g.
introducing doubts, muddying the real issues and reducing confidence. It's
also a waste of time having thought about a glitch and then just ignore it
- it'll probably be forgotten. Not to mention if you have several
developers wondering about the same minor glitch.
<snip>
I'm glad this is not the case. Oh, wait... you didn't literally state
that you _don't_ think code review is bullshit... ;-)
I always welcome comments but most of the time I only receive requests
for explanation.
Let me explain the impression I've partially gotten over time regarding
your attitude towards reviews of your code.
* You don't _really_ think _your_ could needs to be reviewed
* Only minor mistakes typically occur
* These are easily dealt with
* In the worst case, you just revert
Thus I'm asking you for clarification by simply asking:
Do _you_ think that generally speaking _your_ code need and should
be reviewed before it is comitted to trunk?
It's of course a question to be asked of other developers as well, so I'll
also answer.
Personally, having starting with professional software development in the
end of the 80's, I still think my code in general needs to reviewed.
Additionally, having worked alone on relatively large/complicated code
--and despite finding critic difficult to take at times-- I still consider
it a boon and a privilege that others review my code as this indirectly
allows me to keep improving myself.
You can't really master the C++ language _completely_ on your own, you can
just keep learning - especially as the lanugage keeps on evolving.
Best regards,
Christian
--
Christian Ridderström, +46-8-768 39 44 http://www.md.kth.se/~chr
