Re: Fix quilt merge error in acpi-cpufreq.c

[Theodore Tso]

On Thu, Apr 16, 2009 at 09:56:51AM +0200, Ingo Molnar wrote:
> * Rusty Russell <rusty@xxxxxxxxxxxxxxx> wrote:
> > Anyway, Impact: had lead me to think harder about my messages than 
> > the free-form commit style did.  Perhaps it's too rigid, but it 
> > helped.
> 
> btw., and i think this is the crux of the matter, Rusty was quite 
> sceptic about impact lines in the beginning, and did not like them 
> _at all_. We had discussions (months ago) about it with Rusty and he 
> had a similar position to other "read only" participants in this 
> thread.

Hmm, I guess for me what I consider ideal, and what I consciously try
to do, is to include (at most) 2-3 words that describe the impact in
the patch summary line.  Writing a good patch summary line is _hard_;
in 70-75 characters you need to describe both *why* and *what*; it
needs to be something which is both succinct, but which, several
months later, is enough so that someone scanning the patch summaries
has a fighting chance to pick out the relevant patch amongst a sea of
thousands of other patches.

And at least for me, something mechanical just isn't likely to work.
It reminds me of a story when, over 30 years ago, someone at the MIT
AI Lab wrote a proposal to create "programming for non-programmers";
it proposed creating templates so that people who didn't know how to
do things could use as a starting point, and then have expert systems
that would help fill in the rest.  Shortly after this paper was
circulated, a parody was sent out mocking the first paper, "thinking
for non-thinkers".  It suggested creating template idea for people who
weren't smart enough to create their own original thoughts, etc. 

So I'd really like to encourage try challenging people to try to write
a good patch summary line.  It may be that forcing someone to
constrain themselves to 70 characters (75 if they must) and which must
explain both the impact of the patch as well as the what the patch
does, is enough rigidity or a constraint that it might force people to
think.  Because at the end of the day, that's what we really need
people to do.

              					- Ted
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/