At 06:27 AM 2/11/00 , Coy A Hile <hile@cse.psu.edu> wrote:
>Mark Hahn sez....
> >
> > the PCness of this sends chills up my spine. I suppose that suits buying
> > Linux means that expletives must be 'cleansed' from kernel sources?
> >
> >
>You actually think the suits will be reaing the source? If they don't like
>our choice of comments/debugging messages, that's their problem.
I've recently become aware that some companies looking to adopt Linux into
their organization are using existing code-review processes to "get
acquainted" with LInux. Unlike the commercial vendors, Linux is available
in source form so it becomes possible to use it.
Now, the procedures for code review vary all over the block. My experience
with code review practices over the past 30 years indicates that very
little attention would be paid to expletives, unless the evaluator was
VERY easily offended (rare in big business). Attention will be paid
instead to the quality of the code (which is good), quality of the comments
(which is all over the block), and the quality and quantity of the external
documentation (which the CoriolisOpen series is being identified as one
such source of external documentation.)
There is also the problem of the perception by the evaluators of the Linux
community and its ability to "support the product" -- IBM has seen the
light, but other companies are having problems believing that the
development and support method is really going to work. How many of them,
for example, are aware that the linux-kernel list even exists?
What code evaluators want to see is professionalism. Dirty words (in
moderation) doesn't detract from professionalism.
If anyone asked for my recommendation (and no one has, so you can stop
reading this now) my suggestion would be to improve the quality and
accuracy of comments. In particular, when you do something that would look
strange to an "average" person who has been coding C for five years, and
you have a damn good reason for coding it in exactly that way, drop a short
comment explaining what you are doing, and why. The rationale for this
suggestion is simple: the second-rate coders that usually get stuck with
code review aren't system-level people, and rarely do they know assembler
well, let along how to read GNU assembler.
The bad news is that adding comments will increase the bloat of the kernel
source even more, which makes it tougher on people with PSTN modems who
wish to download the source. That just means that adding comments should
be done with thought.
Finally, this observation: by putting the source out there under the GNU
copyleft, we are BEGGING the suits to read our code. Now, the majority of
the suits (marketeers, executives, salesmen, bean counters) won't take the
time, but the ENGINEERING suits will...and do. In other words, the
management most likely to read our code are the ones who are used to the,
er, attitudes of the gear-heads and propeller-heads.
Bottom line: GOOD comments, and try to keep it at or under R-rated.
Satch
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Feb 15 2000 - 21:00:21 EST