Re: [ossig] Secret Windows Code Leaked On Internet

[Julian Gomez <kluivert@xxxxxxxxx>]

On Mon, Feb 16, 2004 at 04:10:22PM +0800, Uwe Dippel spoke thusly:

(snip)

>With the best of intention, once you know a solution to a given problem it
>is very hard to 'forget' that solution on purpose, in order to write one
>that is not derived from the first one.

You are making the assumption that there is *only* one solution.

>The programmer hired by me and knowing about 'that secret sys_call' will
>be less productive than the one untouched; because the latter doesn't need
>time to forget and won't while time away thinking: 'what a bunch of
>buggers not to permit me to use this.'

Again, I think you're making the assumption that said secret sys_call is
the better method in which to solve the problem. But I haven't followed
this thread closely.

>who's still waiting for the first reports of code quality. Would be fun 
>to know that they have trunks of dead code lying around for the simple
>purpose that the compilation dies once it is removed !

So? A number of major OSS projects have got dead code lying around. Samba,
Linux kernel, Apache etc. A source untar of samba-2.2.8a, with a trivial
find of all *.c files; including comments == 250,000+ lines. Assuming
150,000 lines are comments, that's 100,000 lines of code.

When you take into account that code lying around can be due to:

o  Ex-programmers quitting a project.
o  Certain portions being too hard to read/parse.
o  Making changes causing unwanted breakage in critical portions.
o  Scheduled code overhaul/rewrite in the near future.
o  Backwards compatibility.

I think its completely understandable, if not completely desirable.


---------------------------------------------------------
To unsubscribe: send mail to ossig-request@mncc.com.my
with "unsubscribe ossig" in the body of the message