Re: OFFTOPIC: e2fsprogs and +2Gb partitions

ak@muc.de (linker@nightshade.ml.org)
Sat, 13 Jun 1998 14:21:42 -0400 (EDT)


On Sat, 13 Jun 1998, Alan Cox wrote:

<snip>
> Crap. _BSD_SOURCE exists to emulate BSD behaviour on other platforms.
> _LINUX_SOURCE therefore exists to emulate Linux behaviour on platforms
> supporting it.
>
> > People complained that Linux programs are so poorly portable. The
> > reaction from those who wrote these programs was that they didn't know
> > better since they don't have other machines to test.
>
> There are also a lot of people who dont care, or who do care but will
> address the issue in time, not because someone wielding the software
> equivalent of a large hammer has a go at them
>
<snip>

What it comes down to is this: Do the authors of GLIBC have the right to
make me jump through hoops to make my program portable, when I only want
it to run on Linux.

This is just a lame attempt to force a package's author to code things he
doesn't feel like coding right now. There is a group of people on other
Unix systems who see lotsa fun Linux only code and say geewiz wouldn't it
be nice if it ran here. Yet they don't port it themselves. They go and
whine to the glibc authors 'Please get them to make their code portable'.

Writing portable code is a GOOD thing and ultimatly, the LIBC should
encourage people to write portable code. However, since the people who
wanted the portable code (the whiners) arn't the ones doing the porting
(they are passing off the problem to the orignal authors who may have no
intrest in portability) they dont care if the transition goes smoothly.

Infact, having a 'transition flag' (_LINUX_SOURCE) that allows old code to
run without extensive modification would be determental to their get all
linux code portable cause. This is a flagrant disregard to the needs of
the people writing the code.

It is downright disrespectiful.

If someone has written non-portable code that you want to run the correct
course of action is make it portable and submit patches. Most likely
they will include them. (who wouldn't?)

If lots of people are writing nonportable code because the 'programming
system' lends to that kind of programming then it should be changed to
encourage portable programming.

*HOWEVER* when you make a change, you must also help people make the
transition. Why should my work have to stop so that I can fix some stupid
portability problems that I dont care about (and the only people who do
care about them havn't cared enough to send in patches).

Furthermore, some code doesn't need to be portable! Many pieces of
software for Linux dont need to (or could never) run on other platforms.
What would I do with insmod on Hurd?? :)

By forcing code to be portable, and by making it hard to include system
include files, you've made the life of people writing system tools very
much harder..

If I want to make a program for Linux that's non-portable: Thats my right.
If I release it under something like the GPL, and you want to port it,
then you can. I dont have to port it for you. If you send me patches, and
I'm feeling nice, I can include them and make the code portable for
everyone. That's my choice.

Have some respect for software authors, they wrote the code: Let the code
in peace!

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu