Re: Why Linux is doomed (was: Re: FENRIS (nwfs) 1.4.2 Source Code Available)

Arvind Sankar (arvinds@mit.edu)
Wed, 23 Jun 1999 22:11:25 -0400


On Wed, Jun 23, 1999 at 08:11:53PM -0000, leitner@convergence.de wrote:
> I don't think it is too much to ask that before you release a kernel,
> you try to compile it. If it is a stable kernel, then compile it twice,
> once with everything as a module, once with everything linked it. That
> should be the bare minimum that just has to be done before you release a
> kernel. Should Linus really be unwilling to do that, then he should
> just post an URL to a /private/ directory here so someone can grab it
> and do both compiles.

I don't get it. Why private? So less people compile it? So less configurations
get tested? You would be happy if the kernel just compiled, but didn't work?

>
> > > 4. FAT. 2.3.7 does not compile with FAT file system. Excuse me?
> > > This is the single most frequently used file system besides ext2,
> > > and 2.3.7 does not compile with it? Because of a syntax error?!
> > Details? Details are the difference between a rant and something
> > helpful, useful, and constructive.
>
> It does not compile.

Of course it doesn't. What's your point? Everybody who bothers to read
linux-kernel knows that it doesn't, and knows why it doesn't. If you want to
run a kernel that is on a development branch, that was moreover expressly
earmarked `Don't run this unless you're sure you can afford to trash your
system', you _do_ have to at least bother to read the mailing list.

> > Also, does it really surprise you that a development kernel in
> > general, and one in which *major* changes have been made and well
> > publicized, in particular, might have some problems in the areas
> > directly involved in the changes.
>
> Sorry, but if the second most frequently used file system does not
> compile, then don't release it. I am not counting the pseudo file
> systems proc and pts here, obviously.

You don't get it, do you? This isn't a release. Your complaints about
breakages in 2.2.x are perfectly valid. It sucks when something earmarked
stable starts corrupting your disk. Something's wrong with 2.2.9-10. And this
btw, shows why your model doesn't work very well: none of the `core developers'
can reproduce this problem on their hardware.

>
> > > Management summary: stuff like this sucks. I am but a programmer with a
> > > SMP box that likes to run the latest kernel. And yes, I expect all the
> > > kernels to compile out of the box. I don't think that this is too much
> > > to ask.
> > It is not too much to expect other people to give you something that you
> > want, without your incurring even minimal responsibilities?
>
> Huh?
> Please think your answer over.
> It is people like me who do the quality assurance for Linux.
> People like me who complain. And if there is just _one_ goof like the
> FAT problem, there are literally thousands of people who will run into
> this problem and be discouraged to upgrade their kernel in the future.

This is a problem with the development model of the kernel. Some people do
prefer other styles. I do myself: I'd like read-only access to a CVS tree.
But Linus likes this one. End of argument.

-- arvind

-
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/