> Sure, you can do things the MS-Windows way and arbitrarily break
> compatability from one version of Linux to the next (certainly this
> would be nothing new -- software engineering died of a buzzword
> overdose early in the 1990s -- but it would be nice to be able to
> recommend Linux in terms more glowing than "well, it's Unix, I
> guess.")
So we gradually let the project die under the weight of years of accumulated
cruft instead? This is, after all, part of the problem MS are having with
Windows. It's good to clear out (some of the) cruft in a major revision; if you
don't like it then may I kindly point you at 2.2.10, or 2.0.37, or 1.2.13.
It's not like we're removing syscalls or ioctls or breaking function semantics
here (indeed, AIUI we'll be *fixing* some syscall semantics). Any program which
breaks CR/LF translation goes away, was broken anyway. CRLF files can come from
anywhere, not just FAT filesystems.
> So CR/LF is a horrible hack that should never have been done in
> the first place. Good, it's a better lesson to kernel developers
> that when you design a published interface you need to do it well
> or else it will haunt you forever.
We'll here's a thought: FAT doesn't work at all at the moment. So let Al fix
it without the CR/LF stuff, and those that desperately want it back can code
it themselves, hmm?
-- If it's tourist season, why can't we shoot them?- 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/