On Sun, Sep 24, 2006 at 06:01:54PM +0000, Christos Zoulas wrote:
> 
> I think that the response to your change was very harsh and instead
> of having a constructive discussion on how this should be handled
> the end result was that you ended up feeling bad, decided it was not
> worth it, and backed it out.

I think, to be blunt, that this is a crock of, um, stinky dung.

The response to Hubert's *proposed* change was polite, but overwhelmingly
negative.  Many developers responded by asking -- note, *asking* -- that
the change not be committed, all citing essentially the same reason: that
adding synonyms for POSIX features like this leads to a decrease in the
portability of scripts written for the sh interpreter.

It's not just a good reason, it's one that was shared by every developer
who responded, AFAICT.  The only difference in response was that two
developers (of 11) appeared to prefer the "implement and emit warning"
approach to the "do not implement" approach.

However, rather than accepting the consensus of his peers that the change
should not be implemented, Hubert committed the change nonetheless, and
then sent to the list an ad-hominem attack on the people who objected to
it.

If people who behave in such a way should not end up feeling bad, it is
very, very hard for me to understand how they might ever learn not to
behave in such a way.

And if changes against which there is overwhelming consensus among the
developers who respond -- almost two dozen, if you count those who
responded privately -- should remain in the tree, it is hard to see why
there is any point to ever having discussions at all.

In sum: Hubert misbehaved severely, both on the list and with his commit.
The result that "Hubert feels bad and the commit is backed out" is not an
undesirable result, it is the correct result; after all, we all screw up,
feel bad, and learn to do things differently, and we have all (or almost
all) mistakenly committed bogus things and had to back them out at one
point or another.

Thor