Calling a generator multiple times

[Tim Peters]

[Courageous]
> What's bad about that is pointing people to the PEP and expecting
> that it will settle a marked difference of opinion.

Nobody (rational) expects that.  A BDFL Pronouncement is called for
precisely in those areas where extended debate has proved that consensus
will never be reached.  The choices then are eternal paralysis, or, if Guido
wants something enough, he breaks the deadlock by Decree and life moves on.

> Furthermore, I would suggest that if over time I am not alone in
> raising my exasperation here in comp.lang.python

If you feel alone, it's only because you're years late coming to this
debate.

> that the benevolent powers that be consider that for every poster
> there exist a hundred in agreement who don't post,

That cuts both ways, though, and there a hundred who agree with Just who
haven't posted too.  That's what "proved that consensus will never be
reached" means, and this incarnation of this debate only confirms that we
got an accurate reading about that from the last N incarnations.

> and likewise that perhaps "language designer's intuition" can sometimes
> be clouded by the myopathy that sometimes exists in those who are very
> close to what it is they are passing judgement on.

Guido isn't an idiot, and generators for Python have been debated since
1991.  Trust me:  he wasn't close to them at all <wink>.

> ...
> Something that was contentious enough to be discussed "1000
> times" in the mailing list is certainly something you are going
> to be hearing more about again.

The obviousness of that is *why* there's a BDFL Pronouncement:  you're not
going to convince anyone, and nobody else is going to convince you.  Some
people have convinced *themselves* to switch positions, though, based on
experience with the new feature.  I've only seen that switch go in one
direction, though, tending to confirm that Guido's intuition was correct.