The REALLY bad thing about Python lists ..

[Tim Peters]

[Tim]
>> ISO/ANSI C doesn't fully define what happens when (roughly)
>> integer division or % are handed negative numbers.

[Glyph Lefkowitz]
> The foibles of standards commitees will never cease to amuse me.
> Thank you for this little tidbit.

The next iteration of C has threatened to "fix" this, by mandating Fortran's
original mistake (truncation, as opposed to Python's floor).

BTW, have you "served" on a stds committee?  It's, umm, educational.  Most
officially work by consensus, and after a decade of repeating the same
arguments at every mtg, eventually person A will vote for person B's
marginal idea, just so A can get B's vote on their own marginal idea; and
vice versa.  I was on an ANSI committee once tasked with designing a
high-level cross-language parallelism model, partly paralyzed by one member
who refused to approve of anything that required any dynamic allocation (no
heap, and not even a bloody *stack*!).  The committee eventually did the
honorable thing:  in the end, it voted to dissolve itself.  That's one case
where consensus triumphed <0.9 wink>.  Long life to benevolent dictators.

>> ... note that Windows is Python's most heavily used platform
>> now -- I personally doubt Python would have survived if it only
>> had its Mac and Unix base since '91.

> Do you actually have a basis for that statistic?

Mostly divine revelation.  There was a surreal discussion about this on the
Python-Help group about a year ago, with people mostly trading anecdotes
even lamer than yours about Red Hat's use of Python being some sort of
evidence <wink>.  Not that I have better evidence!  The only *hard*
statistic Guido has is the download stats from python.org, where the number
of Windows downloads routinely beats the number of all others combined (last
stats I saw were from last August, where Windows was "winning" by an easy
factor of 2).  You can point to Red Hat today as not counting in the
python.org stats, but that's relatively recent (note my "since '91"), so did
nothing to help Guido sell Python's case when Python really needed the help.
Someone else can point to PySol on Windows, which is also popular and
doesn't boost the python.org Windows stats either.  I leave it to c.l.py to
figure it out for certain <wink>.

not-admitting-that-a-little-known-part-of-python-startup-sends-
    email-to-python.org-whenever-python's-invoked-ly y'rs  - tim