Damnation! (was: Re: Python3k extended grammar)

[Dirck Blaskey]

> > >  Roy Katz <katz at Glue.umd.edu> wrote:
> > >
> > > Here's a list of Unqualified Rumors we've been compiling:
> > > 1. Python 3000 may be case-insensitive.
> > > ...
> > > 4. dividing integers may yield a float.

> > François Pinard <pinard at IRO.UMontreal.CA> wrote:
> >
> > Horror, hell, and damnation!  April 1st should be far behind us by now!
> > ...
> > So, how do I register my vote?  Where do I leave my tears? :-)

> Guido van Rossum <guido at python.org> wrote:
>
> Python is not a democracy.  Voting doesn't help.  Crying may... :-)
> ...
> This is all taken way out of proportion.
> ...
> (and maybe the case insensitivity will be in the tools alone).

(Don't panic yet, there will be plenty of time to panic later.)

Since there have been inumerable oh-no's posted on this topic,
I'm not posting to contribute my vote, since Guido doesn't care :),
but here are some other things to think about:

Any significant change threatens the status quo, and
current Python users have a vested interest in the status quo.

Mr. Guido, as HMFIC, chooses the goals of the development process.
If his priority is to make Python easier to learn,
he will do whatever he believes is necessary to make it so.
If you disagree with his priorities or choices, convince him otherwise.
Arguing for the status quo because it's
the status quo isn't going to be effective.

It is difficult to ignore a large investment in an existing system,
but backwards-compatibility can be Very Destructive -
(see:Windows|C++|HTML|etc.).  Sometimes a split is the best way.

Another point:

A programming language is a computer application.
Users of computer applications tend to fall into three groups:
Beginning, comfortable, and expert users.

A system written solely with the beginner in mind will quickly become
annoying to the comfortable or expert users (paper-clip animations?).
A system without expert-level functionality will eventually be abandoned
in favor of a more powerful system. If the system can adapt or be
configured for the level of the user, it will have a larger audience,
and it will remain useful longer for each user.

This directly conflicts with Mr. Guido's One-Obvious-Way philosophy;
resolving the conflict successfully will take serious, thoughtful effort -
the kind of effort that produced Python in the first place.

Powerful programming languages tend to be expert-level systems,
and configurable in a variety of ways (#include "/dev/tty").
I believe that a language designed to be easy to learn and teach
is not necessarily a good choice for professional engineering.
This opinion is entirely subjective - based on my experience with
Pascal vs. C - and may sound silly given the origins of Python.

Don't get me wrong - the CP4E project is Completely Worthwhile -
but I believe that Computer Programming isn't for Everyone -
at least not until computers get a whole lot smarter.

Just my (long-winded) opinions

d