[Python-Dev] extended print statement, uPre-PEP

[Paul Prescod]

Andrew Kuchling wrote:
> 
> On Sat, Jul 22, 2000 at 05:18:13PM +1000, Mark Hammond quoted:
> >> Ummm...between this and Hyauiu's proposal, I'm afraid we're bloating
> >> Python way too much.
> and then wrote:
> >I've gotta agree here - I really am quite concerned about all the
> >cool-but-not-really-necessary stuff being proposed.  I fear that Python
> 
> Seconded from me. 

I think that this class vague complaints are a little unfair. We've had
a wide range of proposals from extra functions to minor new syntax to
new magical methods to radically new control flow features.

It isn't helpful to lump them altogether and condemn them because Barry
broke the cardinal rule of never using an at-symbol in a feature syntax.
(shame on you Barry!) List which proposed features you like and don't
like and why! That's the only way your concerns could really be
addressed.

> IMHO it's only worth adding a new feature if it
> multiplies power by a significant amount; 

I think it is also worthwhile to recognize "conventions" that could be
made clearer with first-class syntax. List comprehensions replace the
map/lambda convention (and would IMHO, allow map/filter, at-least, to be
deprecated). Range literals replace the for i in range(...) convention
and so forth.

Those of us who have already internalized the conventions are more
likely to see new syntax as an intrusion rather than as a long-term
clarification.

> The urge to add little tiny things that make microtask X convenient is
> what made Perl -- and Sendmail, and DTML, and ... -- what they are
> today.

Many languages (including some of the ones you mention) also suffer from
the syndrome where a first-class feature has been left out so you have
to memorize some stupid and non-intuitive "convention" rather than doing
things in a straightforward way. I strongly feel that map/lambda and in
fact most uses of map fall into this category. The concept is fine but
the syntax sucks.

-- 
 Paul Prescod - Not encumbered by corporate consensus
New from Computer Associates: "Software that can 'think', sold by 
marketers who choose not to."