morning in Python
inhahe
inhahe at gmail.com
Sun May 18 06:37:30 EDT 2008
"inhahe" <inhahe at gmail.com> wrote in message news:...
>
>>It is not clear that the first (cheapest best) human->computer language
>>is a computer language, though if two were orthonormal >in comparison
>>to life, Python's fine. Not my first.
>
> The utterly dry, closed, logical, definitive, hierarchical, consistent,
> determinate nature of a computer language is the only thing that will
> facilitate anything useful on something as utterly stupid (and not to
> mention logical, definite and determined) as a computer.
>
> I mean it, computers are /really/ stupid. They're literally
> stupider than a bug. We just like things we can control.
>
> The requisites I have for a computer language are:
>
> Efficiency (speed)
> Elegance of syntax
> Powerful (conceptual-wise) abstractions
>
> Python has delicious abstractions that make doing a lot of things really
> easy and fun to think about.
> Stackless Python adds even more to that with continuations.
> Also Python's dynamic (another aspect of being powerful conceptual-wise)
> But most of all, I love its syntax. Guido is the awesome.
> (BTW, I won't even use any language that uses := for assignment. I just
> refuse. I don't care what the language has.)
>
> The speed/efficiency issue depends on the task at hand. For most things I
> use Python. But assembly isn't out of the question, and it's fun to code
> in. I also find C/C++ an elegant language. Most things just don't need
> that speed. And Python is 50 times easier to code in than C/C++ and 1000
> times easier to debug in.
>
> I also like C#.
>
> My ideal language would be a natively compiling cross between C++ and
> Python. Objects declared with a type would be statically typed, objects
> not declared with a type would be dynamically typed. There would also be
> keywords to declare that class names won't be reassigned and class
> attributes won't be deleted. Those attributes would be referred to by
> offset, not hast table keys. f
hast = hash
More information about the Python-list
mailing list