Python and the need for speed

bartc bc at freeuk.com
Sun Apr 9 14:34:31 EDT 2017


On 09/04/2017 04:57, Chris Angelico wrote:
> On Sun, Apr 9, 2017 at 10:20 AM,  <breamoreboy at gmail.com> wrote:
>> I've an idea that http://www.mos6581.org/python_need_for_speed is a week late for April Fool's but just in case I'm sure that some of you may wish to comment.

> In fact, extreme dynamism is baked deep into the language. You'd have
> to make some fairly sweeping language changes to get any real benefits
> from restricting things. There are better ways to improve performance,
> which is why statickification isn't really pursued.

I have my own interpreted language which I call 'dynamic', but compared 
with Python, code in it might as well be set in concrete.

The only thing dynamic about it is the types of variables. But, it does 
work extremely well. So I think the dynamism of Python is over the top.

Yes, you can make a highly restricted version for which it is easier to 
make faster implementations, but it would be a different language that 
just uses Python syntax. (With considerable problems interfacing to 
existing dynamic Python code.)

 > Bye bye namedtuple.

I'm not completely sure what namedtuples are but they sound like some 
sort of record type. Presumably there is some reason why general 
attributes can't be used. The latter sound inefficient and so do the 
former if they depend on advanced dynamism to work.

Now, my language has records built-in, and they work extremely well, 
without using any of the dynamic features of Python other than dynamic 
variables as I said.

So such things can be made viable with the proper language support 
rather than relying on layers of what I consider unnecessarily esoteric 
features. Get the basics working first.

-- 
bartc


More information about the Python-list mailing list