Python and the need for speed

[bartc]

On 10/04/2017 03:40, Rick Johnson wrote:
> On Sunday, April 9, 2017 at 1:34:39 PM UTC-5, bartc wrote:

>> I have my own interpreted language which I call 'dynamic',
>> but compared with Python, code in it might as well be set
>> in concrete.
>
> Is this a personal toy, or something that you can share a
> link to?

It's a personal thing but not a toy. Some more info here:

https://github.com/bartg/langs/tree/master/qlang

Click on the 'Q Features' link, this includes a selection of built-in 
features that have been easily and efficiently implemented 
(enumerations, records etc) which Python seems to have trouble with.

(And it can't have simple enumerations such that cat=1, dog=2, but needs 
to have every conceivable embellishment. In other words it doesn't know 
how to keep things simple.)

(To actually run this, there still appears to be a file 'pcc.c' at the 
top level of that github project. This is an old C version of the 
interpreter (yes, just one file) and it seemed to work when I tried it 
just now. You need a 32-bit C compiler. I had to remove a binary version 
(for Windows) when someone said github doesn't allow binaries.)

> For all the hype about GvR's "supposed" time machine, we can
> see that Python has suffered some major design flaws. And
> since Python3000, feature creep has been churning into
> overdrive.

Yes. I suppose when it gets to a point where there are so many features 
anyway, adding a few more makes little difference!

-- 
bartc