On Fri, Jan 13, 2012 at 8:01 AM, ilkosta
<[email protected]> wrote:

> today updating my debian I saw a new package called nuitka and I began to 
> wonder if it could improve performance of Leo.

Thanks for the link.  As many of you know, I have been fascinated, off
and on, for many years by the tantalizing possibility that some way
could be found to speed up Python via translation.

There are good reasons to be skeptical that such a project could
improve Leo, but that hasn't stopped me from wondering...

The essence of the situation, imo, is that we Python programmers
program *as if* we did, in fact, know the types of all (or almost all)
of the objects/variables we use.  This is kinda correct, and kinda
not: the RPython language (part of the PyPy project) shows *exactly*
where this assumption breaks down.

I suppose it is possible that some truly brilliant person will
discover a way around the stupendous amount of analysis that the PyPy
toolchain does.  Maybe Kay Hayen is that person...

Anyone have any other comments?  This is a subject near and dear to my heart.

Edward

P.S. There is another dimension to the problem.  It is imperative that
any tool that speeds up Python should be identical to Python from
Leo's point of view.  For example, a C-based approach, of whatever
kind, would restrict Leo's flexibility.  This would be a huge step
backward, and so is essentially out of the question.

EKR

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/leo-editor?hl=en.

Reply via email to