Re: [Mailman-Developers] python is slow

From: Bernhard Kuemel (darsie_at_gmx.at)
Date: 12/29/03 

Date: Mon, 29 Dec 2003 18:55:38 +0100
To: Barry Warsaw <barry@python.org>

Barry Warsaw wrote:
> On Thu, 2003-12-18 at 14:52, Bernhard Kuemel wrote:
>
>>You name the reasons why I will not produce the 'convincing
>>evidence' Richard Barret is interested in. I do however believe
>>ESR is called authority rightly and assume he is has sufficient
>>knowledge of and communication with experts in the languages he
>>writes about so his opinion adds to my experience that python is
>>slow. Some complex things are hard to prove so I thought you
>>might be interested in the opinion of a neural network trained in
>>the subject.
>
> Saying "Python is slow" is pretty much a value-less statement.

Ohh, I think it makes a good subject header as it describes well
what this thread is about.

> Slower than what? Slow when doing what? Too slow? Which
version of Python?

It seems you lost track of the thread. Ok ... I said, I did't
like python, because it burnt up so much CPU time on our old P60
server. Mailman was started via cron once a minute and took about
15 s CPU time each time. 25% average CPU usage just to find out
that there is nothing to do appears much to me. There was also a
flat mate who used python and i noticed the processor load of
that, too.

Someone on mailman-developers@python.org then was interested in
"convincing evidence" that python is slower than perl. But as we
agree that such evidence is difficult to produce and does not
mean much I cited Eric Stephen Raymond's book "The art of unix
programming" where he says ...

"In fact it's generally thought to be the least efficient and
slowest of the major scripting languages, a price it pays for
runtime type polymorphism." I originally quoted the whole
paragraph to be fair.

> Besides that, this thread is more appropriate on python-list@python.org
> or some other advocacy list.

Ok, I post this message there.

> I firmly believe Python is both fast enough and the right
> choice for Mailman, and that's not going to change.

Allright. I trust you know what you are saying.

> -Barry

Bernhard 

-- 
Webspace; Low end Serverhousing ab 15 e, etc.: http://www.bksys.at
Linux Admin/Programmierer: http://bksys.at/bernhard/services.html

Relevant Pages

  • Re: [OT] Isotope decay chains
    ... Angelo wrote: ... to realise that there is a different weigh to light evidence ... >> once you protect a superfertz idea that is a red light litted ... the strange job to make a lot of effort in convincing ...
    (sci.physics)
  • Re: Aint It Time to Kiss Evolution Goodbye??????????????????????
    ... to find the evidence in its favor to be convincing, ... evidence to be unpersuasive. ... convincing the owner of the same thing. ... Like a lawyer, the human mind wants ...
    (sci.physics.relativity)
  • Re: Nick whacks
    ... Not seen him so convincing since he sold us WMD. ... > objective evidence sine that's what they am gnhtug to do. ... Peeps wot has ...
    (uk.rec.sheds)
  • Epistle to a believer
    ... about set theory, so I thought I might have missed something ... Which is a feeling that my life is complete with God in it ... basically simply "a convincing argument". ... 'evidence' applies to these subjects. ...
    (talk.origins)
  • Re: Lance The Hacker
    ... we'll even be so timid as to put a jury and judge in-between the ... completely convincing, he must have known that it was completely ... the evidence being inadmissible. ...
    (rec.bicycles.racing)