Efficiency debates

[Alan Gauld]

On 15/02/2021 02:26, Avi Gross via Python-list wrote:
> I think we have discussed this a few times.

Indeed, many times!

And there is a natural tendency for a group focused on a
programming language to fixate on language improvements. But
it's worth while to back up and look at real world scenarios
too where choice of programming language is such a minor
influence on performance that it is rarely even considered
significant.

Much more important are things like network speeds,
database speeds, disk access speeds and overall architecture
and design. These are the areas that give orders of
magnitude performance improvement.

Next comes algorithm design, caching models and the like.
These give multiple factors of improvement.

Finally, we come to language tweaks and choice of language
where a doubling of performance might be attainable. (Language
X being "10 times faster" than language Y rarely translates
to a non-trivial application being 10 times faster!)

Of course, it depends on the type of project and if its a high
computational scenario things will likely be different. But for
the big projects that run our lives - air traffic control,
taxation, payroll, billing, banking and so on, choice of
language comes way down the pecking order.

What's that quote about premature optimization?...

-- 
Alan G
Author of the Learn to Program web site
http://www.alan-g.me.uk/
http://www.amazon.com/author/alan_gauld
Follow my photo-blog on Flickr at:
http://www.flickr.com/photos/alangauldphotos