Python 2 to 3 conversion - embrace the pain

[Steven D'Aprano]

On Wed, 18 Mar 2015 08:11 pm, Ned Deily wrote:

> In any case, of the two problems noted with Python itself, there is only
> one that appears to be Python 3 related.  That's still not good but I
> think it would be fairer to ascribe a good chunk of the pain you've
> experienced to the more common pain of upgrading any major software
> system that depends on multiple third-party components.


I think what often goes on is rather similar to this process:


"Jim ran a red light and side-swiped a police car. Jim's a bad driver."

"Fred nearly hit a pedestrian. Fred's a bad driver."

"George crashed into a tree because he was texting. George is a bad driver."

"Susan reversed into the fence and broke it. Women are lousy drivers!"


The precise details of every major upgrade differ, of course, and sometimes
things run more smoothly, or less smoothly, depending on the project.
John's earlier comment:

"All this has cost me about two weeks of work so far."

sounds about normal to me :-) Never mind two weeks late, IT projects running
*two years* late sometimes seems like par for the course.

But as soon as something "different" is involved, like Python 3, some people
(and I'm not pointing the finger at John here, just making a general
observation) make the same exaggerated over-generalisation as they do
with "lousy woman drivers". The two weeks we lost upgrading from Python 2.6
to 2.7 is just the normal upgrade pains you always have to expect from any
major project, but the two weeks we lost going from 2.7 to 3.3 is a sign
that Python 3 is a broken mistake!

Anyway, thanks John for persevering.


-- 
Steven