hi Kevin,
orry, I'm no programmer, this is far beyond my capabilities, I'm just a simple user.
And as a user I need things that are not yet there,
so I can only offer to use, extend, maybe create widgets 
I was not asking you in particular,
Oh than its' perfectly ok with me 
just making a general suggestion in case someone wanted to give a little gift to the project. 
I do want to ask though, why do you think it is beyond your ability?
Ok I admit it's a bad excuse,
so let's be honest,
I just don't like testing,
I want to make big steps in front, instead of standing still.
e.g. Instead of writing test code,
I'm much more interested in writing the code for a real autocompletion of STC,
not the one that is the wxPython demo (although yoy did a really great job Robin !!)
And if you're doomed to work with Windows, you must accept bugs 
Anyone who can write wxPython code should have all the skills needed to write tests, and all programmers should learn how to write automated tests, IMHO. It allows you to write widgets and programs much faster, and really decreases the amount of time you spend debugging. After all, the entire purpose of computers is to automate tedious tasks, so why not automate as much of the development cycle as possible? 
Yes and No,
I agree that we should automate much more,
but by now (after the computer exists for about 50 years,
I would think the computer should be able to understand our human language 
So I don't agree that we still should be occupied with syntax but much more with semantics (of the knowledge domain).
That's why I have written already FSM (Finite State Machine) programs,
to generate code for several "languages" (Delphi, JAL, hardware),
that's also why I'm now working on "Pylab_Works".
And it's very interesting to see the different attitude of users and programmers for these kind of programs 
The (end-)user finds them very welcome, while the programmer sees it as a threat to his job.
Did the introduction of the computer reduce the amount of paper used 
So that's why I think we should more concentrate on the outside user-interface,
develop new user-interface technics (ever tried Dasher, Mindmappers ),
than on the inside of the programs.
Let me finish with a large/big/? thanks to this group,
without the help of this group I would never been able to create a program like PyLab_Works.
cheers,
Stef