>> If unit tests are a good idea, then shouldn't we try to generate the
>> method code directly from them?
>>
>> In other words, there could be/should be a language in which one
>> programs in terms of the criteria to be achieved, and the system
>> tries to come up with code that meets those criteria. (The last one
>> of these that seemed really interesting was ART by Inference Corp
>> (out of the CMU context).)
>>
>> Anyone know of any interesting current work along these lines?
>>
> Ward Cunningham seems to do some work in that direction with
> XGP - eXtreme Genetic Programming: http://www.neocoretechs.com/

I think one key aspect of genetic programming is co-evolution.
See for example Christopher Rosin, Richard Belew
"New Methods for Competitive Coevolution (1996)"
http://citeseer.nj.nec.com/rosin96new.html

What makes me really wonder is the fact, that unit-testing of
class-based object-oriented languages is not very balanced/ pedagogical:
The tests talk about the fitness of certain instances, and the program
can only talk about the behavior of _all_ instances.

So I believe this smell of unit-tests shows us something:
We need an environment, where we can co-evolve our program with the
tests from specific instances and specific unit-tests to class-based
instances together with more abstract assertions, while keeping the
examples/scenarios for executing everything.

Markus