Could Python supplant Java?

Peter Hansen peter at engcorp.com
Thu Aug 22 06:57:10 CEST 2002


Greg Ewing wrote:
> 
> FISH wrote:
> 
> > And how do I test the tests?  (If your test software is there to
> > ensure your program works - what is ensuring your test software
> > works?
> 
> The tests and the program being tested serve as checks
> on each other. If they don't match, something is wrong
> somewhere, and it's up to you to decide whether it's
> the program or the test that's wrong.
> 
> It's possible that there could be errors in both
> which cancel out, but if the tests are reasonably
> thorough, it's pretty unlikely.

With XP, there are always "acceptance tests" which test
the overall functionality at a higher level than the unit
tests (which work on individual functions and methods).

It would be much less likely for all three - code, unit
tests and acceptance tests - to be wrong.  In fact, most likely 
if that occurred it would be because of a misunderstanding
of the actual requirements, not because of bugs.

> Having the tests written by someone different from
> the program should make it even less likely.

And that is generally how the acceptance tests are supposed
to be developed - by the Customer, not the developers.
(Customer refers to somebody who works closely with the 
developers and who defines the requirements, can change scope
on the project, and make other such key business decisions.
It does not always, or perhaps even often, refer to the
end user.)

-Peter



More information about the Python-list mailing list