[TIP] testing: why bother?

[C. Titus Brown]

On Wed, Mar 23, 2011 at 12:50:08PM -0400, Tim Lesher wrote:
> On Wed, Mar 23, 2011 at 12:40, C. Titus Brown <ctb at msu.edu> wrote:
> > Of course, given that I'm still in the backwaters of scientific programming
> > trying to convince people that *version control* is a good idea, I clearly have
> > other things to work on!
> 
> Interesting.
> 
> Personally, I learned the value of test suites in exactly the same way
> I learned the value of version control: by suffering through a
> catastrophe that could have been easily avoided.
> 
> So if you're in an environment that, for other reasons, is unlikely to
> have those kinds of catastrophes, it's harder to see the value for
> yourself. Remember how on Dune they scoffed at the concept of
> "drowning"?

Oh, we have these catastrophes too in science, although of course they
don't end up being very costly, since students are cheap and there's little
pressure to get the right answer (...another set of beer stories).  The
question is, can we use teaching to *avoid* the part where the students learn
about drowning by actually drowning?

After talking with some folk at PyCon, I think one of the biggest problems is
the nearly complete lack of experienced programmers in academic "management"
positions (aka professors).  My students are using version control even without
my prompting, because everyone else in the lab is doing it...

--t
-- 
C. Titus Brown, ctb at msu.edu