Bob La Quey wrote:
I suggest then that the core problem becomes, "How does one make test
and bug fixes fun?"
It is not obvious to me that they cannot be fun.
Who knows? However, not everything has to be fun.
We often pay for the fun parts with the non-fun parts. That's okay.
I didn't enjoy practicing foul shots an hour every day on a basketball
court. I did enjoy banging down free throws in the face of the other
team with better than 95% accuracy.
The issue in programming, as in all things, is where to handle the balance.
I tend to hold that you write a unit test when you go on a "bug hunt".
Unit tests are my breadcrumbs. If it has a unit test, I can trust it
(somewhat). If it doesn't have a unit test, it is suspect until proven
otherwise. I then write unit tests to narrow down the field of
uncertainty until I can see the bug.
This has the advantage that the areas that are bug-prone gain more tests
than those that are relatively bug-free.
-a
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
