Hi everyone. The most recent version of the RB makes available, and easy
to use, the Smalllint tool (it is now in the open menu). There is also
now a test server running on my machine, which runs the SUnit tests
quite often. I think of both as "getting maximal power from the tools we
have to improve Squeak code".

Recently I've had other ideas in the same vein, but I thought I'd
consult with everyone before I start working on them.

* We could automatically run the test runner on every patch that comes
out, and tag it with "fixes 4 tests" "breaks 2 tests" and combinations
thereof. 
* We could automatically run Lint (using some subset of rules) on every
patch that comes out.

Then there's the question of what to do with the results. One option is
to post the results on the BFA, so that they fill the role intended for
the appropriate QA tags, automatically. This could be done with or
without the results appearing on the list. The results could also be
mailed to the author, or some combination of the two.

Anyone else think this is a good idea? bad idea? have no idea what Lint
is? want to suggest alternatives/elaborations? anybody want to help? 

Two open technical questions are:
- Producing useful text for Lint results is not obvious...
- If we compare the test/lint results before/after a patch, do we update
the results whenever the Squeak base changes?

Daniel