StubToolkit must go!

[Richard Bair]

>> So my thought here is that if we can run 30,000 tests in < 5 min by allowing nearly all of them to be run in the same VM, then that is probably better than running 3,000 tests in separate VMs for the sake of our pre-commit testing. For the nightly run, we can certainly run them all in their same VM so that the analysis phase of failures is much simpler.
> 
> This sounds fine, so long as at the pre-commit level we have the option of running the full test suite in the "slow and steady" mode to isolate test instances.

Yes, in fact *all* the tests (except security tests and big apps) will be both in our repo and open source, so that you can run any of the SQE tests, etc. Also you can run the full exact same build configuration etc. that SQE / RE uses. We'll all be operating off the exact same build scripts so that there is no mystery as to what is finally produced at the end. A developer *should* run the appropriate level of tests before pushing. For example if I'm making a little change, I'll just run the pre-commit and go. But if I'm doing a major amount of work on Region (for example) then I'm also going to make sure I run the SQE tests before I push so that I know I'm not going to fail a bunch of tests. All of that will be available to you.