Well. He wasn't entirely wrong. The OODA loop for Beaker has always been a bit clunky.
If I'm developing a test suite and there's something that isn't quite coming together right, I want to be able to:
• Provision the SUT set
• Test, hack, test, hack, test, hack, until the test starts acting correctly
• Maybe re-provision the SUT set if I did something boneheaded that contaminated the node state
Beaker's monolithic design makes that difficult. There are sub-commands that can split those apart, but they all require a coordinated paragraph of flag options, which is clunky.