A lot fewer links today. Yesterday, by the way, the most clicked on link was the “Don’t do this” like to the method_missing nil post.
Handed another draft of the Rcov and Style/Test Quality chapters in. Expecting that to be the next beta next week, but we’ll see.
Kent Beck has a screencast series on TDD from Pragmatic. I linked to the rough version of this some time back. Haven’t gotten the time on this yet, but expect a fuller review later on.
BJ Clark has an essay on whether acceptance tests should be written by developers or by interaction designers. Classic XP is pretty clear that customers should be involved in this process, but that’s not often feasible in reality.
Tim Bray proudly announces that he has never learned the precedence rules in any language he’s used. I am in complete agreement with this – the only reason I know the Ruby ones is from writing about them.
Ryan Davis asks about having a special refactoring mode to change Autotest behavior after failing tests pass. My first impression is that this proposal might be too fiddly, but I think there’s something in the basic idea.
Part 2 of the Teach Me To Code Rails 3 screencast is up.
Finally, the Phusion team has another road map article, including a Passenger light that can be used in development from the command line as you use mongrel or thin. And other deployment features, as well. Cool.