--On Thursday, March 31, 2005 11:27 AM -1000 Aaron Kagawa <[EMAIL PROTECTED]> 
wrote:

Ok.. I actually didn't mean total time interacting with an IDE. I meant IDE
interaction with a file. And I'm pretty sure that we didn't intend for it
to be a effort proxy.

OK. As you see, it's important when proposing a new metric to explain the underlying goal (Hmm, where have we heard that before? :-)

Clearly, different tasks require different kinds of events.  As you note, 
Hongbing
collects all sorts of non-state change data because his goal is not to 
operationalize an
effort metric, but to model a family of styles of test-based development.

It may well be that Hongbing's enhancements to Activity already collect all of 
the
primitive events you need. Perhaps the next step is to specify the goal of your 
analysis,
the kind of model that needs to be developed to support the goal, and 
how/whether the set
of events currently captured can be used to instantiate the model?

Cheers,
Philip

Reply via email to