We would like to discourage students running the “Spec-test” over and over again: they should test their code themselves.

A first step would be to be able to show to both the student as well as to the assistant how often the student used this button (for a given submission).

Submitted by Mathijs de Weerdt on 19 February 2015 at 20:57

Yes, I support this. In my course students are developing too much against the specification test.

Note that instructors can also run a student’s specification test, which is useful to answer questions. However, this should not be counted against their number of runs.

In case a student runs the spec-tests multiple times against the same solution code, should we then count this as 1 spec-test run?

It will now count the number of executed spec-tests instantiated by the student. It will increase by one only when the solution code has changed, i.e. running the spec-tests repeatedly without changing the solution won’t count as additional attempts.