anyway to work this?
can't specify all incorrect responses as they are virtually infinate. all i want to do is give incorrect (but not correct) feedback to an open ended response.

best,
baruch