>I was thinking of the same approach, but then remembered that some things are used annually (e.g. inventories) and some only when the new boss demands them :). ...
That is right; of course, we have the same situation.
The statistics should serve merely as a starting point; once it is seen that a form wasn't used in over a year, further analysis, including consultations with the users, must be done.
By the way, this statistical system can also give false
positives: a user may enter a form to see what it does, perhaps even give a report (my reports always start with a form for the user options) a try run, and then decide that it is useless.
So, the analysis must not concentrate on forms that aren't used at all, but rather on forms that are used fairly seldom.
Also, forms that aren't required don't get deleted, but moved outside the main system - moved to a folder labelled "Obsolete". However, both taking a form out and bringing an "obsolete" form back into the system may require some additional work.
Difference in opinions hath cost many millions of lives: for instance, whether flesh be bread, or bread be flesh; whether whistling be a vice or a virtue; whether it be better to kiss a post, or throw it into the fire... (from Gulliver's Travels)