v15.1 - Can't delete "orphan" stereotypes

[Paolo F Cantoni]

The recent "tightening up" of the stereotype creation process - whereby the stereotype must be in the defined list (whether general or profile based) is to be applauded.  But, there is a downside. If the item has a stereotype that is not on the defined list (perhaps because it was created in the "mists of time" or imported from another repository) it CANNOT be deleted!

This is particularly problematic for us as we "evolve" metatypes as our modelling evolves to handle the needs of the enterprise.  We create/modify/delete stereotypes as we compare our model with the real world.  In that process, we may leave accidentally level such orphans behind.

The list of stereotype names needs to be enhanced to include such "orphan" stereotypes - preferably clearly marked as "Not defined" (and certainly/preferably? not for adding to other items) so that they can be deleted via the user interface.

Reported,
Paolo

[qwerty]

I just swipe that list reqularly. It's only a lookup table with no real reference. And I run a (part of a large) script to find out rubbish stereoypes to smash them into the owner's face. They shall only use stereotypes from our profiles.

q.

[Paolo F Cantoni]

I just swipe that list regularly. It's only a lookup table with no real reference. And I run a (part of a large) script to find out rubbish stereotypes to smash them into the owner's face. They shall only use stereotypes from our profiles.

q.

We already have that kind of thing, but that wasn't the point of the report.  The punters need to be able to fix up stuff through the UI.  Also, if you have multiple stereotypes, getting stuff out of t_xref is more problematic.

Paolo

[qwerty]

Getting people used to EA's UI turns out to be counter productive. Not funny to have training courses each year so people find last year's dialog back again. So simply avoiding that with a defined list: take care of element x with stereotype y. That's difficult enough to establish that.

q.