revUnplacedGroupIDs has both false positives and negatives?

[Dr. Hawkins]

I know that it's undocumented, but has anyone looked seriously at
revUnplacedGroupIDs?

As I went hunting for crud, missing menus, and whathaveyou,

As I stepped through (after blowing a foot off), I found that some of the
ids covered by revUnplacedGroupIDs were, in fact, placed.  In one case, it
was most of the groups on a card.

I also find it *not* consistently finding all of the actually unplaced
groups, although this is much harder to track down.

Has anyone else experimented with this?

And, I suppose more importantly, can anyone reproduce it?

-- 
Dr. Richard E. Hawkins, Esq.
(702) 508-8462