David,

Yes, you are right.  Once the bug was tracked down, it was not too hard to
work around.  My code does not depend on getting this fixed now. 

I do consider this yet another reason that improving the type inference
system in ACL should be very high on the priority list.  It is a sad state
of affairs when something as basic as a failure of the type system (!) to
meet the ANSI spec cannot be fixed for fear of introducing bugs.

Of course, my primary reason for wanting better type inference is for
compilation warnings and automatic and manual code optimization (see my
correspondence with Jim Veitch about how effective I found the CMUCL type
inference system).

Larry

-- 
Lawrence Hunter, PhD.
National Library of Medicine               phone: +1 (301) 496-9303
Bldg. 38A, 9th fl, MS-54                   fax:   +1 (301) 496-0673
Bethesda. MD 20894 USA                     email: <nlm.nih.gov at hunter>