Barton--
Could you enter this in the bug database? Even if orderlessp weren't
supposed to work in this case (and why not?), it is a bug that it is
giving a useless error message.
Note that any function that disreps or totaldisreps (including
$orderlessp) is killing some of the efficiency gained by using a special
representation.... But this will only matter if you're planning to have
huge sets involving MRATs or whatever.
By the way, my guess is that there are other packages which fail in the
case of special representations within lists and matrices, because they
assume that special representations are contagious all the way to the
top level.
-s
> The orderlessp function often halts with a (non-descriptive) error
message if
one or more of its arguments are lists or matrices containing CRE
expressions.
The Maxima documentation for orderlessp doesn't mention this
limitation.