I don't know enough about the details of the Maxima "assume" database to make a definitive recommendation, but
the assume database needs to be able to be easily extended & contracted in a manner similar to a Lisp "a-list".
Thus, we would like to do something like the following:
(let ((new-db (extend-db ***newassumptions*** ***existingdb***)))
... evaluate expressions in context of new-db ...
)
After the end of the "let" expression, we're back to using the ***existingdb***, and "new-db" disappears & is garbage-collected.
At 08:36 AM 2/8/2013, Robert Dodier wrote:
>On 2013-02-06, Rupert Swarbrick <rswarbrick at gmail.com> wrote:
>
>> putting logic about real and complex numbers into the feature system
>> is probably Doing It Wrong.
>
>Agreed.
>
>> My proposal is that we allow features to be declared to be
>> incompatible.
>
>I'm against it. I'd like to move any inference out of the declaration
>system. Also, I agree w/ a comment by Stavros that multiple declarations
>should be treated as a conjunction. It is certainly possible that the
>conjunction is identically false, but that isn't a problem in the
>declaration system.
>
>best,
>
>Robert Dodier