[lisp-hh] My MOP talk, some further remarks...

[Pascal Costanza]

Hi,

Sometimes, you know better ways to answer questions that have come up 
during a talk after you've returned home. So in this case.

There was a question wrt to the LispWorks MOP and its introspection 
capabilities. It seems to me now that I have left the impression that 
one cannot inspect a class for its (declared or effective) slots. This 
is not true. In fact, all the slots that you define as usual in a 
defclass form are still available as slot-definition objects and can be 
queried, and so on.

The only difference between the AMOP specification and the LispWorks MOP 
in this regard is that the slot-xxx-using-class generic functions are 
passed a slot-definition metaobject as the third parameter in AMOP, but 
the slot name in LispWorks.

Of course, one cannot query a class for the slots that happen to have 
been added to single instances after they have been created in the 
hash-table-based classes - after all, the purpose of those is to add 
potentially unknown slots after the fact. However, it's possible a) to 
query single instances for slots that they know about and b) to add more 
functionality to the hash-table-based class so that it keeps track of 
all the slots that have been added to instances thereof, and then it 
would be possible to query that (potentially growing) set. You will just 
not be able to query them as slot-definition metaobjects, that's all.

On a different note: There was one remark about 
slot-makunbound-using-class being too liberal with regard to an object's 
hash table. We have been drawn into a discussion whether one way or the 
other is better. However, what I have been missing and is more important 
here is the fact that we don't have to agree here. The cool thing about 
the MOP is that everyone can have the object model that suits them most 
for the given task at hand. I think it's important to stress this once 
more because it takes a little getting used to the fact that there is no 
one single right way to do things.


Pascal

-- 
Pascal Costanza               University of Bonn
mailto:costanza at web.de        Institute of Computer Science III
http://www.pascalcostanza.de  Römerstr. 164, D-53117 Bonn (Germany)