[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[no subject]


KMP writes:

`` - I feel funny about the use of the word ``contains'' in ``generic
function.'' ''

The reason to use it is that some people have the idea that a
generic function is really just a name and that there are
a set of methods that are associated with that name. A test of this
misconception is to see how they react when you say you want to
save the old definition of a generic function G by grabbing #'G,
dork around with a new definition of G, and then restore the old one.

That is, the word ``contain'' is intended to make you think that when
you pick up a generic function object, the methods ``come along too.''

The word ``contain'' probably implies slightly too much about implementation,
but a generic function acts as if the methods and the method combination were
part of it. 

KMP, would you feel that some phrasing that said that a generic function
was simply a function and could be used exactly the same way would preclude
people from thinking generic functions were amorphous?

Also, note that the current glossary definition is pretty much excerpted from