Python scoping

[Quinn Dunkan]

On Fri, 27 Oct 2000 12:07:25 +0200, Alex Martelli <aleaxit at yahoo.com> wrote:
>"Quinn Dunkan" <quinn at pfennig.ugcs.caltech.edu> wrote in message
>news:slrn8vi15h.j5e.quinn at pfennig.ugcs.caltech.edu...
>    [snip]
>> >    Foo: BEGIN
>> >        /* whatever */
>> >    END Foo;
>>
>> Same for dylan, although it's optional there too:
>>
>> define me a new generic method called foo(x is an <integer> type please)
>>     ..
>> end define me a new generic method called foo(x is an <integer> type
>please);
>
>Terminology question: isn't it "generic function" (with "method" being
>a concrete implementation of a g.f. for certain argument types)?  My
>knowledge of Dylan is limited (the "Dylan Programming" book and little
>more) and I'd like to ensure I have understood its terms & architecture...
>thanks!

Yeah, it's part of the new experimental dylan spec.  There was a lot of debate
about whether the new "define me a new ..." form was really preferable to
plain "define", especially given grammatical concerns, but some guy at FunOb
really liked it, so it stayed.

As far as what a generic method is, you'll have to wait until the spec is
finalized.  It's used in the new (rather complicated) 'fuzzy specialization'
concept and involves the 'please', 'maybe', and 'sorta' keywords.

Don't worry about it, it'll be a long time before a working compiler for the
new spec comes out.