Re: Clean Object Class Design -- Circle/Ellipse
Date: Sat, 1 Sep 2001 23:16:22 -0400
Message-ID: <Wvhk7.642$JS1.127412264_at_radon.golden.net>
I am not thinking about implementation details at all. Smalltalk exposes too many of them by making insufficient distinction between values and variables, but that is Smalltalk's flaw.
James A. Robertson wrote in message <3B9067B1.4DDA4ACF_at_mail.com>...
>Bob Badour wrote:
>>
>
>You are thinking way to much about implementation details that <don't
>matter>
>
>
>> James A. Robertson wrote in message <3B8A67F0.CA3FDC42_at_mail.com>...
>> >Bob Badour wrote:
>> >>
>> >> >Since Smalltalk *has* instances that are values (the number
>> >> >5 comes mind)
>> >
>> >This is trickier in Smalltalk. Why? because instances of classes are
>> >objects, and classes (and metaclasses) are also objects. The bottom
>> >line is that in Smalltalk, <everything> is an object.
>>
>> I don't see how that changes anything. Classes are instances of
metaclasses
>> and classes can change over time, which make them variables of
metaclasses.
>> At any point in time, a class has a metaclass value.
>>
>> Not everything in Smalltalk is a variable and not everything in Smalltalk
is
>> a value.
>>
>> >> Hmmm... I have been looking for Smalltalk's definition of the term
>> >> "instance". So far, all of the definitions seem to equate the term
>> instance
>> >> with the concept of variable:
>> >>
>> >> From http://www.wi2.uni-erlangen.de/sw/smalltalk/glossary.html
>> >>
>> >> instance: An object that is a single occurrence of a particular class.
An
>> >> instance exists in memory or external media in persistent form.
>> >
>> >Variables in Smalltalk are nothing more than references to some object.
>> >That object can be anything
>>
>> Named variables in Smalltalk are nothing more than references to
variables.
>> The referenced variables have values, but they cannot *be* values.
>
>--
>James A. Robertson
>Product Manager (Smalltalk), Cincom
>jarober_at_mail.com
><Talk Small and Carry a Big Class Library>
Received on Sun Sep 02 2001 - 05:16:22 CEST