Oracle FAQ Your Portal to the Oracle Knowledge Grid

Home -> Community -> Usenet -> comp.databases.theory -> Re: Clean Object Class Design -- Circle/Ellipse

Re: Clean Object Class Design -- Circle/Ellipse

From: Richard MacDonald <>
Date: Mon, 10 Sep 2001 04:13:19 GMT
Message-ID: <z5Xm7.18708$>

"Bob Badour" <> wrote in message > "Richard MacDonald" <> wrote in message news:<kyym7.15569$>...
> > "Bob Badour" <> wrote in message
> >
> > > "Richard MacDonald" <> wrote in message
> > news:<fOkm7.15779$>...
> > > > "Bob Badour" <> wrote in message
> > > >
> > > > > "Richard MacDonald" <> wrote in message
> > news:<r1il7.7908$>...
> > > > >
> > > > > > You're just bashing OO
> > > > > > languages.
> > > > >
> > > > > If that is the prejudice with which you entered and the conclusion
> > > > > with which you leave, I can accept that.
> > > >
> > > > I formed my prejudice while reading your posts.
> > > > Is this a dismissal or are you going to address your error.
> > >
> > > What error? Ever assuming you had any intention of communicating?
> > >
> > > Declaring class "VALUE" does not make any variable a value.

> >

> > Meaning what? The above does not address your original error,
> > or if it does, its not precise enough for a single interpretation.
> >

> > You said that all class instances are variables.
> Yes, all instances are variables.

> > Well, there
> > are Smalltalk objects that are values.
> Yes, but object values are not instances.

Well, there are Smalltalk objects that are instances that are values.


> > Does the above mean you still think they're actually variables?
> I never confused the two; although, I know Smalltalk does.

Quite successfully.

> That was my
> principal criticism, if you recall.

I don't recall it like that. I just recall you being incorrect, although reading ahead I see you choose to make your definitions to suit your taste.


> > Are you trying to say that the Smalltalk Integer instance 5 is
> > actually a variable, even though it acts just like (is
> > indistinguishable from ) a value?
> The integer value 5 is a value. An instance, however, is a variable.
> It does not act like a value because one can reference it, and unless
> restricted, one can change it. An instance of type integer, or integer
> variable if you prefer, can contain the value 5.

One cannot change the Smalltalk instance 5. There is no containment.

> Since all values are self-identifying, one can only reference a value > by its value.

As in the Smalltalk instance 5.

> Of course, that does not make the value 5 an instance.

There is a leap here that isn't justified. Care to fill in how you get to the "Of course". If it is "easy to show that. . .", please show that.

> Even if
> Smalltalk confuses a variable with a value, a variable is not a value.

Duh. But irrelevant.

The Smalltalk 5 is a value *and* an instance. If that is confused, it is a very successful confusion. You have shown no reason why this cannot be. You're just restating that it doesn't fit *your* paradigm.

BTW, what does a relationist know about instances anyway? If all is value or variable, where did instance come from?


> > If so, that would be a little
> > awkward, since value/variable is either/or, as you so adamantly
> > claim.
> When one understands the concepts of value and variable, one finds
> nothing awkward about it at all.

One does. One also understands that instances can be values, since one uses a language that allows this with no problems whatsoever. One is wondering when someone else will address this, rather than simply saying one is confused.


> > And you've chosen to ignore my question again: What do you call
> > the supertype of both Variable and Value?
> Do you mean in the fatuous example where you used the names
> syntactically without regard for any semantics?

Smalltalk would be fine. Superclass Object has a subclass called Integer. Instances of Integer are values. Or perhaps I need to be precise: These instances are *indistinguishable* from values. Object has other subclasses that are variables. Since Object is the superclass of both Variables and Values, I'm wondering what you would call Object?

> While a value has types, the concept of value is not a type, and it > has no supertype.

Why did you meta-step into the "concept of"? Actually, never mind. I'd be happy for a single reason (not circular which is all I've seen so far) why an instance cannot be a value.


> > > If you want to drag shapes, I suggest you simply change the value of a
> > > shape variable.
> > >
> > > You brought your prejudices with you before you ever read one of my
> > > posts.
> >

> > Oh I have lots of prejudices like we all do. I'm just exercising a
> > particular set now since your high horse is flimsy and you
> > deserve to be taken down a notch.
> What you call my high horse is not at all flimsy. You expose much
> about yourself and your motives in the above statement. You expose,
> for instance, that you have no interest in learning or in
> communicating -- instead, you have an interest in attacking and
> destroying. I find little merit in your goals.

I'll apologize the second I see you actually acknowledge you've learned anything from anyone else.


> > But I'm genuinely interested
> > in the answer
> You have already admitted that you have no such interest. You have
> explicitly stated that you want to "take [me] down a notch".

The first sentence does not follow from the first.

> You
> cannot succeed because my position is well thought, learned, valid and
> supported by thousands of years of incremental developments in logic.

And in my puny few decades if there is one thing I know for sure its that people who can actually write such a ludicrous sentence like you just did have a huge gaping hole somewhere somehow. I wonder where it is. . .is it that your ideal world simply doesn't exist?

> It is extremely difficult to take down we who sit on the shoulders of
> giants. You certainly won't succeed by trying to destroy what little
> meaning remains in the words you use so haphazardly.

This is *really* funny. I think I'm getting to know you personally.

> Someday, somebody might topple the giants in a single stroke of
> genius, but such events are exceedingly rare in science and
> mathematics. While still rare, more frequently another giant climbs on
> top, and we get to scramble to an even higher perch.

Yup, but sometimes the people who were sitting up there in the clouds often look really stupid for missing something that was right under their noses but they refused to look because they were so sure they were unassailable.

> > though becoming less hopeful you'll say anything
> > useful.

> I don't think you will find anything particularly useful for your
> stated goal. Better men than you have tried and failed.

I'm sure that is so. Perhaps a better man will be actually able to phrase the question so you cannot wiggle out of answering it: I'm really dying to see you *prove* that an instance cannot be a value. Your arguments suck so far. Received on Sun Sep 09 2001 - 23:13:19 CDT

Original text of this message