Thank you for trying to help, Lex.

But really, it's not the '@' form that we're having problems with.

It's the whole "moduleA ModuleB" form that is confusing. Hannes' first
message in this thread was saying that this form was add to:

http://minnow.cc.gatech.edu/squeak/2252

... and that it is very confusing. I agree completely.

Let's come up with another example in this form:

aFlap := Etoy Flaps Flaps new.

In this example, Etoy is a module, the first Flaps is a module, but the
second Flaps is a class! This is an abomination!

Honestly, at this point I'd rather see this form go away completely. Too
many forms will lead to confusion.

Chris Becker


-----Original Message-----
From:	squeak-dev-admin at lists.squeakfoundation.org
[mailto:squeak-dev-admin at lists.squeakfoundation.org] On Behalf Of Lex Spoon
Sent:	Thursday, February 21, 2002 5:59 PM
To:	squeak-dev at lists.squeakfoundation.org
Subject:	RE: [Modules] Upper case message names for accessing modules

"Chris Becker" <chb99 at msn.com> wrote:
> How about this alternative for direct module references ...
>
> ellipse := ModuleReference morphic core basic EllipseMorph new.
>
> Even without an understanding of the underlying classes you get a good
idea
> of what is going on.
>

I don't know if it's been mentioned, but if we map messages to variable
names we are using messages in a different than normal.  For example, I
doubt you'll be able to browse to some class and see either "Morphic" or
"morphic" in the list of messages that are available.  For this reason
alone, I'd prefer  having a separate syntax for qualifying a name
lookup.

Note, you would still be able to talk to the module objects if you want
to.  You'd just use regular messages of the kind that will show up in a
browser, e.g.:

	Modules @ #Morphic @ #Core @ #Basic @ #EllipseMorph

Now you can look up #@ and see what is happening, where before you had
to lookup up #doesNotUnderstand:.


-Lex