From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,b0d68c502c0ae6
X-Google-Attributes: gid103376,public
From: Matthew Heaney <matthew_heaney@acm.org>
Subject: Re: Printing Enum Variable Re: Linux World
Date: 1999/03/08
Message-ID: <m3lnh729s6.fsf@mheaney.ni.net>#1/1
X-Deja-AN: 452618696
Sender: matt@mheaney.ni.net
References: <7bfc2n$jl9@dfw-ixnews5.ix.netcom.com> <7bhh26$r7c$1@remarQ.com>
 <36DCAC1F.430E2C5E@aasaa.ofe.org> <7bk4v8$kl8$1@remarQ.com>
 <36DDA761.7B4E8099@aasaa.ofe.org> <7bkrmm$ao1$1@nnrp1.dejanews.com>
 <36DE0007.5236CEA2@aasaa.ofe.org> <7bmmu2$n0h@news1.newsguy.com>
 <m3k8wufayz.fsf@deneb.cygnus.stuttgart.netsurf.de> <7btj81$j0d$1@remarQ.com>
 <m33e3gzw5n.fsf@mheaney.ni.net> <7bvufh$eel$1@remarQ.com>
NNTP-Posting-Date: Mon, 08 Mar 1999 09:00:27 PDT
Newsgroups: comp.lang.ada
Date: 1999-03-08T00:00:00+00:00
List-Id: <comp.lang.ada>

"Michael Young" <nobody@all.org> writes:

> > The advantage of the Ada syntax is that binary operations are
> > symmetrical.  You don't want to have a distinguished receiver, because
> > there isn't one -- there are two.  Indeed, there may be more.
> 
> Please explain the part about symmetry and two or more receivers. Do you
> mean something like:
> "Joe, meet Jane. Jane, meet Joe."
>   versus
> "Meet(Joe, Jane)"

That more or less captures it. 

The most obvious example is a test for equality:

  procedure Op (O1, O2 : T'Class) is
  begin
     if O1 = O2 then ...;


Equality is a binary operation, and dispatches on the tag of O1 and O2.
This is a lot nicer on the eyes than

  if O1.Equal (O2) then ...;


> I might be missing a whole bunch, but none of this seems really
> earth-shattering in terms of expressing a design.

Ada95 is a revision of Ada83.  There was an existing syntax and mind-set
for Ada83 that the designers of Ada95 wanted to preserve.

The C++ approach of adding an object-oriented "corner" to the language,
with a different syntax from other parts of the language, is exactly the
way they _didn't_ want to go.

 
> > The other benefit of the Ada way is that the syntax for manipulation
> of
> > user-defined types is consistent.  You don't want to have Yet Another
> > Notation just because the type happens to be tagged.
> >
> >   Push (Item, On => Stack);
> >
> > Is stack a tagged type or not?  I don't know -- but that's the point.
> 
> Yes, I see your point regarding notations, but what exactly does that
> mean? 

It was intended as an example of what I alluded to above.  There was an
existing syntax for manipulation of user-defined types, and the
designers wanted to preserve this.

> Could you be alluding to: aNumber.PushOnto(aStack)?

No.  Push is a primitive operation of Stack_Type.  It is not a binary
operation.


> One has to ask why Number needs to be aware of Stack at all. 

No, it doesn't.  I think I have only confused you.  (Sorry.)


> In terms of reader-friendliness, symmetry in notation seems more a
> liability than an asset when symmetry does not truly exist. In fact,
> few useful examples come to mind for symmetry in object designs. A
> uses B, but B is oblivious to A.

Push is not a symmetrical operation, and wasn't intended as an example
of one.  "Symmetrical operations" just means that

  A op B 

and 

  B op A 

have the same meaning.  Just like

  Meet (Joe, Jane);

means the same as

  Meet (Jane, Joe);

or

  Marry (Joe, Jane);

means the same as

  Marry (Jane, Joe);

Meet and Marry are operations that apply to two objects, and both
objects have equal importance as far as the operation is concerned.

Saying

  Joe.Marry (Jane);

is misleading.