Re: JGNAT and ObjectAda and ...

[Brian Rogoff <bpr@shell5.ba.best.com>]

On Wed, 20 Dec 2000, Robert Dewar wrote:
> In article
> <Pine.BSF.4.21.0012200806260.20240-100000@shell5.ba.best.com>,
>   Brian Rogoff <bpr@shell5.ba.best.com> wrote:
> > Well, congratulations anyways! This topic was so heavily
> discussed a few
> > years ago (where is John Volan now?) that it seems a bit odd
> that when
> > the "withing problem" is finally vanquished (OK, almost
> vanquished :) that
> > there is so little fanfare.
> 
> 
> Note that a form of "WITH TYPE" has been implemented in GNAT
> for a long time now (it was in version 3.13)

Of course. GNAT leads the way as usual. I'm just glad that soon this
capability will be standard Ada.

> > So, are there any ARG proposals for more direct support for
> Java-style
> > interfaces? I seem to remember some discussion of this, and
> as long as
> > we're letting Java drive Ada development ;-)
> 
> Not that I am aware of!

Too bad. My memory (correct me if I'm wrong) is that some of the players
in the Ada 9X process have said that they don't anticipate an Ada 0X and 
that language improvements would come about through the ARG and "de-facto" 
standardization. While I'm not an OO fanatic it seems to me that one area 
where Ada OOP capabilities could be enhanced would be in the addition of 
something like interfaces, or (OBE :) GNU C++ signatures. 

> By the way, WITH TYPE is equally useful for the C++ case.

Even more important to me, it is also useful for the "pure Ada" case. 
Anyways, great news, its good that Ada isn't standing still, and this 
removes one of my big annoyances with standard Ada. I was never happy
about the "with type" syntax (I hate "use type" as well) but now that the 
decision is made it is time to put aside our differences and rally around 
Ada. Oops, sorry about that... :-P

-- Brian