TEAM-ADA Archives

Team Ada: Ada Programming Language Advocacy

TEAM-ADA@LISTSERV.ACM.ORG

Options: Use Forum View

Use Proportional Font
Show Text Part by Default
Condense Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Sender:
"Team Ada: Ada Advocacy Issues (83 & 95)" <[log in to unmask]>
Subject:
From:
"(No Name Available)" <[log in to unmask]>
Date:
Sun, 5 Oct 1997 00:20:59 -0400
Reply-To:
Parts/Attachments:
text/plain (59 lines)
[log in to unmask] quoted and then wrote:

>> Almost all the reasons Java isn't being adopted are addressed by Ada.  But

we
>need to convince people that Ada solves these problems.

>We must be careful not to over-promise. What's giving companies
>(and educators, BTW) such a hard time with Java is the under-achievement
>of the write-once/run-anywhere ideal. It was overpromised, as is the
>habit in this industry; it doesn't work yet. Maybe it never will,
>because Sun and Microsoft are not about to let each other win.

Certainly one benefit Ada gets from its non-dominance is that it has
_not_ been selected as a battleground between two giants.  Lesser firms
such as Aonix, ACT, etc. are still locked into that old model of adding
features to the software rather than spending all the money on strategic
marketing initiatives :-).

>The BIG thing that Ada does not accomplish (yet) is the existence of
>somewhat thicker API bindings that are tested, readily-available,
>and as "standard" (I use the term with a grain of salt) as the
>language is.
>
>So far, we have a rather fragmented set of thin bindings to platform-
>specific APIs - X, Mac, OS/2, Win 95/NT, Java AWT. This is, obviously,
>a step in the right direction but we have a long way to go. Ada has in
>its favor that it is a standard starting point, with no one company
>dominating the development. Ada 95 has a far better set of standard
>libs in the Annexes than Ada 83, but leaves us hanging when it comes
>to putting GUI-based apps together.

Like a broken record I will again claim that for those who require full
compliance to platform-specific standards (in order to "sell" the result,
be that literally or figuratively) there likely will _never_ be an
adequate
thick binding possible.  A binding which supports MacOS recordability,
Microsoft OLE, OS/2 compatibility back to V2 and Motif V2 custom widgets
when running on each of those platforms would be a massive effort in
search
of a funding source.

The most likely funding is charging Big Bucks to use the binding, which
typically has meant building in dependence on some special executable at
run-time, meaning the result will not run on _somebody_'s favorite Unix,
or the special OSR2 version of Windows 95 that is unavailable in stores,
or the non-public beta version of another operating system I have
installed
on the machine next to me.

=============

For those who do _not_ require absolute compliance to platform-specific
presentation details thick bindings are indeed possible.  Perhaps CLAW
is it (haven't tried it, since I am in the other camp).  But as Mike
warned, do not overpromise the universal utility of Ada.

Larry Kilgallen

ATOM RSS1 RSS2