comp.lang.ada
 help / color / mirror / Atom feed
From: Paul Whittington <paul@srv.net>
Subject: Re: Ada - State of the art tools... lacking
Date: 1997/12/28
Date: 1997-12-28T00:00:00+00:00	[thread overview]
Message-ID: <34A74195.46994669@srv.net> (raw)
In-Reply-To: 01bd1274$01939f40$652da8c0@steved_home


> ... Exactly.  Going with a single vendor solution can cause severe long term
> problems that don't arise with a standard language that is supported by
> multiple vendors (yes optimistic).

Trust me, experience has shown that unless you vigilantly make sure that
yoursource code compiles and executes properly on all potential
compilers/platforms you'll not realize this dream.  It turns out that although
Ada is by far the best language for developing large and complex systems for
multiple platforms do so is FAR FROM AUTOMATIC.  Things to watch out for are
compiler/platform specific pragmas, representation clauses (legal boundries and

element sizes), and various and sundry semantic interpretation differences.


> If we wrote our code for Delphi, there
> is no reason to think that Borland won't make major changes to the language
> next year that breaks our system.

Oh I think there is!  How about 600,000 or more loyal users that depend on this

kind ofthing not happening, and the $24,000,000.00 ($39.99*600,000) in revenue
that this represents to Borland.  I've used TurboPASCAL since 1.0 and Borland
has NEVER done anything even close to what your talking about here.  They would

have to be completely stupid to even consider it!  I think that Delphi is hands

down the best way to develop Windows software, period.  The only thing that
would make it better is if it used Ada instead of ObjectPascal.

What we've done is built our systems using a Model-View-Controller type of
architecture that allows us to implement our long-term, potentially platform
mobile, code in Ada, and use Delphi to get access to easy-to-use user interface

components or very platform specific stuff.  By doing so we can both get a
state-of-the-art product out the door fast, and make a reasonably intelligent
investment in code for the future.

>  If something happens with ObjectAda we
> can move our source code to GNAT.  If something happens with GNAT we can
> move our source code to DDC-I (when they finish their NT Ada 95 product).
> If something happens to DDC-I we can move to Green Hills... and so on.
>
> Oh, and I think the development tools will mellow with age, but right now
> I'm just frustrated.
>
> SteveD

--
Paul Whittington
Whittington Engineering
260 South Woodruff Avenue
Suite 201
Idaho Falls, ID 83401
paul@srv.net






  reply	other threads:[~1997-12-28  0:00 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <01bd102d$fcfbb910$652da8c0@steved_home>
1997-12-24  0:00 ` Ada - State of the art tools... lacking Robert Dewar
     [not found] ` <01bd108a$12194860$baeb649b@freeman>
1997-12-24  0:00   ` Jon S Anthony
1997-12-24  0:00   ` Robert Dewar
1997-12-27  0:00     ` Steved
1997-12-28  0:00       ` Paul Whittington [this message]
1997-12-29  0:00         ` Michael F Brenner
1997-12-30  0:00         ` Jean-Pierre Rosen
1997-12-31  0:00           ` Paul Whittington
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox