TEAM-ADA Archives

Team Ada: Ada Programming Language Advocacy


Options: Use Forum View

Use Monospaced 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
"Team Ada: Ada Advocacy Issues (83 & 95)" <[log in to unmask]>
Michael Feldman <[log in to unmask]>
Fri, 24 Oct 1997 17:35:20 -0400
Richard Conn <[log in to unmask]>
text/plain; charset=us-ascii
text/plain (136 lines)
I think the Mike raised some valid points, but a basic point
(which is a positive for Ada) is that the theme is shifting
a little from "the language has to be C/C++/Java because
eveyone already knows it" to "the capability we want to have
in a language is platform independence" ... note that by
"language" I also mean "environment" to include GUI interface
development, libraries, databases, and related issues.

Ada may or may not be on the right track for the answer, but
if the masses are starting to be willing to listen to a
capability requirement resolved through some language and
environment as opposed to a language requirement that provides
some capability, the right kind of mind-shift is taking place.
A software engineering theme is emerging, in other words, where our
Operational Concept Description discusses existing facilities
and how they do/do not meet our needs and then looks to desired
capabilities and starts to suggest how to achieve them.

The Steelman requirements, which were recently discussed in Ada
Letters and on the net, are a starting point, but I think they
missed some of the "human factors" capabilities and, in general,
need to be updated to reflect the current state of the industry.
By "human factors" I refer to both the GUI interface issue and
the learning curve issue.


Michael Feldman wrote:
> [said Rick]
> >
> > This is a second opinion, also published with the article
> > I mentioned earlier.  I wonder how this editor would react
> > if he knew about Ada?
> Let's not be too quick to overreact. Java the _language_ is
> reasonably standard; the big fight is not over the language but
> the libraries. The basis of Sun's suit against Mickeysoft is
> (as far as I can tell) over whether Gates will support the full
> set of libs, or substitute some of his own.
> [here Rick quotes Rafe Needleman]
> >
> >              Java has to happen--the world needs a
> >              universal programming language, a lingua
> >              franca of application programming. Visual
> >              Basic is never going to catch on as a universal
> >              programming language, and C isn't suited to
> >              the Web. Just as HTML is supported on every
> > Web client everywhere, Java programs (at least basic ones) will
> > eventually run on any computer, anywhere.
> How about all those HTML variants supported only under version
> p.q.r of one of the clients but not all? Needleman is either
> terribly ignorant of the real state of HTML, or blowing smoke.
> And just watch where HTML goes when the battle between Explorer
> and Navigator gets _really_ fierce.
> We can argue with a straight face that Ada (the language) is
> currently more standard and platform-independent than Java, but
> we are still miles away from the ideal expressed above. We have
> no platform-independent Ada UI APIs beyond the standard libs called
> out in the RM. Even some of the intersting RM stuff - the IS annex,
> for example - is not supported by all Ada 95 compilers. Applications
> dealing with _money_ are where it's at in the commercial sector,
> and we don;t even have a universally implemented money type!
> And in terms of the platform-dependent APIs, Ada is at
> the mercy of Mickeysoft, Sun, Apple, Motif, etc., just as much
> as any other language.
> Let's be realistic re: "write once/run anywhere". For any Ada
> program requiring user interaction beyond Text_IO.Get's and Puts,
> Ada _per se_ has nothing at all to offer but the various thin bindings,
> which are highly platform-dependent. RR's CLAW is somewhat thicker, of
> course, but it is far from platform-independent; it is unabashedly
> for Win 95/NT. AMAF is somewhat thicker, too, but it is Mac-only.
> The Inmet/Aonix Java back end (AppletMagic, ObjectAda) _still_
> depends on version k.l.m of the Java virtual machine. So for this
> to work even minimally, Inmet/Aonix wkill have to commit to
> keeping the back end up to date. And even if they do _that_,
> there is still the question of the Java _libraries_, which
> even the Java folks cannot agree on (see above).
> > Sure, Java will probably never be used to create
> > platform-independent programs that are at once rich and deep,
> > or cross-platform programs with all the features of Microsoft
> > Office and the real-time blitz of, say, Quake. But a
> > programming language that every computer supports has been
> > the dream of application developers as long as there have been
> > application developers. Java will succeed in its mission of
> > "write once, run anywhere" for that reason alone: because the
> > people who write programs will insist on it.
> I think this author is absolutely right that such insistence
> is a precondition for Java's long term success. But with Sun
> and Mickeysoft duking it out in court over the libraries, I am
> not holding my breath for this to happen. The developers will,
> I think, shrug and say "so what else is new?", and write the
> same system-dependent stuff they always have. Or, they will
> just write for the biggest market, leaving other platforms
> unsupported, which means Bill Gates takes over he world.
> I don't mean to sound pessimistic, but if the potentially
> gigantic Java marketplace is having trouble making this happen,
> what do you think the chances are of the Ada community making
> it happen?
> So far, the highly-trumpeted "Ada Common Environment" announced
> with great fanfare by ARA, has come to essentially naught.
> Friends inside the ARA have clued me to the low probability of
> it ever coming to more than naught. No doubt our own vendor
> companies have defensible business reasons for not following through
> on ACE, but the fact remains that we don't have a helluva lot to
> offer beyond our language.
> Mike Feldman
> ------------------------------------------------------------------------
> Michael B. Feldman -  chair, SIGAda Education Working Group
> Professor, Dept. of Electrical Engineering and Computer Science
> The George Washington University -  Washington, DC 20052 USA
> 202-994-5919 (voice) - 202-994-0227 (fax)
> ------------------------------------------------------------------------
> "Cooperate with those who have both know-how and integrity."
>    Fortune cookie, Wu Dynasty, Bethesda, MD, Spring 1996.
> ------------------------------------------------------------------------
> Ada on WWW: or
> ------------------------------------------------------------------------

Richard Conn
mailto:[log in to unmask]       
Opinions expressed are my own and not necessarily those of anyone else.