The Battle of the Cultures

The Battle of the Cultures

[cis.ohio-state.edu!news.sei.cmu.edu!ajpo.sei.cmu.edu!wellerd]

After having seen more of Greg's rantings, I begin to ask myself
if we need a little sanity check here.  Greg's comments are
inarguable.  More _people_ are turning to C++ for writing software.
The demand for Ada software is decreasing with the "Peace Dividend",
leaving Ada to fend for itself in the cold, heartless commercial
(or as Greg would say: Non-Mandated :-) world.

The problem can be summarized in three key points (apologies to
any real estate sellers): Culture, Culture, Culture.   Ada invariably
seems to be the language no single developer would pick over C++ without
being forced to (I'm generalizing here.  I'd chose Ada ANY DAY over
C++, but I'm a sick puppy :-).  Further, Ada seems doomed as a "niche"
language, specialized for large applications and obscure hardware
platforms (and, as Greg would point out, until Borland or Microsoft
start selling Ada, it will never be taken seriously in the PC world.
Then again, as I would say: "Why take the PC market seriously?" :-).

Now, Ada 9X comes along.  Will it save our hides?  Probably not.
It has features and power beyond what C++ could ever hope to support
(qualification: A lot of features, like tasking for instance, were
left out of C++ deliberately.  This doesn't make Ada "superior" to
C++, but emphasizes the directions and cultures of the two different
language domains).  However, C++ enjoys a tremendous market dominance
amongst individual programmers (and, lest we forget, large software
development teams are made up of these "individual" programmers).  
Further, 9X is quite casually washed away by the noephyte because
it lacks multiple-inheritance.

One interesting phenomenon I have witnessed in the last 4 years is that,
as a programming culture, we tend to tinker at home also.  My experience
has been that developers chose more "accessible" languages, like
Borland C++, for development at home (never mind that Meridian is
competetively priced against BC++, it still isn't as accessible).
This experience at home impacts our choices at work.  Our culture,
primarily composed of programmers (and I mean a different context
from "software engineers".  Probably not the word "hacker", more
like "casual programmer"), is driven more by what empowers THEMSELVES,
rather than what empowers a project (perhaps it would be different
if we were required to stick to a software project for it's life? :-).
Thus, project choices are generally driven by a technical "leader"
serving their own agenda (it is quite possible that that person's
"agenda" is good software engineering, thus leading them toward Ada).

So, are we missing the real point?  That the self-empowering nature
of our commercial market is grinding Ada into dust?  Whether deliberate
or not, I feel we are looking at an unstoppable change in our culture.
We will begin to accept index errors and living inside of debuggers
as casually as ever.  We will howl that Ada is a "better" language, 
and grouse in frustration as more C++ compilers become available.
Some of us, myself included, will not give up hope.  We've used both
languages, we know the advantages and disadvantages of both, and
we realize that, as a language that supports development of reliable
software, Ada is a better choice.  Perhaps in five years, this will
all change.  Perhaps GNAT will change it sooner.  Perhaps.

In any case, you must begin to step outside your fold.  Run with
the wolves.  Engage in heated arguments about the virtues of Ada over
C++ with a C++ organization.  A little "In your face" debate never
hurt anybody.  Besides, I guarantee both sides will learn something.
Reading Greg's posts can get a little depressing (and sometimes 
tiresome :-), so do something about it!  

-- 
-Comments above aren't neceessarily the opinion of the SEI, AJPO, or CAE-Link-
David Weller  |  Have you hugged your DRAGOON lately?
----I'm the Ultimate International Masochist: I speak Ada AND Esperanto!-----

Re: The Battle of the Cultures

[david.c.willett]

In article <1993Jun1.124047.16410@sei.cmu.edu>, wellerd@ajpo.sei.cmu.edu (David
 Weller) writes:
> After having seen more of Greg's rantings, I begin to ask myself
> if we need a little sanity check here.  Greg's comments are
> inarguable.  More _people_ are turning to C++ for writing software.
> The demand for Ada software is decreasing with the "Peace Dividend",
> leaving Ada to fend for itself in the cold, heartless commercial
> (or as Greg would say: Non-Mandated :-) world.
> 
> The problem can be summarized in three key points (apologies to
> any real estate sellers): Culture, Culture, Culture.   Ada invariably
> seems to be the language no single developer would pick over C++ without
> being forced to (I'm generalizing here.  I'd chose Ada ANY DAY over
> C++, but I'm a sick puppy :-).  Further, Ada seems doomed as a "niche"
> language, specialized for large applications and obscure hardware
> platforms (and, as Greg would point out, until Borland or Microsoft
> start selling Ada, it will never be taken seriously in the PC world.
> Then again, as I would say: "Why take the PC market seriously?" :-).
> 
	Sorry, but I've listened to just about all the "Ada is a niche
language" drivel I care to.  I don't understand why people think that 
unless you're using, selling, and evangelizing McLanguge, (one size
fits all... coming to a store near you...) there's something 
wrong.  It happens that Ada is very good at building the kinds of systems it
was designed to build.  Specifically, it is well suited for building complex,
large, platform-independent, long-lived, mission-critical software *SYSTEMS*.
That is a very tough job, requiring a language with particular capabilities.

	Not everyone needs a language like that.  Okay, for them there are
other options.  For the military (U.S. and other nations) there are not.
It happens that Ada emphasizes and enforces discipline that is accepted as
good software engineering.  Non-military Ada users have demonstrated that 
these software engineering discipline has general benefit.  Ada has been 
shown to be successful in building systems similar to the military ones
it was designed for (e.g. Air Traffic Control).

	That's all there is to it, folks.  You build a language to solve a
particular set of problems or to work within a specfic problem domain, and
guess what?....it does well there and in domains close to it.  Does this make
it a "niche" language?  Hardly.  It simply means that the designers did their
job well.  It should be obvious by now that Fred Brooks was right, there is 
no "magic bullett".  In other words, there is no "general purpose" language,
like the one Greg seems to want.

-- 
Dave Willett          AT&T Federal Systems Advanced Technologies
A Theoretical Physicist is one whose existence is postulated to make
the numbers balance, but is never observed in the laboratory.

Re: The Battle of the Cultures

[agate!howland.reston.ans.net!math.ohio-state.edu!uwm.edu!caen!saimiri.pri]

In article <1993Jun1.124047.16410@sei.cmu.edu> wellerd@ajpo.sei.cmu.edu (David 
Weller) writes:
>
>One interesting phenomenon I have witnessed in the last 4 years is that,
>as a programming culture, we tend to tinker at home also.  My experience
>has been that developers chose more "accessible" languages, like
>Borland C++, for development at home (never mind that Meridian is
>competetively priced against BC++, it still isn't as accessible).
>This experience at home impacts our choices at work.  Our culture,
>primarily composed of programmers (and I mean a different context
>from "software engineers".  Probably not the word "hacker", more
>like "casual programmer"), is driven more by what empowers THEMSELVES,
>rather than what empowers a project (perhaps it would be different
>if we were required to stick to a software project for it's life? :-).
>Thus, project choices are generally driven by a technical "leader"
>serving their own agenda (it is quite possible that that person's
>"agenda" is good software engineering, thus leading them toward Ada).

I have both Meridian Ada and Think Pascal on my Mac at home.  They were
(order of magnitude) the same price.

So far I just "hack" (er, that's casually program; yeah, I like that).

But, if I were to attempt to develop a software product, I would much
prefer Ada as a programming language, but would choose Pascal.
Why?  Because of the Think Class Library which gives me a huge headstart
on programming the Mac's interface and access to system calls.
(Meridian has an Ada binding to the Mac toolbox and System 7, but its
not the same as what the Think Class Library provides.)
I suspect a similar situation exists for MS Windows on the PC.

If a near-equivalent library were included with the Ada compiler,
I wouldn't ever have considered the Think Pascal/C family.
For my personal use, it isn't "culture" but tool availability and
maturity which determines my choice of language.

Whether this has anything to do with Mandates or not, I couldn't say. :-)

>-Comments above aren't neceessarily the opinion of the SEI, AJPO, or CAE-Link-
>David Weller  |  Have you hugged your DRAGOON lately?
>----I'm the Ultimate International Masochist: I speak Ada AND Esperanto!-----

In that case, I'm one of the Ultimate Computer Masochists:
I speak Ada on a Mac.

-- 
Terry J. Westley, Principal Computer Scientist
Calspan Corporation, P.O. Box 400, Buffalo, NY 14225
westley@calspan.com
Let's hear it for smart mailers that cut off long signa

Re: The Battle of the Cultures

[Gregory Aharonian]

>If a near-equivalent library were included with the Ada compiler,
>I wouldn't ever have considered the Think Pascal/C family.
>For my personal use, it isn't "culture" but tool availability and
>maturity which determines my choice of language.
>
>Whether this has anything to do with Mandates or not, I couldn't say. :-)
>
>Terry J. Westley, Principal Computer Scientist
>Calspan Corporation, P.O. Box 400, Buffalo, NY 14225

   I'll say it since I am tiresome bore.  The Ada Mandate shelters
the Ada vendors from competition, diminishing the market and driving
up compiler prices, driving down users.  With such a small potential
market, it is uneconomical for anyone to produce Ada bindings for
all of the software technology that you can access through C.  It's
also uneconomical to produce pure Ada libraries - the market is so
small you will never recoup your investment, especially selling to
defense contractors who hate buying software technology.
    If only the comrades setting Ada policy paid attention to these
important free market principles, Ada might be able to be revived from
its coma.

Greg Aharonian
" .  ... .. . .   .... "
-- 
**************************************************************************
 Greg Aharonian
 Source Translation & Optimization
 P.O. Box 404, Belmont, MA 02178