Re: Anybody in US using ADA ? New language competition? (long)

["Marin David Condic" <mcondic.auntie.spam@acm.org>]

Wes Groleau <wesgroleau@despammed.com> wrote in message
news:PDhS9.214$Z74.885@dfw-service2.ext.raytheon.com...
>
> > sway someone away from what they already use. Also, I've never liked Ada
> > bindings to C libraries because while it is technically possible, itr
always
> > feels like an unnatural act and demands that the Ada developer think
like a
> > C programmer. It would be better if Ada could go down its own path and
do
>
> Like we did with POSIX--declared _one_ exception, and after it's
> raised you examine a global variable to find out what actually
> happened.
>
That may be one technique for hiding the C-ishness of some underlying
binding, but that is by no means covering *all* of the C-ishness one
encounters in bindings. If an effort was being made to generate bindings to
all of the usual things, I'd prefer that they be thought out and made thick
enough to be Ada-ish.

But then, bindings won't save Ada - its just a necessary condition and one
that Ada would forever be playing "catch up" to. If Ada had its own OS, its
own GUI, its own database, etc. - or possibly some operating environment
that simulated these things and hid the underlying implementation
(portability?) - it would be a better way to go.


> I see your point about "leapfrogging the competition" _but_
> anything new has to actually be not only new but self-contained.
>
Maybe yes and maybe no. What I'm suggesting is that we possibly come up with
a whole bunch of brainstormed ideas about what could be done with Ada that
made something *new* - or at least repackaged an old idea in a way that made
it generally useful in a new way.

Example: Java stole the idea of pseudo-code and came up with the "virtual
machine" targeted towards the Internet. That made it possible to build apps
that were executed by web browsers to do some useful things that weren't
possible with web pages alone. People jumped on it because it filled a need.

What if Ada did something similar? Suppose Ada were to go down a path where
the objective was to provide a whole standard GUI environment that was
portable across platforms? Here's something Java can't do very well because
its interpreted: Execute high-res graphic animations. What if a model was
dreampt up that allowed you to download apps from the internet and they were
auto-recompiled for your machine on the fly and they executed in some
high-res graphics environment? (You're already half way there with Gnat) It
might be a natural for the computer game business - perhaps reinventing
their whole business model in the process. (Download games free and run them
uninterrupted for 10 minutes, then get a 2 minute commercial break - just
like TV?)

This idea or some other one of a similar bent would be a good direction for
Ada - provide something *new* that makes Ada the attractive choice for doing
some particular domain of apps.


> If you make a new web server, it still has to interoperate with
> most of the things other webservers work with, or it won't get
> used.  And it isn't being used, is it?  Sure, by Ada fans, but
> that's not the point.
>
Yes, if you pick a domain you want to address and that domain already has
certain standards for data or content or interfaces, you've got to play in
that ballgame. But that still doesn't mean you will find acceptance if
you're just another programming language that isn't the definer of the
domain. To take over that domain, you'll need something *better* in some
significant way that makes people willing to put up with your fundamental
incompatibility in order to get the *extra* that you provide.

> How about TopGraph'X - is it a better X-server?
> How many of them have been installed compared to
> the number of X-servers in (the) other language(s)?
>
I don't know. I'm not familiar with the product. But if all it is is another
X-server, it isn't really providing anything *different* and it will force
the users into incompatibility with all the "real" X-servers written in C,
right? Why do I want that headache - unless perhaps someone is forcing me to
use Ada by mandate and I still want to use an X-server?


> If they're better and they do interoperate,
> why are they not being used?  Anonymity?
>
No mandate? Costs too much? A "me too!" product with no compelling reason to
use it versus the "industry standard" product? You'll have to tell me.

I suspect that one of the reasons Ada has not been gaining widespread
acceptance is that too many of the people involved with Ada think like
engineers rather than think like businessmen. Perhaps if we shift our focus
a little and think like businessmen, we might persuade some businessmen to
adopt Ada and increase their profits.

MDC
--
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jast.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "I'd trade it all for just a little more"
        --  Charles Montgomery Burns, [4F10]
======================================================================