Re: NOACE- End of the road for Ada?

[Marin David Condic <nobody@noplace.com>]

Alexander E. Kopilovich wrote:
> 
> 
> Well, this sounds all right, except that it raises a disturbing question:
> why that thing was not done already? Or it was done, perhaps several times,
> but was not published and used silently and internally only? It seems that
> the thing can be really useful and really belongs to one of natural Ada
> domains, so it is hard to believe that it was not done... and if it wasn't
> then there should be not-too-obvious obstables or resistances. Perhaps there
> are some problems with patents, licenses etc.? Or something else?
> 
I'm not sure what you are referring to here. The style in which the 
Simulink (and other) tools operate?

Keep in mind that the original intent of things like Simulink was NOT to 
go produce software. It wasn't supposed to be a software design language 
like UML. It was intended for describing things like plants and controls 
for those plants. It was a tool that would let you model things like the 
changes in temperatures, pressures, positions, etc of things in a plant 
and then describe a control that would look at those inputs & apply the 
appropriate control laws. It was NOT originally intended to design code 
to go into a control computer and doesn't tend to support modern 
software design methodologies.

The guys doing this job are essentially thinking in terms of signals 
coming in and signals going out. They're more in line with designing 
some electronic circuits. They came up with techniques back in the days 
of Fortran controls that made things work kind of like an electronic 
circuit. It probably never occurred to anybody working in the field that 
even though they were dealing with essentially an analog kind of thing, 
ultimately they were buiilding digital software and that there may be 
techniques in that field that might be helpful in what they were doing.

I can't describe the entire situation or an entire solution in this 
space. I'm merely picking on this as an example of how there is a 
problem domain out there that has built up a bunch of tools around it 
and that the tools are not especially Ada-ish, yet it is a market that 
Ada ought to want to address (embedded systems). If there were more 
Ada-ish tools out there, perhaps it would be a much more natural 
tendency to opt for using Ada in the corresponding embedded control.

> 
> I think that mechanical CAD area isn't good for Ada - mostly because of 
> already established practice there. They have their "geometry engines" and
> too little remains for Ada.
> 
But perhaps there is some aspect of that which could be better addressed 
in some way by an improved tool? If the improved tool (an end product!) 
was in some way significantly better at helping these folks get to their 
ultimate solutions and the tool was oriented towards Ada, might it not 
find some acceptance.

I don't know because I'm not an ME (I only play one at work ;-) and I'm 
not familiar with their tools. But for those who ARE familiar with the 
domain - perhaps they could think of a better mousetrap?

> I don't think that math and statistics are natural domains for Ada...
> especially statistics, where APL seems to be the most suitable programming
> language.
> 
Ada has fantastic math potential - look at the precise definitions of 
mathematical types & operations and the prefabricated mathematical 
functions built in. What makes APL in any way BETTER? Could it be that 
it just simply provides some intrinsic statistical operations? Is there 
some reason Ada cannot provide some (semi)standard library full of 
statistical functions? (Its going to provide some standard vector and 
matrix functions) Is it lack of capability or lack of will?



> 
> I don't see any particular advantages of Ada for networking.
> 
> 
Perhaps because you're not thinking of it in the same way I am. If Ada 
is a better general purpose programming language than is C, wouldn't 
network applications be better built using Ada? The problem is that 
network apps by their nature don't sit out there all alone. There's all 
sorts of related stuff (like an OS, libraries and other network apps 
that must be communicated with) that's already out there in C or C++ or 
Java. Ada comes to the table late. But IF there were a bunch of Ada 
and/or Ada-ish apps, development tools, surrounding infrastructure, 
etc., then wouldn't it be more natural to develop additional apps in Ada?

It might be difficult to introduce Ada into an existing domain, but 
think in these terms: The uses of networks are expanding and there might 
be some new subset of that in which Ada could become dominant. Java was 
new at one time and it carved out a niche for itself with "portable GUI 
apps". Look at things like Bluetooth - its defining a whole new world. 
What if Ada had been the technology behind Bluetooth? But forget the 
past and look to the future. What might some smart network-geek dream up 
and implement in Ada that might create some new capability and for which 
the surrounding tools would all define an Ada niche?

I guess I'm saying that the Ada world has been thinking a little too 
introspectively - seeing Ada just from a software developer's 
perspective rather than seeing software as inherently a means to some 
other end. Take that technology and apply it to developing some kind of 
products BEYOND software development tools and see if it can carve out a 
money-making niche for itself.

Unless somebody uses Ada to develop some real-world, money-making 
product, Ada is forever condemned to being a "hobbyist" language - by 
definition. Maybe we're too used to counting on the DoD to generate 
those projects and give us that revenue so we're not used to thinking 
about where the money comes from to support our favorite language. Those 
days are gone.

MDC


-- 
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jsf.mil/NSFrames.htm

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

     "'Shut up,' he explained."

         --  Ring Lardner
======================================================================