Re: NOACE- End of the road for Ada?

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

I personally think your criticism is well thought out and makes some 
valid points. Irrational exuberance and rose colored glasses will not 
save Ada or make it more relevant. I work in a DoD related field and I 
can see the customers I have packing their bags and moving on to other 
languages. I can try to influence that decision towards Ada, but they 
are not in a position to spit into the wind and utilize a language 
without much following in the general computing world unless there is 
some compelling reason. It is difficult to find compelling reasons to 
offer them when all the economics tend to get stacked against Ada.

That said, let me offer this: It doesn't help to be negative about it, 
nor does it help to spend hours worrying about whether or not someone 
likes you. If one gets stuck in a rut of saying "Its all hopeless!!!" 
then ipso facto, it becomes hopeless. If one sits around all day 
thinking "Why doesn't anyone like me? What can I do to get people to 
like me?" it is similarly self defeating. You'll never get everyone to 
like you and trying will only expend your efforts in a bunch of futile 
dodges. While we're at it, being a Pollyanna about it ("Everything with 
Ada is WONDERFUL in my little pastel colored, unicorn infested, rainbow, 
gumdrop world!") doesn't help either. One denies the obvious problems 
and refuses to take action to make it better.

Some suggestions that might actually help:

1) Do things in Ada that you want to do and ignore those who keep saying 
its going to hell in a handcart. Make as much Ada code as possible. Make 
it as useful as possible. Make it as available as possible. The more Ada 
there is out there, the more likely Ada has a sound future.

2) Quit thinking about making more software technology or remaking 
things that already exist in other languages. Dream up things to make 
out of Ada that aren't already done and that address some bigger need. 
We keep thinking in terms of "Here's this cool app someone wrote in C. 
Let me rewrite it in Ada..." Hint: NOBODY CARES THAT IT IS WRITTEN IN 
ADA OR ANYTHING ELSE. They care that it does some job. Reinventing 
network tools or software development tools or any other batch of stuff 
that programmer-geeks like to build doesn't really help if there are 
thousands of them out there already and you have nothing new & 
innovative to offer. Its also a small market compared to the wider world 
of general computer users. Think about it this way: Build a better 
mousetrap. What about a better office suite? What about a better 
accounting package? What about a better statistics tool? What about a 
better structural analysis tool? What about a better "Simulink"? (I'd 
like to see one - and one that generates Ada instead of C) Make some 
better mousetrap that has usefulness beyond the interest of a few 
programmer-geeks.

3) Think about starting a business that makes some useful product with 
Ada as part of its technology. If Ada has so many advantages, it ought 
to be a competitive edge. If you build some sort of commercial software 
or embedded system or other useful product with Ada as a component, then 
you create a market for Ada tool vendors and a job market for Ada 
programmers. The people who program in C or C++ generally are not so 
concerned about the language, per se. They're busy building some cable 
TV network or computational fluid dynamics analysis tool or automotive 
control & diagnostic computer. They sell that stuff and hence have money 
to spend on stuff like compilers and programmers.

4) Don't worry if the DoD guys want to abandon Ada. Their motivation is 
one of economics (primarily). Make Ada economical and they'll come back. 
It was and is a mistake to rely on them to create the market for Ada. 
Ada has to have a utilization in the greater world and not just rely on 
the DoD. If the DoD contractors find that some commercial sector that is 
doing something similar to what they want to do are using Ada as part of 
their toolset, they'll follow.

Think about this for a minute: Say I'm a DoD contractor and I have an 
application that involves graphics in some regard. They look at what 
guys in the private sector are using - the GUI building tools, the 
graphics libraries, etc., and they go do the same. Why? Because they can 
readily get the tools and readily get the people who know how to use 
them and since it is technology out there in the field, it is low risk 
to their project. If Ada had the same tools and libraries & skilled 
people out there in real-world projects, they'd go for that. But their 
objective is not to use Ada, but to get a graphics job done. If some Ada 
fan(s) were building the world's coolest video games in Ada and making 
money doing so & employing people to do it and generating/licensing the 
technology, wouldn't DoD contractors go follow suit?

In the world I live in, I see a bunch of tools that are variations on 
Simulink for designing plant models & control systems. Pretty much 
across the board, these tools are designed to work in a style akin to 
60's era Fortran programs. They pretty much suck stylistically in that 
they don't support most of the software engineering kinds of features 
we've developed since the 60's. But they basically do a job: Someone can 
model a plant and model a control and test it out on a workstation. Then 
the pressure becomes to use the C code (few if any still output Ada) 
they generate to be the actual control code. That has problems, but 
hopefully you can understand that pressure: the model already exists and 
it already works and there is already a test suite, so why not dump it 
into the control & scab up some more C code around it to run the real 
time control?

I can imagine a MUCH superior design & modeling tool that might utilize 
lots of Ada concepts like packages & tasking and sophisticated data 
types and all sorts of stuff. I can imagine a MUCH superior simulation 
environment that would buy numerous improvements in flowing the design 
into the actual box & testing it with greater efficiency. If such a 
system got built in Ada and generated Ada and was based on Ada concepts 
and IF IT HELPED DO SOMEONE'S JOB BETTER than the existing technology, 
it might worm its way into the control software market. Perhaps finding 
users in the automotive and aerospace industries. It might secure a 
niche for Ada. This would be an example of something that was being 
built for reasons other than just to use Ada or make Ada popular. It 
would be getting built to make a better mousetrap and might have the 
beneficial side effect of promoting more Ada use. That kind of thinking 
might get Ada somewhere.

MDC



svaa wrote:

> Stephen Leake <stephen_leake@acm.org> wrote in message news:<mailman.19.1110679175.23655.comp.lang.ada@ada-france.org>...
> 
> Denying reality is not a way to solve problems
> 
> 
>>AdaCore is also growing.
>>
> 
> 
> So borland does, so sun does, so C++ does, so Java does, so others do.
> 
> The fact that AdaCore is growing may only mean that AdaCore is
> collecting all potential Ada custumers that doesn't have any other
> company. Perhaps AdaCore is growing not because a new Ada golden age,
> but at expenses of companies that doesn't work with Ada anymore. The
> market of Ada is so small the there is only room for a few companies.
> When a company stops developing with Ada, the rest of companies, that
> still use Ada, grow a little.
> 
> 
>>>You shouldn't need to read this article to realice that Ada is almost
>>>irrelevant, and that's the trend.
>>
>>Hmm. Perhaps _you_ need to read some _other_ articles :).
>>
>>
>>>Why? How Ada has reach that point of irrelevance? what can be done to
>>>change the trend?
>>
>>Pay attention to what's really going on.
> 
> 
> You live in bubble. You should read another articles too. Not only
> those that tell that Ada is lingering, but those about Java, about
> C++, about C, about PHP about Perl, about Ruby, about pyton...
> 
> This look like Esperanto. I played a little with Esperanto. Thanks to
> internet Esperanto is growing. So what?. If you live inside esperanto
> movement, the Esperanto has a lot of associations, literature etc. You
> see esperanto everywhere, and you conclude that esperanto is quite
> alive. If you look esperanto from outside, esperanto is irrelevant.
> 
> If you program most of time with Ada, work on a company/organization
> that works with Ada, you read articles that support Ada, you go to
> conferences about Ada, accept good news about Ada, but filter bad news
> about Ada. You will conclude that Ada is quite alive.
> 
> If you look Ada from outside, you see that Ada is lingering, that it's
> difficult to find a job for Ada, and if you find it, 99% will be to
> support legacy systems, and probably until they move to another
> language. You can find a thousand tools and libraries for any language
> and choose. For Ada you must go to half a dozen sites/companies and
> take what you find there.
> 
> NOACE movement is a good show of what's going on related to Ada. For
> each new project in Ada with a big hype in Ada related conferences,
> congresses, and websites, you can find 100 projects that are giving up
> Ada silently. In demography, more deaths than births is called
> negative growth.

-- 
======================================================================
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
======================================================================