Re: NOACE- End of the road for Ada?

["Alexander E. Kopilovich" <aek@VB1162.spb.edu>]

Marin David Condic wrote:

> ... letting your competition define the whole game.

I like this expression, it is quite good description of a widespread sin.

> One can argue endlessly about the relative value of this or that 
> syntactic feature, but think about it this way: If you change Ada syntax 
> enough to make it look like the "more popular girl" (Java) then it just 
> becomes indistinguishable from the more popular girl - except it doesn't 
> have her name. For the folks who love Java, Ada can never be enough like 
> Java and attempting to make it so will only thwart more productive 
> efforts. I think Ada just has to say "What do *I* want to be good at?" 
> and concentrate on being the best addresser of that particular market.

That's all right, but this metaphor can be explored another way. Ada can
think: "What are the features that people really appreciate in Java? Perhaps
I have some of them no less than Java and possibly even better, but I just
do not pay enough attention to them and do not exhibit them properly.
And perhaps accentuation of those my features is not harmful at all for my
integrity and style, but is just necessary adaptation to today's enviromnent,
needs and viewpoints."

(One feature of this kind comes immediately - it is area of tasking and
synchronization.)

> ... If Ada wants to 
> get introspective about what might make her better for this market, I'd 
> think she would have to concentrate on stuff external to the language 
> proper - like libraries and development tools.

Keeping my eye on Ada over 10 years as a distant observer, I gradually became
convinced that Ada is not (and can't be) particularly good for libraries that
are intended for general use. And that contrary to the popular (here in CLA)
opinion, it isn't native Ada libraries that can attract users to Ada.

First, there will be much more C++ and Java libraries anyway. Second, quite
often the design of application or its components, and not availability of
libraries is the critical thing.

While it is often not too expensive to write a library for specific purpose
or provide bindings to existing C/C++ library, the proper design with Ada 
may be a challenging task. Ada provide many opportunities for design, but if
one isn't a skilled and experienced Ada programmer and the application isn't
too simple, s/he has little chances to make best use of those opportunities.

So it appears that one of the most interesting and powerful sides of Ada
often isn't benefit user, and even becomes a burden instead of being an aid.

Therefore I think that what is needed is not libraries, but open source
generic whole applications or generic complex components, which provide
reusable designs. (But please don't tell me about Design Patterns - they
do not demonstrate _entire_ design... like promoting saints, but not faith.) 

As for development tools... well, just curious, which tools except IDE, GUI
builders and test tools like AUnit you imagine for Ada? That is, do you
imagine any development tools that aren't just adapted for Ada variants of
tools long used for, say, C++?