Why is Ada not gaining ground

[knji]

Any Embedded Software Developers on board?

Can someone explain why Ada is not gaining ground despite being recognized twice as an international standard for embedded computing? All literature surveys comparing Ada and C or C++ for embedded systems development praise the former for implementing best software practices.

Also, developing applications in Ada is more cost effective than in C, the surveys say.

Is this because most programmers are 'used' to C?

 

[itsmoked]

I too think it would be great if it replaced C and C++.

I think the main reason is because it is so rigorously proven and tested that a seat ends up costing an obscene amount of money. Last compiler I looked at a seat was like $12k. Shesh.

Then you have inertia. Why do we all use crappy Windoz?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[knji]

I hear you about expensive cross compilers for embedded Ada. Also it looks like very few targets support Ada executables.

However, GNAT GPS is free for native applications.

 

[IRstuff]

Ada was only really popular among the military. Ada promised reusability because the code was strictly typed and documented.


However, the reusability of Ada code isn't really that much more than any other code. Readability and maintainability is rather irrelevant if the code isn't really reusable in the first place.

Code is very different that logic components. A NAND gate is easily definable and testable. Code requirements are quite different. The same piece of code can be specified in very different ways, and there is no way to postively prove that one set of requirements are identical to another. Without that possbility, there's no way to prove that a certain piece of code perfectly satifies another set of requirements.

TTFN

 

[VisiGoth]

Reasons:
1) Easy to use Tools
2) Affordable tools
3) Efficient tools (low memory footprint)
4) A large educated populace in Ada

Small shops and consultants just can't afford the cost of compilers. It is nice to have a language that crosses into the small uP embedded world.

Ada can allow a mass of mediocre programmers to a large chunk of the work if the overall program has super brains setting thing up right at the top.

 

[xwb]

You can get free ada compilers from

http://www.thefreecountry.com/compilers/ada.shtml

Official Ada compilers have to be certified. There are a few hundred tests which they have to pass or fail to get certified. It is to ensure that nothing outside the standard creeps into the language. This leaves big players like MS out of the game because they are forever bastardizing the languages with 'special' features. See where playing with Java got them.

I don't know if it is popular with military apps - it is rather slow because of all the safety checks. So slow that it is quite unusable. I work on military apps and we use C++ and Java.

I know some ideas in other languages have been pinched from Ada but as I haven't looked at Ada since 1980 I can't remember most of the features.

-- for comments in SQL

generics AKA templates in C++, C#, Java

private parts/members in all class based languages: this caused quite a lot of laughter when it was first published

 

[IRstuff]

The whole question of Ada "gaining ground," is really quite moot. Ada hasn't really been invoked for military products in more than 10 years.

At this point, Ada is essentially dead, other than as a academic curiousity.

TTFN

 

[itsmoked]

Interesting! I actually thought they were still specing it for military stuff!

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[xwb]

If you've ever heard the term "Can't get the staff", it applies to Ada. It is difficult to get staff who know Ada since there are not many jobs out there which use it. If the company collapses, it isn't easy finding another job if Ada is the only language you know.

 

[IRstuff]

There were never that many Ada programmers to begin with. As with the (in)famous MIL-STD-1750A processor architecture, and the VHSIC initiative, the attempt by the military to chart their own technological course in the 1980's proved to be a dead end. At that specific time in history, military electronics had fallen to less than 5% of the total electronics market, and programmers concentrated on things that would afford them the best advantage in the overall marketplace. Being an Ada programmer meant that you would have been stuck in aerospace and defense, since none of the commercial companies saw much benefit in Ada.



TTFN