Re: How to get Ada to “cross the chasm”?

[Norman Worth <nworth@comcastNOSPAM.net>]

Dan'l Miller wrote:
> http://cppdepend.com/blog/?p=994
> 
> “According to Moore, the marketer should focus on one group of customers at a time, using each group as a base for marketing to the next group. The most difficult step is making the transition between visionaries (early adopters) and pragmatists (early majority).”
> 
> Do we all agree that Ada has never quite crossed the chasm (although it almost did during the late 1980s)?  How can Ada cross the chasm?
> 
> How will Ada use one base of programmers/software-engineers as a basis to move onto convincing another group of programmers/software-engineers of Ada's advantages?  Or is the purpose of Ada to preach to its current choir without trying to fill the pews with new parishioners?
> 
I think one of the reasons Ada did not catch on in a really big way was 
the lack of a sponsor.  Yes, DoD was a sponsor in the early days, and 
they did a great job of defining the philosophy and launching the 
language.  But they gave up and mostly abandoned the language.  This was 
in large part because contractors and the corporate world refused to use 
Ada (in part because the programmers thought being labeled as an Ada 
programmer would mean their skills would be unmarketable).  Also, DoD 
sponsorship is a limited niche, you need a few corporate biggies to sign 
on and support the language.  (DoD also sponsored Jovial, an excellent 
language for its time, that persisted into the 1990s but was never 
available outside DoD.)  Several big companies gave it a brief look, but 
decided that the market was too limited or the language was too 
complicated for practical use.  Several universities adopted Ada as 
their primary language for teaching programming, but lack of support 
caused most of them to abandon it.

There are many objections to Ada.  It is a very large language, and is 
quite fussy.  People think that makes it hard to learn and hard to use. 
Those who use it know that you first learn a practical subset and use 
it, learning and adopting additional features as you encounter problems 
that need them.  This is particularly effective in a team programming 
environment where you get peer reviews of your code, a great leaning 
tool.  I have heard complaints about Ada's wordiness.  This is true - 
Ada is almost as verbose as COBOL.  But it is readable, and that 
wordiness also contributes to its reliability.  People think it 
generates poor and inefficient machine code.  Wrong, at least for the 
couple of compilers I've used.  It is particularly more efficient than 
the interpretive languages that are currently popular.  People say it 
makes the development cycle too slow, that it takes too much time to 
write a working program.  In today's get it out the door today, without 
much testing, world, that may be true.  But reliability does count for 
something on the bottom line.  People complain that modern tools that 
make for a productive shop are not available.  They are, but maybe not 
in the style and variety some want.

In general, Ada is just not popular.  It was never given the chance to 
gather a large following.