Jamie Jamison (niteowl@u.washington.edu) wrote:
: 	I am taking an introductory computer science class at the UW and 
: we are learning the Ada language. So far it's pretty nice.  The language
: seems to lend itself to easy readability, the fact that the language is 
: case insensitive is also a nice factor and the language is incredibly 
: powerful. We're in the second quarter of this class and we're already 
: writing ADT's and generic functions. Ada seems to be pretty neat, strong 
: typing, bounds checking and generic functions are all nice language 
: features, and from what I've seen Ada is a lot better than C.

Yes, Ada is a wonderful language.

: So why is 
: it that the real world, such�as it is, programs in C. Sure, C compilers 
: are cheaper, but it seems to me that the labor that you put into writing 
: and maintaining the programs is the real cost, and I know that, at least 
: for me, writing something in C takes a lot longer than writing something 
: in Ada. So why hasn't Ada caught on? Why aren't people developing 
: applications for it?

You've directed your question to comp.lang.ada.  We love the language.
I'm cross-posting this to several groups for a broader range of opinions.

Most large, commercial customers use Ada in the embedded and safety
critical world.  (See the safety-critical survey posting.)  However,
I think you're addressing the usage of Ada for MIS.

Most who read this group recognize certain structural factors that have
impeded the use of Ada in MIS but generally are wondering when large,
commercial companies with get the idea that Ada is great for MIS systems.

IMHO, the structural factors include (in order):

    1.	operating systems and windowing software were written in C

	 MS-Windows and X/Xt/Motif are written in C.  Thus, it was
	 easier to write software in C the environment.  In addition,
	 Ada 83's built in multi-threading wreaked havoc on the
	 non-reentrant O/S and windowing software.
	 
	 The O/S's and windowing systems are finally catching up
	 to Ada's multi-threading capabilities. In addition,
	 Ada 9X's improved non-Ada interfacing features and the
	 advent of Fresco for Ada 9X Win NT and X11R6 windowing
	 software and CORBA for Ada 9X should remove these problems.

    2.	high price of Ada compilers compared to C/C++/BASIC products

	 I agree with your assessment of the real cost of software,
	 but most companies must realize their return on investment
	 in the first year or two to get project funding.  Thus,
	 they won't pay four times the price of a C++ compiler for
	 an Ada compiler.
	 
	 Fortunately, Ada compiler vendors are pricing their new
	 Ada 9X products competitively with C++ products.  In
	 addition, the free GNAT compiler allows one to learn the
	 language without any up front cost.

    3.	lack of flexibility in the first version of Ada (Ada 83)

	 The lack of program pointers, polymorphism, and inheritance
	 precluded the introduction of extensible tools in Ada 83.
	 Ada 83 was designed for ultimate reliability and
	 maintainability only.
	 
	 Ada 9X adds incredible flexibility.  Thus, Ada 9X addresses
	 these requirements and much, much more. 

None of these reasons exist anymore.  Let's see how large
corporations react in the next couple of years.

... Bill

-- 
e-mail: Bill.Beckwith@ois.com       |    Team Ada
Objective Interface Systems, Inc.   | dist, full O-O
1895 Preston White Drive, Suite 250 | multithreading
Reston, VA  22091-5448  U.S.A.      |    built in