comp.lang.ada
 help / color / mirror / Atom feed
From: "Paul H. Whittington" <phw@sage.inel.gov>
Subject: Re: How come so much traffic on comp.lang.ada??
Date: 1 Feb 1995 06:27:29 GMT
Date: 1995-02-01T06:27:29+00:00	[thread overview]
Message-ID: <3gn9kh$h4c@mica.inel.gov> (raw)
In-Reply-To: hbaker-3001951716250001@192.0.2.1

hbaker@netcom.com (Henry Baker) wrote:
>
> I've constantly been amazed by the volume of traffic on comp.lang.ada
> relative to comp.lang.c and comp.lang.c++.  On many days the ratio is
> about 1:3, meaning 1 Ada message for every 3 C messages or every 4 C++
> messages.  I have a hard time believing that this is the same as the
> ratio of the number of Ada programmers to C/C++ programmers.  (One
> glance at computer selection at Barnes & Noble bookstore tells a
> completely different story.)
> 
> I can only speculate on the reasons for this:
> 
> 1.  The Ada language itself is so complex that no one can understand it
> without asking a lot of questions.
> 
> 2.  Ada people work on harder problems which require more consultation.
> 
> 3.  Usenet is getting only a fraction of the C and C++ questions -- the
> Compu$erve and Airheads Online forums handle most of their questions.
> 
> 4.  Ada people are lonely, and talk to each other to cheer one another
> up (ditto for comp.lang.lisp).
> 
> etc.
> 
> Anyone else have any thoughts as to why this is?
> 
> (Other than because of stupid questions like this one :-) :-)

Being well versed in C, C++, Ada, PASCAL, and MODULA-2 and programming
in all of them on a regular basis I can honestly say that there are
at least four reasons why Ada has not caught on and become more
commercially popular in the USA (It is quite popular in other
countries):

1) Although Ada was one of the first Object based languages to be
standardized internationally, in 1983, the 1983 version of Ada lacked
dynamic polymorphism and inheritance.  

The 1995 version of Ada, now being approved as an international
standard, includes full object-oriented programming support including
encapsulation, polymorphism, inheritance and templates (called 
generics in Ada), not to mention a complete integrated multi-tasking
standard and several other standard support libraries.

2) The available Ada compilers have been expensive to purchase and
lacking in both their execution performance and code generation
quality.

Thanks to the wise use of a million or so of our hard earned US tax
dollars, the quality and capabilities of academic persons the
world over, and the communications capabilities of the Internet,
their is now a GNU Ada95 compiler available for FREE! (ftp://cs.nyu.edu)
The GNAT compiler, as its called, shares the same high quality back
end used by other GNU compilers like C and C++, and runs quite a bit
faster on my P-90 running NT 3.5 than some of the commercial offerings
I've used.

3) The Ada83 version of the language lacked the language support 
required to enable the integration of commercially available application
development support tools like debuggers, linkers, and component 
libraries, as well as the machinery required to interface to todays
modern GUI based OSes (e.g. UNIX/MOTIF, Windows, NT, MAC OS).

Again, the Ada95 version of the language solves all of these problems
and it is now almost a cake walk to interface Ada programs with
support DLLs written MODULA-2, C, C++ and PASCAL written internally 
and purchased from third party vendors.

4) Maybe, and I don't mean to punch anybody in the nose here, until
now program size and complexity has not over stressed the limits of
C and/or C++.  Ada was designed from the outset to support the DoD's
requirements to develop very large and reliable software systems.
Studies done (search for AJPO on the Internet) have shown a marked
imporvement in software quality in Ada projects vs. C and C++.  The
point is well made in the most recent issue of Software Development
magazine (There's an article on using Ada for Windows programming)
where the author suggests that Ada may just be the answer to managing
the proliferation of APIs in the Windows programming arena.

My suggestion is take a good long look at Ada.  Don't dismiss it 
because some liberal C hacker CS prof. you had in the 80's said "Ada 
is a piece of garbage language developed by the bomb builders."
Get a copy of GNAT for your favorite OS, download one of the several
Ada95 tutorials available on Internet, and spend a few hours (days?)
learning the capabilities of Ada.

TTFN <;-|)




  parent reply	other threads:[~1995-02-01  6:27 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
1995-01-31  1:14 How come so much traffic on comp.lang.ada?? Henry Baker
1995-01-31  9:20 ` Jahn Rentmeister
1995-01-31 15:08 ` David Emery
1995-02-01 23:26   ` David Weller
1995-02-02  1:27   ` David O'Brien
1995-02-01  6:27 ` Paul H. Whittington [this message]
1995-02-01 13:49 ` Robert Dewar
1995-02-01 15:35 ` Jules
1995-02-01 23:57 ` Samuel Mize
1995-02-07 18:29   ` Richard G. Hash
1995-02-11 13:48     ` Robert Dewar
1995-02-02 13:13 ` Mats Weber
1995-02-07 15:47 ` Fergus Henderson
     [not found] ` <3gtifn$m2l@theopolis.orl.mmc.com>
1995-02-08  2:53   ` DEAN RUNZEL
  -- strict thread matches above, loose matches on Subject: below --
1995-01-31  6:24 tmoran
1995-02-01  1:27 ` gpetrey
1995-02-01 14:53 CONDIC
1995-02-07 15:32 CONDIC
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox