ADA to replace UNIX?

[l...@denelcor.uucp]

I've just read an article in the "Data Base Monthly", a newspaper concerning
Data General people (but not an organ of DG itself). Its concerns the "next
five years" of hardware/software trends, and is written by Doug Kaye,
president of Rational Data Systems, and DG Software firm (as far as I can
tell). To quote:

"I expect ADA to replace UNIX and C in the multi-user multi-task
column for two reasons. First, the profit motivations surrounding
ADA-based products are very strong and growing, and second, I think
that within 5 years the world will finally wake up to the fact
that UNIX isn't all its cracked up to be. ...Pretty soon people
will just have to accept that UNIX has serious flaws in the areas
of reliability and security. If anyone tries to solve these
problems, UNIX will respond by exhibiting lousy performance.
UNIX can be small and fast because it is basicly a reckless operating
system.

"UNIX fans also claim that it has a high degree of portability. In
fact, this is true only because of its great lack of structure
which will, in the long run, cause its downfall. ADA, on the other
hand, provides a vastly superior form of portability via packages.
From day one, the whole language and development and execution
environments have been designed for portability. Remember, "C" is
really just a high-level assembler language for the PDP-11."

Well, has he convinced you? It seems to me that he is confusing ADA (a
language and "environment") to UNIX (an operating system). What is this
"structure" he speaks of? Lack of rigid structure in C data (as opposed
to ADA data structure)?

I attended a conference in December of '68 (I believe) where the ADA
environment was discussed for three days. The consensus (it seemed to me)
was that everone there wanted a UNIX environment, perhaps with ADA in
C's place as THE language (the perhaps was real - even then a lot of
attendees weren't sure whether ADA was what they really wanted). Has
that mood changed? I haven't been able to keep up with ADA. Has some new
concept of an operating system for ADA's environment been developed?

As for the intrinsic reliability of ADA, just ask Edsger Dijkstra what he
thinks.

Finally, the swipe (if indeed its a sneer, as it seems it was intended) about
C right at the end leaves me wondering. Shouldn't the tool fit the job?
Who needs a 4 page data declaration to write a simple program to do
some small task, when it can be done in C in 15 lines and then thrown away?

Oh, well, I thought I would put this out and see what everyone else
thinks. I'm interested in what ADA holds for the UNIX community
future. The profit incentive he mentions is certainly real. Comments?

He promises a future article on both UNIX and ADA. I'll digest when it
appears.

Lyle McElhaney
...hao!csu-cs!denelcor!lmc
...brl-bmd!denelcor!lmc