From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.5-pre1 (2020-06-20) on
	ip-172-31-74-118.ec2.internal
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=3.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.5-pre1
Date: 1 Jun 93 20:52:19 GMT
From: emery@mitre-bedford.arpa  (David Emery)
Subject: Re: How to Make Ada more widely used?
Message-ID: <EMERY.93Jun1155219@goldfinger.mitre.org>
List-Id: <comp.lang.ada>

A subset of Ada like you describe is very close to Modula-2.  (so why
not just use Mod-2?)  And, it makes the resulting subsetted language
very hard to use.  In particular, the behavior of Ada programs without
exceptions is almost impossible to describe.  Removing generics
severely weakens the type model, and makes it almost impossible to do
I/O on user-defined types. 

"Sequential" Ada (Ada without tasks) is a feasable subset, but there's
not that much cost in compiling tasks.  There is a fair amount of cost
in the runtime system.   Removing tasking from the language would
seriously cripple the language for use in embedded applications...

The SIGAda ARTEWG people have been discussing the notion of a standard
interface between a compiler and a RTS for a long time.  Part of the
problem is that this interface is very compiler-specific, and has a
serious impact on the code generation, etc.  And, this interface is
often also very dependent on the target architecture, O.S., etc.  It's
not clear what applications programmers would gain from this
interface, as they should not be calling these services directly.
Instead, they write code that the compiler then translates into calls
to the RTS.   

I am unconvinced that there is a market out there for Ada RTS
developments.  Some people have discussed companies to do this, but
there is not a large set of compiler-independent Ada RTS developers
out there.  In part this is due to the very compiler-dependent aspects
of a good, efficient RTS (for *any* language, not just Ada.)

				dave