Re: Help with book, "Ada for Pyjama Coders"

[Robert A Duff <bobduff@shell01.TheWorld.com>]

Dennis Lee Bieber <wlfraed@ix.netcom.com> writes:

> On Thu, 9 Aug 2012 13:50:09 -0700 (PDT), Patrick
> <patrick@spellingbeewinnars.org> declaimed the following in
> comp.lang.ada:
>
>  
>> I also think it's sad that there are "new" languages like Go that don't really do much more then Ada 83 + half 95. If Google just picked an Ada subset and supported that, Ada would be the new and coolest thing.
>
> 	One problem -- unless the controlling entities have loosened the
> standard...
>
> 	As originated, the Ada standard mandated NO subsets, and NO
> supersets.

Standards don't actually mandate things.  They pretend to, by saying
"shall" all over the place, but there's no police to come and
arrest you if you implement super- or subsets.  In fact,
gnat does both, in various modes, and so do other Ada compilers.

Standards just define what it means to conform to that
standard.  Conformance to standards is optional.

>...Any compiler system wanting to call the language "Ada" had to
> implement the language as defined in the standard documents..

Sure, it was and is incorrect to call something Ada if it's
not Ada.  The gnat docs, for example, are careful to describe
which switches are needed to make it a pure Ada compiler.

>... (some of the
> "annexes" are optional, but for the most part, they do not define the
> language itself but how the language can be used in advanced
> architectures).

- Bob