Re: Anybody in US using ADA ? New language competition? (long)

["Randy Brukardt" <randy@rrsoftware.com>]

Marin David Condic wrote in message ...
>SIGAda is the right forum for establishing all sorts of things as being
>"Conventional Ada" and you're 100% right about its advantages compared
to
>the vendors or standards body. However, I think it could benefit if it
had
>some acceptance & endorsement by the vendors and standards body so that
>everyone saluted the flag concerning the end results. As a member of
SIGAda,
>I'd support the effort and would even be willing to donate some time if
the
>effort was well focused and the relevant players were signed up to
accept
>the end result. Without those conditions, its chances of succeeding are
>slim.


I can't speak for the vendors, of course. As far as the standards body
goes, that's where the idea originated (inside of WG9). SigAda
volunteered to try to implement it, because it is out of scope for WG9.
The hope is that some of the packages pseudo-standardized by SigAda
could be picked up and added to the standard in the future.

I suspect that getting the vendors to do anything will be a hard sell.
The idea as I recall it (I wasn't at either the recent SigAda meeting or
the most recent WG9 meeting, so the plan may have changed) was to set up
a place where "certified packages" could be accessed. Such software
would already have been tested on a variety of compilers, so it ought to
work out of the box for you. Java supposedly has something similar.

The problem with most of the stuff on the net is that it has only been
tested with whatever compiler the author has. And it often doesn't work
anywhere else. Our experience with Claw is that you really have to
compile and test it on each compiler if it does anything interesting at
all, because some things won't work, and some compilers have bugs that
allow illegal code to be compiled, and so on. If this stuff works well
and is popular, the vendors will pick up on it.

                   Randy.