Is there any requirement for handling programs of a particular size that
must be met for a compiler to be validated? i.e. could the Aonix
compiler with its limitations on package size and number of units get
validated anyway?

IMO the prohibition on commercial usage that's in the Aonix package is
vague, ambiguous and probably not needed.  They've already crippled the
horse, why do they have to forbid us from putting it in the Derby? It
would be better/simpler if they had some less ambiguous restriction,
like no distribution of compiled code or whatever.  In my work, I never
know what's going to turn out to be commercial or vice versa.

Al


Michael Feldman wrote:
> The README on the PAL is quite poorly worded. I got a note from a
> contact at Aonix who apologized for the confused wording and said
> it would be fixed.
>
> The compiler in question was validated (but watch out for missing
> annexes!), but the version on the net is obviously a somewhat
> crippled student/trial version.
>
> Indeed, the Win 95 version is nearly (or precisely) the same as
> ObjectAda Special Edition 7.1, which is on the CD in Simon
> Johnston's book. My own book contains 7.0; the pub;isher will
> soon start including 7.1 instead.
>
> There are no restrictions on _distribution_ of this system (AFAIK)
> but there are technical and legal restrictions on its _use_.
>
> As Robert Dewar always points out, there is a big difference
> between a public/free packaging of a validated compiler, and
> a validated compiler. Validation is a legal, as well as a
> technical, issue.
>
> I hope this helps - I speak, obviously, neither for Aonix nor
> ACT nor anyone else but me.:-)
>
> Mike Feldman