AdaZ Validation?

AdaZ Validation?

[Walter Dence]

     AdaZ does not appear on the Nov90 or Dec90 validation lists.  In
addition it does not appear as being in line to be validated.  I for
one am extremely unhappy about unvalidated Ada compilers.  Why have
Ada if there is no discipline?

Re: AdaZ Validation?

[MUNTS PHILLIP A]

In article <9012201150.AA05505@NSWC-WO.ARPA>, wdence@NSWC-WO.NAVY.MIL (Walter Dence) writes...
> 
>     AdaZ does not appear on the Nov90 or Dec90 validation lists.  In
>addition it does not appear as being in line to be validated.  I for
>one am extremely unhappy about unvalidated Ada compilers.  Why have
>Ada if there is no discipline?

     The first page of the manual, "Summary of Changes" lists as the last
item: "Validated under ACVC version 1.11."

     On the other hand, it didn't come with a validation certificate like 
another compiler I bought.  On the third hand, I have only examined a small
sample (2) of compilers so I don't know whether the certificate is normal or
not.

     You won't want to use AdaZ for production anyway.

Philip Munts N7AHL
NRA Extremist, etc.
University of Alaska, Fairbanks

Re: AdaZ Validation?

[Richard Pattis]

In article <9012201150.AA05505@NSWC-WO.ARPA>, wdence@NSWC-WO.NAVY.MIL (Walter Dence) writes:
> 
>      AdaZ does not appear on the Nov90 or Dec90 validation lists.  In
> addition it does not appear as being in line to be validated.  I for
> one am extremely unhappy about unvalidated Ada compilers.  Why have
> Ada if there is no discipline?

AdaZ is a product that uses the standard Meridian compiler, so I'm nor sure
if AdaZ (the product) must be validated independently of the compiler it
uses.  Am I missing something?

Rich Pattis

Re: AdaZ Validation?

[Ashok P. Nadkarni]

In article <1990Dec21.132834.3779@ims.alaska.edu>, ftpam1@acad3.alaska.edu (MUNTS PHILLIP A) writes...
>     You won't want to use AdaZ for production anyway.

Why not ? I'm thinking of buying this compiler for serious use (not just a
learning tool). Is there something I should be aware of :-?

/Ashok Nadkarni

Re: AdaZ Validation?

[Michael Feldman]

In article <14229@june.cs.washington.edu> pattis@cs.washington.edu (Richard Pattis) writes:
>
>AdaZ is a product that uses the standard Meridian compiler, so I'm nor sure
>if AdaZ (the product) must be validated independently of the compiler it
>uses.  Am I missing something?
>
>Rich Pattis

No, Rich, you haven't missed anything. I've got it straight at this point:
the problem is that AdaZ is not the compiler but rather the "package."
The compiler and linker are in fact Meridian Ada 4.1, which appears indeed
on the validation list for December 1990. One of the certificate numbers is
900909W1.11034. The name AdaZ has been up in the air anyway, as regular
readers of this group know. Have no fears: the compiler inside AdaZ is
duly validated under ACVC 1.11.

Perhaps everyone's difficulty comes from the fact that AdaStudent, Meridian's
plain vanilla version 4.0 for students, was not validated and couldn't be
because of missing chap. 13 material. At that time, this was Meridian's
way of distinguishing the products. No more. Their way of distinguishing
current products is that AdaZ is real-mode only. 

Jim Smith of Meridian tells me that indeed they haven't been including a
photocopy of the validation certificate, but will start doing so again as
its lack is causing folks to switch on their flamethrowers.

I will post a copy of the December 1990 list. Note that the list is
shorter than the November one, because it contains only 1.11-validated
compilers. An odd turn of events is that the expiration of 1.10
certificates occurred all at once (for the first time); I am told by
reliable sources that the Wright-Pat AVF and the vendors are working madly
to finish a bunch of pending 1.11 validations.  Recall that under the new
rules, ALL validation certificates for ACVC version K expire one year after
the expiration of that version. 1.11 went into effect 2 months late, but
1.10 still expired on Dec. 1, 1989. So all the 1.10-validated compilers
turned into pumpkins on Dec. 1, 1990. Ah, bureaucracy...

Hope this helps to damp the flames. Back to technical matters.

Mike Feldman

Re: AdaZ Validation?

[MUNTS PHILLIP A]

In article <18443@shlump.nac.dec.com>, nadkarni@ashok.dec.com (Ashok P. Nadkarni) writes...
> 
>In article <1990Dec21.132834.3779@ims.alaska.edu>, ftpam1@acad3.alaska.edu (MUNTS PHILLIP A) writes...
>>     You won't want to use AdaZ for production anyway.
> 
>Why not ? I'm thinking of buying this compiler for serious use (not just a
>learning tool). Is there something I should be aware of :-?

     I should have qualified my comment.  If you are going to do any kind of
tasking or real time work, you won't want to use it.  I can't yet say what its
basic code efficency is, but the size of the executables generated is not
encouraging.

Philip Munts N7AHL
NRA Extremist, etc.
University of Alaska, Fairbanks