Re: ISO Rules on upgrading standards

Re: ISO Rules on upgrading standards

[Larry Kilgallen]

In article <mailman.19.1168757044.18371.comp.lang.ada@ada-france.org>, "Robert Leif" <rleif@rleif.com> writes:

> This is a multi-part message in MIME format.

> color:black'>Is paragraph (e) above relevant to the <st1:City =
> w:st=3D"on"><st1:place
>  w:st=3D"on">Ada</st1:place></st1:City> standard? If so, would it permit =
> small
> annual or biannual additions to <st1:City w:st=3D"on"><st1:place =
> w:st=3D"on">Ada</st1:place></st1:City>?&nbsp;
> I believe that one of the reasons for the lack of market acceptance for =
> <st1:City
> w:st=3D"on"><st1:place w:st=3D"on">Ada</st1:place></st1:City> has been =
> the present
> 10 year cycle for updating the standard.<o:p></o:p></span></font></p>

There has to be _some_ language available for those who prefer stability
to trendiness.

Re: ISO Rules on upgrading standards

[Ludovic Brenta]

Robert Leif writes:
[about multi-part ISO standards]
> Is paragraph (e) above relevant to the Ada standard?

At present, Ada is a single standard, ISO/IEC 8652/1995(E) with
Technical Corrigendum 1 and Amendment 1.  Since Amendment 1, this
single-part standard incorporates and supersedes another standard, ISO
13813:1997 (linear algebra).  So, the answer to your question is no.

See however the recent call for participation in a possible update to
ISO 14519, the Ada POSIX binding.  POSIX is a multi-part standard.
The Rapporteur for the Ada POSIX bindings said that one option would
be to incorporate the Ada POSIX bindings into the Ada standard under
Ada.Interfaces, but he did not say whether splitting ISO 8652 into
several parts was desirable.

> If so, would it permit small annual or biannual additions to Ada?  I
> believe that one of the reasons for the lack of market acceptance
> for Ada has been the present 10 year cycle for updating the
> standard.

In the past, despite being a monolithic standard, Ada has achieved the
effects of multi-part standards by placing ASIS and the linear algebra
in separate standards (ISO/IEC 15291:1999 and ISO/IEC 13813:1997
respectively).  Apparently, that has not helped shorten the revision
cycle for the language definition, so I think that the problem lies
elsewhere.

I think that multi-part standards work best when the working group is
so large that it can split into sub-groups effectively.  I don't think
that that's the case for WG9.  In fact, the main problem in WG9 is
lack of manpower; that's why they decided to publish Ada 2005 as an
Amendment to, rather than a new Edition of the standard.

-- 
Ludovic Brenta.