comp.lang.ada
 help / color / mirror / Atom feed
From: oec@ocsystems.com (Oliver E. Cole)
Subject: Re: object Ada
Date: 1997/02/12
Date: 1997-02-12T00:00:00+00:00	[thread overview]
Message-ID: <1997Feb12.153749.23705@ocsystems.com> (raw)
In-Reply-To: JSA.97Feb10174258@alexandria


   The Intermetrics bindings are currently up for ACE adoption.  They
are thin, on purpose.  Thin bindings need less documentation, as the 
original documentation can be re-used.  

   ACE approved bindings are portable across Ada95 compilers.  As Ada95
compilers get more mature, some things get caught that require
changes to be made, but over time, they will become more stable.  Many
people are already using the Intermetric Motif bindings successfully, 
so such problems as these are isolated in the ACE experience.

   As soon as the ACE (and the ARA) endorses these bindings, they will
be maintained and CM'd a little more aggressively.  I myself would
like things to move faster, but for some rason, I don't always get
my way :-)

   Note that there is no committment from the ARA members to drop
other bindings.  As the post indicates, others bindings may
be used by particular vendors as well, perhaps to support
backwards compatability.

--oec

Jon S Anthony (jsa@alexandria) wrote:
: In article <32FB8031.18D@ccis.adisys.com.au> Steve Gibson <gibson@ccis.adisys.com.au> writes:

: > Jon S Anthony wrote:
: > > 
: > > 
: > > ????  What's wrong with the Intermetrics X11/Motif bindings for Ada95.
: > > I think these things are even up for "ACE" adoption.
: > 
: > 
: > We made some attempt to compile the Intermetrics X bindings with
: > ObjectAdaV7.0 on HP-UX10.10. Had a significant number of problems not to
: > make it worthwhile. In any case the bindings, so I'm told, are
: > significantly different to what we currently use. (since we are porting
: > our 83 code, that's important). Passed the problem back to the vendor,

  [snip]

: This is big time bad news.  These bindings should be pretty much
: standard Ada95 and so should go through anything supporting the
: Interfaces.C hierarchy.  Admittedely I have not looked that closely at
: them and have not had the need yet to use themm under ObjectAda.  But,
: having Aonix moving over to this AXI binding is a disaster in the
: making.  Why?  Simple: Begins again a myriad set of incompatible
: bindings from a number of sources.  Ada95 can solve this problem and
: there is no reason why there should be more than one of these
: bindings.  It _may_ make some sense to have one _thin_ binding and one
: _thick_ binding (I'm not even convinced of this - there should be
: _one_ _thin_ binding and that's it).  Having more than one after all
: the past lessons "learned" from how _bad_ this is, at this point,
: absolutely INSANE.  It would just be a reason for potential new users
: to _not_ use Ada.

: So, ACE members - what's the deal here?

: /Jon
: -- 
: Jon Anthony
: Organon Motives, Inc.
: Belmont, MA 02178
: 617.484.3383
: jsa@organon.com

-- 
------------------------------------------------------------------------------
Oliver E. Cole                                               oec@ocsystems.com
OC Systems, Inc.          http://www.ocsystems.com/             (703) 359-8165




  reply	other threads:[~1997-02-12  0:00 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
1997-01-20  0:00 object Ada Wasiq Hasan
1997-01-20  0:00 ` Larry Kilgallen
1997-01-27  0:00   ` Juanma Barranquero
1997-01-28  0:00     ` Jerry Petrey
1997-01-28  0:00     ` Kevin Radke
1997-02-10  0:00   ` Jon S Anthony
1997-02-12  0:00     ` Oliver E. Cole [this message]
1997-01-25  0:00 ` Dave Wood
1997-01-27  0:00 ` Craig E. Ransom
  -- strict thread matches above, loose matches on Subject: below --
1998-01-05  0:00 Object ADA AA HOLMES
1998-01-05  0:00 ` Larry Kilgallen
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox