SEIC News Brief, Week Ending November 29, 1996

SEIC News Brief, Week Ending November 29, 1996

[SEIC]

Software Engineering News Brief
Week Ending:  November 29, 1996 

****************************************
ADA VS. THE COMMERCIAL WORLD
Topic: Ada

John Stein Monroe reports in the latest issue of Federal Computer Week 
that Ada remains a vibrant language despite a smaller market share than 
some commercial languages.  Commercial languages are proliferating 
throughout the Department of Defense (DoD), but Ada is still preferred 
for many applications and is second only to COBOL in its use in DoD 
information systems.  In fact, the DoD and its contractors have developed 
more than 50 million lines of Ada code.

Recent developments may prove fruitful to the future of Ada.  Java, the 
immensely popular Web tool from Sun Microsystems, Inc., is highly 
interoperable with Ada.  This commonality has DoD and NASA users 
intrigued about the possibility of "Web-enabling" Ada applications.

Despite the prospect of eventually losing ground to commercial languages, 
Ada's outlook looks bright with new opportunities such as the WWW keeping 
interest in Ada high.

SOURCE:  Monroe, John Stein.  "It's Ada vs. the commercial world - and 
the lady's still slugging."  Federal Computer Week. November 18, 1996. 
v.10, n.34, pp. 32-33.

****************************************
NASA REUSES WITH S.O.R.T.
Topic: Reuse

The Software Optimization and Resue Technology (SORT) is a tool to 
facilitate the adoption of reuse throughout NASA.  SORT objectives 
include the performance of domain engineering on selected domains at 
various NASA centers, the establishment of model-based domain-specific 
reuse for NASA, the transfer and adoption of reuse technologies into 
NASA, and provision of metrics, lessons learned, and other feedback to NASA.

SORT is providing workshops on software reuse and domain engineering 
technology transfer.  Visit the SORT home page at <http://sort.ivv.nasa.gov>.

SOURCE:  ACM Ada Letters.  November/December 1996, v. XVI, n.6, p.39.

**********************************************************************
RPM OFFERS REUSE ENVIRONMENT
Topic: Reuse

Reuse Process Manager (RPM) offers a configurable reuse-based software 
development environment with reuse-based methodologies, tools and 
training.  It is a user-friendly, windows-based environment 
integrating reuse-based methodologies, tools, training and project 
management.

RPM, a formal, reuse-based methodology, includes tools launching, project 
management, customization, reporting, and multimedia training.  Among many 
other features, RPM shows how to evaluate, select, create and reuse the 
following:  application architectures and templates; data and process 
models;  prototypes; utility functions; and objects.

For more information, contact Dr. Carma McClure, Extended Intelligence, 
Inc., 25 E. Washington St., Ste. 600, Chicago, IL 60602, t.312/346-5245, 
f.312-372-7762.  

SOURCE:  ACM Ada Letters.  November/December, v.XVI, n.6, p.39.

********************************************************************
The Defense Information Systems Agency (DISA) Software Engineering 
Information Center (SEIC) "Software Engineering News Brief" is a  
compilation of summaries from software engineering-related   
articles in trade magazines, newsletters and press releases. The DISA 
SEIC welcomes suggestions for and pointers to software
engineering-related articles.  Contact the DISA SEIC at:    
    
info@sw-eng.falls-church.va.us   
    
To subscribe to the "Software Engineering News Brief" electronic 
mailing list, send a message to:    
        listproc@sw-eng.falls-church.va.us    
In the body of the message, write:    
        subscribe newslist <your name>    
To unsubscribe, write:    
        unsubscribe newslist    
No signatures please.

Re: SEIC News Brief, Week Ending November 29, 1996

[Keith Thompson]

In <57ngcv$3c4@ns1.sw-eng.falls-church.va.us> seic@sw-eng.falls-church.va.us (SEIC) writes:
> ADA VS. THE COMMERCIAL WORLD
> Topic: Ada
> 
> John Stein Monroe reports in the latest issue of Federal Computer Week 
> that Ada remains a vibrant language despite a smaller market share than 
> some commercial languages.  Commercial languages are proliferating 
> throughout the Department of Defense (DoD), but Ada is still preferred 
> for many applications and is second only to COBOL in its use in DoD 
> information systems.  In fact, the DoD and its contractors have developed 
> more than 50 million lines of Ada code.

So Ada isn't a commercial language?  I doubt that my employer would agree.

-- 
Keith Thompson (The_Other_Keith) kst@aonix.com <http://www.aonix.com> <*>
TeleSo^H^H^H^H^H^H Alsy^H^H^H^H Thomson Softw^H^H^H^H^H^H^H^H^H^H^H^H^H Aonix
10251 Vista Sorrento Parkway, Suite 300, San Diego, CA, USA, 92121-2706
"SPOON!" -- The Tick

Is Ada a commercial language ? (was: SEIC News Brief...)

[Larry Kilgallen]

In article <E1u040.Ks7@thomsoft.com>, kst@aonix.com (Keith Thompson) writes:
> In <57ngcv$3c4@ns1.sw-eng.falls-church.va.us> seic@sw-eng.falls-church.va.us (SEIC) writes:
>> ADA VS. THE COMMERCIAL WORLD
>> Topic: Ada
>> 
>> John Stein Monroe reports in the latest issue of Federal Computer Week 
>> that Ada remains a vibrant language despite a smaller market share than 
>> some commercial languages.  Commercial languages are proliferating 
>> throughout the Department of Defense (DoD), but Ada is still preferred 
>> for many applications and is second only to COBOL in its use in DoD 
>> information systems.  In fact, the DoD and its contractors have developed 
>> more than 50 million lines of Ada code.
> 
> So Ada isn't a commercial language?  I doubt that my employer would agree.

For years Ada has been priced such as not to be commercially viable.
In the Microsoft Windows arena your employer has broken that trend,
but it is not yet certain whether your employer and other vendors
will "see the light" on other platforms.

Commercial acceptance requires a competitive cost-per-seat figure.
But it also requires a market awareness to include the features of
interest to customers, not all of which are from the Ada95 LRM.
I note with interest that the ACT GNAT web pages indicate the
forthcoming Alpha (only) VMS port requires VMS V7.0.  Although
I use it as a developer, I have never heard of anyone using VMS
V7.0 (yet) in production.  I know a large C-shop using VMS V5.4A
on a cluster of 9 VAX systems for development.

I realize the contract ACT got from DEC may have specified 7.0,
but if that is the basis then this certainly looks like the
civilian equivalent to DoD contracts for a compiler -- build
what one large customer specifies rather than what will be
needed by large numbers of customers.  The fact that one
will eventually be able to get access to the sources and
modify it to be compatible with VMS versions earlier than
those supported by the primary support vendor still does
not particularly encourage widespread customer acceptance.

Of course this same issue of commercial viability is why
the Aonix Windows NT offering is only on the most popular
of the three hardware platforms at present, and why even
GNAT is not scheduled to be ported to VAX/VMS.  Those
who build products for VMS just don't build them Alpha-only,
as much as DEC might like to forget it.  Even DEC doesn't.
Choosing GNAT over DEC Ada for commercial products is
just out of the question at present except for specialized
application domains which require 64-bit addressing or
kernel threads (available only on Alpha).  Choosing Ada83,
which is what DEC Ada offers, for _new_ commercial products
is unlikely.

Larry Kilgallen

Re: Is Ada a commercial language ? (was: SEIC News Brief...)

[Robert Dewar]

Larry says

"Commercial acceptance requires a competitive cost-per-seat figure."

  That is not clear at all. For example Focus and similar 4GL systems are
  FAR more expensive than COBOL development systems on a cost per seat
  basis, often by a factor of 10. Statements like the above are often
  made as though they are tautological truths, but they are not!

Larry says

"I note with interest that the ACT GNAT web pages indicate the
 forthcoming Alpha (only) VMS port requires VMS V7.0.  Although
 I use it as a developer, I have never heard of anyone using VMS
 V7.0 (yet) in production.  I know a large C-shop using VMS V5.4A
 on a cluster of 9 VAX systems for development."

  There are two separate issues here which are a little mixed up. First,
  is there really a market for a VMS port for VAX. It would not take much
  of a market to make it worth our while, but it is not clear that even
  this small market exists. So far, of the possibly VAX interested
  customers, none has been sure of whether they would be interested.
  Some VAX projects are base lined on Ada 83 anyway, and there is no
  point in changing. Others are planning on moving to Alphas soon anyway.
  If there is a sufficient customer base to support VAX/VMS, it is certainly
  a technical possibility

  The other issue is VMS 7.0 on the Alpha. Actually, we have had more
  trouble with not (yet) supporting 7.1, than with users who are not
  planning to switch to 7.x yet. There are good technical reasons for
  not choosing 6.3, since getting the tasking to work on 6.3 would be
  very much harder, due to the less developed state of the threads
  libraries.

  The users who have been interested in this product are generally planning
  to use 7.x for future production (which is what is interesting for a future
  product not yet released!)

  It would certainly be possible to generate a limited capability 6.3
  release if there are interested customers. As always we find we are
  more interested in messages that say "if you had x, we would use it",
  or "we need x, please do it", than the style of messages which say
  "I am sure that if you did x (which I would like to have free), I
  am sure others would buy it" :-)

P.S. we are planning an Alpha/NT release of GNAT, but so far we do not
have firm plans for PowerPC/NT, since we have had no customer interest
in that port.

A status report on the VMS port incidentally is that it is about to go
into field test phase two in a week or two, so the product release
is probably a couple of months off (we plan a full 2.0.1 validation
first).

GNAT-VMS and OS version support

[Larry Kilgallen]

(list of newsgroups cut for this narrower derived topic)

In article <dewar.849989159@merv>, dewar@merv.cs.nyu.edu (Robert Dewar) writes:

> Larry says
> 
> "I note with interest that the ACT GNAT web pages indicate the
>  forthcoming Alpha (only) VMS port requires VMS V7.0.  Although
>  I use it as a developer, I have never heard of anyone using VMS
>  V7.0 (yet) in production.  I know a large C-shop using VMS V5.4A
>  on a cluster of 9 VAX systems for development."
> 
>   There are two separate issues here which are a little mixed up. First,
>   is there really a market for a VMS port for VAX.

Sorry, I did not mean to raise any discussion of GNAT on VAX (although
I was interested in your response).  I merely indicated that a shop was
at V5.4A to show how far back some will stabilize, as that was well
before there even _was_ an Alpha.  The same shop has at least some
machines at Alpha VMS V1.5, which is quite backward.

The reason I care about what version others may run is that it is
hard to sell them software which will not run on their machine.
The good thing about the GNAT approach is that in theory I could
make a version which could build programs compatible with earlier
versions of VMS.  The bad thing is that I prefer to avoid being a
toolmaker unless the issue is forced.  Luckily DEC Ada 83 exists
so I have the option of avoiding Ada 95 features in programs for VMS.

>   There are two separate issues here which are a little mixed up. First,
>   is there really a market for a VMS port for VAX. It would not take much
>   of a market to make it worth our while, but it is not clear that even
>   this small market exists. So far, of the possibly VAX interested
>   customers, none has been sure of whether they would be interested.

It may have been obvious to everyone else, but I was surprised to learn
at TRI-Ada that ACT will not sell support contracts in single-seat sizes.
I understand the business reasoning, and it removes a potential expense
for me, but it does narrow the field of potential supporters for GNAT-VAX
just when used VAXen are becoming quite affordable due to being replaced
by Alpha.  (I believe the presence of a GNAT-LINUX variant admits to the
possibility that someone might pay money for compiler support even though
they scrimped on initial purchase of the underlying infrastructure.)

>   It would certainly be possible to generate a limited capability 6.3
>   release if there are interested customers. As always we find we are
>   more interested in messages that say "if you had x, we would use it",
>   or "we need x, please do it", than the style of messages which say
>   "I am sure that if you did x (which I would like to have free), I
>   am sure others would buy it" :-)

And again, the no-single-seats policy limits participation at that
level, although the potential Alpha market for GNAT is not buoyed
by a plethora of used machines as with VAX.

> P.S. we are planning an Alpha/NT release of GNAT, but so far we do not
> have firm plans for PowerPC/NT, since we have had no customer interest
> in that port.

OK, just to register one comment, from my perspective support for 3.51
as at least a target is critical in any NT development tool, but again
that is from a single-seat prospect, and the ACT folk told me three
seats on separate operating systems does not count :-(.

Larry Kilgallen

Re: GNAT-VMS and OS version support

[Robert Dewar]

Larry said

"The reason I care about what version others may run is that it is
hard to sell them software which will not run on their machine.
The good thing about the GNAT approach is that in theory I could
make a version which could build programs compatible with earlier
versions of VMS.  The bad thing is that I prefer to avoid being a
toolmaker unless the issue is forced.  Luckily DEC Ada 83 exists
so I have the option of avoiding Ada 95 features in programs for VMS."

GNAT for VMS is very specifically targetted at users who need Ada 95,
both for new development, and for upgrading existing Ada 83 code. DEC
is still supporting their Ada 83 product, so as long as you stick to
Ada 83, you can use the DEC compiler.

We certainly understand that it is very often the case that projects
base line early versions of the operating system, but typically these
same projects have also base lined early versions of the compiler, and
are not typically the sites that are interested in moving into the use
of Ada 95 anyway -- a number of VAX users of Ada have the same profile,
they are considering Alphas and considering Ada 95, but do not see a use
for Ada 95 on their existing VAX systems.

There definitely ARE some users (e.g. Ken Garlington) who have indicated
a possible interest in CLA and elsewhere in GNAT for VMS on VAX, so it
may well be that this makes sense, but first we are concentrating on getting
the Alpha version into good shape. It is coming along nicely. For those of

you who were not at Tri-Ada and missed the VMS Demo (not to mention "The
Maiden and the Mandate" :-), we demo'ed a flight simulator written by
Bevin for DEC Ada on VMS, which used all kinds of DEC specific features.
He ported it in a few hours of work, changing 50 lines of code out of
27,000 -- and indeed almost all of those 50 lines would NOT need changing
now (they were mostly address clauses, and version 4 of GNAT will generalize
the acceptance of address clauses).

P.S. Ken mentions our new pricing policy on single seats. We now have a 
base price for 1-3 seats of support. This change was based on our
experience that single seat support contracts were generating a
proportionately higher level of support requirements, and we need to
tune our prices to the time we need to spend to give people good support!

Re: Is Ada a commercial language ? (was: SEIC News Brief...)

[Richard Kenner]

In article <1996Dec3.082243.1@eisner> kilgallen@eisner.decus.org (Larry Kilgallen) writes:
>I realize the contract ACT got from DEC may have specified 7.0,
>but if that is the basis

If you'd asked this question when we were talking at the ACT booth
yesterday (or was it Thursday?), I would have told you that the 
reason was technical.  I don't know the details (but people who do
were at the booth), but it has something to do with threads packages.

However, it isn't surprising that any product being newly developed
will be based on the then-current version of the OS.  After all, it
certainly has to work with that version anyway, so developing it for
both that and an older version can be significantly extra work.

>Of course this same issue of commercial viability is why
>the Aonix Windows NT offering is only on the most popular
>of the three hardware platforms at present,

That sounds like precisely the the right decision to me!  The usage of
the other two NT platforms is *very* small.  At ACT, we haven't
received any inquiries about GNAT on PPC/NT and only a very slight
level of interest in Alpha/NT.  I imagine Aonix has heard likewise.

>and why even GNAT is not scheduled to be ported to VAX/VMS. 

"scheduled" is a peculiar word to use in this context.

Perhaps you missed it, but the issue of GNAT on VAX/VMS was discussed
at the GNUG meeting at Tri-Ada last Wednesday.  As Robert said, all
the pieces needed from a technical point of view are present.  The
question is the extent of commercial interest.

There are hundreds of different targets out there.  If you restrict
consideration to just those already supported by GCC (including
VAX/VMS), there are still nearly 300.  It's not feasible to support
more than a small fraction of that number, so choices need to be made.

Though there have been a number of inquiries about GNAT on VAX/VMS, so
far everybody (including you, by the way) has come up with a very
small number when asked about the number of seats of support they'd
need.  That may not add up to enough to be commercially viable.  On
the other hand, the cost of this port is significantly lower than many
others, since most of the pieces exist.  My own view is that this is
"right on the edge" of what's worthwhile for ACT to do, but this is
clearly something that ACT needs to further discuss internally, as
well as with Digital and its customers.

On a more general note, the question of "ophan systems" (like "orphan
drugs") is a tough one.  The 6811 has been discussed at length in this
group and other processors (such as those commonly used by segments of
the military) were discussed on the floor of Tri-Ada.  In the present
fiscal realities, it seems unlikely that any one customer would want
to foot the bill for such development and it would be foolhardy for
any company to create such a product "speculatively" in the hope of
attracting customers for it.  The result of this will be that people
who need to use such processors will have to be granted waivers since
no Ada compilers for such will exist.  This is a problem I think AJPO
will have to come up with a solution for over the next couple of
years.

Re: GNAT-VMS and OS version support

[Larry Kilgallen]

In article <dewar.850019210@merv>, dewar@merv.cs.nyu.edu (Robert Dewar) writes:

> There definitely ARE some users (e.g. Ken Garlington) who have indicated
> a possible interest in CLA and elsewhere in GNAT for VMS on VAX, so it
> may well be that this makes sense, but first we are concentrating on getting
> the Alpha version into good shape. It is coming along nicely.

My experience, and I believe that of most people who have ported some
"foreign" program into VMS is that "one architecture first" indeed is
the best and quickest approach.  I do not know the intricacies of the
threading issues, but for most OS issues dealing with VAX-Alpha changes
is absolutely the smaller part of the task.

Larry Kilgallen

Re: Is Ada a commercial language ? (was: SEIC News Brief...)

[Laurent Gasser]

In article <58cd0h$jqj@news.nyu.edu>, kenner@lab.ultra.nyu.edu (Richard Kenner) writes: > However, it isn't surprising that any product being newly developed
> will be based on the then-current version of the OS.  After all, it
> certainly has to work with that version anyway, so developing it for
> both that and an older version can be significantly extra work.

Let me play the devil's advocate.  This attitude is very popular in the
computer market and goes badly against the interest of customers.  

As a customer, willing to get a new application solving one of my needs,
I have to upgrade to the version of the system developers had chosen.  
In many cases, this implies an upgrade of the hardware to keep up with
the extra burden of the new system.  In some cases, the new hardware 
only comes with a more recent version of the system, which is no more
compatible with the software originally selected by the customer... ;-)

I know no other industry so able to drive consumers to consume.

From the producer standpoint, new version of system simply make 
things possible.  A full backward compatibility requires to write
in the application an equivalent of the system functionality.
Not quite worth a second thought.

-- 
Laurent Gasser (lga@sma.ch)
Computers do not solve problems, they execute solutions.

Re: Is Ada a commercial language ? (was: SEIC News Brief...)

[Larry Kilgallen]

In article <32AF0396.BB6@thomsoft.com>, Dave Wood <dpw@thomsoft.com> writes:
> Laurent Gasser wrote:

>> As a customer, willing to get a new application solving one of my needs,
>> I have to upgrade to the version of the system developers had chosen.
>> In many cases, this implies an upgrade of the hardware to keep up with
>> the extra burden of the new system.  In some cases, the new hardware
>> only comes with a more recent version of the system, which is no more
>> compatible with the software originally selected by the customer... ;-)
>> 
>> I know no other industry so able to drive consumers to consume.
>> 
> 
> It seems to me this is just the nature of a rapidly evolving
> technology.  When hardware or software vendors are able to 
> present a persuasive argument that their latest product is
> vastly superior to the one on your desk, you feel compelled
> to buy.  When enough people buy, the preceding product 
> becomes obsolete, requiring everyone else to buy in as well.

The product becomes "obsolete" only if it no longer serves the
purpose of a user.  I am quite happy with the PIM I use, even
though the vendor just went out of business.  The fact that they
could not get enough others to buy does not reduce the usefulness
to me.

The software vendors who must support older versions are those
whose product is _not_ the basis for buying a computer.  Nobody
upgrades their OS just to support a product which is a minor part
of their use of the machine.  Vendors of such products must be
adaptive and look forward to many years of supporting Windows 3.1,
DOS and MacOS System 6 if they want to maintain popularity with
customers.

> In a few years, you will feel compelled to buy new TV sets
> simply because the software (broadcast signal), while still
> backward compatible with your current sets, will be horribly
> constrained in comparison to its digital decendent.

Not based on the programming currently available !!!

> An analogy might be that you are a great fan of ObjectAda
> for Windows (blush!), and even though you think MacOS or
> OS/2 are superior to Windows 95, you aren't willing to
> change your OS because you can't bear to live without
> your ObjectAda.
> 
> Hey, it could happen.

That is where Standards come in.  If ObjectAda provides a
great development environment (the jury is still out, but
at least it doesn't think it is Unix), we can do the bulk
of our work there and then switch to the Tenon environment
for the _real_ builds (so long as we don't need 68K for
now).  Someday perhaps someone will build a MacOS Ada IDE
the equal of Think Pascal.

Hey, it could happen :-)

Larry Kilgallen

Re: Is Ada a commercial language ? (was: SEIC News Brief...)

[Dave Wood]

Laurent Gasser wrote:
> 
> In article <58cd0h$jqj@news.nyu.edu>, kenner@lab.ultra.nyu.edu (Richard Kenner) writes: > However, it isn't surprising that any product being newly developed
> > will be based on the then-current version of the OS.  After all, it
> > certainly has to work with that version anyway, so developing it for
> > both that and an older version can be significantly extra work.
> 
> Let me play the devil's advocate.  This attitude is very popular in the
> computer market and goes badly against the interest of customers.

I agree in the short term, but in the long term everyone is 
served by this kind of rapid evolution.

> As a customer, willing to get a new application solving one of my needs,
> I have to upgrade to the version of the system developers had chosen.
> In many cases, this implies an upgrade of the hardware to keep up with
> the extra burden of the new system.  In some cases, the new hardware
> only comes with a more recent version of the system, which is no more
> compatible with the software originally selected by the customer... ;-)
> 
> I know no other industry so able to drive consumers to consume.
> 

It seems to me this is just the nature of a rapidly evolving
technology.  When hardware or software vendors are able to 
present a persuasive argument that their latest product is
vastly superior to the one on your desk, you feel compelled
to buy.  When enough people buy, the preceding product 
becomes obsolete, requiring everyone else to buy in as well.

I don't think this is unique to computing, it's just more
evident due to the breathtaking rate of change.  For example,
music lovers who disdained CDs and felt they were inferior
to top-quality vinyl, nevertheless have been compelled to
buy CDs and CD equipment just from sheer overwhelming market
tides.  Enthusiasts of BetaMax, convinced their format was
superior, had to give it up and go with the flow.  Same for
8-track.  Having been forced to move to new hardware (CD
players), the customer naturally had to upgrade all software
(CDs) or be left in the dust.

Horse lovers were compelled to buy gasoline once
automobiles became the dominant form of transporation and
quickly made saddles, buggies, and hay obsolete.

In a few years, you will feel compelled to buy new TV sets
simply because the software (broadcast signal), while still
backward compatible with your current sets, will be horribly
constrained in comparison to its digital decendent.

These indirect marketing effects only happen when truly 
compelling improvements are made in one area that require
an upgrade of related components and don't present any
serious drawbacks.  Video disks never really took off 
because while the quality was clearly superior to a VCR, 
the lack of a recording capability meant that people 
could not give up their VCR, and most people aren't 
willing to carry two variants of the same software (a 
problem Mac enthusiasts, or worse, Amiga enthusiasts, 
will fully appreciate.)  

An analogy might be that you are a great fan of ObjectAda
for Windows (blush!), and even though you think MacOS or
OS/2 are superior to Windows 95, you aren't willing to
change your OS because you can't bear to live without
your ObjectAda.

Hey, it could happen.

-- Dave Wood
-- Product Manager, ObjectAda for Windows
-- Aonix - "Ada with an Attitude"
-- http://www.aonix.com

Re: Is Ada a commercial language ? (was: SEIC News Brief...)

[Robert Dewar]

Larry said

"That is where Standards come in.  If ObjectAda provides a
great development environment (the jury is still out, but
at least it doesn't think it is Unix).

As we explain to our potential customers, the issue of whether the
development environment looks like Unix is indeed a significant one
to many people *in both directions*, i.e. some people really do NOT
want a development environment that looks like Unix, but other people
really do.

The new NT GNAT version, based around the Cygnus tool set, indeed works
to make the development environment look very like Unix, so that for
example, Unix makefiles can be used without any change. Of course you
can still generate Windows applications which are not unixy in any way,
we are talking about the *development* environment here.

Larry is no friend of Unix :-) so he won't like that approach, but that's
fine, that's why we have more than one compiler on the market! What we
(ACT) are interested in doing is providing a full featured Ada 95 compiler
on windows NT (as with all supported versions of GNAT, we will support the
whole language, and our NT validation will be 100%, including annex D [no
other compiler has passed all the annex D tests yet for example]) that has
a development environment that makes it as easy as possible to move from
a unix based environment to an NT based envrironment, or to maintain
parallel development efforts on both Unix and NT. This is something that
a lot of people want.

But it is not for everyone ....

Re: Is Ada a commercial language ? (was: SEIC News Brief...)

[Olivier Devuns @pulsar]

In article <32AF0396.BB6@thomsoft.com> Dave Wood <dpw@thomsoft.com> writes:

[snip]

>   An analogy might be that you are a great fan of ObjectAda
>   for Windows (blush!), and even though you think MacOS or
>   OS/2 are superior to Windows 95, you aren't willing to
>   change your OS because you can't bear to live without
>   your ObjectAda.
>

If you are a great fan of ObjectAda are dissatisfied with window$, then
you can switch to ObjectAda for Unix. Hey, that could happen.

--
Olivier Devuns -- odevuns@aonix.com     |          Aonix
   "Views expressed are my own"         |      San Diego, CA
                                        |  http://www.aonix.com

Re: Is Ada a commercial language ? (was: SEIC News Brief...)

[Dave Wood]

Larry Kilgallen wrote:
> 
> In article <32AF0396.BB6@thomsoft.com>, Dave Wood <dpw@thomsoft.com> writes:
> > Laurent Gasser wrote:
> 
> >> As a customer, willing to get a new application solving one of my needs,
> >> I have to upgrade to the version of the system developers had chosen.
> >> In many cases, this implies an upgrade of the hardware to keep up with
> >> the extra burden of the new system.  In some cases, the new hardware
> >> only comes with a more recent version of the system, which is no more
> >> compatible with the software originally selected by the customer... ;-)
> >>
> >> I know no other industry so able to drive consumers to consume.
> >>
> >
> > It seems to me this is just the nature of a rapidly evolving
> > technology.  When hardware or software vendors are able to
> > present a persuasive argument that their latest product is
> > vastly superior to the one on your desk, you feel compelled
> > to buy.  When enough people buy, the preceding product
> > becomes obsolete, requiring everyone else to buy in as well.
> 
> The product becomes "obsolete" only if it no longer serves the
> purpose of a user.  I am quite happy with the PIM I use, even
> though the vendor just went out of business.  The fact that they
> could not get enough others to buy does not reduce the usefulness
> to me.

Sure, but I'm talking about the broader market, not specific
individuals.

> The software vendors who must support older versions are those
> whose product is _not_ the basis for buying a computer.  Nobody
> upgrades their OS just to support a product which is a minor part
> of their use of the machine.  

Typically, not for a single product.  However, when the 
preponderance of new and interesting products are released
only for the new OS, there is a compelling interest for
the broad market to move to the new OS.  The cost of the
move finally is offset by the cost of sticking with all 
the old software.

> Vendors of such products must be
> adaptive and look forward to many years of supporting Windows 3.1,
> DOS and MacOS System 6 if they want to maintain popularity with
> customers.

Vendors are adaptive only to the extent that they can 
make a profit being so.  The cost of supporting the 
old software increases rapidly as more and more people
move to the new OS, while less and less people buy the
old product.  At some point, the vendor will offer to
provide support only at a higher cost, which will drive
even more people to the new product/OS.  Ultimately, 
the cost of maintenance will become more than either
the vendor or the customer are willing to bear, and
the product will be retired by the vendor.  Of course,
the customer may continue to use it.

This happens all the time, and not just in the software
industry.  You'd be hard put to find a Ford dealer 
willing to sell you a service contract on your Model T,
or manufacture spare parts, but you may maintain it 
yourself or find a third party who specializes in 
classic cars for a hefty fee.  Still, for every Model T
on the street you will find millions of Escorts 
(progress being a fleeting and subjective concept.)

You will find exceptions everywhere (I know someone
who still uses a Sinclair ZX-80, for crying out loud!), 
but the real point of interest is the broad trend of 
the market.

> > In a few years, you will feel compelled to buy new TV sets
> > simply because the software (broadcast signal), while still
> > backward compatible with your current sets, will be horribly
> > constrained in comparison to its digital decendent.
> 
> Not based on the programming currently available !!!

Come now, just imagine the advantages of a digitally-
enhanced, wide-screen Geraldo show!

-- Dave Wood
-- Product Manager, ObjectAda for Windows
-- Aonix - "Ada with an Attitude"
-- http://www.aonix.com