Ada 95 for an ARM-based bare board?

Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

Hi!

Is there any vendor?

--
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de

Re: Ada 95 for an ARM-based bare board?

[Steffen Huber]

"Dmitry A.Kazakov" wrote:
> Is there any vendor?

Not that I know of. However, as GCC can target ARM, it should be "easily"
possible to build at least a cross compiler.

If you have something like that, please tell me about it ;-)

I am currently still developing software for an ancient operating
system called "RISC OS", and I use an equally ancient port of GNAT
(version 3.03!). I would need a modern GNAT quite soon (based on
GCC 3.1 if possible, because this is the only version ported to
RISC OS) that generates code suitable for the later ARM processor
modes (the 32bit ones - the 3.03 port I use generates code for
the old 26bit modes). Unfortunately, I have not yet the skills
to do a port on my own...

Steffen

-- 
steffen.huber@gmx.de               steffen@huber-net.de
GCC for RISC OS  - http://www.arcsite.de/hp/gcc/
Private homepage - http://www.huber-net.de/

Re: Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

On Fri, 16 Aug 2002 11:14:26 +0200, Steffen Huber
<steffen.huber@gmx.de> wrote:

>"Dmitry A.Kazakov" wrote:
>> Is there any vendor?
>
>Not that I know of. However, as GCC can target ARM, it should be "easily"
>possible to build at least a cross compiler.

Well, building a cross GCC was relatively easy, last time I did it
(5-6 years ago (:-)). Even then one should:

1. build GNAT front-end
2. port its run-time to bare board

I don't feel myself ready for such an adventure. In any case our
customer will not pay for that. Believe me, it was hard enough even to
convince the customer to consider Ada as an option.

>If you have something like that, please tell me about it ;-)
>
>I am currently still developing software for an ancient operating
>system called "RISC OS", and I use an equally ancient port of GNAT
>(version 3.03!). I would need a modern GNAT quite soon (based on
>GCC 3.1 if possible, because this is the only version ported to
>RISC OS) that generates code suitable for the later ARM processor
>modes (the 32bit ones - the 3.03 port I use generates code for
>the old 26bit modes). Unfortunately, I have not yet the skills
>to do a port on my own...

Anyway, if GNAT is an option, then I would turn to GNAT Pro, to have
ACT support. The current public version of GNAT (3.14p) has bugs in
the essential for our design parts of tasking.

--
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

Steffen Huber <steffen.huber@gmx.de> wrote in message news:<3D5CC272.629E243@gmx.de>...
> "Dmitry A.Kazakov" wrote:
> > Is there any vendor?
> 
> Not that I know of. However, as GCC can target ARM, it 
> should be "easily"
> possible to build at least a cross compiler.

Please contact sales@act-europe.fr to learn the latest
status of a GNAT port to the ARM.

Robert Dewar
Ada Core Technologies

Re: Ada 95 for an ARM-based bare board?

[Richard Riehle]

"Dmitry A.Kazakov" wrote:

> Hi!
>
> Is there any vendor?

I would suggest taking a look a the option of
a C-Path Ada compiler such as that provided by
Irvine Compiler Corporation (ICC).  ICC is
experienced in building bare-board compilers
and might be able to respond to this kind of
requirement with modifications to existing technology.

No.  I do not work for ICC and have no money invested
in them.

Richard Riehle

Re: Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

Robert Dewar wrote:

> Steffen Huber <steffen.huber@gmx.de> wrote in message
> news:<3D5CC272.629E243@gmx.de>...
>> "Dmitry A.Kazakov" wrote:
>> > Is there any vendor?
>> 
>> Not that I know of. However, as GCC can target ARM, it
>> should be "easily"
>> possible to build at least a cross compiler.
> 
> Please contact sales@act-europe.fr to learn the latest
> status of a GNAT port to the ARM.

Thank you. I am already in contact. I do hope to get it from ACT.

[ I do not know the background, but for people outside, like me, the 
situation with Ada for ARM looks indeed very strange. AFAIK, customers are 
actively starting to use ARM as a new embedded platform. Noticeably that 
many are ready to drop old code to have a TCP/IP option etc. So it is 
really a good time to say - look we have something better (Ada) for you. 
But ... ]

-- 
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> wrote in message news:<ajqde0$1d7l8m$1@ID-77047.news.dfncis.de>...
> [ I do not know the background, but for people outside, like me, the 
> situation with Ada for ARM looks indeed very strange. AFAIK, customers are 
> actively starting to use ARM as a new embedded platform. Noticeably that 
> many are ready to drop old code to have a TCP/IP option etc. So it is 
> really a good time to say - look we have something better (Ada) for you. 
> But ... ]


What's strange? If we developed ports of GNAT everytime
someone said to us: "You really should develop a port of
GNAT for xxxx, I am sure people would be ready to buy
that product," we would have GNAT ports for all sorts of
strange machines and environmemts, but very few actual
customers for these targets :-)

The situation with ARM is no different from any other GNAT
target, there will be an Ada Core Technologies port of GNAT
to the ARM if and only if there is sufficient commercial
interest to provide the necessary funding for the technical
work involved.

Robert Dewar
Ada Core Tecnologies

Re: Ada 95 for an ARM-based bare board?

[Larry Kilgallen]

In article <ajt1ae$1ebok8$1@ID-77047.news.dfncis.de>, Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> writes:

> Yes, it looked to me as if neither you nor other vendors are convinced that 
> [strange] ARM platform will have any considerable success. Your answer 
> shows that it is true, at least in case of ACT. Exactly this is strange 
> from my perspective. There are lots of Ada ports for PowerPC, there is one 
> even for a radiation-hardened something (:-)). And there is no one for ARM. 
> At the same time no less than 80% of the customers we are dealing with, are 
> using or are planing to use ARM.

> Robert Dewar wrote:

>> The situation with ARM is no different from any other GNAT
>> target, there will be an Ada Core Technologies port of GNAT
>> to the ARM if and only if there is sufficient commercial
>> interest to provide the necessary funding for the technical
>> work involved.
> 
> Yes of course. However note that almost none of our customers will ever 
> show any interest. They know nothing about Ada and even less about ACT.

So why should a vendor believe there is a market for Ada on ARM ?

Certainly there _could_be_ such a market, but vendors must prioritize
on that which will sell.

Re: Ada 95 for an ARM-based bare board?

[Darren New]

"Dmitry A.Kazakov" wrote:
> What is ACT expectance of sharing
> the embedded market between PowerPC, x86 and ARM? X:Y:0?

FWIW, ARM sells more processors every year than Intel does, or so I hear
from qualified sources. :-)

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
   ** http://images.fbrtech.com/dnew/ **

Try our EbolaBurgers...
      So tender they melt in your mouth.

Re: Ada 95 for an ARM-based bare board?

[Richard Riehle]

"Dmitry A.Kazakov" wrote:

> Richard Riehle wrote:
>
> > "Dmitry A.Kazakov" wrote:
> >
> >> Is there any vendor?
> >
> > I would suggest taking a look a the option of
> > a C-Path Ada compiler such as that provided by
> > Irvine Compiler Corporation (ICC).  ICC is
> > experienced in building bare-board compilers
> > and might be able to respond to this kind of
> > requirement with modifications to existing technology.
>
> I sent them a request. So far I have got only a response from their E-mail
> robot that the request is in processing. It looks like a collective
> vacation. (:-))

I just received an email from Dan Eilers at ICC.  I think you may be
hearing from him in the near future.    It is my understanding that
ICC is a small compiler company with a highly focused market.
As with most small companies, they must choose their opportunities
carefully.  Resources are scarce and it does no one any good to
chase wishful thinking.   If they determine that there really is
a market for this kind of product, I suspect they will be happy
to make money from it.

It is always a chicken and egg problem, I suppose.   If the Ada
mandate were still in effect,  there might be more incentive to
build compilers for potential customers.   Now that Ada is
dependent on a commercial model,  the market will dictate
what is available.    The ARM marketplace does not seem
to be clamoring for an Ada compiler.   What is the potential
return on investment for the development of such a compiler?

One advantage we have is that there are avenues for migration
of Ada to new platforms.  GNAT, with its open source, is one
avenue.  The C-Path compilers from ICC and others is another.

When Ada is rediscovered, there will be an urgency to find
compilers for some of the platforms not currently targeted.

"Ah yes, there'll be a brighter day tomorrow."    :-)

Richard Riehle

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> wrote in message news:<ajt1ae$1ebok8$1@ID-77047.news.dfncis.de>...
> Yes, it looked to me as if neither you nor other vendors are convinced that 
> [strange] ARM platform will have any considerable success. Your answer 
> shows that it is true, at least in case of ACT. Exactly this is strange 
> from my perspective. There are lots of Ada ports for PowerPC, there is one 
> even for a radiation-hardened something (:-)). And there is no one for ARM. 
> At the same time no less than 80% of the customers we are dealing with, are 
> using or are planing to use ARM.

We are not in the guessing game, and also not in the business of
believing
those who say without hard evidence that platform xxx will be
important for
Ada. It's quite simple, if we have some real customers who are
seriously interested, then a port will appear. There is some sign that
this may happen
for the ARM, but nothing definite enough to announce yet :-)

> Yes of course. However note that almost none of our customers will ever 
> show any interest. They know nothing about Ada and even less about ACT.

OK, so your customers are indeed irrelevant to us. And therefore what
they
do or do not plan to do is not significant. Yes, sure I know there are
those
who say "you should make a port for xxx, maybe that will generate
interest".
Well that kind of speculation is not for us, but the nice thing about
an
open technology like GNAT is that if someone thinks that, and is
confident
that Ada Core Technologies is missing the boat, then there is nothing
stopping
anyone from building the missing port :-)

> I can undestand that Java platform is not so damn promising to support 
> JGNAT. But just out of curiosity, does ACT really believe that ARM will 
> never ever used as an embedded platform? What is ACT expectance of sharing 
> the embedded market between PowerPC, x86 and ARM? X:Y:0?

The embedded market in general does not necessarily reflect likely Ada
usage.
Right now, most embedded Ada work is either on the power PC or the 68K
with
some limited amount of x86 work.

Robert Dewar
Ada Core Technologies

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> wrote in message news:<ajt1ae$1ebok8$1@ID-77047.news.dfncis.de>...
> Robert Dewar wrote:

> I can undestand that Java platform is not so damn promising to support 
> JGNAT.

I am not quite sure what has got you so upset here that you indulge in such
excessive language :-)

But for the record all *sorts* of people were absolutely convinced that the
JGNAT arena would be an important commercial opportunity for the Ada community :-)

Re: Ada 95 for an ARM-based bare board?

[Florian Weimer]

Darren New <dnew@san.rr.com> writes:

>> What is ACT expectance of sharing
>> the embedded market between PowerPC, x86 and ARM? X:Y:0?
>
> FWIW, ARM sells more processors every year than Intel does, or so I hear
> from qualified sources. :-)

That's a funny statement because Intel sells ARM processors, too. ;-)

(I doubt that ARM Ltd. itself sells any processors, maybe besides
prototypes.  AFAIK, they are more in the licensing business.)

Re: Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

Richard Riehle wrote:

> "Dmitry A.Kazakov" wrote:
> 
>> Is there any vendor?
> 
> I would suggest taking a look a the option of
> a C-Path Ada compiler such as that provided by
> Irvine Compiler Corporation (ICC).  ICC is
> experienced in building bare-board compilers
> and might be able to respond to this kind of
> requirement with modifications to existing technology.

I sent them a request. So far I have got only a response from their E-mail 
robot that the request is in processing. It looks like a collective 
vacation. (:-))

-- 
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de

Re: Ada 95 for an ARM-based bare board?

[Darren New]

Florian Weimer wrote:
> That's a funny statement because Intel sells ARM processors, too. ;-)

You know, as soon as I hit send, I knew someone would call me on my
sloppiness...

I should have said "ARM sells more processor cores than Intel sells
Pentiums". (The processor core being the VHDL description in someone's ASIC,
not a separate chip.)

Sorry. 

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
   ** http://images.fbrtech.com/dnew/ **

Try our EbolaBurgers...
      So tender they melt in your mouth.

Re: Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

Robert Dewar wrote:

> Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> wrote in message
> news:<ajqde0$1d7l8m$1@ID-77047.news.dfncis.de>...
>> [ I do not know the background, but for people outside, like me, the
>> situation with Ada for ARM looks indeed very strange. AFAIK, customers
>> are actively starting to use ARM as a new embedded platform. Noticeably
>> that many are ready to drop old code to have a TCP/IP option etc. So it
>> is really a good time to say - look we have something better (Ada) for
>> you. But ... ]
> 
> What's strange? If we developed ports of GNAT everytime
> someone said to us: "You really should develop a port of
> GNAT for xxxx, I am sure people would be ready to buy
> that product," we would have GNAT ports for all sorts of
> strange machines and environmemts, but very few actual
> customers for these targets :-)

Yes, it looked to me as if neither you nor other vendors are convinced that 
[strange] ARM platform will have any considerable success. Your answer 
shows that it is true, at least in case of ACT. Exactly this is strange 
from my perspective. There are lots of Ada ports for PowerPC, there is one 
even for a radiation-hardened something (:-)). And there is no one for ARM. 
At the same time no less than 80% of the customers we are dealing with, are 
using or are planing to use ARM.

> The situation with ARM is no different from any other GNAT
> target, there will be an Ada Core Technologies port of GNAT
> to the ARM if and only if there is sufficient commercial
> interest to provide the necessary funding for the technical
> work involved.

Yes of course. However note that almost none of our customers will ever 
show any interest. They know nothing about Ada and even less about ACT.

I can undestand that Java platform is not so damn promising to support 
JGNAT. But just out of curiosity, does ACT really believe that ARM will 
never ever used as an embedded platform? What is ACT expectance of sharing 
the embedded market between PowerPC, x86 and ARM? X:Y:0?

-- 
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de

Re: Ada 95 for an ARM-based bare board?

[Ole-Hjalmar Kristensen]

Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> writes:

> Larry Kilgallen wrote:
> 

<snip>

> > Certainly there _could_be_ such a market, but vendors must prioritize
> > on that which will sell.
> 
> Surely. My question was why the priorities look so. As far as I understood 
> the strategy is simple - sit and wait until customers come to us.
> 

And preferably pay up front for the develoment of the compiler....  OK
strategy if you sell few expensive development systems, probably not a
good strategy to penetrate the mass market.

> -- 
> Regards,
> Dmitry Kazakov
> www.dmitry-kazakov.de

Re: Ada 95 for an ARM-based bare board?

[Ole-Hjalmar Kristensen]

Ole-Hjalmar Kristensen <oleh@vlinux.voxelvision.no> writes:

> Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> writes:
> 
> > Larry Kilgallen wrote:
> > 
> 
> <snip>
> 
> > > Certainly there _could_be_ such a market, but vendors must prioritize
> > > on that which will sell.
> > 
> > Surely. My question was why the priorities look so. As far as I understood 
> > the strategy is simple - sit and wait until customers come to us.
> > 
> 
> And preferably pay up front for the develoment of the compiler....  OK
> strategy if you sell few expensive development systems, probably not a
> good strategy to penetrate the mass market.
> 
> > -- 
> > Regards,
> > Dmitry Kazakov
> > www.dmitry-kazakov.de

P.S. I'm not thinking of any specific vendor here, just experiences
from the past.

Re: Ada 95 for an ARM-based bare board?

[Larry Kilgallen]

In article <7vlm70cqcz.fsf@vlinux.voxelvision.no>, Ole-Hjalmar Kristensen <oleh@vlinux.voxelvision.no> writes:
> Ole-Hjalmar Kristensen <oleh@vlinux.voxelvision.no> writes:
> 
>> Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> writes:
>> 
>> > Larry Kilgallen wrote:
>> > 
>> 
>> <snip>
>> 
>> > > Certainly there _could_be_ such a market, but vendors must prioritize
>> > > on that which will sell.
>> > 
>> > Surely. My question was why the priorities look so. As far as I understood 
>> > the strategy is simple - sit and wait until customers come to us.
>> > 
>> 
>> And preferably pay up front for the develoment of the compiler....  OK
>> strategy if you sell few expensive development systems, probably not a
>> good strategy to penetrate the mass market.
>> 
>> > -- 
>> > Regards,
>> > Dmitry Kazakov
>> > www.dmitry-kazakov.de
> 
> P.S. I'm not thinking of any specific vendor here, just experiences
> from the past.

Of course in the case of GNAT, anyone can be a vendors, so some of those
who feel the opportunities abound should put their own money at risk in
such a venture.

Re: Ada 95 for an ARM-based bare board?

[Randy Brukardt]

Ole-Hjalmar Kristensen wrote in message
<7vptwccqfs.fsf@vlinux.voxelvision.no>...

>And preferably pay up front for the develoment of the compiler....  OK
>strategy if you sell few expensive development systems, probably not a
>good strategy to penetrate the mass market.


The companies that tried to penatrate the mass market with Ada are
either out of business (Supersoft, Artek, Meridian), have changed
product lines to move away from the mass market (RR Software), or
changed business models (Aonix). A lot of people lost a lot of money
trying to do that, and it has always failed. Here, I have to agree with
Robert: if you think there is some way to make a viable business out of
mass market Ada, feel free to take your own money and start a company to
do it. Otherwise, stop whining.

I've spent 22 years trying to sell Ada to the masses, and primarily all
I have to show for it is a substantial debt. You can argue that RRS has
done this or that wrong (and that's probably true), but I dare anyone
here to do it better on the money we had available. Unless some Ada
stalwart wins a $100 million Powerball jackpot, I very much doubt that
will change.

            Randy Brukardt
            R.R. Software, Inc.

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

Ole-Hjalmar Kristensen <oleh@vlinux.voxelvision.no> wrote in message news:<7vlm70cqcz.fsf@vlinux.voxelvision.no>...
> > And preferably pay up front for the develoment of the compiler....  OK
> > strategy if you sell few expensive development systems, probably not a
> > good strategy to penetrate the mass market.

Not necessarily, ACT has developed many ports of GNAT at our expense, but
only when we know there is a real market out there (well a couple of times
we have skipped that step and got burned).

Indeed we are not trying to "penetrate the mass market". That's not our
business area. Our expertise is in providing high level support from high
level competent engineers. We don't see that as scaling to a mass market.

The world seems full of people who are happy to spend lots
of energy in trying to get Ada Core to move in that direction, but sadly
none of these people seem willing to spend their own energy persuing a
market that they are sure we are missing!

Robert Dewar
Ada Core Techonologies

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> wrote in message news:<ajvqd1$1eh7ep$2@ID-77047.news.dfncis.de>...

> True, but should not it? From my point of view it could be a very good time 
> for Ada there. People are starting to use multitasking, wireless 
> communication etc in their embedded systems. To have a 32-bit controller is 
> no more cosidered too expensive. Shouldn't we use this situation?

Sounds like you believe this is a great business opportunity 
So why not persue it?

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> wrote in message news:<ajvqd1$1eh7ep$2@ID-77047.news.dfncis.de>...
> Robert Dewar wrote:

> [How somebody could have a "hard evidence" that something will be?  (:-))] 
> Anyway to be important /= to be important to Ada. AFAIK, ARM is already 
> important. A minimal requirement for making it important for Ada is to have 
> a compiler for it.

Very simple, if someone comes to us and says: "I want/need an Ada
compiler
and I would like to see you guys do a GNAT port and I am willing to
pay for
the product," then that will get our attention.

If someone says "Well I am sure that if you write a compiler and sell
it inexpensively, then masses of people will flock to buy it," then
frankly
we won't listen. It's not our business area. Companies can't do
everytyhing.

It may be that there is a real business opportunity there, who knows?
At Ada
Core, we are dubious (partly because we have seen more than one
company try
this approach and fail), but competition is all about people having
different
views. When Tom tried to sell low cost copies of the Windows version
of GNAT,
we were happy to cooperate with him, but unfortunately he just did not
find
enough customers to make it worth while.

So if you folks really think ARM is important and that you can take
the embedded world by storm with a GNAT port for the ARM, go ahead. No
one is
stopping you!

Robert Dewar
Ada Core Technologies

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> wrote in message news:<ajvqd1$1eh7ep$2@ID-77047.news.dfncis.de>...

> Unfortunately it is not an option in our case. We simply have not enough 
> resources for that.

Well if you are so sure it is a good business opportunity, you should be
able to get people to provide the resources.

Re: Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

Larry Kilgallen wrote:

> In article <ajt1ae$1ebok8$1@ID-77047.news.dfncis.de>, Dmitry A.Kazakov
> <mailbox@dmitry-kazakov.de> writes:

>> Yes of course. However note that almost none of our customers will ever
>> show any interest. They know nothing about Ada and even less about ACT.
> 
> So why should a vendor believe there is a market for Ada on ARM ?

To sell Ada to them! People I am talking about are not programmers. They 
are working as if they were. (:-)) They are using and planning to use C. I 
was amazed to know how much and how readily some of them pay for a 
pseudo-ANSI C compiler with a disgusting IDE. The situation will not change 
as long as: 

(1) there is no ready-and-easy-to-use products for fairly ignorant people.

(2) these products are incompatible with all sorts of [fancy] tools they 
are accustomed to use (MatLab, LabView etc).

(3) nobody [they respect] talks to and convinces them.

> Certainly there _could_be_ such a market, but vendors must prioritize
> on that which will sell.

Surely. My question was why the priorities look so. As far as I understood 
the strategy is simple - sit and wait until customers come to us.

-- 
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de

Re: Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

Robert Dewar wrote:

> Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> wrote in message
> news:<ajt1ae$1ebok8$1@ID-77047.news.dfncis.de>...

>> Yes, it looked to me as if neither you nor other vendors are convinced
>> that
>> [strange] ARM platform will have any considerable success. Your answer
>> shows that it is true, at least in case of ACT. Exactly this is strange
>> from my perspective. There are lots of Ada ports for PowerPC, there is
>> one even for a radiation-hardened something (:-)). And there is no one
>> for ARM. At the same time no less than 80% of the customers we are
>> dealing with, are using or are planing to use ARM.
> 
> We are not in the guessing game, and also not in the business of
> believing those who say without hard evidence that platform xxx will be
> important for Ada.

[How somebody could have a "hard evidence" that something will be?  (:-))] 
Anyway to be important /= to be important to Ada. AFAIK, ARM is already 
important. A minimal requirement for making it important for Ada is to have 
a compiler for it.

> It's quite simple, if we have some real customers who are
> seriously interested, then a port will appear. There is some sign that
> this may happen for the ARM, but nothing definite enough to announce yet 
> :-)
> 
>> Yes of course. However note that almost none of our customers will ever
>> show any interest. They know nothing about Ada and even less about ACT.
> 
> OK, so your customers are indeed irrelevant to us. And therefore what
> they do or do not plan to do is not significant. Yes, sure I know there
> are those who say "you should make a port for xxx, maybe that will
> generate interest".

No. Even so, these customers would not come! Ada should come to them. They 
will use C, C#, Java whatever but Ada. It is a really hard work to get 
them. Yes they are irrelevant in the sense that you cann't get them with no 
effort. Maybe an exaggeration, but remember how DEC considered home users 
irrelevant. Where is DEC now?

> Well that kind of speculation is not for us, but the nice thing about
> an open technology like GNAT is that if someone thinks that, and is
> confident that Ada Core Technologies is missing the boat, then there is
> nothing stopping anyone from building the missing port :-)

Unfortunately it is not an option in our case. We simply have not enough 
resources for that.

>> I can undestand that Java platform is not so damn promising to support
>> JGNAT. But just out of curiosity, does ACT really believe that ARM will
>> never ever used as an embedded platform? What is ACT expectance of
>> sharing the embedded market between PowerPC, x86 and ARM? X:Y:0?
> 
> The embedded market in general does not necessarily reflect likely Ada
> usage.

True, but should not it? From my point of view it could be a very good time 
for Ada there. People are starting to use multitasking, wireless 
communication etc in their embedded systems. To have a 32-bit controller is 
no more cosidered too expensive. Shouldn't we use this situation?

> Right now, most embedded Ada work is either on the power PC or the 68K
> with some limited amount of x86 work.

-- 
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de

Re: Ada 95 for an ARM-based bare board?

[BSCrawford]

In article <5ee5b646.0208211140.53d594e7@posting.google.com>, dewar@gnat.com
(Robert Dewar) writes:

>When Tom tried to sell low cost copies of the Windows version
>of GNAT,
>we were happy to cooperate with him, but unfortunately he just did not
>find enough customers to make it worth while.

Tom who?  When? 
Did he announce this on CLA or the Team Ada list?

Bard Crawford
Stage Harbor Software

Re: Ada 95 for an ARM-based bare board?

[Larry Hazel]

BSCrawford wrote:
> In article <5ee5b646.0208211140.53d594e7@posting.google.com>, dewar@gnat.com
> (Robert Dewar) writes:
> 
> 
>>When Tom tried to sell low cost copies of the Windows version
>>of GNAT,
>>we were happy to cooperate with him, but unfortunately he just did not
>>find enough customers to make it worth while.
> 
> 
> Tom who?  When? 
> Did he announce this on CLA or the Team Ada list?
> 
> Bard Crawford
> Stage Harbor Software
> 
> 

I don't remember his last name, but I bought a copy about 1995 (I think). 
Shortly after that it was no longer being offered for sale.

Larry

Re: Ada 95 for an ARM-based bare board?

[SteveD]

"BSCrawford" <bscrawford@aol.com> wrote in message
news:20020821200840.11247.00000111@mb-df.aol.com...
> In article <5ee5b646.0208211140.53d594e7@posting.google.com>,
dewar@gnat.com
> (Robert Dewar) writes:
>
> >When Tom tried to sell low cost copies of the Windows version
> >of GNAT,
> >we were happy to cooperate with him, but unfortunately he just did not
> >find enough customers to make it worth while.
>
> Tom who?  When?
> Did he announce this on CLA or the Team Ada list?

As I recall it was Tom Griest (not sure I have the spelling right) that
first made a Win32 port of GNAT available.  Before it was available from
ACT.

SteveD

>
> Bard Crawford
> Stage Harbor Software
>
>

Re: Ada 95 for an ARM-based bare board?

[tmoran]

> As I recall it was Tom Griest (not sure I have the spelling right) that
> first made a Win32 port of GNAT available.  Before it was available from
> ACT.
  That sounds right.  The CDROM I have says
   Labtek Corporation
 Entry Level Ada95 SDK
for Microsoft Windows95
    and Windows NT
Version 3.04A July 1996
(c)1996 Labtek Corp. All rights reserved
(c)1985-1996 Microsoft, All rights reserved.
  IIRC the Labtek CDROM also contained all the MS Windows library stuff
necessary to make a legal program with Gnat.
  I forget what it cost - not enough, I guess. ;)
Looking at the Walnut Creek Ada CDROM dated March 1994, there is a GNAT
directory containing subdirectories "VER_167", "VER_173", and "VER_174",
but the README doesn't sound like those were production releases.

Re: Ada 95 for an ARM-based bare board?

[Richard Riehle]

Robert Dewar wrote:

> Not necessarily, ACT has developed many ports of GNAT at our expense, but
> only when we know there is a real market out there (well a couple of times
> we have skipped that step and got burned).

The list of  dead Silicon Valley companies who spread themselves too
thin by failing to concentrate on their core business and their paying
customers is long and depressing.

ACT, and other Ada compiler publishers can do more for Ada by
providing excellent products and service to a few satisfied
customers.   This appears to be the philosophy of ACT, and we
see some wisdom in it.

In time, when an inventory of Ada success stories is available for
public release, those of us involved in tool building, or compiler
building, or training, or consulting, or applications building will
have the satisfaction of our contribution, big or small, to those
successes.

Richard Riehle

Re: Ada 95 for an ARM-based bare board?

[tmoran]

> The list of  dead Silicon Valley companies who spread themselves too
> thin by failing to concentrate on their core business and their paying
> customers is long and depressing.
  OTOH, Intel went off their core semiconductor RAM business to build
a chip for somebody's calculator, and then overengineered that.  Apple
forgot its core business was hobbyists, and Microsoft forgot its core
business was BASIC.  Sticking with your core business leaves you with
nothing more than your core business.

Re: Ada 95 for an ARM-based bare board?

[Juha Valimaki]

"Robert Dewar" <dewar@gnat.com> wrote in message
news:5ee5b646.0208211132.6c283ef0@posting.google.com...
> Indeed we are not trying to "penetrate the mass market". That's not our
> business area. Our expertise is in providing high level support from high
> level competent engineers. We don't see that as scaling to a mass market.
>
> The world seems full of people who are happy to spend lots
> of energy in trying to get Ada Core to move in that direction, but sadly
> none of these people seem willing to spend their own energy persuing a
> market that they are sure we are missing!
>

> Robert Dewar
> Ada Core Techonologies

Robert, I think you could say I am one person trying to push Ada to mass
market. I understand it's nobody's business to say what ACT should do, but I
also hope you see why people are asking ACT to do something: GNAT simply
seems so close to "perfect" product for many tasks, if just there was
business interest. I think many people feel GNAT is property of ACT (no
matter what the copyright says). People simply expect ACT to do something
(It's not any fault of ACT, but just the way things are). Nothing may happen
until ACT does something and same time the free version version makes it
harder for other companies to target the low-end market. For not-so-realtime
and not-so-high-reliability applications Ada sure has advantages over
C++/Java, but the advantage is not large enough to pay for expensive
tools/support.

I still haven't decided if Ada really is worth all the effort... I mean, Ada
sure has nice qualities and I am sure choosing Ada would make me more
productive, but the tool/support costs are so high I really have to
calculate if I will gain something. (Money, time and productivity are of
course related here: with money I can buy productivity and to get money I
have to work to get income and so productivity will decrease.)

The situation would be very different if there was  "GNAT like" tool
available for, lets say, 2000 USD or less for each major platform. No
project/Ada support, just binaries, manuals and bug fixes. That would be an
easy choice! Even better if the company developing that tool said they are
actually interested in supporting "mass market" use of Ada... (To actually
get the "mass market" there would have to be IMO more visibility for cool
Ada projects and maybe 1000 USD price tag (?))

Juha

P.S. On March 2002 Alias|Wavefront lowered the price of their Maya animation
tool for $7500 to $1999, effectively transforming their company from premium
high-end to mass market. I am not suggesting Ada companies should do the
same and I not even sure if the move was good, but it shows the world is
changing.

P.S.II. Games are these days quite heavy in (soft) real time activities and
multitasking. I personally think Ada would be very well suited for games...
if just someone had the muscle to create/sell interesting tools for game
programmers!

Re: Ada 95 for an ARM-based bare board?

[Ole-Hjalmar Kristensen]

"Randy Brukardt" <randy@rrsoftware.com> writes:

> Ole-Hjalmar Kristensen wrote in message
> <7vptwccqfs.fsf@vlinux.voxelvision.no>...
> 
> >And preferably pay up front for the develoment of the compiler....  OK
> >strategy if you sell few expensive development systems, probably not a
> >good strategy to penetrate the mass market.
> 
> 
> The companies that tried to penatrate the mass market with Ada are
> either out of business (Supersoft, Artek, Meridian), have changed
> product lines to move away from the mass market (RR Software), or
> changed business models (Aonix). A lot of people lost a lot of money
> trying to do that, and it has always failed. Here, I have to agree with
> Robert: if you think there is some way to make a viable business out of
> mass market Ada, feel free to take your own money and start a company to
> do it. Otherwise, stop whining.

I'm not in the business of making compilers, and it wasn't whining,
just an observation that from a customers point of view, it has made
sense to seek other solutions. As a customer, I don't really care
about the vendor's motivation for his strategy.  Of course any vendor
is free to pursue his preferred market.

> 
> I've spent 22 years trying to sell Ada to the masses, and primarily all
> I have to show for it is a substantial debt. You can argue that RRS has
> done this or that wrong (and that's probably true), but I dare anyone
> here to do it better on the money we had available. Unless some Ada
> stalwart wins a $100 million Powerball jackpot, I very much doubt that
> will change.
> 
>             Randy Brukardt
>             R.R. Software, Inc.

I do not doubt that it's very difficult to penetrate the mass market
without very deep pockets, and I do not accuse you of having done
anything stupid. It probably is an indication that there is no mass
market appeal in Ada at the moment.

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

"SteveD" <nospam_steved94@attbi.com> wrote in message news:<LmX89.23476$aA.7706@sccrnsc02>...
 
> As I recall it was Tom Griest (not sure I have the 
> spelling right) that first made a Win32 port of GNAT 
> available.  Before it was available from ACT.
 
Yes that's right. And of course there may be someone who
is sure that the conditions are different, or that the
price was wrong, or that the product was premature, and
that everything is different now. Perhaps. Just because
Ada Core does not market something does not necessarily
mean it can't be marketed.

After all, no doubt one can make money selling halloween
costumes, but Ada Core Technologies is not in that business
last time I looked :-) And more seriously, we can't even expect to
cover all viable aspects of the commercial Ada
market. Successful companies specialize in what they are
good at. We consider our strength to be in providing high
level support for mission critical projects, since that's where we see
the major business opportunity. That's a totally different business
from going after a mass market. Even if the mass market is viable,
going after such a market is not something that fits our expertise and
business model.

This kind of specialization is definitely a key to success.
2001-2002 was a difficult period for many (almost all?) high tech
companies, but Ada Core Technologies has prospered during that period.
We have experienced steady growth for this period in both companies
(Ada Core Technologies and ACT/Europe) and we have a dedicated group
of people, who may not be getting rich with stock options,
but they are paid respectable and steady salaries :-)

I have run into a few people who are puzzled by our financial model,
because they don't see how the founders
expect to make big money out of the company. It's a sad
sign of the times that people assume this is the motive
for founding a company, and the founders, who are not
getting Enron-style-rich, but are also paid respectable and steady
salaries are quite content with the situation thank you :-)

Once again, the nice thing about Free Software is that we
don't lock up the market by stopping anyone else from using
or exploiting our technology. If someone thinks they can be
successful selling an inexpensive ARM port of GNAT to the
mass market, then I say go for it!

Robert Dewar
Ada Core Technologies

Robert Dewar
Ada Core Technologies

Re: Ada 95 for an ARM-based bare board?

[Marin D. Condic]

"Juha Valimaki" <juha.valimaki@xdsloy.com> wrote in message
news:ak21k5$lfj$1@phys-news1.kolumbus.fi...
>
> The situation would be very different if there was  "GNAT like" tool
> available for, lets say, 2000 USD or less for each major platform. No
> project/Ada support, just binaries, manuals and bug fixes. That would be
an
> easy choice! Even better if the company developing that tool said they are
> actually interested in supporting "mass market" use of Ada... (To actually
> get the "mass market" there would have to be IMO more visibility for cool
> Ada projects and maybe 1000 USD price tag (?))
>

But then they would likely be canibalizing their existing business. Do you
think that perhaps there might be companies out there reasoning like this:
"I need an Ada compiler and I need support in the form of regular updates
and bug fixes. I don't really need telephone support or rapid turn around on
bug reports, but I really need at least the full, latest&greatest package
with updates. The only way I can get this is with ACT's premium support
package, so I cough up the cash." That pays the freight to keep ACT's head
above water and development of Gnat actively going on.

Now suppose there were a lower tier product out there from ACT. Might the
reasoning be this: "Let some other company pay the premium support package.
I don't really need telephone support or rapid turn around on bug reports.
I'll just take the monthly or quarterly updates & bug fixes that I get with
the low end package..." Pretty soon some big sector of the market has
dropped their premium support contracts for economy packages and there's not
enough money coming in to keep ACT alive. What happens to Gnat at that
point?

There is a lot more that goes into effective business decision making than
just guessing that someone will buy a product at a given price. You have to
do some strategic analysis of the market to base your business on a
sustainable model.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com

Enabling Digital.
Our Vision is to be the biggest supplier worldwide of digital gateway
technology.
www.pacemicro.com

Re: Ada 95 for an ARM-based bare board?

[Richard Riehle]

tmoran@acm.org wrote:

> > The list of  dead Silicon Valley companies who spread themselves too
> > thin by failing to concentrate on their core business and their paying
> > customers is long and depressing.
>   OTOH, Intel went off their core semiconductor RAM business to build
> a chip for somebody's calculator, and then overengineered that.  Apple
> forgot its core business was hobbyists, and Microsoft forgot its core
> business was BASIC.  Sticking with your core business leaves you with
> nothing more than your core business.

Thanks Tom.  Of course it is important to know when your core business
is no longer relevant.    It is also important to take well-reasoned risks.
I suppose I did oversimplify.   However, when a company has limited
resources and its core business is still emerging, or trying to re-emerge
(as with Ada),  there is some virtue in not straying too far from the
core business.

As to your examples, Intel did take risks, but those risks did not detract
from its core business.   Apple is a different story, as you know.  They
had an opportunity to win the marketplace and threw it away through
their arrogance.  Microsoft seized an opportunity by deceiving a
friend, Tim Patterson, and through a series of evil deeds that no one
in the technological part of our industry admires.  Their core business
was, and still is, marketing, not software innovation.

Richard Riehle

Re: Ada 95 for an ARM-based bare board?

[Randy Brukardt]

Ole-Hjalmar Kristensen wrote in message
<7vhehncnm1.fsf@vlinux.voxelvision.no>...
>"Randy Brukardt" <randy@rrsoftware.com> writes:
>>... Here, I have to agree with
>> Robert: if you think there is some way to make a viable business out
of
>> mass market Ada, feel free to take your own money and start a company
to
>> do it. Otherwise, stop whining.
>
>I'm not in the business of making compilers, and it wasn't whining,
>just an observation that from a customers point of view, it has made
>sense to seek other solutions. ...

Sorry, I wasn't responding personally to you, but rather the tenor of
opinions around here. And I didn't find the right word...

                Randy.

Re: Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

Robert Dewar wrote:

> Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> wrote in message
> news:<ajvqd1$1eh7ep$2@ID-77047.news.dfncis.de>...
> 
>> True, but should not it? From my point of view it could be a very good
>> time for Ada there. People are starting to use multitasking, wireless
>> communication etc in their embedded systems. To have a 32-bit controller
>> is no more cosidered too expensive. Shouldn't we use this situation?
> 
> Sounds like you believe this is a great business opportunity
> So why not persue it?

Because we are not a compiler vendor firm. Don't you agree this area 
requires some know-how, which cannot be gained in 4 months [an average 
duration of our projects]?

If *you* do not believe that Ada is [best] suitable for embedded 
applications, then, well, Ada is lost.

-- 
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de

Re: Ada 95 for an ARM-based bare board?

[tmoran]

> ACT, and other Ada compiler publishers can do more for Ada by
> providing excellent products and service to a few satisfied customers.
  It appears, unfortunately, that after the cream has been skimmed off
the top, and the public version of Gnat has satisfied its users, the
remaining Ada market won't support any compiler/tool vendors.

Re: Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

Robert Dewar wrote:

> Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> wrote in message
> news:<ajvqd1$1eh7ep$2@ID-77047.news.dfncis.de>...
> 
>> Unfortunately it is not an option in our case. We simply have not enough
>> resources for that.
> 
> Well if you are so sure it is a good business opportunity, you should be
> able to get people to provide the resources.

I am afraid that it is technically almost impossible. It is not a high-end 
market. The available financial resources are scattered among many small 
and middle sized projects. The amount is potentially immense, but it is out 
of reach. Add here, that people controlling these resources have prejudice 
against Ada [and anything except C or Java]. They cannot be convinced by 
any technical arguments, until the project collapses, which is then too 
late. So we must be better, cheaper and quicker.

-- 
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de

Re: Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

Richard Riehle wrote:

> Robert Dewar wrote:
> 
>> Not necessarily, ACT has developed many ports of GNAT at our expense, but
>> only when we know there is a real market out there (well a couple of
>> times we have skipped that step and got burned).
> 
> The list of  dead Silicon Valley companies who spread themselves too
> thin by failing to concentrate on their core business and their paying
> customers is long and depressing.
> 
> ACT, and other Ada compiler publishers can do more for Ada by
> providing excellent products and service to a few satisfied
> customers.   This appears to be the philosophy of ACT, and we
> see some wisdom in it.
> 
> In time, when an inventory of Ada success stories is available for
> public release, those of us involved in tool building, or compiler
> building, or training, or consulting, or applications building will
> have the satisfaction of our contribution, big or small, to those
> successes.

Also, the bottom line is - what is possibly good for Ada is likely not good 
for vendor X. I have no information to argue with that. Let it be true, but 
the problem here is, what will be with that "core business" when Ada will 
finally disappear? To my view embedded is a *core business* for Ada. So far 
nobody argued that. The PC-world was effectively lost for Ada. And we all 
saw how impossibly hard it is to get back there. This scenario repeats with 
embedded. The new (32-bit) embedded market will be definitely greater than 
one of PC's. Every device from a refrigerator to a hammer (:-)) will be 
"intelligent". Game market was already mentioned, etc.

-- 
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de

RE: Ada 95 for an ARM-based bare board?

[Robert C. Leif]

From: Bob Leif
To: Richard Riehle et al.
Firstly, I do not have any information to argue with you about
"Microsoft seized an opportunity by deceiving a friend, Tim Patterson,
and through a series of evil deeds that no one in the technological part
of our industry admires." I should point out that SUN deceived virtually
the entire software industry into believing that Java is a new, modern,
safe technology. This act will hurt many individuals as opposed to one. 
Frankly, we could loose many American military because of this. To add
to the problem, I suspect most of the readers of Comp.Lang.Ada have, as
I do, a great respect for the work of Dewar et al. on GNAT and would
recommend it for mission critical uses including weapons. Wearing my
microbiologist-biomedical engineer hat, I have heard intelligent talks
about the inadvisability of publishing certain material. I hope that the
wrong people are not benefiting from availability of Ada technology
including GNAT. Frankly, I do not have a solution to this problem.
Parenthetically, if Martin Carlisle's work on "Weaving Ada95 into the
.NET Environment works, then the ARM processor should run under Windows
CE and probably eventually on Windows XP embedded. These operating
systems, hopefully, could be a very large Ada market. 
Each outside compiler vendor on the Microsoft 3rd party page has a
paragraph to describe their product. I suspect that Microsoft people
might be amused if someone posted a truthful comparison of Ada versus
Java.
 
http://www.microsoft.com/windows/Embedded/ce.NET/evaluation/hardware/pro
videdbsp.asp

-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Richard Riehle
Sent: Thursday, August 22, 2002 10:30 AM
To: comp.lang.ada@ada.eu.org
Subject: Re: Ada 95 for an ARM-based bare board?

tmoran@acm.org wrote:

> > The list of  dead Silicon Valley companies who spread themselves too
> > thin by failing to concentrate on their core business and their
paying
> > customers is long and depressing.
>   OTOH, Intel went off their core semiconductor RAM business to build
> a chip for somebody's calculator, and then overengineered that.  Apple
> forgot its core business was hobbyists, and Microsoft forgot its core
> business was BASIC.  Sticking with your core business leaves you with
> nothing more than your core business.

Thanks Tom.  Of course it is important to know when your core business
is no longer relevant.    It is also important to take well-reasoned
risks.
I suppose I did oversimplify.   However, when a company has limited
resources and its core business is still emerging, or trying to
re-emerge
(as with Ada),  there is some virtue in not straying too far from the
core business.

As to your examples, Intel did take risks, but those risks did not
detract
from its core business.   Apple is a different story, as you know.  They
had an opportunity to win the marketplace and threw it away through
their arrogance.  Microsoft seized an opportunity by deceiving a
friend, Tim Patterson, and through a series of evil deeds that no one
in the technological part of our industry admires.  Their core business
was, and still is, marketing, not software innovation.

Richard Riehle

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

tmoran@acm.org wrote in message news:<2y%89.3154$wW7.204551567@newssvr14.news.prodigy.com>...
> Sticking with your core business leaves you with
> nothing more than your core business.

Which suits us just fine. We did not choose our corporate
name without quite a bit of thought :-)

Robert Dewar
Ada Core Technologies

Re: Ada 95 for an ARM-based bare board?

[Richard Riehle]

"Robert C. Leif" wrote:

> From: Bob Leif
> To: Richard Riehle et al.
> Firstly, I do not have any information to argue with you about
> "Microsoft seized an opportunity by deceiving a friend, Tim Patterson,
> and through a series of evil deeds that no one in the technological part
> of our industry admires."

Off topic.

Good source:   "Fire in the Valley" by Michael Swain.    It documents
the personal computer industry in some detail, and with a healthy
irreverence for some of our best known personalities.    His views
on Mr. William Gates are amusing, depressing, and guaranteed
to disillusion anyone who mistakes Mr. Gates for the technological
whiz-kid so often portrayed as larger than life in the popular media.
At least that is how I read Swain's work.  Someone else might read
it differently.

George Carlin once said, "The caterpillar does all the work and
the butterfly gets all the publicity."  There are people in our industry
who really do make innovative contributions without appropriating
the ideas of others.   They quietly go about their daily lives creating
new ideas only to have those ideas promoted by someone else.  There
seem to be more butterflies than caterpillars in the software industry.

Richard Riehle

Re: Ada 95 for an ARM-based bare board?

[Juha Valimaki]

"Marin D. Condic" <not.valid@acm.org> wrote in message
news:ak2q68$mlr$1@nh.pace.co.uk...
>
> There is a lot more that goes into effective business decision making than
> just guessing that someone will buy a product at a given price. You have
to
> do some strategic analysis of the market to base your business on a
> sustainable model.
>

Yes, I agree with you. It's just shame that sometimes all users don't fit in
the same business model. The two bad things I see in ACT business model is
that (1) it makes it hard for other companies to offer cheap, but not free,
Ada tools and (2) it discriminates small development teams.

Even hobbyist people are willing to pay money for software. It may not be a
large sum for individual users, but if there are enough such users it may be
possible. Because of the current situation competing against high quality
product that is free would be hard. I have a feeling that my view of the
siuation may seem too black and white for many. Hobby users surely take free
stuff happily and maybe also the small companies would be happy to use free
tools, but these companies just don't exist because Ada isn't interesting
enough?

The ACT model seems ok for larger companies and I think the price also isn't
too bad for a large company with only one target platform, but their model
excludes and discriminates small teams. I am sure their support is extremely
good, but it's hard to imagine how I would alone get enough advice from them
to justify the cost (especially if I need binaries for N platforms). I am
hoping for alternatives because of my own situation and because I believe
there are other small development teams that can't affort the premium, can't
find alternative tools and so simply choose some other language.

I already changed plans about my product so that I don't necessarily need
Mac or SGI platforms and I also invested in ObjectAda. I still have the
option to use GCC 3.1 or the GNAT public edition, if needed, but I would not
like to depend on something that I can't financially support (in my small
way). (Mac would have been risk anyway as it's not officially supported by
any company).

I think I have said enough questionable/speculative things about ACT/GNAT,
especially when I don't know the whole history of ACT/GNAT, so I will be
mostly in listening mode in the future about the subject. I still welcome
any company that wants to offer high quality Ada tools for small development
teams and hobby users on any platform!

Juha

P.S. I would like to thank people who have made the Mac port of GNAT. When I
say "Mac would be a risk" I don't mean I don't trust the work you have done,
I just prefer to pay for tools so I can say "fix this bug or I will not buy
from you again" :-)

P.S.II Paying for tools is not always necessary, there certainly are
situations where some deep pocket institution feels it's important to
provide free tools and they are even targeting users like me. (GCC C++ on
Linux & Mac, Java, ...)

Software Economics was RE: Ada 95 for an ARM-based bare board?

[Robert C. Leif]

From: Bob Leif
To: Tom Moran et al.
The GNU approach is essentially monopolistic. The ultimate in unfair
competition is to make your product available for free. As opposed to
professional monopolists, such as Microsoft; the GNU approach instead of
maximizing return actually minimized it. Since it saves hardware vendors
considerable expense, the GNU approach is an excellent solution for
them.
Again, I must emphasize that my comments are orthogonal to the question
of making the source code available. I am in favor of the customer
having the information included in the sources. However, I am totally
against anything that would decrease the value of intellectual property.

-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of tmoran@acm.org
Sent: Thursday, August 22, 2002 3:50 PM
To: comp.lang.ada@ada.eu.org
Subject: Re: Ada 95 for an ARM-based bare board?

> ACT, and other Ada compiler publishers can do more for Ada by
> providing excellent products and service to a few satisfied customers.
  It appears, unfortunately, that after the cream has been skimmed off
the top, and the public version of Gnat has satisfied its users, the
remaining Ada market won't support any compiler/tool vendors.

OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Warren W. Gay VE3WWG]

Robert C. Leif wrote:
> From: Bob Leif
> To: Tom Moran et al.
> The GNU approach is essentially monopolistic. 

This seems unfair, and a sad statement to make IMHO.

> The ultimate in unfair
> competition is to make your product available for free. 

So you think it is unfair for your neighbour next door
to create a software module and share it with you
for free?  After all, this competes with "professional
monopolists".

Or are you saying this is OK, as long as he doesn't
share it with the entire world?  After all, the
"professional monopolists" need to make a monopoly,
er, living ;-)

Your statements don't make a lot of sense to me.

> As opposed to
> professional monopolists, such as Microsoft; the GNU approach instead of
> maximizing return actually minimized it. 

You are evaluating the big picture in terms of a "return".
But what type of "return" are you using for a measure
here?  Monetary returns for the authors? Recoginition?
Control? And for whom(s)?

> Since it saves hardware vendors
> considerable expense, the GNU approach is an excellent solution for
> them.
> Again, I must emphasize that my comments are orthogonal to the question
> of making the source code available. I am in favor of the customer
> having the information included in the sources. 

What "information included in the sources" are we talking
about? Comments?  Or are you referring to the source code
itself and its availability?

> However, I am totally
> against anything that would decrease the value of intellectual property.

So you don't use any of the FSF/GNU tools?  If you are, you are
using intellectual property.  Are you adding or decreasing that
intellectual property?

Curious comments, you make. Warren.
-- 
Warren W. Gay VE3WWG
http://home.cogeco.ca/~ve3wwg

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Preben Randhol]

On Fri, 23 Aug 2002 07:52:52 -0700, Robert C. Leif wrote:
> From: Bob Leif
> To: Tom Moran et al.
> The GNU approach is essentially monopolistic. The ultimate in unfair

In what way?

> competition is to make your product available for free. As opposed to
> professional monopolists, such as Microsoft; the GNU approach instead of
> maximizing return actually minimized it. Since it saves hardware vendors
> considerable expense, the GNU approach is an excellent solution for
> them.

This is a very strange statement as Microsoft has released IE and other
programs free of cost. Of course you must buy their dratted OS to use
them.

The GNU model is different. It says that the source code should be free
(as in speech) for everyone. If you want to use some of the source code
for your projects then you have to *pay* with your source code not
money.

> Again, I must emphasize that my comments are orthogonal to the question
> of making the source code available. I am in favor of the customer
> having the information included in the sources. However, I am totally
> against anything that would decrease the value of intellectual property.
                                           ^^^^^
                                           money value

What will decrease it? You mean that people steal part of the code
despite the license of it? Well industrial espionage has always been
around and I guess Echelon is also used for this purpose. Besides you
have reverse enginering etc...

Some of the software patetens have the intellectual level of a 4 year
old or perhaps it is that of an accountant? I don't see why people
should have to pay just because some smock thought "Hey I can make a lot
of money if I just can get a patent on this simple and trivial procedure"

Preben Randhol

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Darren New]

Preben Randhol wrote:
> 
> On Fri, 23 Aug 2002 07:52:52 -0700, Robert C. Leif wrote:
> > From: Bob Leif
> > To: Tom Moran et al.
> > The GNU approach is essentially monopolistic. The ultimate in unfair
> 
> In what way?

In the same way that people complain about giving away IE for free.

> This is a very strange statement as Microsoft has released IE and other
> programs free of cost. Of course you must buy their dratted OS to use
> them.

Well, then, that's the cost, isn't it? :-) They give away their OS for free,
too. You just have to buy their OS to use it.
 
> The GNU model is different. It says that the source code should be free
> (as in speech) for everyone. 

Not "as speech". Free like free beer.
http://denbeste.nu/essays/freebeer.shtml

> What will decrease it? 

OK, put it this way. The latest video game is released. Everyone gets to
copy it for free. Indeed, before it is announced, everyone knows they'll be
able to get free copies. How many copies actually sell? How do you pay the
people who did the voice work? How do you pay for the plane tickets to fly
around taking pictures of textures to be used? How do you pay for the food
you eat while you're spending 2 years working on the game? How do you pay
the artists who draw the logos for the product? How do you pay for the
actors that do the motion-capture?

Now, if you have to pay for each one you use, then the company can
guesstimate that they'll sell 100,000 copies, and charge $20/copy, and
thereby collect $2million before people find a way to pirate it. They can
spend some of this on employees, some on plane tickets, some on computers,
electricity, etc.

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
   ** http://images.fbrtech.com/dnew/ **

Try our EbolaBurgers...
      So tender they melt in your mouth.

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Preben Randhol]

On Fri, 23 Aug 2002 12:26:59 -0400, Warren W. Gay VE3WWG wrote:
V> Robert C. Leif wrote:
>> From: Bob Leif
>> To: Tom Moran et al.
>> The GNU approach is essentially monopolistic. 
> 
> This seems unfair, and a sad statement to make IMHO.
> 
>> The ultimate in unfair
>> competition is to make your product available for free. 
> 
> So you think it is unfair for your neighbour next door
> to create a software module and share it with you
> for free?  After all, this competes with "professional
> monopolists".
> 
> Or are you saying this is OK, as long as he doesn't
> share it with the entire world?  After all, the
> "professional monopolists" need to make a monopoly,
> er, living ;-)
> 
> Your statements don't make a lot of sense to me.
> 
>> As opposed to
>> professional monopolists, such as Microsoft; the GNU approach instead of
>> maximizing return actually minimized it. 
> 
> You are evaluating the big picture in terms of a "return".
> But what type of "return" are you using for a measure
> here?  Monetary returns for the authors? Recoginition?
> Control? And for whom(s)?

This reminds of the 60 minutes program I saw two days ago. I guess it
was aired last week or earlier in the US, but a TV station here in
Norway sends the programs after midnight every Wednesday. Anyway it was
about a textile mill in the US that burned down. I have forgotten the
name but I think their products (winter clothing) are called Polytech or
something like that. Now what happened was that the owner decided he
wanted to rebuild the factory in the US and not move out of the country
as most of this industry does apparently. He also continued to pay the
workers full wages until the new factory was built. For this he got a
lot of credit and recognition. And as he said (I don't remember the
exact words) in the interview: "It doesn't look good for the state of
USA that when you do the right thing (in terms of moral) you get all
this attention and credit. Just for doing the right thing!"

Now, why is it like this? It is because everything is about economy and
greed. If he were to do the right thing in terms of maximizing the
economy of the factory he should have laid off the workers and hired
workers when the factory opened again. This would of course have
resulted in very bad social economical consequences.

Nowadays it is all about giving the stock holders more money and don't
care about the workers. There are numerous examples that factories that
do well and are making earnings are moved over seas just so that the
stock holders can increase their money by some 5-10-whatever%. But what
about all the sacked people? Or that they stock holders have taken out
enormous amounts of money in bonuses and at the same time they sack
people to maximize earnings. What kind of logic is this?

I can understand that you have to do something if a factory is always
loosing money, but when they are not?

So the reality is that the rich gets richer and as a direct result of
that the poor gets poorer.

So I don't know what professional monopolists are doing so much
better, but I guess some nice representatives are: 

   Enron 
   WorldCom
   Qwest
   AOL
   Andersen
   PricewaterhouseCoopers. 

One of the Norwegian newspapers have a special section on this part of
US bussiness headlined: "The marked of the scandals."

Preben

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Preben Randhol]

On Fri, 23 Aug 2002 18:03:40 GMT, Darren New wrote:
> Preben Randhol wrote:
>> 
>> On Fri, 23 Aug 2002 07:52:52 -0700, Robert C. Leif wrote:
>> > From: Bob Leif
>> > To: Tom Moran et al.
>> > The GNU approach is essentially monopolistic. The ultimate in unfair
>> 
>> In what way?
> 
> In the same way that people complain about giving away IE for free.

Ah a monopoly that contains all the people on the planet. I see.

>> This is a very strange statement as Microsoft has released IE and other
>> programs free of cost. Of course you must buy their dratted OS to use
>> them.
> 
> Well, then, that's the cost, isn't it? :-) They give away their OS for free,
> too. You just have to buy their OS to use it.

Oh where can on get a free Windows OS? I would like Windows 2000 instead
of Windows XP where nothing works.

>> The GNU model is different. It says that the source code should be free
>> (as in speech) for everyone. 
> 
> Not "as speech". Free like free beer.
> http://denbeste.nu/essays/freebeer.shtml

This essay contradicts itself.

>> What will decrease it? 
> 
> OK, put it this way. The latest video game is released. Everyone gets to
> copy it for free. Indeed, before it is announced, everyone knows they'll be
> able to get free copies. How many copies actually sell? How do you pay the
> people who did the voice work? How do you pay for the plane tickets to fly
> around taking pictures of textures to be used? How do you pay for the food
> you eat while you're spending 2 years working on the game? How do you pay
> the artists who draw the logos for the product? How do you pay for the
> actors that do the motion-capture?
> 
> Now, if you have to pay for each one you use, then the company can
> guesstimate that they'll sell 100,000 copies, and charge $20/copy, and
> thereby collect $2million before people find a way to pirate it. They can
> spend some of this on employees, some on plane tickets, some on computers,
> electricity, etc.

Exactly the money value. So why not say this in the first place?

Preben

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Darren New]

Preben Randhol wrote:
> > Well, then, that's the cost, isn't it? :-) They give away their OS for free,
> > too. You just have to buy their OS to use it.
> 
> Oh where can on get a free Windows OS? I would like Windows 2000 instead
> of Windows XP where nothing works.

It comes with the OS. It's already there, once you install Windows 2000.
Just like IE. Just like Linux.

> > Not "as speech". Free like free beer.
> > http://denbeste.nu/essays/freebeer.shtml
> 
> This essay contradicts itself.

In what way?
 
> Exactly the money value. So why not say this in the first place?

Um, this statement makes no sense, by which I mean "it isn't english enough
for me to understand it". On the other hand, this is comp.lang.ada.
Followups redirected.

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
   ** http://images.fbrtech.com/dnew/ **

Try our EbolaBurgers...
      So tender they melt in your mouth.

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

"Robert C. Leif" <rleif@rleif.com> wrote in message news:<mailman.1030068181.18207.comp.lang.ada@ada.eu.org>...

About 15 lines of stuff, clipped

> -----Original Message-----

About 35 lines of junk quotes


------------
This is a threaded news environment. It is definitely NOT necessary to
quote
entire articles that you are responding to. Just a couple of lines is
enough.
If people need more they can go look at the original which is right
there.

Please Robert (Leif) adjust your mailer (no doubt from our friends at
Microsoft) to NOT automatically quote the entire message.

I must say that I see an epidemic lately of

a) people including junk HTML attachments that are redundant copies of
messages. This of course serves no purpose, generates junk in many
environments, and constitutes a security risk.

b) people quoting entire messages all the time. The most aggravating
is when you get two people doing it at one another and generating
quadratic amounts of
quoted junk.

I gather that microsoft mailers are prone to encouraging people to
make these
two annoying mistakes. But they can be tamed. It is one thing for
ignorant users of computers to be stuck with inappropriate microsoft
defaults, but we
are supposed to be people who know something about computers :-)

RE: Software Economics

[Robert C. Leif]

From: Robert C. Leif
To: Warren W. Gay et al.
There is no economic difference between the devastating effect on
competition of a zero priced browser and a zero priced Ada compiler. The
first was done in response to a serious business threat; the second was
done because of government support and ideology. The question is: What
is the best economic strategy to maximize the investment of capital and
labor in Ada products? 

The general consensus for technology other than information technology
is to encourage the investors to believe that they will obtain a good
return on their investment. The concept of moderate prices and a large
market has been very profitable for companies, such as Microsoft. This
concept has been successfully applied by software vendors of "free
software", such as Red Hat.
GNAT has had two advantages over the affordable Ada compilers. 1) A
price of zero and 2) Dewar et al. believed in and developed in Ada.
Randy Brukardt et al. heroically tried to follow the same course with a
very moderate priced compiler; but were undercapitalized.

I have described an economic strategy that I believe would give Ada a
technological Advantage. 
R. C. Leif, "SIGAda '98, Workshop: How do We Expedite the Commercial Use
of Ada?." Ada letters XIX, No 1 pp. 28-39 (1999).
R. C. Leif, "Ada Developers Cooperative License (Draft) Version 0.3",
Ada letters XIX, No 1 pp. 97-107 (1999).

-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Warren W. Gay
VE3WWG
Sent: Friday, August 23, 2002 9:27 AM
To: comp.lang.ada@ada.eu.org
Subject: OT: Software Economics was RE: Ada 95 for an ARM-based bare
board?

Robert C. Leif wrote:
> From: Bob Leif
> To: Tom Moran et al.
> The GNU approach is essentially monopolistic. 

This seems unfair, and a sad statement to make IMHO.

> The ultimate in unfair
> competition is to make your product available for free. 

So you think it is unfair for your neighbour next door
to create a software module and share it with you
for free?  After all, this competes with "professional
monopolists".

Or are you saying this is OK, as long as he doesn't
share it with the entire world?  After all, the
"professional monopolists" need to make a monopoly,
er, living ;-)

Your statements don't make a lot of sense to me.

> As opposed to
> professional monopolists, such as Microsoft; the GNU approach instead
of
> maximizing return actually minimized it. 

You are evaluating the big picture in terms of a "return".
But what type of "return" are you using for a measure
here?  Monetary returns for the authors? Recoginition?
Control? And for whom(s)?

> Since it saves hardware vendors
> considerable expense, the GNU approach is an excellent solution for
> them.
> Again, I must emphasize that my comments are orthogonal to the
question
> of making the source code available. I am in favor of the customer
> having the information included in the sources. 

What "information included in the sources" are we talking
about? Comments?  Or are you referring to the source code
itself and its availability?

> However, I am totally
> against anything that would decrease the value of intellectual
property.

So you don't use any of the FSF/GNU tools?  If you are, you are
using intellectual property.  Are you adding or decreasing that
intellectual property?

Curious comments, you make. Warren.
-- 
Warren W. Gay VE3WWG
http://home.cogeco.ca/~ve3wwg

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Robert Dewar]

"Robert C. Leif" <rleif@rleif.com> wrote in message news:<mailman.1030114442.9877.comp.lang.ada@ada.eu.org>...

> However, I am totally
> against anything that would decrease the value of intellectual property.

Bob Leif often writes bizarre things, but this thread is pretty off
the
wall even by that criterion.

First point, ACT is definitely in favor of maximizing the value of its
property. A company the size of ours (Ada Core and ACT/Europe between
them have over 40 full time employees) needs a considerable level of
income to support our activities and the ongoing development of GNAT
and its surrouding tools. 

Second point, the only difference between Microsoft and ACT is the
license
we use. We both sell licensed copyrighted software.

We use the GPL as our license because we think it maximizes the value
of
the product to our customers. We think restrictive licenses such as
those
used by Microsoft (and presumably by Bob Leif if in fact he is in the
business of selling software -- that's not clear to me) are bad
business because they
don't provide the customers with what they need.

We think it's good business to provide the customers what they need.

And the result of providing customers what they need is that we
prosper. So
we are using the GPL *because* it allows us to support our activities.

BL makes the absurd statement that GNAT is a monopoly. This shows a
level of
unawareness of the market that is truly staggering. In fact there is
fierce
competition in the Ada market, as anyone who is in fact involved in
the market
as a vendor or a procurer of Ada products knows perfectly well.

And that's the last you will here from me in this thread, since I
can't imagine
it going in a sensible direction.

Robert Dewar
Ada Core Technologies

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Robert Dewar]

"Robert C. Leif" <rleif@rleif.com> wrote in message news:<mailman.1030114442.9877.comp.lang.ada@ada.eu.org>...

> The ultimate in unfair
> competition is to make your product available for free.

Well one would have to agree that if someone is making a product free,
subsidizing it from some other income source, that can be unfair
competition.

However, Robert Leif seems to have some completely bizarre view of how
Ada
Core Technologies operates that bares not even a slight resemblance to
reality.

Our product, GNAT Pro, is a high quality Ada compiler tool set, with a
direct
guaranteed licensed status from Ada Core, and with high quality
support, including immediate updates to correct problems, continual
enhancements in
response to customer needs, and special customization to meet the
needs of
customers.

Despite Bob's strange statements, this product is not free. In fact
there are
a fair number of threads on CLA from time to time complaining that we
charge
much too much for this product, and urging us to make some low cost
product
available.

Now this is not something we are about to do, but in fact, we are so
far from
a monopoly that if someone wants to compete with us and make some
other product
based around the GNAT core technology, they are free to do so!

Robert Dewar
Ada Core Technologies

I think the confusion is so great, that it is probably impossible to
cure :-)

Robert Dewar
Ada Core Technologies

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

"Juha Valimaki" <juha.valimaki@xdsloy.com> wrote in message news:<ak4vaj$5g6$1@phys-news1.kolumbus.fi>...

> Yes, I agree with you. It's just shame that sometimes all users don't fit in
> the same business model. The two bad things I see in ACT business model is
> that (1) it makes it hard for other companies to offer cheap, but not free,
> Ada tools

No one ever succeeded at doing this who tried in the past, and the
attempts
at this all failed when there *WAS* no freely available version of
Ada. Perhaps it is because there is no mass market for Ada, perhaps it
is because
the people who tried before did it wrong. As to whether the
availability of
freely downloadable Ada systems helps or hurts, it is not at all
clear. We
think that it very much helps all segments of the Ada market to have a
high
quality Ada compiler that students, hobbyists, and indidual
enthusiastic
engineers can access. 

> and (2) it discriminates small development teams.

What an odd word to use. We don't discriminate against anyone, we
simply don't
serve all segments of the market. We succeed because we don't try to
be all
things to all people. Of course all users don't fit in the same
business model!
What get's companies in trouble is when they try to serve multiple
markets at
the same time and the markets get mixed up (consider the airlines
trying to
sell expensive seats to businesses and the same exact seats cheap to
tourists).

If there is a viable business in providing low cost products to small
development teams, then someone should by all means step in, but
please
don't expect Ada Core Technologies to serve all needs of the Ada
community.
We can't and we won't try.

We are Ada enthusiasts, and we immodestly believe that the continued
viability
of Ada Core Technologies, now with a complement of over 30 highly
experienced
engineers dedicated to the continued support and improvement of Ada,
is critical to the continued success of Ada.

We are thus very conservative in how we approach the market, and so
far this
conservative approach has paid off with steady (but modest and
manageable)
growth, through a period in which other less conservative high-tech
companies
have seen a huge boom-and-bust cycle.

We have increased revenues in the last year, and we have a comfortable
but
modest buffer in our bank account. We don't have zillions in stock
options,
and I am sorry to report that the CEO and other founders have not been
able
to cash out for millions of dollars, no doubt making us a failure in
the eyes
of some, but we are quite happy (and proud) to be succeeding in our
own style!

Robert Dewar
Ada Core Technologies

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

"Juha Valimaki" <juha.valimaki@xdsloy.com> wrote in message news:<ak4vaj$5g6$1@phys-news1.kolumbus.fi>...

> I am sure their support is extremely
> good, but it's hard to imagine how I would alone get enough advice from them
> to justify the cost (especially if I need binaries for N platforms).

One interesting piece of experience here is that when we started we
used to
have prices for 1 or 2 seats. We abandoned that. Why? Because it was
eating
us alive. These tiny projects used far far more of our support
services than
larger projects. Some people complain that they have to pay the same
for
one developer as five, but in fact if we priced according to
historical
data, we should charge more for one person than five.

That's actually not so surprising, one person working on their own is
indeed
all on their own, and thus they need help wherever they can get it,
whereas
even a small team will be able to help one another over many simple
problems.

It just maybe that your imagination is not good enough, and that in
fact you
would find that a support contract would easily justify its cost. Many
of our
customers *are* very small developers, and find our support services
extremely
useful and very much worth the cost.

This is of course something you have to figure out for yourself. If
the GNAT Pro product is not for you, have a look at offerings from
other Ada vendors.
You may well be able to find less expensive products that will meet
your
particular needs. There is a lively competition in the Ada market. It
is not
at all the case that GNAT is the only game in town :-)

Robert Dewar
Ada Core Technologies

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Weston T. Pan]

Preben Randhol wrote:
> 

[snip ]

> This reminds of the 60 minutes program I saw two days ago. I guess it
> was aired last week or earlier in the US, but a TV station here in
> Norway sends the programs after midnight every Wednesday. Anyway it was
> about a textile mill in the US that burned down. I have forgotten the
> name but I think their products (winter clothing) are called Polytech or
> something like that. Now what happened was that the owner decided he
> wanted to rebuild the factory in the US and not move out of the country
> as most of this industry does apparently. He also continued to pay the
> workers full wages until the new factory was built. For this he got a
> lot of credit and recognition. And as he said (I don't remember the
> exact words) in the interview: "It doesn't look good for the state of
> USA that when you do the right thing (in terms of moral) you get all
> this attention and credit. Just for doing the right thing!"
> 

This was a story about the Malden Mills textile company. By the way, the
fabric was called Polartec.

--Weston

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Larry Kilgallen]

In article <3D66FE7A.AF0183AA@mindspring.com>, "Weston T. Pan" <wywtpan2@mindspring.com> writes:

> This was a story about the Malden Mills textile company. By the way, the
> fabric was called Polartec.

I prefer to say "is" called Polartec.  My wife buys it to make clothing
for small children.

RE: Ada 95 for an ARM-based bare board?

[Robert C. Leif]

From: Bob Leif
To: Robert Dewar et al.
I quoted a very small part of the article. As a scientist, I have been
programmed to always show my evidence. I am sorry if we differ on this
matter of culture. I always try to supply a reference for my statements.

I do admit that I have made mistakes using Outlook where material was
hidden from my view at the bottom of the document. However, I do not
believe that this was the case. I also try to test the URL in situ to
make sure that it works. This does not add anything to what I see. I
would be surprised if some Microsoft Wizardry made any unseen additions.
Please send me back a copy illustrating what you believe I should have
omitted.
Thank you.
Parenthetically, Since you seem to be curious about what I do, it is
described at www.newportinstruments.com. 

-----Original Message-----
From: comp.lang.ada-admin@ada.eu.org
[mailto:comp.lang.ada-admin@ada.eu.org] On Behalf Of Robert Dewar
Sent: Friday, August 23, 2002 1:36 PM
To: comp.lang.ada@ada.eu.org
Subject: Re: Ada 95 for an ARM-based bare board?

"Robert C. Leif" <rleif@rleif.com> wrote in message
news:<mailman.1030068181.18207.comp.lang.ada@ada.eu.org>...

About 15 lines of stuff, clipped

> -----Original Message-----

About 35 lines of junk quotes


------------
This is a threaded news environment. It is definitely NOT necessary to
quote
entire articles that you are responding to. Just a couple of lines is
enough.
Big Snip

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[tmoran]

> "Robert C. Leif" <rleif@rleif.com> wrote in message news:<mailman.1030114442.9877.comp.lang.ada@ada.eu.org>...
>
> > The ultimate in unfair
> > competition is to make your product available for free.
>
> Well one would have to agree that if someone is making a product free,
> subsidizing it from some other income source, that can be unfair
> competition.
> ...
> Our product, GNAT Pro, is a high quality Ada compiler tool set, with a
> ...
> Despite Bob's strange statements, this product is not free. In fact
  It's hard to believe anyone could actually think Bob Leif was referring
to Gnat Pro as a free product.  Correct me if I'm wrong Bob, but I
understood you to be referring to the public, $0, version of Gnat, which
appears to be the beneficiary of work on (ie, is subsidized by) Gnat Pro.

Re: Software Economics

[Robert Dewar]

"Robert C. Leif" <rleif@rleif.com> wrote in message news:<mailman.1030147922.26232.comp.lang.ada@ada.eu.org>...

> There is no economic difference between the devastating effect on
> competition of a zero priced browser and a zero priced Ada compiler.

There is a no comparison between these situations. In one case, a
company is
making its product available at no cost to drive out a competitor by
subsidizing the effort with other products (Internet Explorer). In the
case of ACT, we absolutely definitely do NOT make our product (which
is a
fully supported Ada compiler, with guaranteed licensing suitable to
the
application area) available at no cost. In fact many of the threads on
CLA (including fulminations by Leif) complain that GNAT Pro is too
expensive!

> the second was done because of government support and ideology

That is complete nonsense, you simply do not know what you are talking
about.
Bob Leif is making wild guesses and wild statements without the
slightest awareness of what is going on.

The reason that we make versions of GNAT freely available has nothing
to do
with government support (which disappeared 7 years ago) or with
ideology.
The reason we do this is because we think it is the best strategy for
the
continued health of Ada, and thus for the continued success of Ada
Core
Technologies. So it's simply good business sense!

When Ada first came out, there certainly was no freely available
version of
Ada, and several companies tried to market low cost products. They all
failed.
So that's a data point that seems completely contradictory to your
thesis.

Second observation. There has always been a high quality free compiler
for C,
and these days for C++ as well. Do you think this has "devasated the
market
for these compiler technologies?" Of course not, to claim this would
be absurd.

The reason that we think it is critical to have a freely available
high quality
Ada compiler around is that it sparks interest in students, hobbyists,
and individual engineers who want to experiment. If no one knows about
Ada, then
who is going to buy *any* Ada technology? - answer no one. 

Yes, occasionally companies may use the public version of GNAT for
mission critical projects. We find that a dubious decision, since the
use of unsupported software with no assurance of correct licensing is
rather risky.
Few companies are willing to take this risk in practice. We don't see
that
as having a significant impact on the commercial market for Ada.

Robert Leif is frustrated that no one will sell him an inexpensive
supported
Ada compiler, and wants to blame the availability of the free version
of
GNAT, but he really has nothing to support his claim here, and all the
evidence
points in the opposite direction.

My own analysis: Inexpensive mass market products are possible only if
there
is a mass market. I don't think there is such a market for Ada tools
in the
current climate. No one has ever succeeded with this approach yet, and
I would
not expect them to have succeeded. 

The scale of things is just not right.

If you charge $1000 for a product, then you need to sell several
thousand
of them a year to support a reasonable development effort, but that's
still
far to expensive for many hobbyists. Now if you reduce the price to
$50, you
have to sell ten's or hundred's of thousands of copies a year, and I
just don't think the market begins to be there.

Furthermore, if you do try to use the mass market model, then you
simply can't
provide any kind of reasonable support, and our experience is that
serious
Ada projects really appreciate and need (and can afford) good support.

Robert Dewar
Ada Core Technologies

P.S. Yes, I know, I said I would not answer any more messages in this
thread,
but Bob Leif has a funny newsreader that keeps starting new threads
for no
obvious reason, so the threads won't stay killed :-)

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Bill Tate]

Preben Randhol <randhol+news@pvv.org> wrote in message news:<slrnamcv5p.g58.randhol+news@kiuk0156.chembio.ntnu.no>...
> 
> This reminds of the 60 minutes program I saw two days ago. I guess it
> was aired last week or earlier in the US, but a TV station here in
> Norway sends the programs after midnight every Wednesday. Anyway it was
> about a textile mill in the US that burned down. I have forgotten the
> name but I think their products (winter clothing) are called Polytech or

Malden Mills in Massachusetts

> something like that. Now what happened was that the owner decided he
> wanted to rebuild the factory in the US and not move out of the country
> as most of this industry does apparently. He also continued to pay the
> workers full wages until the new factory was built. For this he got a
> lot of credit and recognition. And as he said (I don't remember the
> exact words) in the interview: "It doesn't look good for the state of
> USA that when you do the right thing (in terms of moral) you get all
> this attention and credit. Just for doing the right thing!"

a. Malden Mills manufactures an exceptional product called "Polartech"
(sp?) that is raved about by any customer who purchases a product made
from it, up to & including wedding dresses :>)  Consequently, the
product has quite a devoted following of customers whose demographics
indicate they don't tend to shop in the "Everything is a Dollar"
store.
b. Right thing? Sorry but you are applying YOUR notions of what is the
right thing vis.a.vis. economics and free enterprise.  Had this been a
company with inferior or poor quality products and poorly managed,
would you have expected the company's shareholders or owners to have
kept it alive in the immediate aftermath of the fire and after the
repairs were affected?  Simply because a company exists doesn't mean
it should.

> Now, why is it like this? It is because everything is about economy and
> greed. If he were to do the right thing in terms of maximizing the
> ......
> So the reality is that the rich gets richer and as a direct result of
> that the poor gets poorer.

Fact #1. Economics is not a zero sum game.  
Fact #2. Surprise surprise, there are those who engage in fraud, etc.,
and sometimes they get caught and sometimes they don't.

For me, I'm not especially inspired by Northern Europe countries who
have "aggressive" tax policies - I don't know but tax rates greatly in
excess of 50% of an individual's income seems a tad bit high....

> So I don't know what professional monopolists are doing so much
> better, but I guess some nice representatives are: 
> 
>    Enron 
>    WorldCom
>    Qwest
>    AOL
>    Andersen
>    PricewaterhouseCoopers. 
> 
> One of the Norwegian newspapers have a special section on this part of
> US bussiness headlined: "The marked of the scandals."
Ahhh, European newspapers - monuments to objective analysis of the
U.S.

BTW, former Enron employees who saw their retirement savings dwindle
to nothing did so in 401Ks invested almost exclusively in Enron stock.
 Now, you had to have lived in a cave on Easter Island for the last 50
years to have avoided hearing financial advisor after financial
investor advisor telling individual investors to diversify their
portfolios in order to manage the risk of losing money.  So if one
were to have 90% to 100% of one's retirement savings tied up in one or
two stocks, it might be reasonable to assert that said individual is
betting on a fairly risky investment strategy.

A callous individual such as myself would be further inclined to think
of reasons why these "new members" of the "poor" would do such a
thing.  Perhaps they were motivated by something other than being
inspired to do great things to help their fellow man.  Perhaps they
were motivated, only in part of course, by their own "greed" &
aspirations of "wealth."   As a card-carrying member of the
"neanderthal & mean-spirited" set, I'm somewhat inclined to believe
that the "envelop" of responsibility for these individual financial
losses goes beyond just the companies themselves.  That's just my
opinion of course without the trappings of political correctness
thrown in.

Re: Excessive quoting and Outlook (was: Ada 95 for an ARM-based bare board?)

[Robert Dewar]

"Robert C. Leif" <rleif@rleif.com> wrote in message news:<mailman.1030166702.5161.comp.lang.ada@ada.eu.org>...
> From: Bob Leif
> To: Robert Dewar et al.
> I quoted a very small part of the article. As a
> scientist, I have been
> programmed to always show my evidence.

First, to see what you posted, just go look back at the
article, I am certainly not about to repost it again.

Second, as a scientist if you tried to publish a paper
which consisted of one half or less of your own material
and one half or more of a gigantic quotation from a published paper,
then your submission would be rejected
not the least of the reasons would be that this is almost
certainly not fair use :-)

Bob, when you participate in use net, all messages are
threaded. Well actually I should not really say that, since
you seem to manage to break the threading pretty often.

But even when you are posting, there is almost always a
reference line of the form:

"Robert C. Leif" <rleif@rleif.com> wrote in message news:<mailman.1030166702.5161.comp.lang.ada@ada.eu.org>...

that you can see. This is all that is necessary most of
the time, and is like a reference in a scientific paper.

As for Outlook (one of the worst programs ever written IMO, and a
program that is single handedly responsible
for billions of dollars of damage by viruses because of
its complete lack of concern for security), it most certainly does
have defaults that are quite inappropriate.
I suggest you very carefully check the defaults you are
using and make sure they are appropriate. 

Now to be fair, Microsoft has *finally* recognized that
security is worth worrying about (you may remember the
highly publicized period earlier this year when development at MS was
stopped so everyone could worry about security).
So perhaps things may improve in the future

Re: Ada 95 for an ARM-based bare board?

[Michael Bode]

dewar@gnat.com (Robert Dewar) writes:

> One interesting piece of experience here is that when we started we
> used to have prices for 1 or 2 seats. We abandoned that. Why?
> Because it was eating us alive. These tiny projects used far far
> more of our support services than larger projects. Some people
> complain that they have to pay the same for one developer as five,
> but in fact if we priced according to historical data, we should
> charge more for one person than five.

So you don't sell an economy version of Gnat because the support you
give for Gnat Pro would not be possible for the economy version? What
if you would sell boxes with Gnat Pro CDs without the Gnat-Pro-Class
support and maybe a download site where registered users could
download patches? Do you think this would be too much effort for the
anticipated number of customers, or do you think this would hurt the
Gnat Pro business?

Please note that I don't want to complain or push you in any
direction. I'm just curious. I've no problems at all with the
situation as it is using Gnat public for my projects as the single
developer in my company. OTOH we could as well spend 250 to 500 EUR
for a box with a nice printed manual, Gnat (let's say Windows and
Linux version), GtkAda (not sure if this would be legally possible)
and some regular bug fixes if you have them ready anyway. I've also
used (and still do) Visual C++. I've never called Microsoft support
and in fact have no idea what kind of support I would get from them.

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Robert Dewar]

tmoran@acm.org wrote in message news:<FAG99.5493$wr1.465839426@newssvr13.news.prodigy.com>...

>  Correct me if I'm wrong Bob, but I
> understood you to be referring to the public, $0, version of Gnat, which
> appears to be the beneficiary of work on (ie, is subsidized by) Gnat Pro.

Yes, but the public version of GNAT is not a commercial product by an
stretch
of the imagination. So I am not sure what the comparison with
Microsoft meant.

If Microsoft had been giving away a limited version of IE free with
the intention of stimulating sales for the non-free $$ version from
which they
could make lots of money, that's a totally different manner. Free
samples
are a pervasive marketing tactic after all :-)

mailers, quoting text etc. (was Ada 95 for an ARM-based bare board?)

[Robert Dewar]

"Robert C. Leif" <rleif@rleif.com> wrote in message news:<mailman.1030166702.5161.comp.lang.ada@ada.eu.org>...

> I would be surprised if some Microsoft Wizardry made any unseen additions.

Really?

Typical microsoft mailers make the unseen addition of completely useless
HTML by default that is

a) redundant and serves no purpose
b) wastes bandwidth and clutters up databases
c) presents a security risk (HTML attachments are not benign!)

But I guess none of these (especially c) has ever been of any concern to MS :-)

Re: Excessive quoting and Outlook (was: Ada 95 for an ARM-based bare board?)

[Larry Kilgallen]

In article <5ee5b646.0208241200.274bde80@posting.google.com>, dewar@gnat.com (Robert Dewar) writes:

> Now to be fair, Microsoft has *finally* recognized that
> security is worth worrying about (you may remember the
> highly publicized period earlier this year when development at MS was
> stopped so everyone could worry about security).

Certainly they have recognized that it is bringing them adverse
publicity.

> So perhaps things may improve in the future

You may feel that I am unfair, but I am highly skeptical that they
will make significant progress.

Re: Ada 95 for an ARM-based bare board?

[Darren New]

Michael Bode wrote:
> What
> if you would sell boxes with Gnat Pro CDs without the Gnat-Pro-Class
> support and maybe a download site where registered users could
> download patches? Do you think this would be too much effort for the
> anticipated number of customers, or do you think this would hurt the
> Gnat Pro business?

Just out of curiousity, is there any reason why one person can't buy the
10-seat support package and then sell the "economy" package to any number of
users? I mean, this *is* GNU software we're talking about. It's not really
ACT who has to do this, is it?

(Not that I would, even if I thought it would be economical, as it would
likely kill ACT which would then kill my business, but...)

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
   ** http://images.fbrtech.com/dnew/ **

Try our EbolaBurgers...
      So tender they melt in your mouth.

Re: mailers, quoting text etc. (was Ada 95 for an ARM-based bare board?)

[Larry Kilgallen]

In article <5ee5b646.0208241357.39e0956a@posting.google.com>, dewar@gnat.com (Robert Dewar) writes:
> "Robert C. Leif" <rleif@rleif.com> wrote in message news:<mailman.1030166702.5161.comp.lang.ada@ada.eu.org>...
> 
>> I would be surprised if some Microsoft Wizardry made any unseen additions.
> 
> Really?
> 
> Typical microsoft mailers make the unseen addition of completely useless
> HTML by default that is
> 
> a) redundant and serves no purpose
> b) wastes bandwidth and clutters up databases
> c) presents a security risk (HTML attachments are not benign!)

The defects introduced by Microsoft starting with Word 6 (allowing
instructions provided by someone else to execute in the context of
the recipient) were originated by IBM under the CMS system with a
text formatting extension.  IBM retracted that "innovation" and
discussed it at subsequent security conferences for several years.
Only then did Microsoft "innovate" with their own implementation,
apparently choosing not to bother studying anything else that has
happened in the computer industry.

> But I guess none of these (especially c) has ever been of any concern to MS :-)

There have been public statements from Microsoft that the "features"
provided by these loopholes are more important to their customers
than fixing security problems.

Re: Ada 95 for an ARM-based bare board?

[tmoran]

> b) people quoting entire messages all the time. The most aggravating
> is when you get two people doing it at one another and generating
> quadratic amounts of
> quoted junk.
  This is wrong.  It's exponential.

Re: Ada 95 for an ARM-based bare board?

[SteveD]

"Darren New" <dnew@san.rr.com> wrote in message
news:3D681154.AFF57FA6@san.rr.com...
[snip]
> Just out of curiousity, is there any reason why one person can't buy the
> 10-seat support package and then sell the "economy" package to any number
of
> users? I mean, this *is* GNU software we're talking about. It's not really
> ACT who has to do this, is it?

I have thought about the same thing myself.  The problem is that people
paying for the economy package would expect support.

My guess is that many support questions are really just Ada questions.
These questions may be easy to answer, but time consuming ($$$).

Occasionally issues would come up that are in fact defects in the compiler.
After clearly identifying these defects you could go to ACT for support.  If
you attempted to route all questions directly to ACT, I would expect the
cost of your support contract with ACT to go up accordingly.

And the name... APT (Ada Peel Technology), a spin of of ACT (get it?).

> (Not that I would, even if I thought it would be economical, as it would
> likely kill ACT which would then kill my business, but...)

I think this economy package would have almost no impact on ACTs buisness.
They are in the buisiness of providing a product with a high level of
support.  Unless you're able to compete on the support level, this should
have little impact on their buisness.

I would think that ACT would actully encourage such a buisness.

SteveD

> --
> Darren New
> San Diego, CA, USA (PST). Cryptokeys on demand.
>    ** http://images.fbrtech.com/dnew/ **
>
> Try our EbolaBurgers...
>       So tender they melt in your mouth.

Re: Ada 95 for an ARM-based bare board?

[Michael Bode]

"SteveD" <nospam_steved94@attbi.com> writes:

> I have thought about the same thing myself.  The problem is that people
> paying for the economy package would expect support.

And premium support will cost premium money. That's why for an
"economy" package it must be perfectly clear in the first place that
support will only come in the form of maybe a website with a FAQ and
the bug fixes that are produced anyway for the premium support
customers. You can report a bug but don't expect a patch the next
day. If someone doesn't like that he should either use the free
version or the Pro version. Maybe under conditions like this there is
no demand for an "economy" version?

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

tmoran@acm.org wrote in message news:<pG_99.325$kO6.29849783@newssvr21.news.prodigy.com>...
> > b) people quoting entire messages all the time. The most aggravating
> > is when you get two people doing it at one another and generating
> > quadratic amounts of
> > quoted junk.

>   This is wrong.  It's exponential.

Oh dear!

Time for an elementary lesson in math :-)
(well actually in computer science, you need to be able to do this
sort of thing to do elementary complexity analysis of algorithms)

Let's suppose that everyone adds 10 lines and quotes everything before

then we have message lengths that are

10 20 30 40 50 60 70

Each message contains one copy (only) of everything that has been sent before
(not multiple copies). I trust that is obvious

In other words the size of message N is 10N lines

So the size of messages grows only linearly

But, by elementary summing of an arithmetic series, the total amount of space
occupied by the first N messages will be

(10 + 10N)/2 * N  = (10N + 10N**2) / 2

The quadratic term dominates and the result is asymptotically

  5 * N**2

which is called quadratic.

The casual use of exponential to talk of things that grow fast is something
that technical people should avoid!

Re: Ada 95 for an ARM-based bare board?

[tmoran]

> Let's suppose that everyone adds 10 lines and quotes everything before
  Let's suppose that everyone adds 10% to what they are quoting.  Then
the size of message N is 1.1**N * OP.

> Let's suppose that everyone adds 10 lines and quotes everything before
> So the size of messages grows only linearly
  Right.  In that case new messages don't grow exponentially, but
linearly.

> But, by elementary summing of an arithmetic series, the total amount of space
  The reader of a message doesn't see the sum, but rather the latest
individual message (unless he has a really strange mail reader).  Only
someone who saves all the traffic (Google, for instance) would see
quadratic growth.  Google's disk space is not usually what people are
thinking about when they complain about long messages.  An individual
saving old messages would probably be smart enough, in the case of
complete quoting, to just save the latest, which contains all previous ones.

  In a more perfect world, everyone would quote, say, 2 lines for context,
add a concise 3 line comment, and all messages would be a constant 5 lines.

Re: Ada 95 for an ARM-based bare board?

[John R. Strohm]

Robert, where would I find a writeup on what is involved in retargeting
GNAT?

Is it just a matter of doing a GCC target (i.e., writing the machine
description files and the necessary gcc runtime support)?

Are there GNAT runtime routines that must be rebuilt?  What about executive
support?

"Robert Dewar" <dewar@gnat.com> wrote in message
news:5ee5b646.0208211132.6c283ef0@posting.google.com...
> Ole-Hjalmar Kristensen <oleh@vlinux.voxelvision.no> wrote in message
news:<7vlm70cqcz.fsf@vlinux.voxelvision.no>...
> > > And preferably pay up front for the develoment of the compiler....  OK
> > > strategy if you sell few expensive development systems, probably not a
> > > good strategy to penetrate the mass market.
>
> Not necessarily, ACT has developed many ports of GNAT at our expense, but
> only when we know there is a real market out there (well a couple of times
> we have skipped that step and got burned).
>
> Indeed we are not trying to "penetrate the mass market". That's not our
> business area. Our expertise is in providing high level support from high
> level competent engineers. We don't see that as scaling to a mass market.
>
> The world seems full of people who are happy to spend lots
> of energy in trying to get Ada Core to move in that direction, but sadly
> none of these people seem willing to spend their own energy persuing a
> market that they are sure we are missing!
>
> Robert Dewar
> Ada Core Techonologies

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Georg Bauhaus]

Bill Tate <tatebll@aol.com> wrote:
: For me, I'm not especially inspired by Northern Europe countries who
: have "aggressive" tax policies - I don't know but tax rates greatly in
: excess of 50% of an individual's income seems a tad bit high....

The effective percentage largely depends on number of kids,
married or not, self-employed or not, etc etc etc. It is
difficult for a rich and un-divorced person to become poor by
a lawful manner in at least some northern European countries.

-- Georg

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Preben Randhol]

On 24 Aug 2002 06:16:36 -0700, Bill Tate wrote:
> 
> For me, I'm not especially inspired by Northern Europe countries who
> have "aggressive" tax policies - I don't know but tax rates greatly in
> excess of 50% of an individual's income seems a tad bit high....

Well it is no excess of 50% as you say. Besides how much money do you
pay for insurance in case you have to go to the hospital? The system is
different here. Everybody will get medical care no matter if they have
money or not at a hospital. This is only one example. Education is
another. So you cannot simply compare it to the low tax rates you may
have you have to take into account what you have to pay to the different
other service suppliers too.

Preben

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Ted Dennison]

dewar@gnat.com (Robert Dewar) wrote in message news:<5ee5b646.0208241353.5f69b4bf@posting.google.com>...
> If Microsoft had been giving away a limited version of IE free with
> the intention of stimulating sales for the non-free $$ version from
> which they
> could make lots of money, that's a totally different manner. Free

I have seen it argued that they were doing something close to that:
substitute "web server" for "version" above, and you have it. That's
certianly what Netscape was doing (the browser was downloadable and
licensed for free for non-commercial use).

However, I don't think that's accurate either. In the view of the
courts at least, Microsoft gave away the browser to take market share
from 3rd parties (like Netscape), so they could neutralize the threat
posed to their platform monopoly by web browsers. It wasn't giving it
away for free ($$$) that did the trick either. Again, Netscape also
was licensed to home users at no cost, and was doing OK for years
against Explorer. It was the various strong-arm tactics employed by
Microsoft leveraging against their monopoly (eg: integrating it and
distributing it with the OS, pressuring PC makers to not distribute
the competition, etc.), that made the difference.

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Ted Dennison]

Darren New <dnew@san.rr.com> wrote in message news:<3D66792F.F77550E@san.rr.com>...

> OK, put it this way. The latest video game is released. Everyone gets to
> copy it for free. Indeed, before it is announced, everyone knows they'll be
> able to get free copies. How many copies actually sell? How do you pay the

The original ID games (Castle Wolfensein 3D and Doom) were released
shareware, and ID became (and still is) one of the most successful
game companies around. After the first game came out, John Carmack and
the other founders supposedly each went out and bought multiple
imported sports cars. On
http://www.wired.com/wired/archive/4.08/id.html?topic=gaming&topic_set=newmedia
you can read a bit more about their story (including what its like to
ride in John's Ferrari).

Note in particular the paragraph:

The most popular computer game of all time, Doom has sold more than 2
million copies. But that only hints at its popularity. As many as 20
million shareware versions are installed worldwide - unregistered but
perfectly legal copies, distributed free with id's blessing. Myriad
Doom players are creating worlds of their own. By releasing chunks of
their games as shareware, id's marketing strategy turns every player
into a potential distributor and puts the skeleton keys to cyberspace
into millions of hands.

So perhaps you should ask John Carmack this question. :-)

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Marin D. Condic]

Except that when I buy insurance from a private company, I get to make a
bunch of decisions - the first of which is whether or not I want to buy it
at all. I can choose how much coverage I want, what terms and conditions I
am willing to accept, who will provide the medical care, who will provide
the insurance, how much I will pay, etc. If I am unhappy with the coverage I
have, I can make choices and change things.

When a government provides health care - choices go out the window. Quality
goes down. Costs go up. Rationing takes place. etc.

And it is a fundamental law of economics that health care is *never* going
to reach 100% of the population so arguing it needs to come from government
or some people will be left out will violate the laws of supply and demand.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com

Enabling Digital.
Our Vision is to be the biggest supplier worldwide of digital gateway
technology.
www.pacemicro.com

"Preben Randhol" <randhol+news@pvv.org> wrote in message
news:slrnamk70m.2po.randhol+news@kiuk0156.chembio.ntnu.no...
>
> Well it is no excess of 50% as you say. Besides how much money do you
> pay for insurance in case you have to go to the hospital? The system is
> different here. Everybody will get medical care no matter if they have
> money or not at a hospital. This is only one example. Education is
> another. So you cannot simply compare it to the low tax rates you may
> have you have to take into account what you have to pay to the different
> other service suppliers too.
>

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Darren New]

Ted Dennison wrote:
> The original ID games (Castle Wolfensein 3D and Doom) were released
> shareware,

Yes, long after the relatively short timeframe of profitability. Releasing
DOOM as shareware after sales have tapered off to a dribble is a wise
business decision. They're not making any more money at it because it's
already been widely pirated, so you might as well give away the source so
that your competitors get hit as well.

> So perhaps you should ask John Carmack this question. :-)

Carmack didn't release DOOM as shareware. He released it as fee-ware, then
later re-released it as shareware. My question was, if you knew you didn't
have to pay for it ever, even on the first day, how many people would?

> The most popular computer game of all time, Doom 

Considering that Myst has *also* been quoted as being the most popular
computer game of all time, I'm not sure I trust either set of statistics.
How do they know how many people are playing the shareware version? No
citation there.

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
   ** http://images.fbrtech.com/dnew/ **

Try our EbolaBurgers...
      So tender they melt in your mouth.

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Robert A Duff]

dennison@telepath.com (Ted Dennison) writes:

> The most popular computer game of all time, Doom has sold more than 2
> million copies. But that only hints at its popularity. As many as 20
> million shareware versions are installed worldwide - unregistered but
> perfectly legal copies, distributed free with id's blessing. 

I thought "shareware" meant that the software is *not* free (neither in
the beer nor speech sense), but that payment is on the honor system
(try it out, and if you like it and intend to use it long term, send in
some money).

- Bob

Re: Ada 95 for an ARM-based bare board?

[Florian Weimer]

tmoran@acm.org writes:

> Sticking with your core business leaves you with nothing more than
> your core business.

For individuals, this might get boring, but corporations have no
feelings, so it doesn't matter.  And even if you don't change business
every five years (first Visual Basic, then Java, now C# or something
like that), the actual work done by people doesn't have to be boring
at all, and that's what people care about (among other things).

Re: Ada 95 for an ARM-based bare board?

[Randy Brukardt]

Robert Dewar wrote in message
<5ee5b646.0208231819.981a776@posting.google.com>...
>"Juha Valimaki" <juha.valimaki@xdsloy.com> wrote in message
news:<ak4vaj$5g6$1@phys-news1.kolumbus.fi>...
>
>> Yes, I agree with you. It's just shame that sometimes all users don't
fit in
>> the same business model. The two bad things I see in ACT business
model is
>> that (1) it makes it hard for other companies to offer cheap, but not
free,
>> Ada tools
>
>No one ever succeeded at doing this who tried in the past, and the
attempts
>at this all failed when there *WAS* no freely available version of
>Ada.

That's a truly odd thing to say, because it is so obviously False. The
last great attempt (ObjectAda) certainly occurred after GNAT was around.
And we (RR Software) were planning to try again until GNAT destroyed our
core business (inexpensive Ada compilers for students, individuals, and
small teams). Without the core revenue, we simply couldn't support
another push in that direction.

                Randy Brukardt

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Ted Dennison]

Darren New wrote:
> Carmack didn't release DOOM as shareware. He released it as fee-ware, then

This is wrong. You really should have read a bit more carefully. Wolf3D 
and Doom were released as shareware. You are thinking of their GPL 
releases, which are indeed downrev versions of their older titles. Until 
sometime around the release of Quake, ID's sole income sources were 
voluntarily submitted shareware fees and engine licenseing fees.

> later re-released it as shareware. My question was, if you knew you didn't
> have to pay for it ever, even on the first day, how many people would?
> 

And this article answers the question with "if its as good as Wolf3D or 
Doom (in their day), then more than enough for the whole development 
staff to buy multiple ferraris" :-)

>>The most popular computer game of all time, Doom 
> 
> Considering that Myst has *also* been quoted as being the most popular
> computer game of all time, I'm not sure I trust either set of statistics.

Doom came out before Myst. This is an old article.

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Ted Dennison]

Robert A Duff wrote:
> I thought "shareware" meant that the software is *not* free (neither in
> the beer nor speech sense), but that payment is on the honor system
> (try it out, and if you like it and intend to use it long term, send in
> some money).

Exactly. So this is the perfect way to answer Darren's question:

--
OK, put it this way. The latest video game is released. Everyone gets to
copy it for free. Indeed, before it is announced, everyone knows they'll 
be able to get free copies. How many copies actually sell?

Re: Ada 95 for an ARM-based bare board?

[Juha Valimaki]

"Robert Dewar" <dewar@gnat.com> wrote in message
news:5ee5b646.0208231819.981a776@posting.google.com...
> We think that it very much helps all segments of the Ada market to have a
> high quality Ada compiler that students, hobbyists, and indidual
> enthusiastic engineers can access.
>

If some kind of proof is needed why small markets _need_ every customer look
at it this way:
X is number of people using free tools
Y is number of people using low end commercial tools
Z is number of people using high end commercial tools

X+Y+Z is the group of all Ada users. If you take away the free tools then N%
of group X will buy a tool. N%*X+Y+Z > Y+Z, if N>0. Note however: Y+Z is
likely to be smaller than X+Y+Z, but the market will be larger in terms of
money!.

> If there is a viable business in providing low cost products to small
> development teams, then someone should by all means step in, but
> please don't expect Ada Core Technologies to serve all needs of the Ada
> community. We can't and we won't try.
>

For me it seems you are afraid that low end solutions would eat up your
revenue and that Ada would be used more in general purpose programs that
don't focus on mission critical code, which seems to be your main focus.
That's a perfectly legal reason to suppress competition :-)

I really believe Ada tool vendors would understand that there has to be some
way for people to try Ada and do research without large cost. There are
already good solutions for that: Institutions get tools usually for (almost)
free and demo/limited versions help new users to start coding.

> Robert Dewar
> Ada Core Technologies

Juha

BTW: Low end is relative, for many hobby users $500 may max price, but I
think many small companies would easily pay $2000-$4000 per developer (PER
DEVELOPER is the key here).

--------------------------------------
remove x from domain name to reply

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Preben Randhol]

"Marin D. Condic" <not.valid@acm.org> wrote on 27/08/2002 (12:41) :
> When a government provides health care - choices go out the window.
> Quality goes down. Costs go up. Rationing takes place. etc.

This does not apply to hospitals at least. Your american hospitals are
very expensive and little cost efficient as far as I know.

-- 
Preben Randhol ------------------- http://www.pvv.org/~randhol/ --
                 �For me, Ada95 puts back the joy in programming.�

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Dale Stanbrough]

Marin D. Condic wrote:

> When a government provides health care - choices go out the window. Quality
> goes down. Costs go up. Rationing takes place. etc.

<cough, cough, splutter> Australia has a principally government
funded (with 10-20%? direct taxpayer contribution) which is 
regarded as one of the most efficient, effective systems in 
the world.

I have had very good experiences with it. It's not perfect, 
but it's much -much- better than anything in the US.

> And it is a fundamental law of economics that health care is *never* going
> to reach 100% of the population so arguing it needs to come from government
> or some people will be left out will violate the laws of supply and demand.

I'm not sure what % of the population is reached by the Aus.
system, but it's my guess that most anyone who wants it can 
get it. Currently there are problems with the lack of doctors in
rural areas. I'm not quite sure what you are sayig in your last
sentence.

Dale

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Bill Tate]

We can go around & around on this.  Suffice it to say, I prefer to
solve my own problems and make my own decisions and accept
responsibility for those decisions rather than gov't institutions
making those decisions for me.

As far as your analogy the "rich get richer, poor get poorer" - if you
want to select out anecdotal examples fine.  However, I would point
out, again, that the enron employees lost most of their net worth in
401K's invested in a single company's stock.  An idiotic and risky
investment strategy to say the very least.  Regardless of whether an
individuals portfolio loss could be attributed to fraud, bad
management, an ailing economy, lousy products, obsolescence, etc., an
individual investor has a responsibility of understanding how a
significant hit of any kind in ANY area of their portfolio would have
on their net worth.  Once understood, they have an obligation to
consider whether this is something they can live with?  Lastly, they
an obligation to consider how long would it take to recover from such
a loss?  If the risks are unacceptable, then you spread the risk
around.  It ain't rocket science and the math is pretty damn
straight-forward.

I certainly feel compassion for these folks because it was fraud that
caused their loss, but I don't feel an obligation to provide (i.e.,
pay for) a security blanket for someone else's poor financial
decision.  A decision that any high school graduate would have known
to avoid.  Especially when you consider the myriad of other investment
options available to anyone that are substantially less risky and far
more secure than the choice these individuals made.

And as far as the "scandal" plagued U.S.  No one is excusing the fraud
and the nature of the U.S. economy certainly doesn't give license to
it & it is certainly not predisposed to it as the papers would
suggest.  If you think otherwise, then we should talk about some
really interesting european adventures in this area.

Re: mailers, quoting text etc. (was Ada 95 for an ARM-based bare board?)

[Florian Weimer]

Kilgallen@SpamCop.net (Larry Kilgallen) writes:

> The defects introduced by Microsoft starting with Word 6 (allowing
> instructions provided by someone else to execute in the context of
> the recipient)

Huh? Multiplan 3.x or even earlier version supported autoload macros.

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Florian Weimer]

dewar@gnat.com (Robert Dewar) writes:

> Second point, the only difference between Microsoft and ACT is the
> license we use. We both sell licensed copyrighted software.

Oh, is Microsoft offering real support (with certain guarantees) for
their Visual product line of development tools now?

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[tmoran]

> > Second point, the only difference between Microsoft and ACT is the
> > license we use. We both sell licensed copyrighted software.
>
> Oh, is Microsoft offering real support (with certain guarantees) for
> their Visual product line of development tools now?
  In a way - you'll get continued upgrades as long as you continue
paying the rent.

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Bill Tate]

Preben Randhol <randhol+ada@pvv.org> wrote in message news:<mailman.1030445042.8161.comp.lang.ada@ada.eu.org>...
> "Marin D. Condic" <not.valid@acm.org> wrote on 27/08/2002 (12:41) :
> > When a government provides health care - choices go out the window.
> > Quality goes down. Costs go up. Rationing takes place. etc.

 
> This does not apply to hospitals at least. 
   Says who? I'm not sure where you are getting your information but
rationing is for real in both Canada & Britain which hardly strikes me
as being examples of better quality or more efficient. And their costs
are going up....

Regardless, what I don't want is some faceless gov't bureaucrat trying
to tell me that they care more about me or my family than I do.  Your
system means their choices & their decisions on what is in your "best
interest."  Excuse me if I find that whole notion abhorent,
condescending & elitist.

Re: Ada 95 for an ARM-based bare board?

[Michael Bode]

"Juha Valimaki" <juha.valimaki@xdsloy.com> writes:

> I really believe Ada tool vendors would understand that there has to be some
> way for people to try Ada and do research without large cost. 

So you can take your vanilla Linux distributoin and use the Ada
compiler that comes with it. Or download the free Gnat for Windows.

I've seen that there is also Janus Ada + Claw for US$ 295. I don't
know this product but the price competes well with any other
commercial Windows development package I've seen so far. They
seemingly don't have a Linux version, so I would still use Gnat and
GtkAda.

Re: OT: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Bill]

Bill Tate wrote:

> Preben Randhol <randhol+ada@pvv.org> wrote in message news:<mailman.1030445042.8161.comp.lang.ada@ada.eu.org>...
> > "Marin D. Condic" <not.valid@acm.org> wrote on 27/08/2002 (12:41) :
> > > When a government provides health care - choices go out the window.
> > > Quality goes down. Costs go up. Rationing takes place. etc.
>
>
> > This does not apply to hospitals at least.
>    Says who? I'm not sure where you are getting your information but
> rationing is for real in both Canada & Britain which hardly strikes me
> as being examples of better quality or more efficient. And their costs
> are going up....
>
> Regardless, what I don't want is some faceless gov't bureaucrat trying
> to tell me that they care more about me or my family than I do.  Your
> system means their choices & their decisions on what is in your "best
> interest."  Excuse me if I find that whole notion abhorent,
> condescending & elitist.

As opposed to a faceless industry beaurocrat? Any attempt to pool resources to reduce risks and leverage buying
power requires some means of deciding what is best for the participating parties. In some fields that can be
determined directly by a simple contract, but that has never been fully the case in a rapidly changing and diverse
area such as health care.

Health insurance has always required a beurocracy. In the US, employer provided health care plans have dominated
the health care system as a relatively direct means of identifying large groups of people available to pool
resources. While private plans are available which are more flexible than the employer plans, the combined effects
of higher costs of finding people to join the pool and cost control by excluding people with know problems have
made those plans inappropriate for most people.

For many decades the high per capita income in the US, relatively young population, and strong unions mitigated the
effects of this beaurocracry, for most citizens. But the aging of the population has increased the fraction of the
population needing frequent care, and many medical advances have allowed improvements in treatment but at a much
higher incremental cost. These costs have in effect forced all insurance plans to examine cost controls, rationing
in particular. The weakness of the unions means that in employer provided health care plans, the employer is likely
to pay less attention to the desires of the employees in reducing costs. When the US resisted a wider government
provided health care system in the early 90's, the net result was that some of the population lost all coverage as
their employers dropped their plans, and most of the remaining covered population had their coverage changed to
HMOs. If you think HMOs don't impose more rationing than the plans they replaced then I have a bridge in Brooklyn
I'd like to sell you at a very cheap cost.

While HMOs are not yet as bad as the British health care system, they have been significant for only about ten
years, the British system has had over fifty years to get into the state its in.  Lets see what happens to the pool
of doctors, nurses, hospitals, and pharmacies over the next twenty five years.

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

"Juha Valimaki" <juha.valimaki@xdsloy.com> wrote in message news:<akf4bg$454$1@phys-news1.kolumbus.fi>...

> If some kind of proof is needed why small markets _need_ every customer look
> at it this way:
> X is number of people using free tools
> Y is number of people using low end commercial tools
> Z is number of people using high end commercial tools
> 
> X+Y+Z is the group of all Ada users. If you take away the free tools then N%
> of group X will buy a tool. N%*X+Y+Z > Y+Z, if N>0. Note however: Y+Z is
> likely to be smaller than X+Y+Z, but the market will be larger in terms of
> money!.

A completely bogus argument, because it ignores the fact that the
existence
of a free tool may increase the total number of users, so the X/Y/Z
are by
no means constant, and you have no evidence that they are (or rather
would be)


> For me it seems you are afraid that low end solutions would eat up your
> revenue

Nope, that has nothing to do with it, it is just that this end of the
market
is not what we specialize in, and we believe our success (our sales
are substantially up this year) is significantly due to the fact that
we keep
a narrow well defined focus.

> and that Ada would be used more in general purpose programs that
> don't focus on mission critical code

No, we don't believe that at all! If you believe it why not start a
business
aiming at these users, since it sounds like you believe it would be
profitable.
By the way, notice that you are also agreeing that the X/Y/Z may not
be
constant in your above equation (though in the opposite direction from
me :-)

> That's a perfectly legal reason to suppress competition :-)

Nobody is suppressing competition here. We are one of the very few
software
companies that invites you to compete its market, and hands you at no
cost
the technology that you need to compete.

> I really believe Ada tool vendors would understand that there has to be some
> way for people to try Ada and do research without large cost. 

For trying ada and doing research, the public version of GNAT is
definitely
appropriate, and for sure does not have a large cost.


> but I
> think many small companies would easily pay $2000-$4000 per developer (PER
> DEVELOPER is the key here).

Well of course for small teams of people from 3 on up, that's
precisely our
price range, and you are right, many small companies are prepared to
pay this
for high level support.

The one case where we do not cater to this model is for smaller groups
(of
1 or 2 people), but as I said earlier, we did this at first, and found
that
these isolated users of GNAT expected far far more support than larger
groups
and we simply could not afford to give them the high quality support
that we
guarantee at the single seat price. We actually found that the typical
pattern
was that one person teams asked far more questions than a five person
team,
but we really can't charge more, but we do charge the same!

Robert Dewar
Ada Core Technologies

Re: Ada 95 for an ARM-based bare board?

[Dale Stanbrough]

Robert Dewar wrote:

> > That's a perfectly legal reason to suppress competition :-)
> 
> Nobody is suppressing competition here. We are one of the very few
> software
> companies that invites you to compete its market, and hands you at no
> cost
> the technology that you need to compete.


Could I have your non proprietory test suites, and your 
testing software (which you have previously said took you
several months of a couple of people to develop :-) then?


Dale :-)

Re: Ada 95 for an ARM-based bare board?

[Georg Bauhaus]

Juha Valimaki <juha.valimaki@xdsloy.com> wrote:
: X is number of people using free tools
: Y is number of people using low end commercial tools
: Z is number of people using high end commercial tools

Can you describe your idea of why people choose to obtain
development tools? What reasons do they have to choose a vs b?

One possible question to ask vis a vis development tools
might be, "can I by knowledge?" :-)

-- Georg

Re: Software Economics was RE: Ada 95 for an ARM-based bare board?

[Preben Randhol]

On Tue, 27 Aug 2002 18:52:15 GMT, tmoran@acm.org wrote:
>> > Second point, the only difference between Microsoft and ACT is the
>> > license we use. We both sell licensed copyrighted software.
>>
>> Oh, is Microsoft offering real support (with certain guarantees) for
>> their Visual product line of development tools now?
>   In a way - you'll get continued upgrades as long as you continue
> paying the rent.

In advance.

Preben

Re: Ada 95 for an ARM-based bare board?

[Tucker Taft]

Green Hills Software supports both Ada and ARM.  You
should certainly check with them whether they currently
have an Ada offering certified for the ARM.  Even if they
don't, they might be able to create something.

Start at:  http://www.ghs.com

Note also that SofCheck, Inc. has a validated
Ada 95 compiler that uses optimized ANSI C as its
intermediate representation.  This has been successfully
used for relatively quick ports to new architectures which
have a decent C compiler.  If you are interested
in pursuing this, please contact us.

-Tucker Taft  stt@sofcheck.com
President, SofCheck, Inc.

(SofCheck now owns and maintains the AdaMagic(tm) technology 
 developed by Intermetrics, Inc., aka AverStar, AverCom/Titan)

Re: Ada 95 for an ARM-based bare board?

[Dale Stanbrough]

Tucker Taft wrote:

> (SofCheck now owns and maintains the AdaMagic(tm) technology 
>  developed by Intermetrics, Inc., aka AverStar, AverCom/Titan)

Whew! For a moment there I thought they'ld changed their name
yet again!

Dale :-)

Re: Ada 95 for an ARM-based bare board?

[Robert A Duff]

Dmitry A.Kazakov <mailbox@dmitry-kazakov.de> writes:

> Tucker Taft wrote:
> > Note also that SofCheck, Inc. has a validated
> > Ada 95 compiler that uses optimized ANSI C ...
> 
> But one general question, what about run-time? We have a bare board with no 
> OS planned. To my limited knowledge to translate from Ada to ARM is not the 
> main problem here.

The run-time system is written in Ada.  It needs a small amount of
configuration for the target.  If you have an RTOS and want Ada tasks
mapped to RTOS threads, it requires more substantial work (which we also
have done in one case).  But you have no OS, so it's pretty easy.

> > -Tucker Taft  stt@sofcheck.com
> > President, SofCheck, Inc.
> > 
> > (SofCheck now owns and maintains the AdaMagic(tm) technology
> >  developed by Intermetrics, Inc., aka AverStar, AverCom/Titan)
> 
> -- 
> Regards,
> Dmitry Kazakov
> www.dmitry-kazakov.de

- Bob Duff  bobduff@sofcheck.com
  Vice President, SofCheck, Inc.

Re: Ada 95 for an ARM-based bare board?

[Richard Riehle]

"Dmitry A.Kazakov" wrote:

> Tucker Taft wrote:
>
> > Even if they
> > don't, they might be able to create something.
>
> I have no doubt about it. The question is who will pay for the banquet?

Nyeekogda nye cvabodnyee obyed!   (There is no free lunch)

Richard Riehle

data on cla messages, was Re: Ada 95 for an ARM-based bare board?

[tmoran]

> > Let's suppose that everyone adds 10 lines and quotes everything before
>   Let's suppose that everyone adds 10% to what they are quoting.
  Looking at the last 228 messages on CLA, containing 7350 total body
(not header) lines, 2118 of which are quote lines, I find the
length of a body decreased on average by 1.8 lines from the message
it referenced, with standard deviation 23.4, or, taking ratios,
bodies decrease to .995, standard deviation 2.2  For two messages
both containing quoted lines, the number of quoted lines increased
by 1.4, s.d. 12.1, or to 119%, s.d. 3.5  So the ratios have substantially
less variance than the deltas, indicating an exponential fits the
data much better than a linear model.

Re: Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

Tucker Taft wrote:

> Green Hills Software supports both Ada and ARM.  You
> should certainly check with them whether they currently
> have an Ada offering certified for the ARM. 

AFAIK no, for ARM they have only C.

> Even if they
> don't, they might be able to create something.

I have no doubt about it. The question is who will pay for the banquet? 
(:-))

> Start at:  http://www.ghs.com
> 
> Note also that SofCheck, Inc. has a validated
> Ada 95 compiler that uses optimized ANSI C as its
> intermediate representation.  This has been successfully
> used for relatively quick ports to new architectures which
> have a decent C compiler.  If you are interested
> in pursuing this, please contact us.

Thanks. I will.

But one general question, what about run-time? We have a bare board with no 
OS planned. To my limited knowledge to translate from Ada to ARM is not the 
main problem here.

> -Tucker Taft  stt@sofcheck.com
> President, SofCheck, Inc.
> 
> (SofCheck now owns and maintains the AdaMagic(tm) technology
>  developed by Intermetrics, Inc., aka AverStar, AverCom/Titan)

-- 
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

tmoran@acm.org wrote in message news:<klia9.74$Bg4.9214321@newssvr14.news.prodigy.com>...
> > Let's suppose that everyone adds 10 lines and quotes everything before
>   Let's suppose that everyone adds 10% to what they are quoting.  Then
> the size of message N is 1.1**N * OP.

Tom you are digging yourself in deeper here in an attempt
to justify your basic error :-)

If you make the absurd assumption that the size of replies
asymptotically increases by 10%, then the total amount of
material filed increases exponentially WITHOUT quoting.
That's of course obvious but totally irrelevant.

It is true that sometimes the drivel on this newsgroup
in some threads gives the impression of increasing with
every reply, but even the CLA folks can't keep that up
for ever :-)

> > Let's suppose that everyone adds 10 lines and quotes 
> > everything before
> > So the size of messages grows only linearly
>   Right.  In that case new messages don't grow 
> > exponentially, but linearly.

Yes, but my comment was about the size of the thread.

> > But, by elementary summing of an arithmetic series, the 
> > total amount of space
>   The reader of a message doesn't see the sum, but rather 
> > the latest
> > individual message (unless he has a really strange mail 
> > reader).

Someone reading the thread will see the quadratic 
accumulation.

>   In a more perfect world, everyone would quote, say, 2 
> lines for context,
> add a concise 3 line comment, and all messages would be a 
> constant 5 lines.

That's a silly comment, especially without a smiley. Sometimes you
have to quote quite a bit, sometimes
only a little. Sometimes you can say what you are saying
briefly, sometimes it takes longer. Some of my most useful
posts are in the form of lengthy tutorials. Now that Google
has the full archives, I am extracting these into a book :-)

The underlying annoying behavior here is people using
microjunk mailers set to quote the entire message by
default (yes, I realize some other mailers besides microjunk ones can
make this too easy as well -- people
have a nasty habit of following the MJ lead :-)

Re: mailers, quoting text etc. (was Ada 95 for an ARM-based bare board?)

[chris.danx]

Robert Dewar wrote:

> Really?
> 
> Typical microsoft mailers make the unseen addition of completely useless
> HTML by default that is
> 
> a) redundant and serves no purpose
> b) wastes bandwidth and clutters up databases
> c) presents a security risk (HTML attachments are not benign!)
> 
> But I guess none of these (especially c) has ever been of any concern to MS :-)

That is why eveyrone with sense should use a decent mailer like Mozilla 
Mail, Eudora or whatever and avoid MS mailers like Outlook and Outlook 
Express, or use some script/html blocking software with OE or Outlook.


Chris
-- 
for personal replies change 'spamoff' to 'chris'

Re: mailers, quoting text etc. (was Ada 95 for an ARM-based bare board?)

[chris.danx]

Larry Kilgallen wrote:

> There have been public statements from Microsoft that the "features"
> provided by these loopholes are more important to their customers
> than fixing security problems.

What someone should do is use a sandboxed version of (X)HTML that 
doesn't have scripts.

Chris
-- 
for personal replies change 'spamoff' to 'chris'

Re: Ada 95 for an ARM-based bare board?

[tmoran]

> If you make the absurd assumption that the size of replies
> asymptotically increases by 10%, then the total amount of
  Apparently you didn't read my post with the subject changed to
"data on cla messages, was Re: Ada 95 for an ARM-based bare board?"
Asymptotically the size of messages decreases to zero when a thread dies.
In the short term, however, quote growth in CLA is 14.8% (based on 261
messages) from one message with quotes over its referent with quotes.
Total body length shrinks to 97.9% on successive messages.  The
exponential fits the data quite a bit better than a linear model.
  There's a story that an early professor of aeronautical engineering
proved that bumblebees can't fly.  His calculations were probably
correct, so it must have been his model that was wrong.
> Sometimes you can say what you are saying briefly, sometimes it takes
> longer.  Some of my most useful posts are in the form of lengthy tutorials.
  I'm sure that the value of certain people's posts is directly proportional
to their size.  That's usually not the case, however.

Re: Ada 95 for an ARM-based bare board?

[Robert Dewar]

tmoran@acm.org wrote in message news:<_16c9.144872$aA.31015@sccrnsc02>...

Tom, your attempt to use bogus statistics (I won't even
bother to analyze them in detail here) to cover up a basic
mistake is instructive. I will point my students to this
as an example of how statistics can be used to
obfuscate :-)

The fundamental fact remains that under the only reasonable
assumption which is that added text is approximately constant, the
effect of quoting all previous messages adds
a quadratic effect, not an exponential effect.

Re: Ada 95 for an ARM-based bare board?

[Dmitry A.Kazakov]

Richard Riehle wrote:

> "Dmitry A.Kazakov" wrote:
> 
>> Tucker Taft wrote:
>>
>> > Even if they
>> > don't, they might be able to create something.
>>
>> I have no doubt about it. The question is who will pay for the banquet?
> 
> Nyeekogda nye cvabodnyee obyed!   (There is no free lunch)

It is rather like paying for construction and equipment of the whole 
restaurant when you just want a lunch. (:-))

[Besplatnij sir bivaet tolko v mishelovke.]

-- 
Regards,
Dmitry Kazakov
www.dmitry-kazakov.de