I would like write my application with ADA language. Where can I find ADA
compiler for QNX 6
Bwaahaahaa ha ha snort cough…hee…
Umm…sorry about that.
As far as I know we don’t have any plans to make Ada available any time
soon. The next generation of gcc (version 3) has Ada support built in (much
like the current version has Fortran) so, in theory, when we eventually move
to version 3, it might be possible to compile Ada on QNX 6. That being
said, there is a lot more to a language than a compiler which is precicely
the reason we don’t support Fortran. If you’ve got a working compiler, you
still need headers and libs in order to access operating system services,
etc. so supporting a new compiler toolchain is decidedly non-trivial. I
think that someone would need to make a very strong business case for
something like that to happen. (along the lines of, “Hi…we’re from the
Department of Defense and we’d like to spend 10 gazillion dollars on our new
system but it needs to be written in ADA” would probably be about right 
.
So the short answer is, “Sorry, we don’t have your model in stock. May I
interest you in this lovely C++ compiler? It’s only got 150000 kilometers
on it, lady driven…a real beaut”.
cheers,
Kris
“Voncken” <cedric.voncken@acksys.fr> wrote in message
news:a530cd$jkv$1@inn.qnx.com…
I would like write my application with ADA language. Where can I find ADA
compiler for QNX 6
Kris Warkentin wrote:
So the short answer is, “Sorry, we don’t have your model in stock. May I
interest you in this lovely C++ compiler? It’s only got 150000 kilometers
on it, lady driven…a real beaut”.
I would also love Ada support for QNX, however, I realize that there
must be a business case.
OTOH, there is, decidedly, a case for C++ support, and we don’t have
that yet (sorry but 2.95 doesn’t quite qualify as an industrial strength
C++ compiler IMO). When you have a C++ compiler (preferably with high
mileage on it - but low mileage with all the options is OK too 
I’ll
buy. It is rather embarassing to have to admit that VC++ is currently a
better tool than what is available on QNX (and I’m not talking about the
ornamental chihuahua in the back window with the head that bobs up and
down - which VC++ has a lot of, and I don’t give a hoot about - but
basic compiler features).
btw: I think that if QSSL worked with ACT, Ada would actually be lower
cost to obtain than a truly industrial strength 100% standard conformant
C++ compiler (getting a defective C++ compiler is cheap and easy, but
doesn’t do much for anyone). GNATS is industrial strength and exhibits
a very high degree of conformance to the ISO Ada95 standard (I’m not
sure if that can be said even for the C++ support with gcc 3.0). Again,
I would be happy with either Ada or C++, my only concern is compiler
quality.
Well, I’ve often thought that “business case” is a bit of a chicken and egg
problem. After all, just because people aren’t ASKING for an ADA compiler
(or Linux based SDK or …) doesn’t mean they wouldn’t buy one if it was
available. I mean, how many military projects out there have a list of
requirements that might include ADA support so when they look at various
suppliers they go, “Oops…no ADA. Next!”? I agree with your point
however, that it would probably be relatively cheap to just get GNATS or
someone to support ADA on QNX6, rather than doing it ourselves.
cheers,
Kris
“Rennie Allen” <rallen@csical.com> wrote in message
news:3C752CE3.7040801@csical.com…
Kris Warkentin wrote:
So the short answer is, “Sorry, we don’t have your model in stock. May
I
interest you in this lovely C++ compiler? It’s only got 150000
kilometers
on it, lady driven…a real beaut”.
I would also love Ada support for QNX, however, I realize that there
must be a business case.OTOH, there is, decidedly, a case for C++ support, and we don’t have
that yet (sorry but 2.95 doesn’t quite qualify as an industrial strength
C++ compiler IMO). When you have a C++ compiler (preferably with high
mileage on it - but low mileage with all the options is OK too >
buy. It is rather embarassing to have to admit that VC++ is currently a
better tool than what is available on QNX (and I’m not talking about the
ornamental chihuahua in the back window with the head that bobs up and
down - which VC++ has a lot of, and I don’t give a hoot about - but
basic compiler features).btw: I think that if QSSL worked with ACT, Ada would actually be lower
cost to obtain than a truly industrial strength 100% standard conformant
C++ compiler (getting a defective C++ compiler is cheap and easy, but
doesn’t do much for anyone). GNATS is industrial strength and exhibits
a very high degree of conformance to the ISO Ada95 standard (I’m not
sure if that can be said even for the C++ support with gcc 3.0). Again,
I would be happy with either Ada or C++, my only concern is compiler
quality.
“Kris Warkentin” <kewarken@qnx.com> wrote in message
news:a53hed$8jp$1@nntp.qnx.com…
Well, I’ve often thought that “business case” is a bit of a chicken and
egg
problem. After all, just because people aren’t ASKING for an ADA compiler
(or Linux based SDK or …) doesn’t mean they wouldn’t buy one if it was
available. I mean, how many military projects out there have a list of
requirements that might include ADA support so when they look at various
suppliers they go, “Oops…no ADA. Next!”? I agree with your point
however, that it would probably be relatively cheap to just get GNATS or
someone to support ADA on QNX6, rather than doing it ourselves.
That’s brings an interesting issue; how does one evaluate the market for QNX
third party product. Let’s said I’d be interested in doing an ADA port
(just as an example…). How can I get a feel of the possible number of
sales,
how to evaluate return on investment. I have no access to QNX market,
aside these newgroups (which only reaches a VERY SMALL portion of
the market)
Probably some more meat for “advocacy”
cheers,
Kris
“Rennie Allen” <> rallen@csical.com> > wrote in message
news:> 3C752CE3.7040801@csical.com> …
Kris Warkentin wrote:
So the short answer is, “Sorry, we don’t have your model in stock.
May
I
interest you in this lovely C++ compiler? It’s only got 150000
kilometers
on it, lady driven…a real beaut”.
I would also love Ada support for QNX, however, I realize that there
must be a business case.OTOH, there is, decidedly, a case for C++ support, and we don’t have
that yet (sorry but 2.95 doesn’t quite qualify as an industrial strength
C++ compiler IMO). When you have a C++ compiler (preferably with high
mileage on it - but low mileage with all the options is OK too >
buy. It is rather embarassing to have to admit that VC++ is currently a
better tool than what is available on QNX (and I’m not talking about the
ornamental chihuahua in the back window with the head that bobs up and
down - which VC++ has a lot of, and I don’t give a hoot about - but
basic compiler features).btw: I think that if QSSL worked with ACT, Ada would actually be lower
cost to obtain than a truly industrial strength 100% standard conformant
C++ compiler (getting a defective C++ compiler is cheap and easy, but
doesn’t do much for anyone). GNATS is industrial strength and exhibits
a very high degree of conformance to the ISO Ada95 standard (I’m not
sure if that can be said even for the C++ support with gcc 3.0). Again,
I would be happy with either Ada or C++, my only concern is compiler
quality.
“Mario Charest” <goto@nothingness.com> wrote in message
news:a53mvf$5hg$1@inn.qnx.com…
“Kris Warkentin” <> kewarken@qnx.com> > wrote in message
news:a53hed$8jp$> 1@nntp.qnx.com> …
Well, I’ve often thought that “business case” is a bit of a chicken and
egg
problem. After all, just because people aren’t ASKING for an ADA
compiler
(or Linux based SDK or …) doesn’t mean they wouldn’t buy one if it
was
available. I mean, how many military projects out there have a list of
requirements that might include ADA support so when they look at various
suppliers they go, “Oops…no ADA. Next!”? I agree with your point
however, that it would probably be relatively cheap to just get GNATS or
someone to support ADA on QNX6, rather than doing it ourselves.That’s brings an interesting issue; how does one evaluate the market for
QNX
third party product. Let’s said I’d be interested in doing an ADA port
(just as an example…). How can I get a feel of the possible number of
sales,
how to evaluate return on investment. I have no access to QNX market,
aside these newgroups (which only reaches a VERY SMALL portion of
the market)Probably some more meat for “advocacy”
American (and probably Canadian) support for Ada has always been low, I
think you would have to look at, and possibly contact some of the QNX sales
reps in Europe where Ada is seen in a more favorable light. I for one like
the language, as I have said before. It’s possible that supporting Ada
might open a few more doors in the banking industry as well as the military
one. I think I heard a while back that the banking industry in Europe had
embraced it.
Jerry Kirk
Jerry Kirk wrote:
It’s possible that supporting Ada
might open a few more doors in the banking industry as well as the military
one.
As well as civilian aerospace.
Rennie
My wife works at NavCanada (formerly Transport Canada) where they take care
of air traffic in this country. She is working on the 10 year old, billion
dollar CAATS (Canadian Automated Air Traffic System) project which is done
in ADA. From what I understand, other parts of the company have used QNX in
other projects. shrug I personally think that richness of a product
offering is always a good thing but our problem is the amount to do and the
people to do it. Our tools group is small and supporting C and C++ is about
all we can do with the staff we have. We just don’t have the manpower and
expertise right now to support an ADA toolchain. Hey…maybe they’ll hire
my wife.
cheers,
Kris
“Mario Charest” <goto@nothingness.com> wrote in message
news:a53mvf$5hg$1@inn.qnx.com…
“Kris Warkentin” <> kewarken@qnx.com> > wrote in message
news:a53hed$8jp$> 1@nntp.qnx.com> …
Well, I’ve often thought that “business case” is a bit of a chicken and
egg
problem. After all, just because people aren’t ASKING for an ADA
compiler
(or Linux based SDK or …) doesn’t mean they wouldn’t buy one if it
was
available. I mean, how many military projects out there have a list of
requirements that might include ADA support so when they look at various
suppliers they go, “Oops…no ADA. Next!”? I agree with your point
however, that it would probably be relatively cheap to just get GNATS or
someone to support ADA on QNX6, rather than doing it ourselves.That’s brings an interesting issue; how does one evaluate the market for
QNX
third party product. Let’s said I’d be interested in doing an ADA port
(just as an example…). How can I get a feel of the possible number of
sales,
how to evaluate return on investment. I have no access to QNX market,
aside these newgroups (which only reaches a VERY SMALL portion of
the market)Probably some more meat for “advocacy”
cheers,Kris
“Rennie Allen” <> rallen@csical.com> > wrote in message
news:> 3C752CE3.7040801@csical.com> …
Kris Warkentin wrote:
So the short answer is, “Sorry, we don’t have your model in stock.
May
I
interest you in this lovely C++ compiler? It’s only got 150000
kilometers
on it, lady driven…a real beaut”.
I would also love Ada support for QNX, however, I realize that there
must be a business case.OTOH, there is, decidedly, a case for C++ support, and we don’t have
that yet (sorry but 2.95 doesn’t quite qualify as an industrial
strength
C++ compiler IMO). When you have a C++ compiler (preferably with
high
mileage on it - but low mileage with all the options is OK too >
I’ll
buy. It is rather embarassing to have to admit that VC++ is currently
a
better tool than what is available on QNX (and I’m not talking about
the
ornamental chihuahua in the back window with the head that bobs up and
down - which VC++ has a lot of, and I don’t give a hoot about - but
basic compiler features).btw: I think that if QSSL worked with ACT, Ada would actually be lower
cost to obtain than a truly industrial strength 100% standard
conformant
C++ compiler (getting a defective C++ compiler is cheap and easy, but
doesn’t do much for anyone). GNATS is industrial strength and
exhibits
a very high degree of conformance to the ISO Ada95 standard (I’m not
sure if that can be said even for the C++ support with gcc 3.0).
Again,
I would be happy with either Ada or C++, my only concern is compiler
quality.
\
Kris Warkentin <kewarken@qnx.com> wrote:
My wife works at NavCanada (formerly Transport Canada) where they take care
of air traffic in this country. She is working on the 10 year old, billion
dollar CAATS (Canadian Automated Air Traffic System) project which is done
in ADA. From what I understand, other parts of the company have used QNX in
other projects. shrug I personally think that richness of a product
offering is always a good thing but our problem is the amount to do and the
people to do it. Our tools group is small and supporting C and C++ is about
all we can do with the staff we have. We just don’t have the manpower and
“manpower”? you can always outsource to people like Mario
money is probably more of an issue. you need money to increase the
manpower or outsource.
expertise right now to support an ADA toolchain. Hey…maybe they’ll hire
my wife. >cheers,
Kris
<spamme@spam.com> wrote in message news:a55mos$jsg$1@inn.qnx.com…
Kris Warkentin <> kewarken@qnx.com> > wrote:
My wife works at NavCanada (formerly Transport Canada) where they take
care
of air traffic in this country. She is working on the 10 year old,
billion
dollar CAATS (Canadian Automated Air Traffic System) project which is
done
in ADA. From what I understand, other parts of the company have used
QNX in
other projects. shrug I personally think that richness of a product
offering is always a good thing but our problem is the amount to do and
the
people to do it. Our tools group is small and supporting C and C++ is
about
all we can do with the staff we have. We just don’t have the manpower
and“manpower”? you can always outsource to people like Mario >
money is probably more of an issue. you need money to increase the
manpower or outsource.
Money is implied by my statement of course but corporate will has just as
much to do with it. If the powers that be don’t see a market or are
concentrating on other markets, the money/manpower won’t be found. Like I
said before, there needs to be a business case. If Mario thinks that he can
make money with an ADA toolchain on QNX then he should do it. If he
doesn’t, then he won’t. Same goes for us. I DO however think that he had a
good idea about the third party stuff. There definitely should be better
mechanisms for us to open up opportunities for the ‘man in the middle’. We
should help customers and contractors/consultants/VARs/whatever get in touch
with each other, especially in markets that we’re not concentrating on.
Cheers,
Kris
expertise right now to support an ADA toolchain. Hey…maybe they’ll
hire
my wife. >cheers,
Kris
It’s done…just not in the nice way… “done” as in “Done like most of Mike
Tyson’s opponents”. They’ve been hacking this thing for 10 years (a lot of
the work was done by Raytheon) and they still don’t have a thing to show for
it but a couple million lines of obfuscated ADA (yes, it IS possible) and a
lot of frustration. When a software project is 10 years old and they’re
still changing requirements, specifications and design on a daily
basis…sigh. I just feel sorry for my wife getting stuck on a death
march project like that for her first job out of university. Better than
being unemployed I suppose.
cheers,
Kris
“Alex Cellarius” <acellarius@systems104-don’t-you-spam-me!.co.za> wrote in
message news:1105_1014409991@pentiumii…
On Fri, 22 Feb 2002 09:24:56 -0500, “Kris Warkentin” <> kewarken@qnx.com
wrote:
My wife works at NavCanada (formerly Transport Canada) where they take
care
of air traffic in this country. She is working on the 10 year old,
billion
dollar CAATS (Canadian Automated Air Traffic System) project which is
done
in ADA.I thought this project was long done and delivered?
(originally done by a Hughes/MDA/? partnership)Is it still going?
…
“Kris Warkentin” <kewarken@qnx.com> wrote in message
news:a55sru$sv$1@nntp.qnx.com…
spamme@spam.com> > wrote in message news:a55mos$jsg$> 1@inn.qnx.com> …
Kris Warkentin <> kewarken@qnx.com> > wrote:
My wife works at NavCanada (formerly Transport Canada) where they take
care
of air traffic in this country. She is working on the 10 year old,
billion
dollar CAATS (Canadian Automated Air Traffic System) project which is
done
in ADA. From what I understand, other parts of the company have used
QNX in
other projects. shrug I personally think that richness of a
product
offering is always a good thing but our problem is the amount to do
and
the
people to do it. Our tools group is small and supporting C and C++ is
about
all we can do with the staff we have. We just don’t have the manpower
and“manpower”? you can always outsource to people like Mario >
money is probably more of an issue. you need money to increase the
manpower or outsource.Money is implied by my statement of course but corporate will has just as
much to do with it. If the powers that be don’t see a market or are
concentrating on other markets, the money/manpower won’t be found. Like I
said before, there needs to be a business case. If Mario thinks that he
can
make money with an ADA toolchain on QNX then he should do it. If he
doesn’t, then he won’t.
How do I know Of course there is some risk to be taken but the risk has
to
be evaluated!
Same goes for us. I DO however think that he had a
good idea about the third party stuff. There definitely should be better
mechanisms for us to open up opportunities for the ‘man in the middle’.
We
should help customers and contractors/consultants/VARs/whatever get in
touch
with each other, especially in markets that we’re not concentrating on.
Tell me about it…
Cheers,
Kris
expertise right now to support an ADA toolchain. Hey…maybe they’ll
hire
my wife. >cheers,
Kris
On Fri, 22 Feb 2002 09:24:56 -0500, “Kris Warkentin” <kewarken@qnx.com> wrote:
My wife works at NavCanada (formerly Transport Canada) where they take care
of air traffic in this country. She is working on the 10 year old, billion
dollar CAATS (Canadian Automated Air Traffic System) project which is done
in ADA.
I thought this project was long done and delivered?
(originally done by a Hughes/MDA/? partnership)
Is it still going?
…
Kris Warkentin wrote:
It’s done…just not in the nice way… “done” as in “Done like most of Mike
Tyson’s opponents”. They’ve been hacking this thing for 10 years (a lot of
the work was done by Raytheon) and they still don’t have a thing to show for
it but a couple million lines of obfuscated ADA (yes, it IS possible) and a
lot of frustration. When a software project is 10 years old and they’re
still changing requirements, specifications and design on a daily
basis…sigh. I just feel sorry for my wife getting stuck on a death
march project like that for her first job out of university. Better than
being unemployed I suppose.
What is it with air traffic control systems ? They all seem to turn into
dead ends…
Rennie Allen <rallen@csical.com> writes:
Kris Warkentin wrote:
It’s done…just not in the nice way… “done” as in “Done like most of Mike
Tyson’s opponents”. They’ve been hacking this thing for 10 years (a lot of
the work was done by Raytheon) and they still don’t have a thing to show for
it but a couple million lines of obfuscated ADA (yes, it IS possible) and a
lot of frustration. When a software project is 10 years old and they’re
still changing requirements, specifications and design on a daily
basis…sigh. I just feel sorry for my wife getting stuck on a death
march project like that for her first job out of university. Better than
being unemployed I suppose.
What is it with air traffic control systems ? They all seem to turn into
dead ends…
… despite the fact that real-time programming books always use air traffic
control as the example ;->
Yeah, the Americans poured something like 750 million bucks into theirs
before they gave up. Canadians are much more stubborn though and are
probably at the billion dollar mark by now. I think it’s only a matter of
time before the project is scrapped. The thing is, it’s not rocket
science - you’ve just got to track a bunch of planes and make sure they
don’t run into each other. Lots of video games are way more complicated.
The big problem, IMHO, is that its SO cricical because any bug or problem
and people die. It seems like a lot of the complexity comes from the fact
that they have to interface with so many existing systems, not only here but
in other countries as well.
cheers,
Kris
“Rennie Allen” <rallen@csical.com> wrote in message
news:3C76C627.1060108@csical.com…
Kris Warkentin wrote:
It’s done…just not in the nice way… “done” as in “Done like most of
Mike
Tyson’s opponents”. They’ve been hacking this thing for 10 years (a lot
of
the work was done by Raytheon) and they still don’t have a thing to show
for
it but a couple million lines of obfuscated ADA (yes, it IS possible)
and a
lot of frustration. When a software project is 10 years old and they’re
still changing requirements, specifications and design on a daily
basis…sigh. I just feel sorry for my wife getting stuck on a death
march project like that for her first job out of university. Better
than
being unemployed I suppose.
What is it with air traffic control systems ? They all seem to turn into
dead ends…