The Consultant point of View

I’v decided to speak out about various topic that affect me as a QNX
consultant.

As some of you may have notice there is currently 3 QNX consultants seeking
contract that have offer their services via the newsgroups. One of them
having quite an exposure via books and stuff.

First thing I’d like to complain about is the total lack of means to reach
the QNX community. At one time there was the QNX Magazine sent
to QNX users. One could advertise product or services. I remember
the first think I always check in QNXNews was the adds and the “new
products” section. I wanted to keep in touch with what was available,
hardware and software wise. The article were interesting but I was more
concern about what was available for QNX out there. That is no longer
the case today.

Where does one advertise QNX related product or services; answer is nowhere.
I have often heard how QNX has a total disregard or at best no interest in
third-party provide. At that time I though maybe these people were
exagerating.
Today as a service provider I tend to agree with them. How can
Connect Tech, Corman, Astra, get in touch with the QNX community.
I remember a customer who gave me shit because he wasn’t made aware
of the existance of Automap before, isn’t that odd !?!? The only
mean the QNX community can find out about the stuff out there is
mainly via the QNX partner web page, that sucks if you ask me.
I look at my situation (seeking work) trying to find ways to reach out to
new
market and I realised I’m facing a wall.

I’m also facing other challenges. In some ways QNX and I are competitors.
Our target customers overlap. I guess they don’t have much insentive to
help us
out (and I don’t really expect them to) . When there is plenty of work
that’s a
different story as consultants can fill holes QNX can’t, everybody’s happy.
When work is sparse it’s a different story, but that’s ok it’s the nature of
the
business.

That being said, as a consultant finding work is part of the job, if I don’t
like it I can just get out !!! My real gripe is having no means to reach
out to people that don’t read newsgroups or the partner web page…
Maybe I could try spamming, not!

Because of the somewhat competitive nature (with QNX) of being a QNX
consulting I notice how other consulting company that offers “plus-value”
seems to be in a better position. If you look at Astra, Cogent, Nexware,
Pace, Q-TPS, Ruzz, and others (I bet you’ve never heard of some of them…)
they have develop specific area of expertise, setup-box, industrial control,
etc. In practice I can’t really compete with these guys. My background
is industrial control, can I design a setup box, sure I can. Will the end
product be as good as Astra for example, no way. Experience is of
immense value.

The way I’ve position myself is, a company with an expertise in
a specific field (lotery terminal for example) decides to use QNX.
Then I’m there to deal with all the QNX related issues while they
concentrate on their area of expertise which is the lotery. Worked
so far :wink:

Ok time to conclude, why this post in advocacy; I’m officialy complaining
about the lack of tools to get in touch with the professional QNX
communicty. Who’s responsibility is it, I have no clue. What I know
is that every month or so I receive a 4 pages document from WindRiver
with lots of info about available product, even though I never
bought a thing from them (only order an eval kit)…

Cheers!

“Mario Charest” <goto@nothingness.com> wrote in message
news:a2o9qs$1ct$1@inn.qnx.com

I’v decided to speak out about various topic that affect me as a QNX
consultant.

[cut]

Somebody pointed out by email that I didn’t provided my email address
in my post. Kinda stupid when I’m complaining about lact of exposure :wink:))

Here goes:

“Mario Charest” <goto@nothingness.com> wrote in message
news:a2p3cg$jgd$1@inn.qnx.com

“Mario Charest” <> goto@nothingness.com> > wrote in message
news:a2o9qs$1ct$> 1@inn.qnx.com> …


I’v decided to speak out about various topic that affect me as a QNX
consultant.

[cut]

Somebody pointed out by email that I didn’t provided my email address
in my post. Kinda stupid when I’m complaining about lact of exposure
:wink:> ))

Here goes:

Grrrr wrong key:

\

  • Mario



\

While I wouldn’t have put so much effort into saying what Mario has said, I
do essentially agree with him.

Let’s face it, QNX is a nitch paroduct. Everyone does NOT know about it.
Im my opinion, QNX needs us as much as we need them. I must spend at least
5 hours a week explaining the virtues of QNX to people that have never heard
of it. Most of those people, I think, would agree to use QNX if they didn’t
think that QNX was just a flash in the pan. I know QNX isn’t a flash in the
pan, but if you’ve never heard of it before and someone starts pitching it
to you, that’s what it sounds like.

These are rough times. I’m hungry. I’m not getting the work I used to get.
I’m willing to bust my hump to get work but it would be nice if it weren’t
such an uphill battle. It seems to me that it would ultimately help QSSL to
have a public place where anyone can say, “These are the products that we
have to offer.” “These are the services that we have to offer.” “We’ve
been supporting QNX clients for X number of years.”

Let me shift gears just a little. Generally, any ink is a good thing. QNX6
is a great OS. I believe that. But it’s a little disturbing that QNX6 is
clocking in slower across the board (i.e. CPU, disk I/O, compiles, etc.)
than QNX4. And QSSL is not making any encouraging comments about improving
this. Our (us consultants) clients don’t really understand the difference
between real-time and fast. And they don’t know diddly about deterministic.
To them, real-time is supposed to mean fast. And they don’t care about how
portable RTP may be. They are developing their application for a specific
platform adn that’s all they care about.

Like I said, I believe in QNX. I just wish it weren’t so difficult to sell
(the idea of using) QNX to those few people out there that are developing
new products/projects right now.


Bill Caroselli – 1(626) 824-7983
Q-TPS Consulting
QTPS@EarthLink.net


“Mario Charest” <goto@nothingness.com> wrote in message
news:a2p3dh$jgr$1@inn.qnx.com
[lots of interesting stuff deleted]

Mario Charest wrote:


I remember a customer who gave me shit because he wasn’t made aware
of the existance of Automap before, isn’t that odd !?!?

This is an interesting point. I am a Canadian, who has now lived in the
U.S. for 4 years. 4 years ago, I would have thought it odd that someone
would get reamed by a new/potential customer for not having advertised
their product sufficiently. This is one area where American culture is
quite different from Canadian (there aren’t too many of these).

Having being immersed in U.S. culture for 4 years now, I can completely
understand why this would happen. In the American ethic, you (as the
provider of a product or service) are completely responsible for all
aspects of your product. Advertising is considered as integral to your
product as (in your case) the source code. Not having advertised your
product sufficiently is considered as serious a flaw as if your product
SIGSEGV’d in a leap year.

Canadian culture (at least as I have perceived it) views advertising as
separate and distinct from the quality of the product itself, and there
is this cultural “understanding” that those who might benefit from your
product have the responsibility to find out about it. If the product is
of superior quality, there is almost an assumption that advertising is a
secondary consideration; in fact, I think that Canadians tend to view
heavy advertisement as potentially indicative of an inferior quality
product. Americans (on the whole) make no such assumptions about the
underlying quality of a product based on how well advertised it is .

Business can thrive within either cultural bias; but if you wish to sell
from one culture into the other you must be aware of (and accomodate)
these differences. I think that it is often easier for companies from
countries that have radically different cultures to sell into the U.S.,
since they cleary recognize the need to study the U.S. culture and
accomodate it. Since Canadians share so much culturally with the U.S.
they often fail to notice the subtle differences.

ps: Of course, these are wide cultural biases, and there are many
individuals from each culture that perceive things differently.

btw: QSSL used to do a pretty good job of advertising in the old days. I
first heard about QNX in a byte magazine article, where QNX used to have
catchy adds (for the time). I didn’t get a chance to actually use QNX
until a couple of years after having seen these ads many times. When I
finally started using QNX, I couldn’t call Dan up and yell at him for
not having made me aware of QNX (not that I would have anyway :slight_smile:

“Rennie Allen” <rallen@csical.com> wrote in message
news:3C505375.5050906@csical.com

Mario Charest wrote:

[…]
product. Americans (on the whole) make no such assumptions about the
underlying quality of a product based on how well advertised it is .

Just don’t advertise on Home Shopping Network :slight_smile:

That is perceived as sign of scam by americans. In general, they notice
the channel being used for advertisement. Expensive high profile channels
get a lot more respect, as they probably should. Truth is, company making an
inferior product probably would not be able to advertize consistently in
such a channel.

  • igor

Here’s a couple of more or less ramdom thoughts on the subject.

Ken Schumm and I were discussing QNX consulting the other day. We both
have some old QNX4 work, but see no new prospects.

Ken was wondering if the “free” RTP is part of the problem. Did we loose
our wizard hats in the process? After all anyone can download and be an
“expert” now. I initially thought that the program would give much
needed exposure - if so I don’t see it. Hopefully QSSL does.

Another part of the problem is that many of us consultants probably
tended to work on low volume high end projects - and QSSL may be big
enough now to ignore this market. Instead they concentrate on the high
volume projects like Audrey ( really too bad that it died BTW ).

I’m trying to remember the last QNX ad I saw - can’t say that I’ve seen
one during the past couple of months. And the ones I do recall were
pretty marginal IMHO.

And Rennie - I did call QNX (Quantum) after seeing one of those Byte
ads. It looked like the perfect solution for the problem I had -
monitoring a turbine power plant using a 286 box. It turned out to be as
good as advertised and more.

Until a couple of years ago - I’d get at least quarterly calls from
people needing QNX help - often from the QSSL 3rd party program. Then it
dried up. Of course a lot has dried up in the last 18 months…

I’ve also talked to staff from local collage programs. I see no interest
in anything besides .NET type training - but they at least at the
community collage level can’t afford to play with interesting
technologies unless they see student jobs in the short term.

It seems to me that QSSL is still in it’s catch up mode - too many
drivers, tools and so on missing. It would sure be nice if there were a
way for consultants to help out more.

Dave

“David Hawley” <david.l.hawley@computer.org> wrote in message
news:3C505FE1.6000200@computer.org

Here’s a couple of more or less ramdom thoughts on the subject.

Ken Schumm and I were discussing QNX consulting the other day. We both
have some old QNX4 work, but see no new prospects.

Ken was wondering if the “free” RTP is part of the problem. Did we loose
our wizard hats in the process? After all anyone can download and be an
“expert” now. I initially thought that the program would give much
needed exposure - if so I don’t see it. Hopefully QSSL does.

Hum interesting point. I get the feeling that is not a real problem.
Becoming
an “expert” takes years, RTP has been out for over a year now. Not
enough to become an expert IMHO :wink:

Another part of the problem is that many of us consultants probably
tended to work on low volume high end projects - and QSSL may be big
enough now to ignore this market.

I think they are loosing the low-volume high-end project market share
at an incredible rate…

Instead they concentrate on the high
volume projects like Audrey ( really too bad that it died BTW ).

Probably money driven precision. QNX being a business that’s
understantable :wink:

I’m trying to remember the last QNX ad I saw - can’t say that I’ve seen
one during the past couple of months. And the ones I do recall were
pretty marginal IMHO.

Can’t remember seeing an ad either. Maybe their was a conscious
decision made to drop small customer. If it was i think it’s going
to back fire…

And Rennie - I did call QNX (Quantum) after seeing one of those Byte
ads. It looked like the perfect solution for the problem I had -
monitoring a turbine power plant using a 286 box. It turned out to be as
good as advertised and more.

Until a couple of years ago - I’d get at least quarterly calls from
people needing QNX help - often from the QSSL 3rd party program. Then it
dried up. Of course a lot has dried up in the last 18 months…

Some of it may be blame on QNX behavior and some to general

slow economy

I’ve also talked to staff from local collage programs. I see no interest
in anything besides .NET type training - but they at least at the
community collage level can’t afford to play with interesting
technologies unless they see student jobs in the short term.

XPe definitaly does have it’s appeal.

It seems to me that QSSL is still in it’s catch up mode - too many
drivers, tools and so on missing. It would sure be nice if there were a
way for consultants to help out more.

That’s up to them…

Dave

Since Astra is mentioned in this post, I figure I will follow up with some
of my thoughts on the subject.

If I had to pick the one thing that has been an ongoing problem for
consultants in this community, it is the lack of vehicle to get themselves
known to their potential customers. Certainly at one point QSSL even went
as far as publishing a catalog of products and services that were available
to customers. At this point, even the web version of that seems to be not
emphasized enough to be useful. QNX News seems to have been shelved.

The challenge we have faced is that we are trying to be a little larger
than the single person consulting solution. This allows us to work on
slightly larger projects and gives us the ability to share problems amongst
ourselves and use the best person for the problem when we are trying to
solve them. The problem is keeping everyone busy.

I hear about all the design wins QNX gets and everyone is using it, but we
don’t see a lot of people looking for help. In this case, I BELIEVE they
are out there, but just don’t know how to find us. We get most of our work
from referals, so if no-one refers, we run into problems.

Now with the demise of qnxStart as a community web site, I (and Astra in
general) have become somewhat involved in supporting a community effort
called qnxZone (http://www.qnxzone.com) and I think we can make a good
argument for a consultants section on that site. It will becomes a site
which QSSL themselves can point to as a resource for anyone using QNX,
either commercially or not, and with a presense there, it may help all of
us. We can also ensure the data is presented in a way which helps
differentiate us in a way which is most useful to the potential customer.

Do you think that will help? Basically we take care of ourselves cause we
can’t count on anyone else to.

Rick…


Rick Duff Internet: rick@astranetwork.com
Astra Network QUICS: rgduff
QNX Consulting and Custom Programming URL: http://www.astranetwork.com
+1 (204) 987-7475 Fax: +1 (204) 987-7479

Mario Charest wrote:

“David Hawley” <> david.l.hawley@computer.org> > wrote in message
news:> 3C505FE1.6000200@computer.org> …

Here’s a couple of more or less ramdom thoughts on the subject.

Ken Schumm and I were discussing QNX consulting the other day. We both
have some old QNX4 work, but see no new prospects.

Ken was wondering if the “free” RTP is part of the problem. Did we loose
our wizard hats in the process? After all anyone can download and be an
“expert” now. I initially thought that the program would give much
needed exposure - if so I don’t see it. Hopefully QSSL does.


Hum interesting point. I get the feeling that is not a real problem.
Becoming
an “expert” takes years, RTP has been out for over a year now. Not
enough to become an expert IMHO > :wink:

Becoming an expert takes years, becoming an “expert” takes only a couple
of weeks :slight_smile:

I used to be involved in small-fish QNX-related business (training). Now
I am involved with big corporate customer, so I can see both sides. Here
are my few cents…

External consultants for solving QNX-related issues are only useful for
small shops. Bigger ones either have a ‘consultant on payroll’ (which is
what I am, pretty much) or can afford Custom Engineering services
provided by QNX. Or both. In practice, big customers always have
OEM/support contract and if problems appear they love to make QNX do
some work for the money being paid [anyway].

Using consultants for pilot/design is only useful if they have expertise
in the problem area, not just QNX. They also must be capable of doing
the work in accordance with process requirements established in the
organisation. That usually means such consultants are hired as
contractors. Outside serivces can be accepted for some isolated feature,
but from an established vendor, hardly from a one-man LTD.

With economy like we have today, everyone is getting rid of contractors
(if they still have them) and small shops mostly survive by
support/maintenance. Basically, nobody wants to give you guys a piece of
shrinking pie. I think perhaps the only way to have any ‘consultant’
business now would be training. That is of course not easy when QNX
competes with you, but you can beat them by doing tailor-made training
tied to specific needs of customers instead of shrink-wrapped courses
QNX provides. To do that you need reputation and good relations with
local QNX sales channels. Such strategy used to work back in Russia, but
here it might be harder with QNX trying to milk all NA customers
directly.

Life sucks, and then you die :wink:

Now I’ll give a free advice for desert. If you want to survive do not
lock yourself in ‘QNX-only’ business. Become an expert in few other
systems and offer choice. Wider audience = more customers, plain and
simple. I will tell you a very recent story (this winter). We needed to
get DVD-RAM support for Solaris. We got it for QNX through custom
engineering contract but it is quite different story to deal with Sun.
We started looking for 3rd parties and found just what we need.
Unfortunately, they tie their driver with some hardware ID and we could
not agree to it since it makes hardware components non-replaceable in
the field. We tried to call and work out some contract, with prepayment
for future shipments. Believe you or not, they just denied. They said
they sell far too many licenses without us (like tends of thousands per
month), to bother with making special version for us and suffer
consequences from possible piracy(!). And that was just a small shop,
all they do is some SCSI pass-through device drivers, filesystems and
some backup/restore software. Prices start from $500 per license.

The morale is simple. It is a business mistake to tie yourself up as a
consultant to a vendor who does not even promote himself on the market
you wanna play, let alone promotes you. I know where it comes from, it
is passion for inner beauty and enthusiasm. Sadly, those are bad
advisers in business. You end up fighting an uphill battle, instead of
piggybacking on your vendor’s success.

  • igor
    #include <std-disclaimer.h>

Rick Duff wrote:

Since Astra is mentioned in this post, I figure I will follow up with some
of my thoughts on the subject.

If I had to pick the one thing that has been an ongoing problem for
consultants in this community, it is the lack of vehicle to get themselves
known to their potential customers. Certainly at one point QSSL even went
as far as publishing a catalog of products and services that were available
to customers. At this point, even the web version of that seems to be not
emphasized enough to be useful. QNX News seems to have been shelved.

The challenge we have faced is that we are trying to be a little larger
than the single person consulting solution. This allows us to work on
slightly larger projects and gives us the ability to share problems amongst
ourselves and use the best person for the problem when we are trying to
solve them. The problem is keeping everyone busy.

I hear about all the design wins QNX gets and everyone is using it, but we
don’t see a lot of people looking for help. In this case, I BELIEVE they
are out there, but just don’t know how to find us. We get most of our work
from referals, so if no-one refers, we run into problems.

Now with the demise of qnxStart as a community web site, I (and Astra in
general) have become somewhat involved in supporting a community effort
called qnxZone (> http://www.qnxzone.com> ) and I think we can make a good
argument for a consultants section on that site. It will becomes a site
which QSSL themselves can point to as a resource for anyone using QNX,
either commercially or not, and with a presense there, it may help all of
us. We can also ensure the data is presented in a way which helps
differentiate us in a way which is most useful to the potential customer.

Do you think that will help? Basically we take care of ourselves cause we
can’t count on anyone else to.

Rick…


Rick Duff Internet: > rick@astranetwork.com
Astra Network QUICS: rgduff
QNX Consulting and Custom Programming URL: > http://www.astranetwork.com
+1 (204) 987-7475 Fax: +1 (204) 987-7479

Igor Kovalenko wrote:

I will tell you a very recent story (this winter). We needed to
get DVD-RAM support for Solaris. We got it for QNX through custom
engineering contract but it is quite different story to deal with Sun.
We started looking for 3rd parties and found just what we need.
Unfortunately, they tie their driver with some hardware ID and we could
not agree to it since it makes hardware components non-replaceable in
the field. We tried to call and work out some contract, with prepayment
for future shipments. Believe you or not, they just denied. They said
they sell far too many licenses without us (like tends of thousands per
month), to bother with making special version for us and suffer
consequences from possible piracy(!). And that was just a small shop,
all they do is some SCSI pass-through device drivers, filesystems and
some backup/restore software. Prices start from $500 per license.

Correct me if I’m wrong, but it sounds as if your particular problem was
easily solved with QNX, but is still not solved for the Solaris
platform. It sounds as if you’re impressed that these guys can turn
away your business (you are Motorola right ?), but where exactly does
that leave you ? and how is this an example of a failing of QNX (perhaps
it is not supposed to be) ? Perhaps this is only meant to illustrate the
relative size of the markets ? If so, it does go to show that having a
big market share doesn’t necessarily lead to better customer
satisfaction, or more effective solutions.

Damn it Igor! When you’re right, you’re right.


Bill Caroselli – 1(626) 824-7983
Q-TPS Consulting
QTPS@EarthLink.net


“Igor Kovalenko” <Igor.Kovalenko@motorola.com> wrote in message
news:3C5084C0.5B8D5E7A@motorola.com

I used to be involved in small-fish QNX-related business (training). Now
I am involved with big corporate customer, so I can see both sides. Here
are my few cents…

External consultants for solving QNX-related issues are only useful for
small shops. Bigger ones either have a ‘consultant on payroll’ (which is
what I am, pretty much) or can afford Custom Engineering services
provided by QNX. Or both. In practice, big customers always have
OEM/support contract and if problems appear they love to make QNX do
some work for the money being paid [anyway].

Using consultants for pilot/design is only useful if they have expertise
in the problem area, not just QNX. They also must be capable of doing
the work in accordance with process requirements established in the
organisation. That usually means such consultants are hired as
contractors. Outside serivces can be accepted for some isolated feature,
but from an established vendor, hardly from a one-man LTD.

With economy like we have today, everyone is getting rid of contractors
(if they still have them) and small shops mostly survive by
support/maintenance. Basically, nobody wants to give you guys a piece of
shrinking pie. I think perhaps the only way to have any ‘consultant’
business now would be training. That is of course not easy when QNX
competes with you, but you can beat them by doing tailor-made training
tied to specific needs of customers instead of shrink-wrapped courses
QNX provides. To do that you need reputation and good relations with
local QNX sales channels. Such strategy used to work back in Russia, but
here it might be harder with QNX trying to milk all NA customers
directly.

Life sucks, and then you die > :wink:

Or put more scientifically,

“There’s no such thing as gravity. The world just sucks.”


Now I’ll give a free advice for desert. If you want to survive do not
lock yourself in ‘QNX-only’ business. Become an expert in few other
systems and offer choice. Wider audience = more customers, plain and
simple. I will tell you a very recent story (this winter). We needed to
get DVD-RAM support for Solaris. We got it for QNX through custom
engineering contract but it is quite different story to deal with Sun.
We started looking for 3rd parties and found just what we need.
Unfortunately, they tie their driver with some hardware ID and we could
not agree to it since it makes hardware components non-replaceable in
the field. We tried to call and work out some contract, with prepayment
for future shipments. Believe you or not, they just denied. They said
they sell far too many licenses without us (like tends of thousands per
month), to bother with making special version for us and suffer
consequences from possible piracy(!). And that was just a small shop,
all they do is some SCSI pass-through device drivers, filesystems and
some backup/restore software. Prices start from $500 per license.

The morale is simple. It is a business mistake to tie yourself up as a
consultant to a vendor who does not even promote himself on the market
you wanna play, let alone promotes you. I know where it comes from, it
is passion for inner beauty and enthusiasm. Sadly, those are bad
advisers in business. You end up fighting an uphill battle, instead of
piggybacking on your vendor’s success.

  • igor
    #include <std-disclaimer.h

Rick Duff wrote:

Since Astra is mentioned in this post, I figure I will follow up with
some
of my thoughts on the subject.

If I had to pick the one thing that has been an ongoing problem for
consultants in this community, it is the lack of vehicle to get
themselves
known to their potential customers. Certainly at one point QSSL even
went
as far as publishing a catalog of products and services that were
available
to customers. At this point, even the web version of that seems to be
not
emphasized enough to be useful. QNX News seems to have been shelved.

The challenge we have faced is that we are trying to be a little larger
than the single person consulting solution. This allows us to work on
slightly larger projects and gives us the ability to share problems
amongst
ourselves and use the best person for the problem when we are trying to
solve them. The problem is keeping everyone busy.

I hear about all the design wins QNX gets and everyone is using it, but
we
don’t see a lot of people looking for help. In this case, I BELIEVE
they
are out there, but just don’t know how to find us. We get most of our
work
from referals, so if no-one refers, we run into problems.

Now with the demise of qnxStart as a community web site, I (and Astra in
general) have become somewhat involved in supporting a community effort
called qnxZone (> http://www.qnxzone.com> ) and I think we can make a good
argument for a consultants section on that site. It will becomes a site
which QSSL themselves can point to as a resource for anyone using QNX,
either commercially or not, and with a presense there, it may help all
of
us. We can also ensure the data is presented in a way which helps
differentiate us in a way which is most useful to the potential
customer.

Do you think that will help? Basically we take care of ourselves cause
we
can’t count on anyone else to.

Rick…


Rick Duff Internet: > rick@astranetwork.com
Astra Network QUICS: rgduff
QNX Consulting and Custom Programming URL:
http://www.astranetwork.com
+1 (204) 987-7475 Fax: +1 (204) 987-7479

Rennie Allen wrote:

Igor Kovalenko wrote:

I will tell you a very recent story (this winter). We needed to
get DVD-RAM support for Solaris. We got it for QNX through custom
engineering contract but it is quite different story to deal with Sun.
We started looking for 3rd parties and found just what we need.
Unfortunately, they tie their driver with some hardware ID and we could
not agree to it since it makes hardware components non-replaceable in
the field. We tried to call and work out some contract, with prepayment
for future shipments. Believe you or not, they just denied. They said
they sell far too many licenses without us (like tends of thousands per
month), to bother with making special version for us and suffer
consequences from possible piracy(!). And that was just a small shop,
all they do is some SCSI pass-through device drivers, filesystems and
some backup/restore software. Prices start from $500 per license.

Correct me if I’m wrong, but it sounds as if your particular problem was
easily solved with QNX, but is still not solved for the Solaris
platform.

I could tell you how much exactly that easyness was, but it would be
inappropriate.

It sounds as if you’re impressed that these guys can turn
away your business (you are Motorola right ?), but where exactly does
that leave you ? and how is this an example of a failing of QNX (perhaps
it is not supposed to be) ? Perhaps this is only meant to illustrate the
relative size of the markets ?

Right. It is not failing of QNX, it is size of the pool a consultant can
fish in.

If so, it does go to show that having a
big market share doesn’t necessarily lead to better customer
satisfaction, or more effective solutions.

Rushing to conclusions again. Of course it does not necessarily lead
there, but it does not preclude that either. I’d say if it does not lead
there for long enough, then customer satisfaction tends to dry and
market share declines then. Sun is about only Unix vendor right now who
increases their market share, so I would not call their customers
unsatisfied. Well, not until they announced EOL for Solaris/Intel :wink:

  • igor

The morale is simple. It is a business mistake to tie yourself up as a
consultant to a vendor who does not even promote himself on the market
you wanna play, let alone promotes you.

I’m realizing this now :wink:

Igor Kovalenko wrote:

Now I’ll give a free advice for desert.

Here some sweet stuff for your desert :slight_smile:

If you want to survive do not lock yourself in ‘QNX-only’ business. Become an expert in few >other systems and offer choice. Wider audience = more customers, plain and
simple.

Platform independency is the key for such a concept.
(have a look to: http://www.steinhoff.de/SPS2001_PVM_Steinhoff_e.zip
…)

So I would (and will) never write (again) a Photon application using the
native Photon API.
All of our new Photon applications will be recompilable for LINUX,
U*IXES and M$Windows … or
they will be just re-usable if they are written in wxPython.

Also the code for the interprocess communication should be completely
system independent … the PVM/MPI APIs are execelent interfaces at the
library level. (it is also possible
to use the native (local) message passing of QNX for PVM … it’s on our
to-do list)

What you can do with Python shows our Tilcon wrapper for Python and the
nice
implementation recently announced in the QNX4 devtools group … with
these implementations
you are able to handle Photon in a objectoriented manner :slight_smile:

On my to-do list is a port of gtk++ to Photon as a base for the system
independent GUI API
wxWindows/wxPython … just to make sure that our Photon apps are
recompileable for LINUX
or M$Windows.

JAVA is not a solution … it is really a shame that SUN’s JAVA can’t be
used with QNX !!

I will tell you a very recent story (this winter). We needed to

get DVD-RAM support for Solaris. We got it for QNX through custom
engineering contract but it is quite different story to deal with Sun.
We started looking for 3rd parties and found just what we need.
Unfortunately, they tie their driver with some hardware ID and we could
not agree to it since it makes hardware components non-replaceable in
the field. We tried to call and work out some contract, with prepayment
for future shipments. Believe you or not, they just denied. They said
they sell far too many licenses without us (like tends of thousands per
month), to bother with making special version for us and suffer
consequences from possible piracy(!).

No business is better than bad business, which can destroy your future
business!

And that was just a small shop,
all they do is some SCSI pass-through device drivers, filesystems and
some backup/restore software. Prices start from $500 per license.

IBM makes money with the noisy action of the floppy drives … and the
licemse
are under 1$. BTW … a protected M$Windows XP license is in the range
of $500(?)

So what’s your problem with the $500 license?

The morale is simple. It is a business mistake to tie yourself up as a
consultant to a vendor who does not even promote himself on the market
you wanna play, let alone promotes you. I know where it comes from, it
is passion for inner beauty and enthusiasm. Sadly, those are bad
advisers in business. You end up fighting an uphill battle, instead of
piggybacking on your vendor’s success.

Hmm … ‘piggybacking on your vendor’s success’? IMHO … the situation
is reverse, because
every third party has always to do TWO sales … at first QNX and then
their own stuff.
And at the end … if you did these two sales you have to handle very
carefully your customer relationship and your product strategy and
details against QSSL.

QSSL is the only small OS vendor which competes against there own
customer base … it is just
the reality and the reason for a lot of one-shut usage of QNX.

Regards

Armin

“Armin Steinhoff” <a-steinhoff@web_.de> wrote in message
news:3C514341.23172E0@web_.de…

…snip


So I would (and will) never write (again) a Photon application using the
native Photon API.
All of our new Photon applications will be recompilable for LINUX,
U*IXES and M$Windows … or
they will be just re-usable if they are written in wxPython.

I’ve been thinking about wxWindows lately…I wonder how much work a native
Photon port of it would be? It seems like a great way to do multiplatform
gui stuff - do you know if anyone has looked into a Photon version? Do you
think it’s a mature enough product for commercial apps?

cheers,

Kris

…snip

QSSL is the only small OS vendor which competes against there own
customer base … it is just
the reality and the reason for a lot of one-shut usage of QNX.

Regards

Armin

“Armin Steinhoff” <a-steinhoff@web_.de> wrote in message
news:3C514341.23172E0@web_.de…

No business is better than bad business, which can destroy your future
business!

I am not sure what you mean, but sounds funny.

And that was just a small shop,
all they do is some SCSI pass-through device drivers, filesystems and
some backup/restore software. Prices start from $500 per license.

IBM makes money with the noisy action of the floppy drives … and the
licemse
are under 1$. BTW … a protected M$Windows XP license is in the range
of $500(?)

So what’s your problem with the $500 license?

We did not have problem with $500 license. And if you want to continue
analogy, WinXP can be re-installed on different hardware from the same
original media. MS does understand that hardware might need replacements.
Thos folks suggested to contact them every time we need to replace hardware.
Very convinient when replacement must be done on a remote pasific island in
quite different timezone, by people who do no speak english.

MS also supplies ‘coroprate’ version of WinXP to those customers who can not
live with registration headache. They just pay in advance in bulk
quantities. We suggested to do the same, but were denied.

  • igor

Igor Kovalenko wrote:


Rushing to conclusions again. Of course it does not necessarily lead
there, but it does not preclude that either. I’d say if it does not lead
there for long enough, then customer satisfaction tends to dry and
market share declines then. Sun is about only Unix vendor right now who
increases their market share, so I would not call their customers
unsatisfied. Well, not until they announced EOL for Solaris/Intel > :wink:

Sun is the only *nix vendor who increases their market share ?

I bet QSSL has increased their market share over the last 24 months; not
to mention Linux, which technically is not “a” vendor, but
technicalities aside, has definately increased market share.

QNX is not Unix Rennie and it plays on totally different market. Linux is
not Unix either and has no market at all. It is being used by other
products/vendors as a ‘media’ or ‘host organism’, so to speak. In parasiting
sense, yes.

  • igor

“Rennie Allen” <rallen@csical.com> wrote in message
news:3C51A5D1.4020802@csical.com…> Igor Kovalenko wrote:

Rushing to conclusions again. Of course it does not necessarily lead
there, but it does not preclude that either. I’d say if it does not lead
there for long enough, then customer satisfaction tends to dry and
market share declines then. Sun is about only Unix vendor right now who
increases their market share, so I would not call their customers
unsatisfied. Well, not until they announced EOL for Solaris/Intel > :wink:


Sun is the only *nix vendor who increases their market share ?

I bet QSSL has increased their market share over the last 24 months; not
to mention Linux, which technically is not “a” vendor, but
technicalities aside, has definately increased market share.