LINUX flight software

Apparently, there are a few people out there working with Linux for gliding
applications. (Thanks for the links) Maybe it's a good thing.

I keep thinking that the potential user base in this space is so small that
if the PDA based apps split it, there won't be all that much profit to
support a commercial business - especially with a price war. So, we are
back to the hobbyist/enthusiast hacker who tends to like Linux/GPL anyway.
This person just wants some recognition for expertise and contribution made.
Gliding software might make a good resume entry.

Should someone take up the challenge, I'd like to see something along the
lines of an x86 based (lots of small, cheap, powerful, highly configurable,
12V boxes like those from Via) with the in-flight software developed on a
very small, hard real-time Linux kernel with an API like DIAPM RTAI. This
removes all the constraints of the PDA platform. (Somehow, I just don't
think a pocketable device designed for to-do and a contact lists is a great
platform for advanced soaring software, even if they are widely available.)

This approach offers the chance to do the entire panel in a gliding specific
"glass cockpit" with a screen size that's actually big enough to read at a
glance.

Bill Daniels

Bill Daniels wrote:

Apparently, there are a few people out there working with Linux for
gliding
applications. (Thanks for the links) Maybe it's a good thing.
Yes, there are. I am one of them.

I keep thinking that the potential user base in this space is so small
that if the PDA based apps split it, there won't be all that much profit
to
support a commercial business - especially with a price war. So, we are
back to the hobbyist/enthusiast hacker who tends to like Linux/GPL anyway.
This person just wants some recognition for expertise and contribution
made. Gliding software might make a good resume entry.
It does, actually. I am currently in contact with a potential employee who
seems to be very interested in my work. I'm not sure what you are
suggesting though, that I am undermining the commercial glidersoftware
business?

Should someone take up the challenge, I'd like to see something along the
lines of an x86 based (lots of small, cheap, powerful, highly
configurable, 12V boxes like those from Via) with the in-flight software
developed on a
very small, hard real-time Linux kernel with an API like DIAPM RTAI. This
removes all the constraints of the PDA platform. (Somehow, I just don't
think a pocketable device designed for to-do and a contact lists is a
great platform for advanced soaring software, even if they are widely
available.)

This approach offers the chance to do the entire panel in a gliding
specific "glass cockpit" with a screen size that's actually big enough to
read at a glance.
Problem is: these devices a
-too powerhungry
-too hard to install in a cockpit.

I am working on an application called Cumulus on the PDA platform. But, it's
easy to port to x86. In fact, most of the development work is done on x86.

André
projectleader of Cumulus

"André Somers" wrote in message
...
Bill Daniels wrote:

Apparently, there are a few people out there working with Linux for
gliding
applications. (Thanks for the links) Maybe it's a good thing.
Yes, there are. I am one of them.

I keep thinking that the potential user base in this space is so small
that if the PDA based apps split it, there won't be all that much profit
to
support a commercial business - especially with a price war. So, we are
back to the hobbyist/enthusiast hacker who tends to like Linux/GPL
anyway.
This person just wants some recognition for expertise and contribution
made. Gliding software might make a good resume entry.
It does, actually. I am currently in contact with a potential employee who
seems to be very interested in my work. I'm not sure what you are
suggesting though, that I am undermining the commercial glidersoftware
business?

Nope, just pointing out that it's likely to be a labor of love rather than a
serious effort to get rich. Nothing wrong with getting paid for your work.

Should someone take up the challenge, I'd like to see something along
the
lines of an x86 based (lots of small, cheap, powerful, highly
configurable, 12V boxes like those from Via) with the in-flight software
developed on a
very small, hard real-time Linux kernel with an API like DIAPM RTAI.
This
removes all the constraints of the PDA platform. (Somehow, I just don't
think a pocketable device designed for to-do and a contact lists is a
great platform for advanced soaring software, even if they are widely
available.)

This approach offers the chance to do the entire panel in a gliding
specific "glass cockpit" with a screen size that's actually big enough
to
read at a glance.
Problem is: these devices a
-too powerhungry
-too hard to install in a cockpit.

André
projectleader of Cumulus


Really? I see systems that pull less than 750ma. That would go for more
than 8 hours on a 7.5AH battery. Spend a little more for a Li-Ion polymer
battery and 10 - 15 hours are possible. The systems are about the same size
as a CD drive (5.5 inch form factor). I've got plenty of room for that and
the battery. As for the screen, I have dry-fitted a 15" screen over my
panel and I have room to spare. It just needs to hinge at the top so I can
flip it up out of the way if I need to see the old round gauges.


I am working on an application called Cumulus on the PDA platform. But,
it's
easy to port to x86. In fact, most of the development work is done on x86.

I read your web page and I'm really impressed. Waaay cool!

Bill Daniels

Bill Daniels wrote:

I'm not sure what you are
suggesting though, that I am undermining the commercial glidersoftware
business?
Nope, just pointing out that it's likely to be a labor of love rather than
a
serious effort to get rich. Nothing wrong with getting paid for your
work.
I didn't start this project to get rich. I like OpenSource. That concept has
allready made Cumulus far better than I had ever been able to make myself,
because we now have a number of contributors and we are free to use code
from other GPL projects. Cumulus will always remain OpenSource. So yes, It
is a labor of love. Nothing wrong with that either, is there?

Problem is: these devices a
-too powerhungry
-too hard to install in a cockpit.
Really? I see systems that pull less than 750ma. That would go for more
than 8 hours on a 7.5AH battery. Spend a little more for a Li-Ion polymer
battery and 10 - 15 hours are possible. The systems are about the same
size
as a CD drive (5.5 inch form factor). I've got plenty of room for that
and
the battery. As for the screen, I have dry-fitted a 15" screen over my
panel and I have room to spare. It just needs to hinge at the top so I
can flip it up out of the way if I need to see the old round gauges.
And how much power does your (touch) screen draw? I would be really
interested to see such a setup, and I must say I find the concept
intreaging. Maybe you can fit, say, a 10" screen in your pannel. That would
allready be a huge improvement over a PDA screen, and I must say that the
additional CPU power of an x86-based system would give some *very* nice
options indeed...

I am working on an application called Cumulus on the PDA platform. But,
it's
easy to port to x86. In fact, most of the development work is done on
x86.

I read your web page and I'm really impressed. Waaay cool!
Thanks. The webpage is a bit outdated, and the screenshots are pretty old...
Stuff looks better now, and 1.2 will be a lot cooler feature-wise :-)

André

"André Somers" wrote in message
...
And how much power does your (touch) screen draw? I would be really
interested to see such a setup, and I must say I find the concept
intreaging. Maybe you can fit, say, a 10" screen in your pannel. That
would
allready be a huge improvement over a PDA screen, and I must say that the
additional CPU power of an x86-based system would give some *very* nice
options indeed...

André

Some of these screens draw about 650ma @ 12V with the brightness all the way
up. I don't know if this is a good number or if there are better screens.
A narrow viewing angle would reduce the demands on the backlight.

What are your thoughts on Real-Time Linux for this application?

Bill Daniels

Bill Daniels wrote:


"André Somers" wrote in message
...
And how much power does your (touch) screen draw? I would be really
interested to see such a setup, and I must say I find the concept
intreaging. Maybe you can fit, say, a 10" screen in your pannel. That
would
allready be a huge improvement over a PDA screen, and I must say that the
additional CPU power of an x86-based system would give some *very* nice
options indeed...

André

Some of these screens draw about 650ma @ 12V with the brightness all the
way
up. I don't know if this is a good number or if there are better screens.
A narrow viewing angle would reduce the demands on the backlight.

What are your thoughts on Real-Time Linux for this application?
I guess that depends on what you want to do with it. For one, you can't use
it for primairy instruments, at least not here. The greatest power of using
a computer is where it is used now too: navigational aid and flightcomputer
functions. For that, you don't really need a RT system IMHO.

André