Future of Electronics In Aviation

On Jun 20, 6:07*am, Bob Noel
wrote:
In article ,
And what analysis techniques would be applied to prove that the resulting
software intensive system is adequately safe?

The same techniques that employed, in general, by experts to test
software.

I don't care how many "fastidious" people look at an architecture or the
as-built system, if they don't know what they are looking for and how to
find it, the odds of proving *anything useful are pretty small.

Well, assuming they are experts, each in their respective areas, they
would indeed know what to look for. Also, peer-review (by other
experts) is a very good way to check structural integrity of software
(or any system).

-Le Chaud Lapin-

In rec.aviation.piloting Le Chaud Lapin wrote:
On Jun 20, 6:07?am, Bob Noel
wrote:
In article ,
And what analysis techniques would be applied to prove that the resulting
software intensive system is adequately safe?

The same techniques that employed, in general, by experts to test
software.

I don't care how many "fastidious" people look at an architecture or the
as-built system, if they don't know what they are looking for and how to
find it, the odds of proving ?anything useful are pretty small.

Well, assuming they are experts, each in their respective areas, they
would indeed know what to look for. Also, peer-review (by other
experts) is a very good way to check structural integrity of software
(or any system).

Blue Screen of Death (BSOD).

Do I need to say more?


--
Jim Pennino

Remove .spam.sux to reply.

In article
,
Le Chaud Lapin wrote:

On Jun 20, 6:07*am, Bob Noel
wrote:
In article
,
And what analysis techniques would be applied to prove that the resulting
software intensive system is adequately safe?

The same techniques that employed, in general, by experts to test
software.

And exactly what level of reliability do you think you'll need to have?

Note that the cost can rise enormously for fairly small increases in end
product reliability.

And so far we haven't said much about what the lawyers will bring to
your nifty new product. (Trust me, it won't be something to make you
emit small cries of joy.)

I don't care how many "fastidious" people look at an architecture or the
as-built system, if they don't know what they are looking for and how to
find it, the odds of proving *anything useful are pretty small.

Well, assuming they are experts, each in their respective areas, they
would indeed know what to look for. Also, peer-review (by other
experts) is a very good way to check structural integrity of software
(or any system).

You *really* don't know what is involved in developing verifiably
correct software systems, either in time or money.

It's *very* difficult and expensive.

On Jun 20, 12:06*pm, Steve Hix
wrote:
Well, assuming they are experts, each in their respective areas, they
would indeed know what to look for. *Also, peer-review (by other
experts) is a very good way to check structural integrity of software
(or any system).

You *really* don't know what is involved in developing verifiably
correct software systems, either in time or money.

It's *very* difficult and expensive.

Which is why so many researchers the world over spend time trying to
find mechanized approaches to proving that software is good, or that
it satisfies some definition of correctness, etc. Each of these
researchers seek what they regard as the holy grail of software
engineering - a machine that can help good engineers be better by
looking at what they make.

As far as FAA certification, if it turned out that verification were
prohibitively expensive (greater than $100US million), that would be a
problem.

In any case, because the material cost of software is $0, the cost of
verification would have to be very high indeed before a point would
reached, beyond which, it did not make sense to make the software
because the market could not support it.

-Le Chaud Lapin-

Recently, Le Chaud Lapin posted:
[...]

In any case, because the material cost of software is $0, the cost of
verification would have to be very high indeed before a point would
reached, beyond which, it did not make sense to make the software
because the market could not support it.

I tried to resist jumping in, but having read through many of the posts
only to see you wind up where you began is incredible. if one more voice
saying that you are grossly missing the fundamental costs involved in
software development helps to move you from this position, then perhaps it
won't be wasted effort.

GA is a small market. Too small to warrant specialized development of much
of anything, which is why most of the components are either used or
spin-offs from other areas of aviation. Comparing it to the _general_
automotive market is completely off-base, as even a single model of a
single brand in a single year will have more units in the market than all
of GA.

So, to think that a body of expert programmers will somehow collaborate on
systems that, at best will be less reliable than the pulley and wire that
they replace is an unrealistic fantasy.

BTW - if you think that "the material costs of software is $0", let me
know where you're getting your language compilers and hardware to create
and test your code. And, don't tell me about "Open Source" options,
either, unless you want to increase your development costs by a factor of
100 or so.

Neil

On Jun 20, 4:27*pm, "Neil Gould" wrote:
GA is a small market. Too small to warrant specialized development of much
of anything, which is why most of the components are either used or
spin-offs from other areas of aviation. Comparing it to the _general_
automotive market is completely off-base, as even a single model of a
single brand in a single year will have more units in the market than all
of GA.

It's a Catch-22. The FAA, NASA, DARPA, CAFE, and other organizations
are trying to make it not a small market, so the assumption is that,
if a PAV were created, it would be created for a mass market.

So, to think that a body of expert programmers will somehow collaborate on
systems that, at best will be less reliable than the pulley and wire that
they replace is an unrealistic fantasy.

A bit of a stretch.

BTW - if you think that "the material costs of software is $0", let me
know where you're getting your language compilers and hardware to create
and test your code. And, don't tell me about "Open Source" options,
either, unless you want to increase your development costs by a factor of
100 or so.

Accountants define material cost to be the cost of the components from
which the system is synthesized, not from the tools used to design or
create the system.

For example, the material cost of an iPod would include its hard disk,
RAM, ROM, resistors, capacitors, dials, faceplace, battery holder,
wires, mounts, shock absorbers, etc. It would not include
dehumidifier, blower, oscilloscope, spectral analyzer, or other factor
equipment used to manufacture the product.

The material cost of software, if sold in a store, would include the
cost of manual, the disks, and the packaging.

Compilers and hardware do not factor into the material cost of
software any more than an oscilloscope factors into the material cost
of an iPod.

To determine what components are considered "material", move the
product over a large distance. Whatever components move with the
products, those components are considered material. Those that stay
behind are something else.

-Le Chaud Lapin-

In rec.aviation.student Le Chaud Lapin wrote:
On Jun 20, 4:27?pm, "Neil Gould" wrote:
GA is a small market. Too small to warrant specialized development of much
of anything, which is why most of the components are either used or
spin-offs from other areas of aviation. Comparing it to the _general_
automotive market is completely off-base, as even a single model of a
single brand in a single year will have more units in the market than all
of GA.

It's a Catch-22. The FAA, NASA, DARPA, CAFE, and other organizations
are trying to make it not a small market, so the assumption is that,
if a PAV were created, it would be created for a mass market.

So, to think that a body of expert programmers will somehow collaborate on
systems that, at best will be less reliable than the pulley and wire that
they replace is an unrealistic fantasy.

A bit of a stretch.

BTW - if you think that "the material costs of software is $0", let me
know where you're getting your language compilers and hardware to create
and test your code. And, don't tell me about "Open Source" options,
either, unless you want to increase your development costs by a factor of
100 or so.

Accountants define material cost to be the cost of the components from
which the system is synthesized, not from the tools used to design or
create the system.

For example, the material cost of an iPod would include its hard disk,
RAM, ROM, resistors, capacitors, dials, faceplace, battery holder,
wires, mounts, shock absorbers, etc. It would not include
dehumidifier, blower, oscilloscope, spectral analyzer, or other factor
equipment used to manufacture the product.

The material cost of software, if sold in a store, would include the
cost of manual, the disks, and the packaging.

Compilers and hardware do not factor into the material cost of
software any more than an oscilloscope factors into the material cost
of an iPod.

To determine what components are considered "material", move the
product over a large distance. Whatever components move with the
products, those components are considered material. Those that stay
behind are something else.

Therefore you saying "the material costs of software is $0" is about as
usefull and insightfull as saying "watermelon has no bones".


--
Jim Pennino

Remove .spam.sux to reply.

On Jun 20, 4:55*pm, wrote:
In rec.aviation.student Le Chaud Lapin wrote:
Accountants define material cost to be the cost of the components from
which the system is synthesized, not from the tools used to design or
create the system.
For example, the material cost of an iPod would include its hard disk,
RAM, ROM, resistors, capacitors, dials, faceplace, battery holder,
wires, mounts, shock absorbers, etc. *It would not include
dehumidifier, blower, oscilloscope, spectral analyzer, or other factor
equipment used to manufacture the product.
The material cost of software, if sold in a store, would include the
cost of manual, the disks, and the packaging.
Compilers and hardware do not factor into the material cost of
software any more than an oscilloscope factors into the material cost
of an iPod.
To determine what components are considered "material", move the
product over a large distance. *Whatever components move with the
products, those components are considered material. *Those that stay
behind are something else.

Therefore you saying "the material costs of software is $0" is about as
usefull and insightfull as saying "watermelon has no bones".

Not true.

Accountants define material cost as above becasue material cost is a
per-unit cost that cannot be amortized. It is a necessary evil of
selling a product.

Let's take an example:

I can buy a new Sony DVD player for about $50.
I can buy Microsoft Flight Simulator for about $50.
Let us say that the development cost for the DVD player is $2 million.
Let us say that the development cost for MSFS is $5 million.

Sony and Microsoft sell their respective products to make a profit.
Let us assume that the market for each, in terms of number of
consumers, is exactly 1 million in 1 year. In that case, each product
will generate gross revenue of $50 million. But there is a problem: in
addition to the development cost, there is a per unit material cost,
the cost that the Microsoft and Sony must pay for the components that
form the product.

In the case of the DVD player, we assume that the material cost,
including resistors, capacitor, laster, motor, stabilizers, cases,
manuals, and packaging, etc. is $35, yielding a per-unit profit margin
of $15.

In the case of MSFS, the per-unit material cost is due to the manuals
and packaging, which we conservatively say costs $5, yield a per-unit
profit margin of $45.

If the packaging is eliminated, as is often the case, then the
material-cost effectively goes to zero for MSFS as does the
distribution cost. The per-unit profit of the software then becomes
the entire $50. If the packaging is eliminated from the DVD player,
the profit only rises to $20.

If, upon release, 1 million units of DVD player are desired, Sony can
expect $20 million in revenue.

If, upon release, 1 million units of MSFS are desired, Microsoft can
expect $50 million in revenue.

If both companies determine through market research that $7 is the
magic price point for each product, where demand becomes effectively
unsatiable, meaning 100 million units,...

Microsoft can sell 100 million at $7 for $700 million in profit.

Sony will not be able to sell and units because $7 is below the price
they need to sell to avoid a loss.

This is why software companies succeed even with marginally-desirable
products. The material cost and distribution costs become close to
zero, allowing them to test demand/price elasticity over the full
domain of variables. Also, problems with suppliers are almost non-
existent, as the suppliers are only used to supply tools that make the
products, not components of the products themselves. This eliminates
opportunities for the suppliers to "ride the market", where they know
a priori that a component is only used in, say military applications,
and will charge exhorbitant fees for the part simply because they can.

Also, the "manufacturing" cost of software is essentially zero:

To make 1 million DVD players, there is a per-unit manufacturing cost
of operating the assembly machines (and people) is some number greater
than 0.

To make 1 million copies of software, the per-unit manufacturing cost
is essentially zero.

These facts becomes more clear when the software becomes downloadable.

-Le Chaud Lapin-

Le Chaud Lapin wrote:
On Jun 20, 4:27 pm, "Neil Gould" wrote:
GA is a small market. Too small to warrant specialized development of much
of anything, which is why most of the components are either used or
spin-offs from other areas of aviation. Comparing it to the _general_
automotive market is completely off-base, as even a single model of a
single brand in a single year will have more units in the market than all
of GA.

It's a Catch-22. The FAA, NASA, DARPA, CAFE, and other organizations
are trying to make it not a small market, so the assumption is that,
if a PAV were created, it would be created for a mass market.


You just named three government agencies and a non-profit. By all their
very nature they are designed to blow smoke up the publics collective
ass. Winning the X-Prize isn't what motivated SpaceShipOne into
sub-orbital flight. It was a nice bonus though. The $250,000 prize CAFE
is offering won't even buy and fly one copy of what they are trying to
replace.


So, to think that a body of expert programmers will somehow collaborate on
systems that, at best will be less reliable than the pulley and wire that
they replace is an unrealistic fantasy.

A bit of a stretch.

BTW - if you think that "the material costs of software is $0", let me
know where you're getting your language compilers and hardware to create
and test your code. And, don't tell me about "Open Source" options,
either, unless you want to increase your development costs by a factor of
100 or so.

Accountants define material cost to be the cost of the components from
which the system is synthesized, not from the tools used to design or
create the system.

For example, the material cost of an iPod would include its hard disk,
RAM, ROM, resistors, capacitors, dials, faceplace, battery holder,
wires, mounts, shock absorbers, etc. It would not include
dehumidifier, blower, oscilloscope, spectral analyzer, or other factor
equipment used to manufacture the product.

The material cost of software, if sold in a store, would include the
cost of manual, the disks, and the packaging.

Compilers and hardware do not factor into the material cost of
software any more than an oscilloscope factors into the material cost
of an iPod.

To determine what components are considered "material", move the
product over a large distance. Whatever components move with the
products, those components are considered material. Those that stay
behind are something else.

-Le Chaud Lapin-


But those things still have to be paid for and are figured into the cost
of the software so the cost of software is zero.

On Jun 23, 10:49*am, Gig 601Xl Builder
wrote:
Le Chaud Lapin wrote:
It's a Catch-22. *The FAA, NASA, DARPA, CAFE, and other organizations
are trying to make it not a small market, so the assumption is that,
if a PAV were created, it would be created for a mass market.

You just named three government agencies and a non-profit. By all their
very nature they are designed to blow smoke up the publics collective
ass. Winning the X-Prize isn't what motivated SpaceShipOne into
sub-orbital flight. It was a nice bonus though. The $250,000 prize CAFE
is offering won't even buy and fly one copy of what they are trying to
replace.

You cannot blame them for trying. After all, when DARPA allocates $3
million award for a company or organization to solve a problem, and
the problem is not solved, it is the organization's fault, generally.
The alternative is to fund nothing at all, which will not work,
because someone will come up with the brilliant idea that government
agencies should provide stimulus funding for innovation.

The $300,000 being offered by NASA/CAFE is not a huge amount, true. I
regard it as NASA's way of saying, "if you do your part, we will do
ours."

Last year, the entries into the PAV Challenge were embarrassingly
unimaginative, but the funds were still allocated. I suspect that, if
someone were to actually enter something that looked more like a PAV,
NASA would not be the only agency providing funding. DARPA would
join, etc.

They are waiting for innovators in aviation to do more than introduce
slightly-modified LSA's.

-Le Chaud Lapin-