Dr Skip wrote:

> So, as engineers, where is the line drawn on when to integrate some new
> technology? Internet toasters... 

It's the market... :)  People like it and buy it, or they don't. In the
end, what engineers do is to serve someone. Find enough who think it serves
them to make it profitable, and you have a product. 

> There's also the perspective in the decision. Imagine a condition where a
> device (even a car) can enter a 'broken' state. 

This state is not unknown for cars -- even since before they were
electronic :)

> You could design in a soft-fail mode where it continues in limited mode
> until fixed, or it could just stop until it is repaired. 

For example automatic transmissions have a "limp-home mode". When the
conditions indicate an error that's not specifically recognized, the
transmission typically goes into 3rd gear and stays there. (The rationale
is that you can go wherever you need to go to get it fixed in 3rd gear.)

I think there are several types of firmware. There is the more
software-like firmware that's reprogrammable in the field, often by the
user. And there's the more hardware-like type that's like a piece of
hardware once it has left the factory. How to deal with it depends a lot
what type you're thinking about.

> Chances are, the original freezer problem came about this way. Is it
> corporate misconduct? 

I think it's in part the difference between free market theory and actual
market. People don't have all the relevant data in their hands when buying
products. So crappy products get bought all the time -- sometimes people
could have known, sometimes not. And sometimes they know and still buy...
:)

More in the vein you were talking about, I agree that it's not really what
is commonly called misconduct. But it's often probably just plain bad (or
inadequate) organizational structures and people not doing their best to
fill the existing structures with purpose.

Gerhard

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist