Re: Is Linux right for Embedded?

["Marin David Condic" <dont.bother.mcondic.auntie.spam@[acm.org>]

Q: What do you get when you add one plus one?

Mathematician: 2.0

Engineer: 2.0 +/- 0.001

Accountant: What do you want it to be? :-)

The point: If you *want* the answer to be "Program everything in C", you can
probably find a way to stack up a bunch of reasons why it is that "C" is the
right answer. OTOH, if you truly have an open mind and want to investigate
what might be an optimal solution for development of a particluar kind of
system, you might look at some of the studies I cited elsewhere and the
experience attested to by a variety of users and *possibly* come to the
conclusion that it is worth a shot at using Ada (or other languages that
provide more safety) and then think that the one-time transitional overhead
is worth it.

Remember that at one time Java didn't exist, yet a bunch of people didn't
see any big problem in adopting it for a variety of applications. It
required training, investment in compilers, tools, etc. as you pointed out.
But they believed they were going to get something as a result. Not really
different for Ada, would you think?

BTW: As for cost of compilers and tools, depending on your platform the cost
can be as little as $0.00. See: http://www.adapower.org/ for links to free
compilers, free GUI builders, free bindings, free etc.

MDC
--
Marin David Condic
Senior Software Engineer
Pace Micro Technology Americas    www.pacemicro.com
Enabling the digital revolution
e-Mail:    marin.condic@pacemicro.com
Web:      http://www.mcondic.com/


"bgeer" <bgeer@xmission.xmission.com> wrote in message
news:9pcvbn$r52$1@xmission.xmission.com...
>
> Man hours, maybe, if you have a team that already has significant Ada
> experience.
>
> Include costs of compiler, language & tool training, development &
> emulation platforms, & future maintenance plus factor in employee
> turnover, may reveal a different economic interpretation.