"Richard Erlacher" <edick(a)idcomm.com> wrote:
OK, you can remove it by desoldering or whatever, but
what about the
distinction between the unprogrammed part, e.g. FPGA when the power's off,
and when it's on and loaded? It's still hardware, isn't it? It's just
different hardware when the configuration code is in it and powered, or is
it?
Hi
Kind of a gray area but it is a piece of hardware that is firmware configured
by software that was loaded into it.
Now, what about the pseudo hard-disks sold on PCMCIA cards as "flash-disk"
or whatever. Is the stuff stored there firmware or is it software?
It is firmware, the stuff stored on it is still software or data.
I see a change in the generalized definition of "soft" vs. "hard" in
this
context in a much more general sense. People talk about machine-readable
copy, e.g. data on floppy disk, as "soft" copy, and they talk about
purportedly fixed specifications or code segments, or definitions, as being
"hard" implying they're, at least relatively, "etched in
concrete," as
opposed to being etched in "silly-putty" as some spec's are.
I think the problem is the sense of what is meant. Hardware is
things that you can solder in. Yes, that includes PROMs. Firmware
is hardware that contains software. That could also be something
like a PROM. Software is what tells the processor what to do.
So, you can have hardware that is firmware that contains software
that runs on the machine. I think the problem is that you are trying
to make software, firmware and hardware the same thing. It is like making
automobiles and drivers the same thing. We don't have a seperate name
for the auto with a driver in it like the word firmware but we
could have. The automobile is clearly not the driver and the driver
is clearly not the automobile. The combination isn't either of the
two, it is something different ( but not specifically named in
the english language ).
Dwight