RFD: comp.arch.embedded
Tim Tsai
Group Name: comp.arch.embedded
Status: unmoderated
Distribution: world-wide
Summary: Discussions related to embedded systems
Proposed by: Tim Tsai (ts...@iia.org)
This is a formal Request For Discussion (RFD) on the subject of
creating an unmoderated Usenet newsgroup comp.arch.embedded.
CHARTER
The proposed unmoderated newsgroup comp.arch.embedded will be
opened to discussions on topics related to embedded systems,
including embedded programming techniques, tools, hardware
systems, and embedded peripherals such as displays and data
entry equipment. An embedded system can be described as any
application where a dedicated computer is built right into the
system but the end product is not a computer. For example, the
computer system that's built into a typical cable TV box is
considered an embedded system.
Topics not specific to embedded systems should be directed to
the appropriate group. Only topics that involve an embedded
system that can not be better answered elsewhere should be
posted here.
This RFD is cross-posted to comp.os.msdos.programmer,
comp.realtime, comp.robotics, and sci.electronics where I have
noted discussions on embedded systems in the past.
This is NOT a call for votes. A Call For Votes (CFV) will be
posted after the RFD period and will be conducted by an
independent third party.
Note the follow-up to news.groups.
phil dawkins
>
> REQUEST FOR DISCUSSION
>
> Group Name: comp.arch.embedded
> Status: unmoderated
> Distribution: world-wide
> Summary: Discussions related to embedded systems
> Proposed by: Tim Tsai (ts...@iia.org)
>
>
>This is a formal Request For Discussion (RFD) on the subject of
>creating an unmoderated Usenet newsgroup comp.arch.embedded.
>
[interesting stuff snipped, as it's already here]
Very interested in this, but i'm sort of lobbying so that the stuff
that i'm involved in doesn't fall through the cracks (again, *sob*). The
sort of things we do typically involve some sort of scada and may be
termed computers in that they are programmable, but at the same time can
be thought of as embedded in that they need industrial or mil specs, are
completely enclosed, have low-power usage and automotive, truck (more or
less battery) power requirements with normally quite different interface
devices (a barcode reader or J1708 rather than a mouse and a 4x40 lcd or
a touch-screen rather than a regular video screen) and the real
challenges tend to centered around more esoteric subjects such as
consistent RF communications (or, more properly, how to get it).
So we are sort of a cross between a fax machine system and a desktop
computer. we are currently working on a pc-104 range of services (maybe
more correctly isa bus), so one of these boxes could be thought of as a
pc (and could, if you wanted, be used as such), but the stuff we do and
need is in no way covered by os.msdos.* (certainly not os.windows. or
os2.*), at least in the mobiles and the foreseeable future (104
featured prominently in the embedded system conference, and seemed very
much at home). Also, although time critical to a certain extent (say
milliseconds), typically not realtime. So we're sort of half way
between. embedded, i would say certainly because of our requirements and
the sometimes strange and somewhat bizarre peripherals that we talk to
and the fact that sometimes these peripherals are not "normalised" for
our protection (price is still important). however, because of the
massive improvement in computer price / performance and the clear
"software costs" issue, we are, if you like, supplying computers as
embedded systems, taking advantage of the huge variety of tools
available for the pc, but definitely not using them in the same way a
typical computer application would.
so i suppose, i'm saying that it's the phrase "An embedded system can be
described as any application where a dedicated computer is built right
into the system but the end product is not a computer" is for us more
"An embedded system can be described as any application where a
dedicated computer is built right into the system but is not perceived
by the end user as a computer". not trying to nit pick, but this is
significant for me cos i produce tools that run on embedded systems for
programmers and i know that those guys *know* that it's a computer, and
so i would like to see discussions about multi-tasking and superloops,
but at the same time am worried about lock-ups and watchdogs and deadly
embraces and stuff.
is this a useful comment, or can't i see the wood for the trees, or does
this not matter? i'd hate to see someone get flamed for asking why dos
can't be made re-entrant (for example), when this could well be an
important strategic decision for an embedded development.
BTW: i saw a thread on ? (well it was comp.something) talking about the
next microsoft. i think it's embedded microsoft (or could it be embedded
ibm?). wouldn't be surprised if there are more than 1000 cpus in the
average us home by the year 2020. sort of why write the address, program
the stamp.
phil.
KBSmith163
dawkins) writes:
[...would this group cover embedded systems like XXX...]
From your description, I think you are right in the "mainstream" of
embedded systems. They include many DOS/PC based systems, but also
high-end RISC processors (e.g. laser printers) and low-end
microcontrollers (e.g. the 8051 in a microwave oven).
I think an embedded newsgroup would be great! I haven't found any other
groups that even come close to discussing embedded issues.
Kevin
Mark Zenier
: Group Name: comp.arch.embedded
I HATE the name. comp.arch???? I suspect that the group-advice
people don't know what end of a soldering iron to pick up.
Oh well. The rest is a really good idea.
Mark Zenier mze...@eskimo.com mze...@netcom.com
Tim Tsai
>so i suppose, i'm saying that it's the phrase "An embedded system can be
>described as any application where a dedicated computer is built right
>into the system but the end product is not a computer" is for us more
>"An embedded system can be described as any application where a
>dedicated computer is built right into the system but is not perceived
>by the end user as a computer". not trying to nit pick, but this is
>significant for me cos i produce tools that run on embedded systems for
>programmers and i know that those guys *know* that it's a computer, and
>so i would like to see discussions about multi-tasking and superloops,
>but at the same time am worried about lock-ups and watchdogs and deadly
>embraces and stuff.
You raised a very valid point. I think you misunderstood
though. Even though I gave a definition of an embedded system
the group is not limited to the embedded system itself.
Obviously tools, support products, etc. are all valid topics in
the group (as stated in the charter). It'd be incredibly boring
if we are only limited to talking about microwave ovens and fax
machines! I personally would be very interested in hearing
about your products.
BTW, I stole the definition from Jack G. Ganssle's "The Art of
Programming Embedded Systems".
Tim