> > but would it be sufficient for a "smart" terminal, connected to
> > something cheap like a Palm or a Visor?
>
> Perhaps. You do have ethernet, rs232, and 1-wire microlan plus
> general purpose IO pins to play with. It really depends on what
> level of complexity you need and whether you want to work at a
> higher level (java programming) vs lower level (code running on
> a traditional microcontroller). The cool thing about TINI is
Well, this is why I'm talking about a "host computer" like a Palm
or a Handspring Visor...they would give the traditional "computational
power" that we think of, and the terminal would provide a generic
interface tailored to wearing. It could connect such things as a
keyboard (twiddler, or whatever), display systems (HMD, sleeve
LCD, Braille display), and a generic interface to such things as the
1-wire peripherals that you mention further down in your original
message.
As handhelds become cheaper and more powerful, the generic
terminal will still allow a wearable interface to whatever system
you connect it to. Standard serial or ethernet connection...
Of course, I assume that there is a JVM that runs on these
handhelds...simply instruct the software to output, rather than
to stdout:, to com:. The wearable terminal would provide the
remainder of the specifics for the interface.
Is this a realistic approach, Doug? You've had a wearable...I
haven't.
> > and a separate video "module" which could grow and
> > change as necessary. We need to come up with a
> > specification for the port
>
> We would be crazy to not use the existing ISA or PCI
> busses for connection to the core CPU. As for the output
What about something like AGP? It is designed for high
performance video interfaces...I know nothing about its
power requirements or its specifications.
> What we really need to do is make a list of all of
> the various parts of the infrastructure, and then ask
An excellent idea.
> > Theoretically, there is no reason that someone could not hack
> > an old Amiga, or MAC, and use it as a wearable base...with an
> > appropriate VM it could run all the software we come up with,
> > and still provide a low cost (this is an important factor to
> > me) starter system.
>
> Well, just as Edward Keyes cautioned not to go FPGA crazy, I
> also caution not to go JVM crazy. Both have advantages in
I just mentioned this, because (in theory) JVMs allow for platform
independence. I'm a firm believer in "starter" systems...low end
machines that people can actually afford. There is no reason why
someone should have to sink $2K into an experimental collection
of parts, when they could take the rather small motherboard out
of a MAC, or even a Wintel laptop, and hack it together themselves.
Frequently, however, this approach locks them out of the use of
much of the software out there.
I'm suggesting JAVA because this way, regardless of what is cost
effective for each wearable user, if a JVM exists they can run the
base of software which will undoubtedly exist at some point in the
not too distant future. Of course I wouldn't expect a 386 to run
as quickly as a PIII, or a 68K based MAC to run as quickly as
a G4 equipped system. But, if we keep the software virtually
hardware independent, then all of these become viable wearable
platforms.
I have a feeling that this type of approach could help wearables
reach critical mass rather quickly.
> certain areas and disadvantages. With Jave you also take a
> performance hit but sometimes it's worth it (for example I
I would expect that. As I understand it, doesn't the JVM
translate bytecode into native language instructions?
> the JVM is not as magically portable as Sun touted, there
> are problems in compatibility across architectures (esp in
> thread handling)
Threads are a problem? That's disappointing. Threads are such
a basic function.
> > I wouldn't object to a basic JAVA capable system that could be
> > mass produced on such a scale that it could be stuck in that
> > $50 wearable after certain quantities had been reached.
>
> I don't think this is realistic by today's standards.
Of course not...but it's also a nice goal. With a basic and very
flexible hardware core, that was mass produced on a sufficient
scale, I don't think this is unreasonable.
Look at it this way. The Motorola T-900 pagers are based on an
embedded 386 core, as I understand it. This is a wearable sized
386 computer, with a keyboard and LCD display, and wireless
2-way communications, which is mass produced on a scale
sufficient to sell it for <$200.
That's retail. I've seen them for <$100 locally, which means that
the main board probably costs ~$50 to manufacture, at most.
With sufficient quantities and a "cheap to build" design, we could
probably meet my $50 goal. I don't think it's unrealistic.
> > Possible way to increase market share
>
> Market share? I would prefer to call it mindshare. I would
Fair enough...I prefer to look at it as a potential growth industry.
Example, ISA bus specifications. IBM released them publically,
and a whole industry was built around them. Because of this,
ISA based IBM machines enjoyed market dominance for 20
years, and technically still do, though it's no longer IBM
specifically.
LINUX is the same way -- free OS, with a huge industry (RedHat,
Yggdrasil, LINUX approved hardware specs, etc) beginning to grow
around it. Market share for LINUX is beginning to grow, and eat
into windoze and other alternatives...it's also establishing itself as
an embedded system, like on the internet terminals that so many
companies are beginning to develop.
> like to think of this as a way to make wearable tech avail
> to many people, and a way to advance the field through
> collaborative efforts among many wearable hackers ...
Of course. But...and you'll probably agree with this...for this to
take off and reach critical mass, business will have to become
involved. Churning out cheap SBCs suitable for experimentation
will likely be the first step.
That way, more than just a relative handful of list members will
be able to get involved in the revolution.
-- Chuck Knight
________________________________________________________________
GET INTERNET ACCESS FROM JUNO!
Juno offers FREE or PREMIUM Internet access for less!
Join Juno today! For your FREE software, visit:
http://dl.www.juno.com/get/tagj.
--
Subscription/unsubscription/info requests: send e-mail with subject of
"subscribe", "unsubscribe", or "info" to 
Wear-Hard Mailing List Archive (searchable): http://wearables.blu.org
please, Please, *PLEASE* don't subscribe through a forward/false domain
From Wear-Hard Mailing list Archive (WH)
Maintained by R. Paul McCarty
Archive created with babymail