RE Interface:
------------
I agree with Victor ... we need an interface for wearables that is
designed for wearables. The output is OK (HMD, flat-panel, etc.), but
the inputs are the challenge.
I've wondered about vocal control of a pointing device.
For example, humming higher or lower to move a pointer, etc.
Does anyone know of applications of human-generated audio for computer
input other than voice command/control/recognition?
RE Powering off devices:
-----------------------
That won't be an issue in the future.
Even though many devices still have rotating media today, smart SW can
shut them down when not in use.
Displays and audio amps can be SW controlled as well (even our 1 watt
computers can disable Video to drop 1/3 off the power draw, and turn off
the audio amplifiers, USB host power, etc. ... and eventually go into
"sleep mode" at ~ 1% of a watt ... actual figure TBD)
But as long as people are using X86 CPUs and hard-disks you will have to
shut down devices to get decent battery utilization.
But devices based on "modern" CPUs (such as ARM, or IBM's 405LP) will
dramatically extend run times. For example, the "preliminary version" of
IBM's 405LP CPU manufactured in their 0.18um, CMOS SA-27E process
consumes 1/2 watt at 380 MHz or 577 MIPs ... the equivalent of Pentium
III 750 MHz roughly for compute power, and it drops to only 53
milli-watts at 152 MHz. When this CPU is migrated to their .13 micron
technology in Q2 2003, the power will be reduced in half again!
And it has a dedicated front-end processor for voice-recognition that
can be shut-down when not needed to save power.
Our target is to build computers that will run a week on a couple of
flashlight batteries with lap-top-like performance. Is anyone else
working on this?
RE storing data on the network:
------------------------------
With proper security you don't have to worry about a thing.
We are considering using technology that we implemented for a NATO
secure fax device years ago for forward error correction and
technology. Its based on BCH codes.
In the future, your local storage will be solid-state, and it will
simply be a cache of data that is located elsewhere -- maybe spread over
dozens of locations, from your cell-phone to your home gateway to the
library of congress.
With intelligent algorithms, the data you use *the most* will be
retained, rather than just the most recently accessed as per current
caching schemes. Your personal "data bots" will make sure data you need
is always available, and will also spread all your personal and business
stuff over a variety of repositories to secure against storage and
communications failures, all totally encrypted.
Brian
~~~~~~~~~~~~~~~~~~~~~~~~
Brian Empey, P. Eng.
President
Technical Solutions Inc.
~~~~~~~~~~~~~~~~~~~~~~~~
Victor Alexander Debuque wrote:
>
> The point is not whether the wearable UI is part of the OS or not. The
> point is that the desktop metaphor (i.e. the so-called WIMP interface -
> Windows, Icons, Menus, Pointer) doesn't match the ergonomics of the
> wearable paradigm. True, you can run software on top of the desktop and
> hide it, but on a wearable that just means you have an extra layer of
> software that you don't need, wasting valuable resources. What we need
> is something to replace the desktop that facilitates usability for
> wearables.
>
> > Another question: It occurs to me that wearables might be always-on
> > devices (or at least, that's an option I might want). So do we plug
> > them in overnight to reload their batteries, but they're in standby
> > mode (in order to take phone calls, etc.)? Or just let our network
> > ISP/Web host store our calls, etc., until we ourselves wake up and
> > "wake up" our wearable?
>
> I would think it's a matter of personal preference - no reason why you
> couldn't choose between both options depending on your needs or on how
> you're feeling that particular day. Personally, I wouldn't mind either
> of the alternatives you mentioned.
>
> Not directly related -- it would be nice if they could be recharged
> wirelessly. Anyone know if such technology is anywhere near becoming
> feasible?
>
> > Finally, I do NOT want my primary data store to be on any network,
> > unless by "primary data store" you mean things like all the works of
> > Shakespeare and maps and other raw data already (pretty much) on the
> > Web. Any information I have massaged or created (including papers
> > ABOUT said works of Shakespeare, etc.) I want stored locally, for too
> > many reasons to mention. (Your faith in your government's respect for
> > your privacy is touching, but naive. They've been after mere library
> > reading records for years, and used the terrorist attack on WTC &
> > Pentagon to get that far.)
>
> Believe me, I have no faith whatsoever in any government's or
> corporation's respect for individual privacy. I was thinking more of a
> secure data server based at home or at the office, to which you can
> have access over the network. As Robin mentioned, you can connect over
> SSH -- no need to let someone else act as gatekeeper to your private
> data.
>
> Victor
>
--
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* don't subscribe through a forward/expander/false domain
From Wear-Hard Mailing list Archive (WH)
Maintained by R. Paul McCarty
Archive created with babymail