Focused Effort Title: Integration of Palm-Top Devices into Education
                      and Computing Portals

Thematic Areas:
       1. Dod User Productivity

PI Name:  David Bernholdt
PI E-mail Address  bernhold@npac.syr.edu
PI Telephone: 315 443 3857
PI Fax: 315 443 1973

Project Description:

Personal digital assistants (PDAs) or "palm-top" computing devices,
such as the 3Com Palm Connected Organizer and systems running Windows
CE, have come into widespread use as personal organizers.  Because of
their portability, low cost, and programmability, they also have
significant potential in an interconnected environment to link users
to information sources and to each other.

These capabilities can be applied to avariety of areas, including
computing and education.  In an educational setting, palm-top devices
can provide individual students with a number of useful capabilities
at a lower cost-per-student than a full-blown computer:
o Supporting simple quizzes or voting
o Sending key data to students (reference URLs, notifications, etc.)
o Note taking and annotation
o E-Mail
o Chat tool
o Simple sketchpad
When used to to access computing portals, PDAs could provide
notification and summary details of final or intermediate results of a
computation, initiation of jobs, input of "key" parameters, monitoring
of batch queues, and other functionality.  As the communications and
computing capabilites of palm-tops improve, we can expect to see a
growing suite of capabilities.

We propose to use the infrastructure of collaborative environments (in
particular Tango Interactive) to provide this integration, as such a
system is designed event messaging, which is natural communication
mechanism for palm-tops.  Presently the Tango Interactive only supports
synchronous messaging, and extension to asynchronous event messaging will
be required.  Events will be serialized in terms of XML and rendered in
device-dependent fashion using either a specialized application such as a
chat or whiteboard or more generally a "microbrowser" (essentially similar
to, but much simpler than, a workstation browser).  It will be possible to
filter events so that only those appropriate to micro Tango clients are
actually transmitted.  Event processing and filtering will occur in a
"gateway" between the standard Tango system and the palm-top device. This
filtering is equivalent to applying "palm-top style sheets" to XML Web
pages which can be viewed on conventional displays with different
style-sheets.

This architecture has some similarities with the current "JavaScript
Shared Browser" (JSSB) Tango application.  One JSSB window queues all
incoming Tango event messages, which are looked at in 0.5 second
intervals and either executed, deferred, or discarded.  Note that the
architecture also naturally supports event archiving. Using an analogy,
we see current Tango having a pure synchronous architecture like
phone conversation; palm-tops seem better supported by a "pager"
architecture with both synchronous and asynchronous events supported.

We expect that Java and Jini will ultimately provide an excellent
environment in which to produce micro Tango clients, though their use
will not be feasible in the initial stages of this project. (A robust
Java infrastructure should be available in a 12 month time frame.)
Jini allows clients with possibly erratic connectivity to register
themselves, which can in turn be a source of events in the Tango
system.

The PDA might be viewed as a standalone Tango client, or might be treated
as an application in conjunction with a regular Tango instance running on
"host" PC.  The palm-top might connect to Tango in a variety of ways
ranging from temporary wired (i.e. Palm cradle) or infrared serial
connections to modems, pagers, cellular packet data services, or other
approaches.  Our initial efforts will focus on the second approach where
support for asynchronous events will be built into the Tango application
(as done already in shared browser) rather than the full Tango system.
This is obviously limited but should fully illustrate the basic ideas even
if it does not have the general power coming from integrating asynchronous
support into Tango. Note that the development of the micro device gateway
(XML serialization, filtering, and rendering) will be designed to be more
or less independent of the specific connectivity.

This initial project will pave the way for later work which could use both
a new architecture for Tango and also applications such as the Whiteboard
which produce XML suitable for either palm-top or PC display. As we expect
the communications capability of palm-top's to greatly improve, we will
not try to optimize this during this first effort but make use of existing
cradle/modem/wireless capability. Note we will try to have a capability by
SC99 but cannot guarantee this. Formally we have listed December 99 as
first demonstration of concept. We see many important technology
developments over the next 6-12 months including the widespread deployment
of Version 5 browsers and much greater use of XML and Jini. we include as
deliverable, a report describing both lessons from the PDA work and the
implications for collaborative systems of these other developments.

Benefits:

This project is intended to investigate how low-cost ubiquitious
palm-top computing devices might be harnessed to provide users with
easier access to each other (collaboration) and information.  While
somewhat speculative, we believe this project could provide tools for
use in education and training, and in operation and use of the MSRC.
In the longer term, as the communications infrastructure expands, and
the capabilities of hand-held devices increase, we can envision
ruggedized wireless PDAs deployed to troops in the field to allow them
rapid access to a wide range of information on the battlefield using
an architecture such as this.

Required Resources:

Total cost: $50,000

Estimated monthly expenditure (in thousands of dollars)
   Mar   Apr   May   Jun   Jul   Aug   Sep   Oct   Nov   Dec   Jan   Feb   Mar 
   0.0   0.0   0.0   0.0   0.0   0.0   8.2   7.5   7.5   7.0   8.4   7.4   4.0


Deliverables:
o Assessment of potential micro Tango platforms and selection of one
  environment for inital development work (Sept'99)
o Initial Very Simple Proof of Principle (Dec'99 -- we will try for
  SC99 but cannot guarantee this)
o Micro Tango gateway based on XML (Jan'99)
o Proof of principle demonstrations in the educational and computing
  areas. (Through collaboration, it might be possible to have
  interesting and non-trivial content for these demonstrations; see
  note below.) (Mar'99)
o Simple "microbrowser" rendering engine for palm-top, possibly based
  on available software (Mar'99)
o Report on Technology for next generation of collaborative systems
  including PDA's (Mar'99)

Opportunities for Collaboration:

We expect to focus primarily on the infrastructure used to process and
render event messages for palm-top devices.  There is ample
opportunity to work with other PET partners who can focus on various
types of content that can make special use of a "micro Tango" system.
Examples could include interfacing with HPC systems to provide
machine, queue, and job status information; or providing tools to
handle various kinds of educational content noted above.

Notes:

The model described here for integration of palm-top devices into
Tango has many similarities to the Wireless Access Protocl (WAP)
developed by the WAP Forum (http://www.wapforum.org).  The emphasis of
the WAP Forum is essentially on providing a limited form of web access
to wireless devices, but we will watch this technology both from a
standards-compliance viewpoint and for anything that might be useful
in the development of micro Tango.

Expected Customers
o PET & MSRC staff
o MSRC users
o Education and Training participants