Ref.
Dr. Mark is vice president of the Information and Computing Sciences
Division of SRI International. (bill.mark@sri.com)
http://www.research.ibm.com/journal/sj/384/mark.html
Cited References http://www.research.ibm.com/journal/sj/384/markref.html
1. D. Englebart, Augmenting Human Intellect: A Conceptual Framework,
AFOSR-3233 (October 1962), available at
sloan.stanford.edu/mousesite/EngelbartPapers/B5_F18_ConceptFrameworkInd.html.
22. A. Cheyer, L. Julia, and J.-C. Martin, "A Unified Framework for
Constructing Multimodal Applications," in Proceedings of the Second
International Conference on Cooperative Multimodal Communication (CCMS
'98), San Francisco (January 1998), pp. 63-69.
This research paper concludes that multiperson speech and gesture
tracking - whom is saying what to whom ... is key for turning pervasive
computing into mediated spaces.
Since the UWB radar microphone can function as a biometric, motion
sensor, and localizer simultaneously, it's capable of tracking the
collaboration between people within ~ 12 ft. radius and can be extended
beyond by an distributed array of multiple network hops. Below are links
for more info, including Dr. Mark's smart and mediated spaces
application notes.
UWB Radar Microphone
http://speech.llnl.gov/ICASSP2000/index.html
Berkeley Wireless Research Center (BWRC) project proposal to DARPA gives
a good overview of the different aspects targeted by a network of
PicoRadio nodes.
http://bwrc.eecs.berkeley.edu/Research/Pico_Radio/documents/990525_IIB_TechRation.pdf
PicoRadio use case scenarios
http://bwrc.eecs.berkeley.edu/Research/Pico_Radio/documents/picoenv.pdf
* Smart spaces
"The relationship between people and pervasive computation that ought to
come into being is a seamless integration of people, computation, and
physical reality: a "smart space." The concept of a smart space has a
long history in computer science. In the early 1960s Doug Engelbart at
Stanford Research Institute (now SRI International) was exploring the
concepts of human-computer systems that could augment human capability,
especially humans working in groups.1 Although this work is most famous
for the mouse interface, its primary contribution is probably the "smart
space" vision that still informs the research community. The "Media
Room" project developed by the MIT (Massachusetts Institute of
Technology) Architecture Machine Group in the mid-1970s2 explored the
concept of users interacting with room-sized computational environments.
The result was a new human-computer interface based on the combination
of speech and gesture input, and text and graphics output."
* The mediated spaces vision
"Mediated spaces will enhance human activities by enhancing the
interaction of people in the space. For example, almost all complex
artifacts are designed by groups, and group design requires group
interaction. A design session is a meeting of a team to work on a
design. Usually it is an informal interaction in which design problems
and solutions are raised and discussed, often resulting in a decision.
The goal is to extend or correct the group's shared representation of
the design. This shared representation is partly on paper, partly in the
minds of the participants, and, increasingly, partly in computer
software. But even with digital desks and smart offices, even with the
great advances in computer-aided design software and the computer form
factors that allow it to play a role in design sessions, much of the
information in a design session is lost. The interpersonal interaction
that takes place during the session does not make it into the updated
design representation. This information, including the rationale for
decisions and the alternatives that were discussed, is only incompletely
and briefly remembered by the participants and is not available at all
to nonparticipants.
In a mediated design session, the computational environment incorporates
the information contained in the speech and gestures of the participants
into the shared design representation. The mediated space in this case
is a smart conference room like the Intelligent Room. The room's
paraphernalia might include a smart whiteboard, people tracking cameras
and microphones, and so on.3,4 Designers may enter the room with
personal notepads or other computational devices that contain
information about the design (see mock-up in Figure 1). As designers
discuss and argue about the design, they point to images on paper or
embodied in the computer-aided design (CAD) software of their notepads,
workstations, or other displays. The difference is that the mediated
space is focused on the participants' interaction with one another, not
just with the devices in the space.
Figure 1
The role of the mediated space is to incorporate interpersonal
interaction into the design representation. The goal is not to simply
record everything that was said,11 but to represent relevant information
in relevant places in the design representation. The space takes a
proactive role, suggesting relevant information from outside the
session, including other designs it knows, along with ancillary
information from the company's intranet and the Web. It proactively
detects inconsistencies with both the current design ("that conflicts
with an earlier decision") and the other designs it knows ("these
similar designs had a higher power budget").12,13 The result is an
enhanced design experience, guided by the space's understanding of the
current and previous sessions, and an enhanced design representation
that incorporates rationale and alternatives."
This archive was generated by hypermail 2.0.0 : Tue Aug 21 2001 - 17:57:48 PDT