CSLI Calendar, 14 May 1997, vol. 12:28

     C S L I   C A L E N D A R   O F   P U B L I C   E V E N T S

14 May 1997                    Stanford                Vol. 12, No. 28

                     A weekly publication of the
       Center for the Study of Language and Information (CSLI)
      Stanford University, Ventura Hall, Stanford, CA 94305-4115

             CSLI ACTIVITIES DURING 14 MAY -- 23 MAY 1997

         4:15pm Seminar on Computational Learning and Adaptation (SCLA)
                Cordura 100
                Knowledge Intensive Concept Learning and Theory
                Cliff Brunk
                Silicon Graphics
                Abstract below

         4:15pm EE380: Computer Systems Laboratory Colloquium
                Gates B03 (NEC Auditorium)
                Defining the Next Wave: Network Computing
                Mark Foster
                Digital Equipment Corporation
                Abstract below

         5:00pm Discourse @ Networks2000
                Humanities Center Annex   
                Software Engineering and the Unintended Objectification
                of Culture
                Jaron Lanier 
                Abstract below

         8:00pm Kant Lectures
                Bldg 200:34 (History)
                The Necessities of Love
                Harry Frankfurt 
                Department of Philosophy
                Princeton University

        11:00am Philosophy 169: Intensional Logic
                room 60:62A
                Dynamic Logic
                Marco Hollenberg
                Abstract below

        12:00pm CSLI CogLunch
                Cordura Hall, Room 100
                The Metaphorical Structure of Mathematics
                George Lakoff 
                UC Berkeley, linguistics

         4:00pm Symbolic Systems Distinguished Speaker Event
                Annenberg Auditorium
                Post Symbolic Systems
                Jaron Lanier
                Abstract below

        12:30am Seminar on People, Computers, and Design
                Gates B03 (NEC classroom)  
                Negotiating Ontologies: bridging the gaps 
                Austin Henderson and Jed Harris
                Pliant computing
                Abstract below
         3:30pm Linguistics Department Colloquium 
                Margaret Jacks Hall 460:146 
                Jennifer Arnold 

         4:15pm Stanford Digital Libraries Seminar
                Gates B08
                Chuck Swenberg, Agentware Inc.

        11:00am Philosophy 169: Intensional Logic
                room 60:62A
                Foundations for Intensional Logic
                Ed Zalta 
                Abstract below

        12:30pm Philosophy 298: Topics in Logic, Language and Computation
                Room 60:61F
                Games in Logic, A Survey
                Johan van Benthem 
                Abstract below

         4:15pm EE380: Computer Systems Laboratory Colloquium
                Gates B03 (NEC Auditorium)
                John Doerr 
                Kleiner Perkins Caufield and Byers

        11:00am Philosophy 169: Intensional Logic
                room 60:62A
                Foundations for Intensional Logic
                Ed Zalta
                Abstract below
         4:00pm Xerox PARC Forum
                George Pake Auditorium, Xerox PARC
                Rapid prototyping
                Paul Jacobs

         4:15pm Symbolic Systems Forum
                Margaret Jacks Hall 460:146
                Thinking, Meaning, and Speaking
                Avrom Faderman
                Philosophy, Stanford
         7:30pm Stanford Phonology Workshop
                Margaret Jacks Hall 460:146
                Anthi Revithiadou
        12:30am Seminar on People, Computers, and Design
                Gates B03 (NEC classroom)
                Towards A New View of Information: Designing the
                Intellectual Workplaces of the Future
                James D. Hollan
                Computer Science Department, University of New Mexico
                Abstract below

         3:30pm Linguistics Department Colloquium 
                Margaret Jacks Hall 460:146
                Henriette de Swart 



CSLI Tutorials will be held on Thursday and Friday, May 22 and 23.


for further information on the Tutorials.  Because of this no CSLI
CogLunch will take place on May 22.

             on Wednesday, 14 May 1997, 4:15pm to 5:30pm
                           Cordura Hall 100

      Knowledge Intensive Concept Learning and Theory Refinement
                             Cliff Brunk
                  MTS Data-Mining and Visualization
                  Silicon Graphics Computer Systems
                      Mountain View, California
Concept learning algorithms have been used to solve difficult problems
in fields ranging from medical diagnosis to astronomy. In spite of
these successful applications, most concept learners are only able to
utilize knowledge about the problem being solved that is expressed as
a set of training examples. Relevant knowledge from other sources can
not be utilized even when it is available. This talk will present an
investigation of techniques that utilize knowledge in the form of an
approximate theory to facilitate learning accurate concept
descriptions. The first half of the talk will present an approach that
allows the learning algorithm to dynamically select the level of
generality at which to express the target concept. The second half of
the talk will focus on techniques that attempt to identify and repair
flaws in the approximate theory that lead to incorrect classification.
A novel technique for utilizing a previously unused form of knowledge
contained in the approximate theory will be presented. This technique
uses lexical information contained in the term names in conjunction
with an ontology of English concepts to prefer repairs that lead to
the most linguistically coherent theory.
             on Wednesday, 14 May 1997, 4:15pm to 5:30pm
  Hewlett-Packard Auditorium (B01), Gates Computer Science Building
                SITN: Thursday, Channel E-3, 8:00-9:15

              Defining the Next Wave: Network Computing
                             Mark Foster
                    Digital Equipment Corporation
The computer industry is on the threshold of a new era:
Internet-centric computing. Digital Equipment is leading this charge
with the development of an important new Network Computer, which will
enable entirely new generations of Internet appliances. Beyond PC
replacement, this new generation of devices will ultimately transform
the way we work (and play!), and will significantly expand computing
beyond today's PC market.
Biography: Mark Foster is the Director of the Internet Appliances
Group at Digital Equipment Corporation. Part of Digital's Advanced
Development organization, the IAG is chartered with developing
Internet-centric reference designs and appliance products. Mark
previously led the development of more than a dozen different portable
computers, including Digital's HiNote Ultra. He has an MBA from the
University of Notre Dame.

                      DISCOURSE @ NETWORKS.2000
                 Stanford Humanities Center Workshop
                on Wednesday, 14 May 1997, 5:00-7:00pm
                       Humanities Center Annex

                         Software Engineering
            and the Unintended Objectification of Culture
                             Jaron Lanier

Mitch Kapor came up with the marvelously compact slogan "Architecture
is Politics." ("Architecture" in this case signifies the technical
structure of a computer network.)

We would be fortunate, in a way, if the significance of architecture
could be limited to what we think of as politics. Architecture, alas,
is so much more than politics, that it is almost impossible to capture
its importance.  Architecture will also be a foundation for the
language, society, and culture of the future. At first, the design of
the network will seem less important than the content that is moved
over it. This will be true only for the first generation or two of
users. After that it will become apparent that the network's design is
like genetic material out of which our culture unfolds, an intimate
and pervasive presence, a thing, like the structure of our spoken
language, whose influence is too great to be isolated or measured.
The influence of network architecture will re-cast every human
endeavor that involves communication across distance or time. We are
about to create the material with which our civilization will be
largely woven for generations to come. The design of the information
infrastructure will form the weave and the flow of its contents, which
will be most of what we what we create together and pass on as a
Unlike the conduct of government, the structural elements of the
network will be facts, not laws subject to interpretation and
refinement. Most importantly, however, the choices we make will
largely be irreversible.
There are principles that must be built into the network resources
that are created in the next ten years that are not likely to change
for generations, if ever. One reason for this is that it would be hard
for us to agree on how to change a network, just as it is hard to
agree on how to change a law. The main reason, however, is a technical
problem: Software and network architectures are built in a lock-step,
puzzle-like way on foundation assumptions that are almost impossible
to undo once a system has become large.

              on Thursday, 15 May 1997, 11:00am-12:15pm
                             Bldg. 60:62A

                            Dynamic Logic
                           Marco Hollenberg

This is an introduction to Propositional Dynamic Logic, a modal logic
for reasoning about programs and other dynamic systems that involve
state change. We give the basic formalism, as well as a sound and
complete axiom system. In particular, we prove decidability via a
(constructive) finite model property.  (Ref.: R. Goldblatt (1992),
"Logics of Time and Computation", 2nd edition, CSLI Lecture Notes
volume 7, Stanford.)

                   on Thursday, 15 May 1997, 4:00pm
                         Annenberg Auditorium

                        Post Symbolic Systems
                           by Jaron Lanier

Virtual Reality (VR) has lead me to explore some of the extremes of
what might be possible in both natural languages and programming
languages, and these two explorations have influenced each other in
surprising ways.

I originally became involved with VR in the hopes of improving user
interfaces for large and complex computer programming tasks.  In the
course of this work, I became convinced that advanced user interfaces
would influence the core as well as the surface of programming
language design.  Some of the most entrenched ideas about computer
programming languages might be understood better as mnemonic devices
to help users cope with text-based or text-influenced user interfaces.
I have explored this possibility by designing a series of user
interface-intensive programming languages that reject seemingly
ubiquitous ideas like parsers and source code.

I also started to wonder about the role of symbols and abstraction in
natural language.  I have postulated a new type of natural
communication, as a thought experiment, that might be at least
theoretically possible at some time in the future.  There would be
excellent modeling and programming tools for networked VR in this
future, and a community of people highly skilled in the fast
construction of shared virtual worlds.  Members of this community
could hypothetically communicate by creating rapidly changing content
in a shared, objective world.  They would create and share content
directly, instead of referring to contingencies indirectly with words
or other symbolic devices.  This is what I call post-symbolic

While it might at first seem that symbols, abstractions, and
categories would be needed to communicate anything substantial, even
in this future, that does not appear to be the case.  For just one
example, instead of abstract categories or platonic ideals, it might
be possible to create a concrete, but very large, collection of
objects that are to be considered as similar.  Such a collection could
be held inside a virtual jar, for instance, that is small on the
outside but big inside, and could be available as conveniently as a

Another way to say this is that concreteness could be as versatile as
abstraction, if it becomes very easy to make and change concrete
things.  If it is at least possible that our understanding of the
range of potential natural languages has been limited by assumptions
based on text, it is certainly worth re-examining our assumptions
about computer languages.

Biography: Jaron Lanier is a computer scientist, composer, visual
artist, and author.

Lanier is probably best known for his work in Virtual Reality. He
coined the term 'Virtual Reality', and founded the VR industry. He
started the first VR company, VPL Research, Inc., which produced most
of the world's VR equipment for many years. He is the co-inventor of
fundamental VR components such as interface gloves and VR networking.

Lanier was also the first to propose and implement a variety of
technologies that have since spawned industries in their own
right. Among his lineup of "firsts" are the first "avatar" for network
communications, the first moving camera virtual set for television
production, and the first performance animation for 3D computer
graphics.  He was the first to propose web-based network
computers. Along with Dr.  Joe Rosen and Scott Fisher he initiated the
fields of real-time endoscopic surgical simulation and telesurgery. As
a computer scientist, Lanier is also known as a pioneer in the field
of visual programming.

Music is Lanier's first love and he has been an active composer and
performer in the world of new classical music since the late
seventies. He is also a pianist and a specialist in unusual musical
instruments, especially the wind and string instruments of
Asia. Lanier has performed with artists as diverse as Philip Glass,
Ornette Coleman, Vernon Reid, Terry Riley, Barbara Higbie, and Stanley
Jordan. He also writes chamber and orchestral music. His record
"Instruments of Change" was released on Point/Polygram in 1994. In the
works are a new album of chamber music for Sony Classics, an
orchestral commission for the St. Paul Chamber Orchestra, and a
Ballet, "The Thinning of the Veil".

Lanier's paintings and drawings have been exhibited in galleries in
the United States and Europe and in the Internet. In 1994 he directed
the film "Muzork" under a commission from ARTE Television. His 1983
"Moondust" is generally regarded as the first art video game, and the
first interactive music publication. In 1996 he presented the "Video
Feedback Waterbed", a large installation at Exit Art in New York
City. Lanier's best known visual art, however, is his work in the
design of virtual worlds, including "The Sound of One Hand", and many

Lanier is also a well known author and speaker. He writes on numerous
topics, including the philosophy of consciousness, internet politics,
and the future of humanism in a technological world. He is a founding
contributing writer for Wired Magazine, and was the guest editor of a
special issue of the magazine SPIN devoted to the future (November
1995).  He appears on national television regularly, on shows such as
"Nightline" and "Charlie Rose", has been profiled in many prominent
publications, such as the Wall Street Journal, and has had his
original research featured on the cover of Scientific American twice.

                 on Friday, 16 May 1997, 12:30-2:00pm
                      Gates B03 (NEC Classroom)
                          (SITN Channel E4)

              Negotiating Ontologies: Bridging the Gaps
                   Austin Henderson and Jed Harris
                           Pliant Research

The last fundamental advance in human-computer interaction -- the
graphical user interface -- was made well over two decades ago; it's
been commercially available for at least 15 years. Should we content
ourselves with incremental improvements, or should we aim to do better
than that?

In this talk we claim that we should do much better: that current
design practices lead to a serious mismatch between human activity --
very rich and flexible -- and computational activity -- very
simplistic and rigid. People constantly negotiate and evolve their
practices, while digital technology typically demands a fixed view of
the world. This inevitably leads to "ontological gaps" between people
and their machines. However, we claim that improved design practices
can help to both bridge these gaps, and to make digital systems pliant
enough so that people can mold them to fit their practices.

Biography: Jed Harris has worked in software architecture and object
oriented technology since 1977. He contributed to distributed object
architectures at Data General and Intel, and was one of the founders
of the OOPSLA conference. At Apple he developed foundations for
software components, and was co-architect of OpenDoc. During the past
two years he focused on research leading to formation of Apple's
Discourse Architecture Laboratory. Currently he is Chief Technology
Officer of Ricoh Silicon Valley, and a co-founder of Pliant Research.

Austin Henderson has been in the field of Human-Computer Interaction
since 1964. He has a B.Sc. in Mathematics from Queen's University,
Canada, an MS in Computer Science from the University of Illinois, and
a Ph.D in Computer Science for MIT. He has built applications in areas
including manufacturing, air traffic control, electronic mail
(Hermes), user interface design tools (Trillium), workspace management
(Rooms, Buttons). He has done research and user interface architecture
for Xerox at both PARC and EuroPARC, and for Apple Computer, and
industrial design with Fitch. During the past two years at Apple, he
managed the Discourse Architecture Laboratory and its precursors.
Currently, he is a principal in Rivendel Consulting, and a co-founder
of Pliant Research. Professionally, Austin has been active in
ACM/SIGCHI since 1983, including as conference chair (1985), and
organization chair (1989-1993).

     on Tuesday and Thursday, 20 and 22 May 1997, 11:00am-12:15pm
                             Bldg. 60:62A

                  Foundations for Intensional Logic
                               Ed Zalta

We will motivate and develop a formal system and, instead of building
models, formulate a precise theory about intensional objects such as
properties, relations, propositions, possible worlds, situations, and
fictions.  The logic will be applied to solve puzzles about various
kinds of "hyperintensional" contexts which occur in natural language.
(Ref.: E. Zalta, "25 Basic Theorems in Situation and World Theory",
Journal of Philosophical Logic 22 (1993) 385-428.)

               on Tuesday, 20 May 1997, 12:30pm-2:15pm
                             Bldg. 60:61F

                       Games in Logic, A Survey
                          Johan van Benthem

Games have a long history in logic. Examples are: Ehrenfeucht games
for model comparison, Hintikka games for evaluation, or Lorenzen games
for winning arguments. These games come from many traditions, ranging
from medieval logic exams to the philosophy of science or language.
We will present a historical survey of the 'game-theoretic paradigm',
discussing its changing fortunes, and current dynamic relevance.
(Ref. J. van Benthem, 'Games in Logic', in J. Hoepelman, ed., 1988,
"Representation and Reasoning", Niemeyer Verlag, Tuebingen, 3-15.)

                 on Friday, 23 May 1997, 12:30-2:00pm
                      Gates B03 (NEC Classroom)
                          (SITN Channel E4)

                  Towards A New View of Information:
     History-Enriched Digital Objects and Dynamic Work Materials
                           James D. Hollan
        Computer Science Department, University of New Mexico

The future promises an ever richer world of computationally-based work
materials that exploit task representations, semantic relationships
explicit and implicit in information and our interactions with it, and
user-specified tailorings to provide effective, enjoyable, and
beautiful places to work. One of the barriers to achieving this vision
is that most current user interfaces employ computation primarily to
mimic mechanisms of older media. While there are important cognitive,
cultural, and engineering reasons to exploit earlier successful
representations, imitating the mechanisms of an old medium strains and
underutilizes the new.

For quite some time I have been involved in a research enterprise to
look beyond imitation as the fundamental strategy of interface design.
This has led to an investigation of:

  * HISTORY-ENRICHED DIGITAL OBJECTS: Records the interaction events
    that comprise the use of digital objects so that on future
    occasions graphical abstractions of the accrued histories can be
    rendered as part of the objects themselves.
  * BEYOND BEING THERE: Questions the efficacy of imitating
    face-to-face interaction as the ideal way of supporting
    computer-mediated communication.
  * PAD++: Explores a dynamic zoomable graphical interface for
    multiscale information access and visualization.

In this talk I give an overview of this work and argue that there is
evidence for the beginnings of a paradigm shift for thinking about
information, one that starts to view information as being much more
dynamic and reactive to the nature of our tasks, activities, and even
relationships with others.

Biography: Professor James D. Hollan received a Ph.D. from the
University of Florida and completed a postdoctoral fellowship at
Stanford University. He was on the faculty of the University of
California at San Diego (UCSD) for a decade. During that period he
directed the Future Technologies Research Group at NPRDC and, in
collaboration with Professors Edwin Hutchins and Donald Norman, led
the Intelligent Systems Group in the Institute for Cognitive Science
at UCSD.

Professor Hollan left UCSD to become Director of the MCC Human
Interface Laboratory and subsequently established the Computer
Graphics and Interactive Media Research Group at Bellcore. In Fall
1993, he became Chair of the Computer Science Department at UNM. His
primary research interests are human computer interaction,
computationally-based media, and computer-mediated communications.

                             END MATERIAL

The CSLI Calendar appears weekly on Wednesdays throughout the academic
year.  Announcements, abstracts, and other information to appear in
the Calendar should be submitted to the editor, who reserves the right
to decide what does or does not go in the calendar

Requests to be added to the mailing list should be sent to
[mailto:csli-request@csli.stanford.edu].  Note that the CSLI Calendar
no longer has a paper version.

The full current issue is at
and the archives at

People on most of the CSLI computers can type 'help csli-calendar' to
see the current issue.  They can also type 'help quip' to see the
Linguistics Department newsletter, the Sesquipedalian,

The CSLI Calendar is also posted each week to

Information about CSLI's research program is available at

For maps to the Stanford University campus see