[Prev][Next][Index][Thread]
CSLI Calendar, 18 February 1998, vol. 13:21
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
______________________________________________________________________
18 February 1998 Stanford Vol. 13, No. 21
______________________________________________________________________
A weekly publication of the
Center for the Study of Language and Information (CSLI)
Stanford University, Ventura Hall, Stanford, CA 94305-4115
____________
ACTIVITIES DURING 18 FEBRUARY TO 27 FEBRUARY 1998
WEDNESDAY, 18 FEBRUARY
12 noon Psychology Department Colloquium
Bldg. 420:050 (Jordan Hall)
Perception and Memory as Related Processes:
Evidence from Early in Development
Scott Adler
University of Denver
3:45pm Psychology Department Colloquium
Bldg. 420:050 (Jordan Hall)
Towards the Neuronal Correlates of Visual Consciousness
Christof Koch
CalTech
4:15pm EE380: Computer Systems Laboratory Colloquium
Gates B03 (NEC Auditorium)
Hamlet on the Holodeck:
The Future of Narrative in Cyberspace
Janet H. Murray
MIT
Abstract below
THURSDAY, 19 FEBRUARY
12 noon CSLI CogLunch
Cordura Hall, Room 100
Executions, Accomplishments, Motivation
David Israel
SRI
4:00pm Xerox PARC Forum
George Pake Auditorium, Xerox PARC
Design and Modeling in the Fourth Dimension
Professor Martin Fischer
Stanford University, Department of Civil and
Environmental Engineering
Abstract below
4:00pm CSLI Talk
Cordura 100
Amalia -- Abstract MAchine for LInguistic Applications
Shuly Wintner
Tuebingen
Abstract below
4:15pm Seminar on Computational Learning and Adaptation (SCLA)
Gates 100
Massive Time-Series Data Mining via Envelope
Learning and Monitoring
Dennis DeCoste
Jet Propulsion Laboratory, Pasadena, CA
Abstract below
4:30pm History of Science Colloquium
History Corner, room 307
Science Studies after Social Construction: The Return of
the Comparative and the Global
Margaret Jacob
History and Sociology of Science, University of
Pennsylvania
7:30pm Phonology Workshop
Margaret Jacks Hall, 460:146
Complex Clusters: New Prosodic Domain-stem (Case study in
Modern Georgian)
Marika Butskhrikidze
Leiden University
Abstract below
FRIDAY, 20 FEBRUARY
12 noon Logic Lunch
Room 380:383N
On the Classical and Parametric Complexity of the
Ehrenfeucht-Fraisse Game
Elena Pezzoli
Stanford
Abstract below
12:30pm Seminar on People, Computers, and Design
Gates B01 (HP classroom)
The Eyes Have It:
User Interfaces for Information Visualization
Ben Shneiderman
University of Maryland
Abstract below
3:15pm Philosophy Department Colloquium
Bldg. 90:92Q
Autonomous Action: Self-Expression in the Passive Mode
Sara Buss
Asst. Professor, University of Iowa
3:15pm Cognitive Seminar
Jordan Hall 420:100
Cognition and the Visual Arts
Bob Solso
U Nevada: Reno
MONDAY, 23 FEBRUARY
3:30pm Computer Science and Economics Seminar
Math 380:380X
Aggregation and Disaggregation of Information Goods:
Implications for Bundling, Site Licensing and
Micropayment Systems
Erik Brynjolfsson
Economics, MIT / Stanford
3:30pm Psychology Social Lab
Jordan Hall 420:100
How Do I Love Thee?
Patterns of Romantic Attachment and Loss
Phillip R. Shaver
UC Davis
4:30pm Stanford Digital Libraries Seminar
Gates B08
Bayou:
Supporting Asynchronous Collaboration through Replicated Data
Karin Petersen
Xerox PARC
Abstract below
WEDNESDAY, 25 FEBRUARY
1:15pm Neurobiology Seminar
Fairchild Auditorium
The Neurobiology of Sex
Dr. Simon LeVay
Host: Dr. Barbara Barres
4:15pm EE380: Computer Systems Laboratory Colloquium
Gates B03 (NEC Auditorium)
To be announced
THURSDAY, 26 FEBRUARY
12 noon CSLI CogLunch
Cordura Hall, Room 100
Agency, Will and Practical Reason: In Search of a Middle
Way
Michael Bratman
Stanford, Philosophy
4:00pm Xerox PARC Forum
George Pake Auditorium, Xerox PARC
Self-Assembled, Self-Ordered, Nanostructures: A Path to
21st Century Electronics?
Ted Kamins
Quantum Structures Research Initiative
Hewlett-Packard Laboratories
4:15pm Seminar on Computational Learning and Adaptation (SCLA)
Gates 100
Tommi Jaakkola
UC Santa Cruz.
7:30pm Phonology Workshop
Margaret Jacks Hall, 460:146
Some Arguments Against Using Underlying Representations
in OT Phonology
Edward Flemming
Department of Linguistics, Stanford University
Abstract below
FRIDAY, 27 FEBRUARY
12:30pm Seminar on People, Computers, and Design
Gates B01 (HP classroom)
AHA: Audio HTML Access
Frankie James
Stanford
Abstract below
3:15pm Philosophy Department Colloquium
Bldg. 90:92Q
Identifying Praise: In Moments of Henry James and Fred
Astaire
Stanley Cavell
Professor, Harvard University
3:15pm Cognitive Seminar
Jordan Hall 420:100
Language Biases in Recognition Memory
Meredyth Krych
3:30pm Linguistics Department Colloquium
Margaret Jacks Hall 460:146
Title to be announced
James Huang
CASBS, Stanford University, and University of California,
Irvine
____________
EE380 COMPUTER SYSTEMS COLLOQUIUM
on Wednesday, 18 February 1998, 4:15pm to 5:30pm
NEC Auditorium (B03), Gates Computer Science Building
http://www.stanford.edu/class/ee380/
Hamlet on the Holodeck:
The Future of Narrative in Cyberspace
Janet H. Murray
MIT Center for Educational Computing Initiatives
http://web.mit.edu/jhmurray/www/
http://web.mit.edu/jhmurray/www/HOH.html
The current moment is analogous to the invention of the movie camera
100 years ago. A new medium of communication, with particular promise
for the creation of narrative has arrived, but we have yet to invent
the new narrative forms that will be the equivalent of the "movies."
We are currently in transition between the additive stage of
"multi-media" (equivalent to the creation of "photo-plays") and the
development of expressive form. This change will happen when we learn
to exploit the intrinsic properties of the computer. This talk will
identify those properties and the narrative pleasures and problems
intrinsic to digital media.
Biography: Janet H. Murray is Senior Research Scientist and Director
of the Program in Advanced Interactive Narrative Technology (PAINT) in
MIT's Center for Educational Computing Initiatives. She is a pioneer
of multimedia computing in the humanities and has won the Educom and
the Gold Cindy awards among others for her work in interactive
design. Her recently published Hamlet on the Holodeck: The Future of
Narrative in Cyberspace (Free Press, 1997) was cited by Library
Journal as one of the two "Best Books of 1998" in the field of
computing, and has been hailed as a "brilliant look at the future of
storytelling," (James Coates, Chicago Tribune),"a landmark book" (Jon
Katz, Hotwired ) and a "Poetics for Cyberspace" (Peter Petre,
Fortune). It is based in part on her course in Interactive Narrative,
which has been offered at MIT since 1992. She lectures and consults
widely on interactive design and applications of new media.
Dr. Murray has been at MIT since 1971, dividing her interests between
the literature of the nineteenth and the twenty-first centuries. She
was born in New York City, and holds a Ph.D. in English Literature
from Harvard University.
____________
XEROX PARC FORUM
on Thursday, 19 February 1998, 4:00pm - 5:00pm
George Pake Auditorium, Xerox
http://www.parc.xerox.com/ops/projects/forum/
Design and Modeling in the Fourth Dimension
Martin Fischer
Stanford University, Assistant Professor
Department of Civil and Environmental Engineering
http://www-cive.stanford.edu/fischer/
http://gaudi.stanford.edu/4D-CAD/INTRO-4DCAD.HTML
Four-dimensional CAD modeling (3D plus time) has the potential to
completely change the way facilities are designed and built.
Combining the 3D perspective of the designer with the temporal view of
the builder, the technology is used to represent the information
required in the life cycle of engineered facilities. The models
generate realistic visualizations of the facility design and its
changes over time, and allow computer-based analysis of
constructibility, cost, productivity, and other project performance
variables dependent on an integrated analysis of time and space. Early
test cases have amply demonstrated that 4D models can help to enhance
schedule, cost, quality and safety, with potential benefits in the
billions of dollars annually for the global construction sector alone.
Unfortunately, the difficulty and cost of creating and using such
models is currently blocking their widespread adoption. Full 4D
models are very time-consuming to generate manually and cannot
currently share their representations with analysis programs. To
support analysis, 4D models must represent time explicitly and cannot
consist of a simple sequence of 3D model views. Since each
construction project is a unique combination of thousands of
components, relatively little theoretical and computational support
exists for concurrent engineering and 4D modeling.
Professor Fischer's research addresses these issues by formalizing a
general, computer-interpretable construction planning vocabulary and
by defining and using customizable software agents that translate a 3D
model into a 4D model. Using the vocabulary, the agents transfer
construction knowledge to activities in the project schedule so that
the activities know when to create themselves, how to calculate their
durations, how to relate themselves to other activities and design
components, and how to get the resources they need. Professor Fischer
plans to discuss his vision of how this research can unlock the
potential of 4D modeling to enhance the facility development process.
He also hopes to generate feedback, not only about his software design
research, but about the potential for 4D CAD to impact product design
and manufacturing.
Biography: Martin Fischer is an assistant professor in the
Construction Engineering and Management Program in the Department of
Civil and Environmental Engineering at Stanford University. Prior to
joining the faculty at Stanford University in 1991, he worked in
engineering and construction firms in Switzerland, the U.S., and
Japan. He graduated, in 1983, from the Swiss Federal Institute of
Technology in Lausanne, Switzerland with a degree in Civil
Engineering. He completed his graduate studies at Stanford
University, receiving an MS in Industrial Engineering (emphasis on
Engineering Management) in 1987 and a Ph.D. in Civil Engineering
(emphasis on Construction Engineering and Management) in 1991.
____________
CSLI TALK
on Thursday, 19 February 1998, 4:00pm
Cordura Hall, Room 100
http://www-csli.stanford.edu/csli/Coglunch/
Amalia -- Abstract MAchine for LInguistic Applications
Shuly Wintner
Tuebingen
Contemporary linguistic formalisms such as LFG or HPSG have become so
rigorous that it is now possible to view them as very high level
declarative programming languages. Consequently, grammars for natural
languages can be viewed as programs; this view enables the application
of various methods and techniques that were proved useful for
programming languages to the domain of natural languages. This work
introduces such an application: an implementation technique that is
common for logic programming languages, namely the use of an abstract
machine, is applied to (a subset of) the ALE formalism, originally
designed for specifying feature-structure based phrase-structure
grammars.
We present Amalia (Abstract MAchine for LInguistic Applications), an
abstract machine specifically tailored for processing ALE grammars. It
is composed of data structures and a set of instructions, augmented by
a compiler from the grammatical formalism to the abstract
instructions, and a (portable) interpreter of the abstract
instructions. The effect of each instruction is defined using a
low-level language that can be executed on ordinary hardware.
Execution of the compiled code amounts to parsing with respect to the
original grammar. A variant of the compiler produces code for
generation.
The advantages of the abstract machine approach are twofold. From a
theoretical point of view, the abstract machine gives a well-defined
operational semantics to the grammatical formalism. From a practical
point of view, Amalia is the first system that employs a direct
compilation scheme for unification grammars that are based on typed
feature structures. The use of Amalia results in a much improved
performance over existing systems.
____________
SEMINAR ON COMPUTATIONAL LEARNING AND ADAPTATION (SCLA)
on Thursday, 19 February 1998, 4:15pm to 5:30pm
Gates 100
http://www-csli.stanford.edu/cll/scla.html
Massive Time-Series Data Mining via Envelope Learning and Monitoring
Dennis DeCoste
Jet Propulsion Laboratory/Caltech
Pasadena, CA
mailto:decoste@aig.jpl.nasa.gov
Automated detection of anomalies in complex real-world time-series
data is typically performed using a combination of knowledge
engineering approaches that suffer from frequent false alarms and/or
inability to detect anomalies before they lead to critical system
failures. The domains of interest often involve thousands of sensors
over millions of time samples, making it very difficult to overcome
the limitations of such knowledge intensive approaches using
straightforward application of standard regression techniques (e.g.
neural networks with error bars). To address such concerns, I
developed the ELMER (Envelope Learning and Monitoring via Error
Relaxation) system. ELMER automatically learns high and low limit
functions ("envelopes") from data. Toward reducing operations costs
while improving reliability and automation, ELMER focuses on learning
envelopes which maintain low false alarm rates. It does so in an
anytime fashion, beginning with relatively wide limits, such as
constant red-lines which are not context-sensitive. As promising
context-defining sensors are identified during learning, it forms
tighter and more input-conditional envelopes. ELMER is being evaluated
for a variety of NASA domains --- including both ground and onboard
operations and both earth-orbiting and deep space missions. In this
talk, I will present examples and discuss some of the technical
advances behind the ELMER work, which includes work on asymmetric
regression cost functions, learning Bayes nets, and feature
selection/construction.
____________
STANFORD PHONOLOGY WORKSHOP
on Thursday, 19 February 1998, 7:30pm
Margaret Jacks Hall 460:146
http://www-linguistics.stanford.edu/Linguistics/pinterest/
Complex clusters:
New prosodic domain - Stem
(Case study Modern Georgian)
Marika Butskhrikidze
Leiden University
Georgian as well as many other languages (e.g., Berber, Bella Coola,
Russian, Polish, etc.) having complex onsets pose a "problem" for the
current theories of syllable structure. Two universal phonological
principles that have been proposed: the Sonority Sequencing Principle
and Prosodic Licensing seem to be violated in these languages.
Two ways might be offered to resolve the problem: (i) To reconsider
the analysis of the language's phonotactic system or (ii) To
reconsider the universal principles themselves. Most of the studies
(at least on Georgian syllable structure) have been done proposing
different analyses of complex onsets, e.g. Headless Syllables (Nepveu
1994); Semi-Syllables (Cho & King 1996); Syllabified Consonants (Bush
1997), Syllabic consonants.
We are proposing a different type of approach for Georgian and, in
general, for the languages having complex clusters not obeying the
universal phonological rules. In our study we do not need to
reconsider neither the phonotactics of languages, nor principles, but
simply look carefully at the morphological constituency of words.
The most salient characteristic of complex clusters is their
distribution (even though it was noticed in almost all the studies
mentioned, it has not been properly appreciated) and hence it is the
crucial in our approach. When analyzing Georgian data it is obviously
seen that the complex clusters not obeying the SSP only occur in the
word initial position. The initial syllable of the word coincides with
the stem initial position. Thus the domain of the distribution of
clusters is the stem. The domain of the Sonority Sequencing Principle
is the syllable (we argue that word initial is not the same as
syllable initial and word final is not the same as syllable final). We
adopt the definition of the syllable as "the smallest unit of
recurrent sequence". Thus, since domains do not match, there is no
problem with consonant clusters. Still there is another universal
phonological principle: Prosodic Licensing. How can we handle the
violation of this principle?
My previous studies (Butskhrikidze 1994, 1995, 1996,1997) examined the
correlations between: 1. Morphological structure of a word and
placement of fixed accent; 2. Placement of fixed accent and final
devoicing; 3. Placement of fixed accent and the distribution of
consonant clusters; 4. Word morphological structure and distribution
of consonant clusters.
The last correlation seems relevant for the discussion.
Hypothesis 1: If a language has a basic word form coinciding
with the stem and the fixed accent on the final syllable,
obstruent clusters are likely to occur word finally rather
than word initially.
Hypothesis 2: If a language has a basic word form: stem+suffix
and the fixed accent is not on the final syllable, obstruent
clusters are more likely to occur word initially rather than
word finally.
The hypotheses have been tested in 50 fixed accent languages from 11
language families.
The languages are divided into two groups:
1. Languages having a word form: stem + affix and a fixed accent not on
the final syllable. The accent falls either on the first syllable
of the word, or on the penultimate or the placement is restricted
to the second and third syllables from the end).
2. Languages having word form coinciding with stem and fixed accent on
the final position.
Type 1: Kartvelian (Georgian, Megrelian, Laz); Germanic (Jiddish,
German, Icelandic); Jeniseyan (Ket); Finno-Ugric (Finish,
Karelian, Livonian, Saam); Baltic (Estonian); Slavic (Polish);
Indic (Sanskrit); Iranian (Old Persian).
Type 2: Iranian (Persian, Beludchi, Tadzhic, Kurdish, Jazghulam,
Ishkashim); Lezgian (Lezg); Manchu-Tungus (Nanay, Itelmen,
Oroch) Armenian (Grabar, Modern Armenian); Turkic (Turkish,
Karaim, Turkmen, Uzbek, Gagauz, Kumukh, Tuva, Khakas);
Finno-Ugric (Udmurt), Romance (French).
In type 1 languages obstruent clusters are likely to occur word
finally and in type 2 languages -- word initially.
To summarize, from the above established relations what is relevant
for our present discussion about complex clusters is that clusters are
related to the stem and word accent. If the stem (namely the
distribution of stem in a word) is the domain for phonological rules,
(i.e. it determines: (i) The word prosodic structure; (ii)
Distribution of consonant clusters in a word; (iii) Final devoicing)
it seems that we can freely introduce a stem level in to the Prosodic
Hierarchy. If a unit (in our case stem) functions as a domain for
phonological rules, then it is a constituent in phonological
representation (Inkelas 1989), 40). Thus the revised version of the
Prosodic Hierarchy will be:
Prosodic Hierarchy:
Prosodic Word
STEM
Foot
Syllable
Now after introducing new prosodic domain in the Prosodic Hierarchy it
seems we can account the "problem" of Georgian complex clusters.
First of all the clusters do not pose problem for Prosodic Licensing,
since distribution of complex clusters is restricted to word initial
(stem initial position). If we recognize the prosodic domain stem,
then to account for the appearance of the clusters in the stem initial
position is easy. After revision of the notion syllable as a smallest
unit of the recurrent sequence, complex clusters do not create problem
for the SSP. The SSP is defined at the domain of the syllable and
complex clusters not obeying the SSP appear in the domain of stem.
Thus the violation is not surprising since the domains do not match.
Bella Coola, Russian, and Berber follow the same pattern: clusters not
obeying the SSP are restricted to the word initial (stem initial)
position. The clusters should not be considered as constituents of
syllables but rather of the stems. The evidence for this is
reduplication. When reduplication applies these clusters undergo
changes: either one of the consonant deletes, or schwa appears between
consonants, etc. Thus the domain of "odd" clusters is the prosodic
domain the Stem and not the Syllable and therefore neither Prosodic
Licensing nor the SSP is violated. Prosodic Licensing is not violated
because we are still dealing with the prosodic domain Stem which
licenses the occurrence of clusters, and the SSP is not violated
because the domain of the SSP is the syllable and domain of "odd"
cluster distribution is Stem.
____________
LOGIC LUNCH
on Friday, 20 February 1998, 12 noon
Math Corner 380:383N
http://math.stanford.edu/upcoming/upcoming.html
On the Classical and Parametric Complexity
of the Ehrenfeucht-Fraisse Game
Elena Pezzoli, Stanford
In the first part of the talk we present some results concerning the
complexity of the EF problem, that is the problem of determining the
winner of the r-moves Ehrenfeucht-Fraisse game on structures A and B
over signature Sigma: we show that the EF problem is hard for NP, and
that the one-sided EFP problem (where the spoiler can choose from
structure A only) and, for fixed k, the k-alternations EF problem
(where the spoiler can alternate at most k times; the spoiler makes an
alternation in some round of the game, if he chooses from a different
structure than in the previous round) are PSPACE complete. In the
second part of the talk we introduce the theory of fixed parameter
tractability of Downey and Fellows and we place the EF problem in the
parametric hierarchy.
____________
SEMINAR ON PEOPLE, COMPUTER, AND DESIGN
on Friday, 20 February 1998, 12:30-2:00pm
Gates B01 (HP Classroom)
http://www-pcd.stanford.edu/seminar/
(SITN Channel E2)
The Eyes Have It:
User Interfaces for Information Visualization
Ben Shneiderman
Human-Computer Interaction Laboratory, University of Maryland
mailto:ben@cs.umd.edu
Human perceptual skills are remarkable, but largely underutilized by
current graphical user interfaces. The next generation of animated
GUIs and visual data mining tools can provide users with remarkable
capabilities if designers follow the Visual Information-Seeking
Mantra:
Overview first, zoom and filter, then details-on-demand. But this is
only a starting point in the path to understanding the rich set of
information visualizations that have been proposed. Two other
landmarks are:
* Direct manipulation: visual representation of the objects and
actions of interest and rapid, incremental, and reversible
operations
* Dynamic queries: user controlled query widgets, such as sliders
and buttons, that update the result set within 100msec.and are
shown in the FilmFinder, Visible Human Explorer (for National
Library of Medicine's anatomical data), NASA EOSDIS (for
environmental data), and LifeLines (for medical records and
personal histories).
As a guide to research, information visualizations can be categorized
into 7 datatypes (1-, 2-, 3-dimensional data, temporal and
multi-dimensional data, and tree and network data) and 7 tasks
(overview, zoom, filter, details-on-demand, relate, history, and
extract). Research directions include algorithms for rapid display
update with millions of data points, strategies to explore vast
multi-dimensional spaces of linked data, and design of advanced user
controls.
Biography: Ben Shneiderman is a Professor in the Department of
Computer Science, Head of the Human-Computer Interaction Laboratory,
and Member of the Institute for Systems Research, all at the
University of Maryland at College Park. Dr. Shneiderman is the author
of Software Psychology: Human Factors in Computer and Information
Systems (1980) and Designing the User Interface: Strategies for
Effective Human-Computer Interaction (1987, second edition 1992, third
edition 1997), Addison-Wesley Publishers, Reading, MA.
Dr. Shneiderman has co-authored two textbooks, edited three technical
books, and published more than 180 technical papers and book chapters.
His 1993 edited book Sparks of Innovation in Human-Computer
Interaction collects 25 papers from ten years of research at the
University of Maryland. This collection includes Dr. Shneiderman's
seminal paper on direct manipulation, a term he coined in 1981 to
describe the graphical user interface design principles: visual
presentation of objects and actions combined with pointing techniques
to accomplish rapid incremental and reversible operations. Ben
Shneiderman received his BS from City College of New York in 1968, his
PhD from State University of New York at Stony Brook in 1973. He
received an Honorary Doctorate of Science from the University of
Guelph, Ontario, Canada in 1996 and was elected as a Fellow of the
Association for Computing (ACM) in 1997.
____________
STANFORD DIGITAL LIBRARIES SEMINAR
on Monday, 23 February 1998, 4:30pm
Gates Building, B08
http://diglib.stanford.edu/diglib/seminars/seminars.html
Bayou: Supporting Asynchronous Collaboration through Replicated Data
Karin Petersen
Xerox PARC Computer Science Laboratory
The Bayou system was designed to support collaboration among users who
cannot be or choose not to be continuously connected. Network
connections may at times be too slow, too expensive, or too faulty for
users to effectively utilize, or it may not be possible to establish a
network connection at all. This can occur, for example, when
collaborators are widely dispersed across the Internet or work on
portable computers. In such a setting, weak consistency replication of
shared data is the key to obtaining high data availability, good
access performance, and good scalability.
Bayou uses weak consistency replication techniques to manage replicas
of shared calendars, electronic mail messages, databases, documents,
and other artifacts that are central to collaboration. To maximize
availability, users can read and write any accessible replica. Bayou's
design has focused on supporting application-specific mechanisms to
detect and resolve the update conflicts that naturally arise in such a
system, ensuring that replicas move towards eventual consistency, and
defining a protocol by which the resolution of update conflicts
stabilizes. It includes novel methods for application-specific
conflict detection and per-write conflict resolution based on
client-provided merge procedures.
In this talk I will present the Bayou system architecture, the
important design decisions and new technology developed for this
system, and some applications built on top of Bayou.
Biography: Karin Petersen is the Manager of the Networked Document
Systems Area in the Computer Science Laboratory at Xerox PARC's. She
has been with PARC since September of 1993. Prior to joining PARC,
Karin received an Engineering degree in Computer Science from the
Simon Bolivar University (Venezuela) in 1988, and a Ph.D. in Computer
Science from Princeton University in 1993. Her research focuses on
distributed and parallel systems, performance evaluation, and user
interfaces for mobile and distributed applications.
____________
STANFORD PHONOLOGY WORKSHOP
on Thursday, 26 February 1998, 7:30pm
Margaret Jacks Hall 460:146
http://www-linguistics.stanford.edu/Linguistics/pinterest/
Some Arguments Against
Using Underlying Representations in OT Phonology
Edward Flemming
Department of Linguistics
Stanford University
In standard Optimality Theoretic phonology, faithfulness to underlying
representations (URs) serves two functions:
i. Requiring contrasts - 'Richness of the base' mandates many
different inputs, and faithfulness requires that these differences
be maintained in the output, yielding surface contrasts.
ii. Constraining alternations - A morpheme is posited to have a unique
UR, and faithfulness requires surface alternants of this morpheme
to be similar to this UR, thus indirectly constraining the extent
of allomorphic variation.
I will present evidence that:
a. These functions need to be distinguished, not conflated -
constraints on alternations are different from constraints on
contrasts.
b. Faithfulness to UR cannot fulfill either of these functions
adequately.
c. Satisfactory accounts of contrast and alternation employ
constraints which are evaluated exclusively on surface
representations (constraints requiring surface contrasts, and
'output-output' correspondence), and thus leave no role for URs.
____________
SEMINAR ON PEOPLE, COMPUTER, AND DESIGN
on Friday, 27 February 1998, 12:30-2:00pm
Gates B01 (HP Classroom)
http://www-pcd.stanford.edu/seminar/
(SITN Channel E2)
AHA: Audio HTML Access
Frankie James
Stanford Computer Science and CSLI
mailto:fjames@cs.stanford.edu
Every day, more and more information is being made available online to
the general public in the form of electronic documents. Since the
advent of the WWW, hypertext (in particular, HTML) has become the
medium of choice for the presentation of these documents. This is
because HTML can be used to present not only the text of a document,
but also much of its structure. The ability to use this structure in
a generic (multi-modal) way would mean that electronic documents could
be accessible to everyone, even non-standard users such as blind users
or users connecting to the WWW via the telephone.
In this talk, I will present my research into the development of audio
interfaces to HTML that can allow blind users (and others) to access
the WWW non-visually. In particular, I will discuss the AHA (or Audio
HTML Access) framework that I developed for choosing the types of
audio markings to use in an HTML interface. This framework, which was
based on user testing, can be used in conjunction with other
information about the users and their tasks to allow interface
designers to select specific sounds for the presentation of HTML in
audio.
Biography: Frankie James is completing her Ph.D. in Computer Science
at Stanford University. Her dissertation is on auditory interfaces to
HTML for blind computer users and others who might need to access the
WWW nonvisually, such as PDA users or people driving cars. She has
developed the AHA framework as a means for choosing sounds to use in
such interfaces, based on empirical evidence gathered from user
testing. Frankie has also worked for the Archimedes Project at
Stanford's Center for the Study of Language and Information (CSLI),
whose goal is to promote equal access to computer technology for
individuals with disabilities.
____________
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
mailto:incalendar@csli.stanford.edu.
Requests to be added to the mailing list should be sent to
mailto:majordomo@csli.stanford.edu. With the lines in the body of
the text of either
subscribe csli-calendar
for the long form or
subscribe csli-short-calendar
for the short form. Problems with subscribing or unsubscribing should
be sent to owner-csli-calendar@csli.stanford.edu.
The full current issue is at
http://www-csli.stanford.edu/csli/Archive/calendar/1997-98/current.html
and the archives at
http://www-csli.stanford.edu/csli/Archive/calendar/.
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,
http://www-linguistics.stanford.edu/Linguistics/.
The CSLI Calendar is also posted each week to
news://nntp-csli.stanford.edu/csli.bboard.
Information about CSLI's research program is available at
http://www-csli.stanford.edu/csli/.
For maps to the Stanford University campus see
http://www.stanford.edu/home/visitors/maps.html.
____________