[Prev][Next][Index][Thread]
CSLI Calendar, 3 October 1996, vol. 12:2
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
______________________________________________________________________________
3 October 1996 Stanford Vol. 12, No. 2
______________________________________________________________________________
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 3 OCTOBER -- 11 OCTOBER 1996
THURSDAY, 3 OCTOBER
4:15 - Seminar on Computational Learning and Adaptation
Cordura 100
Self-Customizing Applications: Principles and Case Studies
Jeff Schlimmer, Washington State University
(abstract below)
4:15 - Symbolic Systems Forum
Bldg. 460:146 (Margaret Jacks Hall)
Symbolic Systems Summer Internship Presentations
(abstract below)
FRIDAY, 4 OCTOBER
12:30 - PCD Seminar
Gates Hall B01 (HP Classroom)
Ensuring the Usability of the Next Computing Paradigm
Jakob Nielsen, SUN Microsystems
(abstract below)
3:15 - Philosophy Department Colloquium
Bldg. 100:101K
Why Be Humean?
Tim Maudlin, Rutgers University
3:30 - Linguistic Department Colloquium
Bldg. 460:146 (Margaret Jacks Hall)
There's more than one way to talk about motion:
Consequences of linguistic typology for narrative style
Dan Slobin, UC Berkeley
(abstract below)
TUESDAY, 8 OCTOBER
4:15 - Logic Seminar
Bldg. 380:381T (Math corner)
The problem of self-applicable structural concepts (cont'd.)
Sol Feferman, Stanford
THURSDAY, 10 OCTOBER
12:00 - CSLI CogLunch
Cordura 100
Title to be announced
4:15 - Symbolic Systems Forum
Bldg. 460:146 (Margaret Jacks Hall)
Topic to be announced
FRIDAY, 11 OCTOBER
10:00 - Philosophy Department Seminar (Philosophy 288)
Cordura Hall, Room 104
Platonism, Formal Ontology, and Naturalism
Edward N. Zalta, Stanford CSLI and Philosophy
[http://mally.stanford.edu/syllabi/platonism-naturalism.html]
12:30 - PCD Seminar
Gates Hall B01 (HP Classroom)
LearningWorks
Adele Goldberg, Neometron
(abstract below)
3:15 - Philosophy Department Colloquium
Bldg. 100:101K
Reasons, Relationships and Special Responsibilities
Sam Sheffler, UC Berkeley
3:30 - Linguistic Department Colloquium
Bldg. 460:146 (Margaret Jacks Hall)
y/X Alternation and the Phonology-Morphology Interface
in Luganda
Larry Hyman, UC Berkeley
(abstract below)
____________
The CSLI Calendar appears weekly on Wednesdays throughout the academic
year. Announcements, abstracts, and other information to appear in
the Calendar can be submitted to
[mailto:incalendar@csli.stanford.edu].
Requests to be added to the mailing list should be sent to
[mailto:csli-request@csli.stanford.edu]
Information about CSLI's research program and past issues of the CSLI
Calendar are available at [http://www-csli.stanford.edu/csli/]. The
CSLI Calendar is also posted each week to
[news://nntp-csli.stanford.edu/csli.bboard].
CORRECTIONS: This week's Linguistics Colloquium was wrongly listed
last week as taking place on Thursday. The correct day is Friday.
Also the PCD Seminar will take place in Gates Hall room B01 (HP
Classroom) instead of Skilling Auditorium as announced earlier.
____________
CSLI COGLUNCH
As in 1995-96, the CogLunch series will continue this academic year,
Thursdays 12:00-1:30 at Cordura Hall 100. The theme for the Fall
quarter, and probably for the rest of the year, will again be
"interdisciplinary approaches to consciousness". And, as usual, we
will continue to provide sandwiches and drinks at the door for a small
fee.
The first CogLunch talk will take place on Thursday, October 9, 1996.
Please monitor the CSLI calendar for details. A complete schedule of
the Fall Quarter will also be posted there as soon as it is complete.
____________
SEMINAR ON COMPUTATIONAL LEARNING AND ADAPTATION
Self-Customizing Applications: Principles and Case Studies
Jeff Schlimmer, Washington State University
On Thursday, 3 October
4:15 pm, Cordura Hall, Room 100
Cost-effective software applications are governed by a pair of
conflicting principles. First, more flexibility allows more tasks to
be addressed, increasing its potential market. Second, more
flexibility requires more end-user configuration, reducing its
potential market. Machine learning technology offers a way to reduce
the required end-user customization, and in doing so, effectively
increase the market for a particular application. A supplemental
piece of learning code watches over the shoulder of the user and
configures the base application. This results in a self-customizing
application. Unlike earlier efforts in automatic programming, the
learning task is tractable because the application addresses a limited
range of tasks. After discussing this general framework, I will
present vignettes of several self-customization applications.
Reflecting on these examples, I will describe design principles for
successful self-customizing applications and point out possibilities
for future research.
____________
SYMBOLIC SYSTEMS FORUM
Symbolic Systems Summer Internship Presentations
On Thursday, October 3
4:15 p.m., Bldg. 460:146 (Margaret Jacks Hall)
Anne Bracy, Erik Flister, Matthew Kodama, Marissa Mayer, and David
Soergel who did work this summer through the SSP Internship Program
will discuss their experiences.
Refreshments will be served.
____________
PEOPLE, COMPUTERS, AND DESIGN SEMINAR
Ensuring the Usability of the Next Computing Paradigm
Jakob Nielsen, SUN Microsystems
[mailto:jakob@eng.sun.com]
On Friday, 4 October
12:30-2:00 pm, Gates Hall B01 (HP Classroom)
The very definition of the concept of "a computer" is changing as is
the source of much of the user experience. User interfaces defined by
traditional software companies will be frameworks for dynamic content
provided by (soon) millions of servers around the world. The battle of
Triceratops (Netscape Navigator) vs. Tyrannosaurus (Internet Explorer)
will be overshadowed by the battle of the mice (Websites) vs. the
elephants (television networks). Unfortunately, this new world is in
dire danger of a usability meltdown: the Net may be brought to its
knees, not by the bandwidth collapse forseen by most commentators, but
by user revolt.
For a complete list of PCD Seminars see
[http://www-pcd.stanford.edu/hci/courses/cs547.html]
____________
LINGUISTICS DEPARTMENT COLLOQUIUM
There's more than one way to talk about motion:
Consequences of linguistic typology for narrative style
Dan Slobin, UC Berkeley
on Friday, 4 October
3:30 p.m., Bldg. 460:146 (Margaret Jacks Hall)
A language can lexicalize the PATH of motion in a verb stem (e.g.,
enter), or in an associated element, such as a verb particle (e.g., go
in). Following Talmy, the first type is referred to as "verb-framed"
and the second as "satellite-framed". In satellite-framed languages,
MANNER can be expressed in the main verb (e.g., run in), while in the
first type, MANNER is expressed in an associated form (e.g., enter
running). A complex path can be expressed by a sequence of verbs in
the first type (e.g., exit the house, cross the field, and enter the
forest) and by a sequence of particles and prepositional phrases in
the second (e.g., go out of the house, across the field, and into the
forest). This collection of typological contrasts appear to have
consequences for the ways in which motion events are described in
narrative. Using crosslinguistic data of picture-elicited narratives,
novels, and translations of novels, it is found that narrative texts
in satellite-framed languages devote more attention to MANNER of
movement and to the physical GROUND elements associated with movement
(source, goal, milestone, medium). By contrast, verb-framed
narratives may devote more attention to SCENE-SETTING (descriptions of
layout, terrain, inner state). The colloquium will present work in
progress, suggesting that lexicalization patterns can influence
attention to aspects of motion events, leading to different types of
narrative style.
------------------
Reception follows.
For directions and a complete list of colloquia, see
[http://www-linguistics.stanford.edu/Linguistics/colloq/colloq.html]
____________
PEOPLE, COMPUTERS, AND DESIGN SEMINAR
The Design of LearningWorks
Adele Goldberg, Neometron
[mailto:adele@neometron.com]
On Friday, 11 October
12:30-2:00 pm, Gates Hall B01 (HP Classroom)
(SITN Channel E4)
Jobs in commercial software development are team jobs. The teams build
systems that bring value to their customers. New hires, whether fresh
out of school or transferring from other commercial projects, are
successful only if they can contribute to the team's goals. Yet the
formal education of these new hires emphasized programming, not system
building, and rarely offered team experience in creating and
maintaining large systems. What might a curriculum focused on teaching
system building contain?
Faced with that question, I set out to design a new system that would
support authoring and delivering such a curriculum. The result is
LearningWorks, available for free on the Internet. The purpose of this
talk is to discuss the user interface design for LearningWorks,
specifically how the simple metaphor of learning books can be used
both to teach software system building concepts and to support teams
in building systems.
For a complete list of PCD Seminars see
[http://www-pcd.stanford.edu/hci/courses/cs547.html]
____________
LINGUISTICS DEPARTMENT COLLOQUIUM
y/X Alternation and the Phonology-Morphology Interface in Luganda
Larry M. Hyman
University of California, Berkeley
on Friday, 11 October
3:30 p.m., Bldg. 460:146 (Margaret Jacks Hall)
In this talk I address the issue of how phonology and morphology
interact in a highly agglutinative language (lu-Ganda) and language
family (Bantu). The issue is whether the phonological facts support an
"interactionist" view of this interface, with progressive
morphology-phonology interleaving, i.e., add an affix, do some
phonology, add another affix, repeat the phonology. Attention will be
on verb structure, which, as shown below, may be quite complex:
(i) Word = prefixes + stem
(ii) Stem = ROOT + suffixes
(iii) Example (stem in brackets, root capitalized):
a-ba-ta-li- [ FUMB-ir-agan-a ] 'they who will not cook for
IV-they-neg-fut cook-appl-e.o.-FV each other'
(IV = initial vowel morpheme; FV = final vowel morpheme)
In previous work on lu-Ganda (Hyman and Katamba 1991) and ci-Bemba
(Hyman 1995), we have shown that the stem-level ("stratum 1") suffixal
morphology and phonology are cyclic, a demonstration that can be
extended to many other Bantu languages. In this paper I shift
attention to the nature of the morphology-phonology interface in
prefixation (hence at the word level, "stratum 2"). The question in
this paper, then, is whether the "slot-and-filler"-type prefixes show
the same kind of cyclic effects as suffixes or whether the prefixal
phonology is non-cyclic, as generally assumed. I will show that here
too there are cyclic effects, though of quite a different type.
Complex facts from root-initial y/X alternation establish not only
the unimportance of "slots" in prefix morphology-phonology
interactions, but also that these, although cyclic in a sense to be
discussed, must not be handled in a "compositional" manner.
Specifically, stratum 1 (stem) and stratum 2 (word) intersensitivities
argue for an integrated interface, where all requirements are checked
at the same time. This result is of course consistent with the move
away from procedural derivationality and towards direct-mapping
approaches to grammar.
References
* Hyman, Larry M. 1995. "Cyclic phonology and morphology in
Cibemba". In J. Cole and Charles Kisseberth (eds.), Perspectives
in Phonology, 81-112. Stanford: C.S.L.I.
* Hyman, Larry M. and Francis X. Katamba. 1993. "Cyclicity and
suffix doubling in the Bantu verb stem". In Special Session on
African Language Structures, 134-144. Berkeley Linguistic Society
17.
___________
CSLI VISITORS
Sei Hangai
Science University of Tokyo
Visiting the CSLI ASTL Lab
(Academic Year 1996-97)
Wen-Lian Hsu
Institute of Information Science, Academia Sinica,
Taipei, Taiwan,
Visiting the CSLI ASTL Lab
(Academic Year 1996-97)
Yasuhiko Kato
The National Language Research Institute,
Tokyo, Japan
(Academic Year 1996-97)
Godehard Link
University of Munchen, Linguistics
(August-October '96)
Louisa Sadler
University of Essex, UK, Linguistics
(till Dec. '96)