Home
Meetings
Seminars
The Real Times
What is GBC/ACM?
This page was last updated on
Sunday, 26-May-2002 17:36:29 PDT
Past GBC/ACM Meetings for 1997-1998
Bob Balaban, Looseleaf Software
Thursday, June 18, 1998
Note: GBC/ACM Officer Elections were held
at this meeting.
Click here for candidate
information.
Special Double May Meeting:
Marc Cooper, Industrial Light and Magic
Thursday, May 14, 1998 and Friday, May 15, 1998
Bartos Theater, MIT Media Lab, Cambridge
Note: The May meetings were joint Meetings with
SIGGRAPH/Boston and Boston Section IEEE-CS
James Glass, Spoken
Language Systems Group, Laboratory
for Computer Science, MIT
Thursday, April 23, 1998, 6:30 p.m.
MIT Laboratory for Computer Science, Room 518, 545 Technology Square,
Cambridge
V. Michael Bove, Jr.,
MIT
Media Laboratory
Thursday, February 19, 1998
Azer Bestravos, Computer
Science Department, Boston University
Thursday, January 15, 1998
Walter Bender,
MIT
Media Laboratory
Thursday, December 11, 1997 (Note: 2nd Thursday)
Andrew Merlino, The MITRE Corporation
Thursday, November 20, 1997
Joint meeting with IEEE Computer Society, Boston Section
Note: The meeting was held at MITRE in Bedford, MA. Click
for directions.
Ennio Mingolla, Boston University
Thursday, October 23, 1997
Michael Steinmann, Tessera Enterprise Systems
Thursday, September 18, 1997
Meeting Details
June, 1998 Meeting
Subject
Groupware Programmability and the Internet: Domino and Java
Speaker
Bob Balaban
Looseleaf Software
Date
Thursday, June 18, 1998.
Time
Presentation starts at 7:00 PM. Refreshments at 6:30 PM.
Location
BBN, Cambridge. See below for directions.
Meeting Overview
Lotus Domino has since 1990 been widely recognized as the best and most
popular groupware applications development platform. Its architecture combines
platform independence, networking, messaging, object storage and forms
development technologies in a client/server framework. Then came the Internet.
In 1995 and 1996 the buzz in the industry was that Notes/Domino was dead,
roadkill on the information superhighway. But that didn’t happen. In 1998
Domino is more popular than ever. One of the factors in the “reinvention”
of Domino as an Internet/intranet development tool has been its embrace
of “open standards”, both in the communications area and in terms of programmability.
The ability to write programs that make use of the robust service layers
in Notes/Domino and make the product do precisely what you want it to in
an automated fashion is one of its great strengths.
Beyond an overview of what Domino is and does, this talk surveys the
Domino programmability landscape, with particular attention to the recently
developed Java APIs. Time, interest and equipment permitting, there will
be a demo or two, and we can even look at some code.
Speaker Biography
Bob Balaban worked as a developer at Lotus Development Corp and at Lotus
subsidiary Iris Associates for over ten years (over four years as an engineer
working on Domino). He left Lotus/Iris last fall to start Looseleaf Software,
Inc., a software consulting and contract development firm. Among other
accomplishments at Lotus, Bob was the principal architect of the agent
subsystem in Domino, and was the author of both the LotusScript and Java
programming interfaces. He lives in Lexington, Mass., and has been a GBC/ACM
member since 1981.
1998-1999 Chapter Officer Candidates
President
James Ganino is the current Vice-President. He has been active with the
GBC’s PDS Committee for the last six years as the financial planner for
the committee and recently as a session chair for two seminars. James works
at TASC, in Reading, MA, as a member of the technical staff both as a software
engineer and as systems programmer supporting a variety of systems development,
distributed simulations, and networking projects.
Vice-President
Jim Byrd has been a volunteer with the PDS committee for four years. Jim
is a programmer, working for Atex Media Solutions, in Bedford.
Treasurer
Stephanie Collins is the current Chapter Treasurer. Prior to becoming the
Treasurer, she was a member of the GBC PDS Committee for many years as
analyst for our seminar feedback forms. Until recently, Stephanie was a
Professor at Northeastern University in the Business School and is now
a Professor at the Graduate School of New Hampshire College. She is also
an independent consultant and developer of distributed client/server applications.
Secretary
Ed Bristol is our incumbent Secretary and a long time IEEE volunteer and
leader. Ed is well known for his work and writing on Process Control Systems
with some ninety papers published and numerous professional awards. Currently
Ed is a Fellow at The Foxboro Co.
Election of Officers took place at the June 18,
1998 GBC/ACM Meeting.
May 1998 Meetings
Subject
May 14: Pin-Face -- How ILM Produces an Animated Commercial
May 15: A Day in the Life at Industrial Light and Magic or How
Did They Do That Visual Effect...
Speaker
Marc Cooper, Industrial Light and Magic
Date
Thursday, May 14, 1998 and Friday, May 15, 1998
Time
Networking time at 6:30pm, announcements and feature presentation at 7:00pm.
Location
Bartos Theater, MIT Media Lab, Cambridge. See below for directions.
Meetings Overview
GBC/ACM, SIGGRAPH/Boston and the Boston Chapter of IEEE-CS are pleased
to present a speaker from Industrial Light and Magic. ILM is the award-winning
(multiple Oscars and Cleos, that is) company owned by George Lucas that
is famous for all the visual effects we see in the movies such as the Star
Wars series.
May 14
We all have seen those pinboard toys where you can press your hand or your
face into the back and leave an "pin-pression" of it on the front. But
how do you simulate that functionality using computer graphics? Marc Cooper,
a technical director at ILM will explain the various technical and artistic
challenges encountered in making over eight thousand virtual pins perform
for a commercial called "CPA Pin-Face" for the Certified Public Accountants
Association. In addition he will review some of his work from the
last Star Trek movie, Star Trek: First Contact.
May 15
Marc will present an overview about ILM and what it takes to produce animation
and visual effects for the movies and TV. He will describe how several
visual effects were produced from such films as Lost World, Dragonheart
and Star Trek. Come to the meeting to see what a day is really like
at ILM. (It might not be a much fun as we all think...)
Speaker Biography
Marc Cooper is a Technical Director at Industrial Light and Magic of
Marin County, California. He has worked on numerous feature films including
Casper,
Dragonheart, and Flubber. He has just completed work on Dreamworks'
next film: Deep Impact.
Marc has been at ILM for 4 years. Prior to working at ILM, he worked
as a research programmer and systems administrator at the NASA Lewis Research
Center in Cleveland, Ohio and as a visualization programmer for the National
Center for Supercomputer Applications (NCSA) at the University of Illinois.
He has a degree in Computer Science from the University of Illinois.
Directions to Bartos Theater, MIT Media Lab
From the West
From the Massachusetts Turnpike: Exit at "Brighton/Cambridge". Follow signs
to Cambridge. The Doubletree Hotel will be on your right. Go straight over
the bridge into Cambridge (on River Street) and take your first right onto
Memorial Drive. The Charles River will be on your right. Go straight on
Memorial Drive, staying to the center of the road and going over the overpass
at the Boston University (B.U.) Bridge (past MicroCenter). Pass under the
next bridge, which is the Massachusetts Avenue Bridge. The first street
after that bridge, to the left, is Ames Street--but it is one way in the
wrong direction. You will have to take the second left (sign says Kendall
Square) onto Wadsworth St. Take the first left onto Amherst Street. On
Amherst Street, take the third right onto Ames Street. The Media Laboratory
is the second building on your right. It is a large contemporary white
tiled building. Bartos Theater is on the ground floor.
By Subway
Please check the subway map at your boarding point. You want to take the
Red Line to the "Kendall/MIT" stop. You can transfer without additional
fees from any other subway line at the appropriate subway station. At the
Kendall/MIT stop, you will surface at Main Street in Kendall Square. Landmarks
include Au bon Pain, the Marriott Hotel, the MIT Coop and MIT Press Bookstore
(across Main Street from the others), and Legal Seafoods Restaurant (at
the west end of the block). Facing Main Street, with the Marriott and the
MIT Coop to your back, proceed right (west) towards Legal Seafoods.
Walk to the first traffic light. This is the intersection of Main Street
and Ames Street. Turn left onto Ames Street. The Media Laboratory is the
second building on the left at 20 Ames Street. Bartos Theater is on the
ground floor.
Parking
There are rarely any on-street parking spaces on the MIT campus. The only
garage that is convenient to the Media Laboratory is the Marriott Hotel
garage. The entrance to this is on Ames Street, between Main and Broadway.
Address
Media Laboratory
MIT Building #E15, The Wiesner Building
20 Ames Street
Cambridge, MA 02139
617-253-0338
April, 1998 Meeting
Subject
Conversational Spoken Language Systems for Human-Computer Interaction
Speaker
James Glass, Spoken
Language Systems Group, Laboratory
for Computer Science, MIT
Date
Thursday, April 23, 1998
Time
Presentation starts at 6:30 PM. Note: Different location than usual. See
below.
Meeting Overview
As the user community for computers rapidly expands from experienced programmers
to naive users, we must seriously address the issue of the human-computer
interface. It is highly desirable to allow a user to access information
and solve problems via spoken input, using their native language. In many
of these applications involving interactive problem solving (e.g., extracting
information from a database), speech recognition technology must be augmented
with natural language processing so that the speech can be understood,
sometimes within the context of a verbal dialogue, and the appropriate,
verbal response generated.
In this talk I will present an overview of recent research at MIT on
conversational system development. I will outline some of the challenges
facing researchers trying to move beyond recognition to complete understanding
of spontaneous-speech input. I will then describe the approaches we are
developing for speech recognition, language understanding and generation,
meaning representation, as well as discourse and dialogue. Our experiences
with issues of data collection, system evaluation, portability, and multilinguality
will also be discussed.
The talk will incorporate both video clips and live demonstrations of
several spoken language systems that have been developed in our group,
including a web-based spoken language interface for on-line information
access in the travel domain, and a telephone-based interface for weather
forecast information.
Speaker Biography
James R. Glass (http://www.sls.lcs.mit.edu/jrg)
obtained his Ph.D. degree in Electrical Engineering and Computer Science
from the Massachusetts Institute of Technology in 1988. He is currently
a Principal Research Scientist and Associate Head of the Spoken Language
Systems Group in MIT’s Laboratory for Computer Science. Dr. Glass is a
-
former member of the IEEE Signal Processing Society Technical Committee
on Speech Processing and is currently an associate editor of the IEEE
Transactions on Speech and Audio Processing.
Directions to MIT LCS
The MIT Laboratory for Computer Science is located at 545 Technology Square
on Main Street in Cambridge near the intersection of Main and Vassar. Parking
is available in front of the building. The closest T-stop is Kendall Square
on the red line.
The talk will be given in room 518. Please see http://whereis.mit.edu/doc/getting-to-mit.html
for additional directions to MIT and http://whereis.mit.edu/bin/map?locate=lcs
for a map showing the location of the MIT LCS.
February, 1998 Meeting
Subject
Responsive Object-Based Media
Speaker
V. Michael Bove, Jr.,
MIT Media Laboratory
Date
Thursday, February 19, 1998 (Note: Postponed from
January)
Time
Presentation starts at 7:00 PM. Refreshments at 6:30 PM.
Meeting Overview
Object-Based Media refers to the representation of audiovisual
information as a collection of objects -- the result of scene-analysis
algorithms -- and a script describing how they are to be rendered for
display. Such multimedia presentations can adapt to viewing
circumstances (e.g. size and aspect ratio of display) as well as to
viewer preferences and behavior, and can provide a richer link between
content creator and consumer. With faster networks and processors,
such ideas become applicable to live interpersonal communications as
well, creating a more natural and productive alternative to
traditional videoconferencing.
In my talk I will present some examples of object-based media
applications developed by my group, and describe several analysis
methods and our scripting language. I will also briefly examine
research we are conducting into hardware and software designs that
meet the high computational requirements of object-based and other
advanced media representations.
Speaker Biography
V. Michael Bove, Jr. holds an S.B.E.E., an S.M. in Visual Studies, and
a Ph.D. in Media Technology, all from the Massachusetts Institute of
Technology, where he is currently head of the Object-Based Media Group
at the Media Laboratory. He is the author or co-author of over 40 journal
or conference papers on digital television systems, video processing
hardware/software design, multimedia, scene modeling, and optics. He
holds patents on inventions relating to video recording and hardcopy,
and has been a member of several professional and government
committees. In December 1995, Boston Magazine named him one
of the
``People Shaping Boston's High-Tech Future.'' He is on the Board of
Editors of the Journal of the Society of Motion Picture and Television
Engineers, and served as general chair of the 1996 ACM multimedia
conference.
More detail on his research, and downloadable documents, can be found
at
http://vmb.www.media.mit.edu/people/vmb/
January, 1998 Meeting
Subject
Taming the WWW: From Web Measurements to Web Protocol Design
Speaker
Azer Bestravos
Computer Science Department, Boston
University
Date
Thursday, January 15, 1998 (Note: Change of Speaker)
Time
Presentation starts at 7:00 PM. Refreshments at 6:30 PM.
Meeting Overview
The increasing use of the World Wide Web as the infrastructure for large-scale
distributed information systems dictate that we achieve a better understanding
of its performance characteristics. This talk reports on a number of research
projects undertaken by the Oceans
Group at Boston University. These projects focus on using empirical
performance measurements to design and tune new middleware services and
scalable server architectures for the World Wide Web.
In the first part of this talk, the design and performance of various
caching, dissemination and prefetching services that capitalize on the
WWW temporal, spatial, and geographical locality of reference properties
will be presented. In the second part of this talk, the findings of a study
on WWW traffic characteristics will be presented. In particular, the evidence
and possible causes of self-similarity in WWW traffic will be discussed
and the relationship between WWW traffic self-similarity and the heavy-tailed
distributions of available, unique, and requested WWW files will be established.
Speaker Biography
Prof. Azer Bestavros obtained his S.M. and Ph.D. degrees from Harvard University
in 1988 and 1992, respectively. Since 1992 he has been on the faculty of
the Computer Science Department at Boston University. Prof. Bestavros'
current research interests are mainly in real-time systems and in large-scale
networked information systems. Prof. Bestavros has authored in excess of
70 refereed publications. He is currently the editor-in-chief of the Newsletter
of the IEEE-CS TC on Real-Time Systems and the PC Chair of the IEEE
Real-Time Technology and Application Symposium. He is a member of both
the ACM and the IEEE Computer Society. His research is supported by NSF,
DARPA, ARO, and GTE.
December, 1997 Meeting
Subject
News in the Future
Speaker
Walter Bender
MIT Media Laboratory
Date
Thursday, December 11, 1997 (Note: 2nd Thursday)
Time
Presentation starts at 7:00 PM. Refreshments at 6:30 PM.
Meeting Overview
The News in the Future (NiF)
research consortium provides a forum for the MIT
Media Laboratory and member companies to explore and exploit technologies
that will affect the collection and dissemination of news. The goals
include enhancing the efficiency of production, the timeliness of delivery,
the convenience of presentation,
and the relevance of editorial and advertising content to the consumer.
NiF focuses on four areas: description of news by and for computer; observation
and modeling of consumer behavior; presentation and interface design; and
application. The consortium develops technologies for managing data, building
linkages between news
providers and consumers, and enabling new approaches to the look and
feel of news content. It investigates the application of these technologies
by means of experiments at MIT and field experiments set up in cooperation
with individual member companies.
Speaker Biography
Mr. Bender is a Principal Research Scientist at the MIT Media Laboratory
and Principal Investigator of the laboratory's News in the Future consortium.
He received his BA from Harvard University in 1977. Mr. Bender joined the
Architecture Machine Group at MIT in 1978. He received his MS at MIT in
1980. A founding member
of the Media Laboratory, Mr. Bender is engaged in the study of new
information technologies, particularly those
which affect people directly. Much of the research addresses the notion
of building upon the interactive styles associated with existing media
and extending them into domains where a computer is incorporated into the
interaction. He has participated in much of the pioneering research in
the field of electronic publishing, including the ``Aspen Movie Map'',
and the ``NewsPeek Electronic Newspaper'', both early examples of personalized
interactive multi-media.
November, 1997 Meeting
Joint meeting with IEEE Computer Society, Boston Section
Subject
Content Based Video Access - Broadcast News Navigator
Speaker
Andrew Merlino, The MITRE Corporation
Date
Thursday, November 20, 1997
Time
Presentation starts at 6:30 pm. Refreshments at 6:15 pm (Note: earlier
than usual GBC/ACM meeting).
Location
Note: The meeting will be held at MITRE in Bedford, MA, not at the usual
GBC/ACM meeting site at BBN. Click for directions.
You will need a photo ID.
Meeting Overview
Access to video based on its content is an important requirement for applications
such as video teleconference archiving, video mail access, and individualized
video news program generation. With the increase of multimedia information,
including broadcast news, video mail, and even surveillance video, it is
becoming more and more difficult to access this information based on its
content efficiently and to derive summaries such as trend analyses from
this information.
Towards a solution for this problem, we in the Multimedia Computing
program at The MITRE Corporation have developed techniques to automatically
capture, annotate, segment, summarize and visualize stories from a large
body of video information taken from the broadcast news media (such as
Jim Lehrer News Hour, CNN PrimeNews, and ABC World News Tonight). Our state-of-the-art
system, which automatically extracts stories and commercials and visualizes
the news, is the first to process audio, imagery, and text streams simultaneously
to provide context based access to broadcast news.
Our system, based on a foundation of commercial relational database
and video server technology, includes a Broadcast News Editor (BNE) component
and an associated video viewer, the Broadcast News Navigator (BNN). With
the Broadcast News Editor, we perform video analysis (Black Frame, Logo,
Anchor booth, Reporter Scene Recognition), audio analysis (Silence Detection,
Speaker Change), and closed caption analysis (proper name extraction and
token detection). We process these correlated detections through a finite
state machine to discover broadcast, commercial and story segments. A summarization,
gist, theme and key frame are generated from each story segment. With this
information in our multimedia database, the user of the Broadcast News
Navigator can perform queries on broadcast video as well as perform trend
analysis queries. Some user evaluations suggest that reductions in search
time by a full order of magnitude might not be unreasonable. Currently,
we are looking into integrating speech transcription tools into our system,
and we are transitioning our technology to "the real world."
In our presentation, we will outline the problem of working with large
volumes of multimedia data (processing the data, storing it, and exploiting
it for the web based end user) and we will explain our solution. We will
also discuss the evaluation techniques we used during the development of
this system, and some of our findings. And we will give a live demonstration
of the system, accessing over a year's worth of news from various broadcast
agencies.
Speaker Biography
Andrew Merlino is a Lead Database Technology Software Engineer at The MITRE
Corporation and the Project Lead for BNN. He received a BS in Computer
and Information Science from the University of Massachusetts at Amherst
in 1985 and an MS in Computer Systems Engineering from Northeastern University
in 1989.
Directions to MITRE
The meeting will be held at The MITRE Corporation, 202 Burlington Rd, Bedford,
MA. Meet in the A Lobby.
From Rt. 95 - Take exit 32 (Rt. 3 north). Take exit 26 (Rt. 62). Right
on Rt. 62. Left at the first light onto MITRE grounds. Follow road to the
Right past the A-Building Lobby to parking on left. Please show a photo
ID to the guard in A-Building Lobby.
Also see http://www.mitre.org/about/location/b-map.html.
Additional Information for the November Meeting
Meeting begins at 6:30 pm (with coffee at 6:15 pm) at The MITRE Corporation.
A no-host dinner follows.
For more information, contact Alan Brooks at (617) 271-6497, abrooks@mitre.org
October Meeting
Subject
Neural Circuits for Perceptual Grouping
Speaker
Professor Ennio Mingolla
Boston University
Date
Thursday, October 23, 1997
Time
Presentation starts at 7:00 PM. Refreshments at 6:30 PM.
Meeting Overview
A neural network model of visual boundary detection and completion and
perceptual grouping will be described. While originally derived from an
analysis of perceptual data, the functions of stages of the model can now
be linked to specific local circuits of the primate brain. The model suggests
possible functional roles for many physiological phenomena, including nonlinear
long-range feedback. Perceptual effects as diverse as illusory contour
formation, hyperacuity, and Gestalt grouping receive a unified treatment
by model circuits. Simulations of the model lead naturally to algorithms
for image enhancement, some results of which will be presented.
Speaker Biography
Ennio Mingolla is an Associate Professor of Cognitive and Neural Systems
and of Psychology and a member of the Center for Adaptive Systems at Boston
University. His professional activities include:
-
co-holder of one of the first patents granted for a neural network architecture
for vision and image processing (Boundary Contour System)
-
co-organizer of the Third Workshop of Human and Machine Vision in 1985
-
member of the editorial boards of journals Neural Networks and Ecological
Psychology
-
member of the 1987 Defense Advanced Research Projects Agency (DARPA) Neural
Network Study Panel
-
session chair or invited speaker at several recent international conferences
on neural networks.
His dissertation research involved using computer graphics displays for
stimulus generation for the first parametric studies of human perception
of shape-from-shading, in which many of the assumptions of then existing
computational approaches to shape-from-shading were found to be at odds
with human performance. Since joining Boston University in 1983, his research
has involved design of neural network architectures for preattentive segmentation,
completion, and grouping of static and moving boundaries, textures, and
shaded regions. His research interests include computational vision, neural
network models, perceptual optics, and visual psychophysics.
September Meeting
Jointly sponsored by the GBC/ACM and the IEEE/CS Boston Section
Subject
Using the TPC-D as a Data Warehouse Performance Predictor
Speaker
Michael Steinmann, Director of Technology, Tessera
Enterprise Systems
Date
Thursday, September 18, 1997
Time
Presentation starts at 7:00 PM. Refreshments at 6:30 PM.
Meeting Overview
This is our second presentation on Data Warehousing. The Transaction Processing
Council, an independent industry standard and testing organization, has
developed the TPC-D Benchmark to measure query performance. Understanding
how to use the TPC-D metric is very useful for planning a Data Warehouse.
This presentation explains a methodology for using the TPC-D benchmark
as a configuration and capacity planning tool for Data Warehouse environments.
Different hardware architectures are compared, based on their TPC-D results.
We will identify new hardware trends, and will see how the benchmark can
be used as a predictor of Data Warehouse performance.
Presentation Outline:
-
Model the Hardware requirements of Data Warehouses
-
Review the Historical Platforms used for DWG
-
Why we need a tool for Configuration and Capacity Planning
-
What are the goals of the TPC-D benchmark?
-
Learn how to use the TPC-D benchmark as a perforormance tool
-
Identify the leading hardware architectures for DWG
-
Identify the future hardware architectures for DWG
-
The future and trends for I/O performance
Speaker Biography
Michael Steinmann is Director of Technology at Tessera, with responsibility
for the management of data warehouse systems architecture and development.
Michael has had over 10 years experience in the Relational Database Field.
Prior to joining Tessera, Michael was a Principal Systems Engineer at Sun
Microsystems Computer Corporation. In this position, Michael was responsible
for creating large scale systems configurations for Customers looking to
implement Data Warehouses and other large scale OLTP installations on Sun
hardware. These architectures included configuration and capacity planning
involving all the major Relational Database Vendors, DB/2 (IBM), Oracle,
Informix, Sybase, etc.), networking infrastructure and supporting technologies.
Before working at Sun Microsystems, Michael was a Principal Technical Account
Manager for Ingres Corporation.