CGL Meeting Agenda

Wednesday, January 24, 1996


Location:
DC 1304
Time:
12:30 PM
Chair:
Don Dragomatz

1. Adoption of the Agenda - additions or deletions

2. Coffee Hour

Coffee hour this week:
Mike McCool
Coffee hour next week:
Don Dragomatz

3. Next meeting

Date:
January 31, 1996
Location:
DC 1304
Time:
12:30 PM
Chair:
Saar Friedman
Technical presentation:
Matthew Davidchuk

4. Forthcoming

Chairs:
  1. Ryan Gunther
  2. Fabrice Jaubert
  3. Rick Kazman
Tech Presenters:
  1. Ed Dengler
  2. Don Dragomatz
  3. Saar Friedman

5. Technical Presentations

Presenter #1:
Wilkin Chau
Title: CRT Display Calibration
Abstract:
Most of the book descrbed the relationship between the CIE XYZ tristimulus values and the monitor RGB values as a linear transformation. However, there are some assumptions behind this formulation. In this talk, I will discuss the monitor calibration and the assumptions have been made in order to use the simple 3x3 transformation matrix to convert the values between the CIE XYZ space and monitor RGB space.
Presenter #2:
Bill Cowan
Title: Imaging with limited means
Abstract:
(This is a talk that was given at the Centre Jacques Cartier Workshop on Imaging last December, which generated quite a bit of controversy.) Much computer graphics imaging, both modelling and rendering, aims for a poorly-defined standard called something like "realism". This talk argues that there is no well-defined concept of realism, that abandoning realism invalidates some activities within the computer graphics research paradign, and also answers some questions by providing new programs of research. Last December this talk was interrupted often, and as a result ran well beyond the allocated time of forty minutes. I will probably break it into two pieces, saving the other piece for my next technical presentation.

6. General Discussion Items

7. Action List

8. Director's Meeting

9. Seminars


DEPARTMENT OF COMPUTER SCIENCE
UNIVERSITY OF WATERLOO
SEMINAR ACTIVITIES

THEORY SEMINAR

                    -Wednesday, January 24, 1996

Professor  Ming  Li,  Dept.  Comp.  Sci., University of
Waterloo, will speak on "On NNI".

TIME:                3:30-4:30 p.m.

ROOM:                DC 1304

ABSTRACT

We answer a major open question, proposed by biologists
more  than  25  years  ago,  in  computational biology:
"nearest  neighbor  interchange" (NNI) distance measure
is NP-complete to compute.

Given  two evolutionary trees, comparing them is one of
the central questions in computational biology. The NNI
distance  measure  has  been studied by many scientists
including  biologists  and  computer  scientists in the
past.  Relatively few results, many wrong results, were
published.  In  this  talk,  we present a comprehensive
study  of  this  problem.  We  will  disprove  more old
results  of Waterman and Smith (1978), improve some old
results  of  Culik  and  Wood  (1982),  answer two open
questions,  give  approximation algorithms for NNI, and
other related results.

This  work represents joint effort of our computational
biology group supported by CGAT program.

Joint work with: B. DasGupta, X. He, T. Jiang, J. Tromp
and L. Zhang.

==================================================================

CP Seminar Series Industrial 
Applications of Information Technology 
Computer Science Department 
University of Waterloo

Intercom Ontario - An Interactive Community
[Ed. note: was titled "The Markham Project - a Networked Community" 
in uw.cs.grad]

By:  Bob Campbell 
Of:    Intercom Ontario

Date:     Thursday,  January  25,  1996 
Time:   2:30 P.M.  
Place:
  Davis Centre, Room 1302

Abstract:

Mr.  Campbell will discuss one of the most innovative information
highway-type trials being undertaken in North America, right here
in  Ontario.   The  Intercom  Ontario interactive broadband trial
will  link  a  new residential community with schools, government
offices,  businesses,  and  cultural institutions in an effort to
determine how consumers are likely to respond to a large array of
new,  high-speed  multimedia  applications.   The  overview  will
cover:

     -    Concepts  behind  the trial     
     -    Technologies being deployed   
     -    Application  delivery  plans    
     -    Intellectual property issues

Biography

Bob  Campbell  joined  Intercom  Ontario as Trial Manager after a
career  with Bell Canada.  As Vice-President, Network Planning at
Bell,  his  team  was  responsible  for planning and implementing
Bell's  digital  switching  and  fibre  optics transport networks
across Ontario.  Among his various assignments, Mr. Campbell also
spent  three  years  as  Bell's  senior executive in Saudi Arabia
managing  the  Saudi  Telecom  network, and several years at Bell
Northern Research Labs at Palo Alto, California.

Everyone is welcome. Refreshments served.  Undergraduate students
are particularly encouraged to attend!

==================================================================

                          Intro to Web Browsing
                                    by
                           by Christina Norman

                        Thursday, 25 January, 1996
                                 4:30 PM
                                 MC 3006

                             All are welcome.

 The talk/demo will introduce attendees to the concept of Web Browsing.
 The following topics will be covered:
 - What is the World-Wide Web?
 - What information is available through the WWW?
 - How do I use lynx?
 - How do I use Mosaic?
 - Hypertext links to other documents
 - What is an URL?
 - Text vs graphical data
 - What information is available through the WWW?
 - Some interesting places to visit
 This  is a non-technical introduction.   Web publishing  will not be  covered.
People  who  have  already  browsed the  WWW  will  not  benefit  from  this
talk/demo.

==================================================================

DEPARTMENT OF COMPUTER SCIENCE
UNIVERSITY OF WATERLOO
SEMINAR ACTIVITIES

MASTER'S ESSAY PRESENTATION

                    -Friday, January 26, 1996

Tongyan  (Tony)  Wang, Dept. Comp. Sci., Univ. Waterloo
will  speak on ``A Proposed SQL Standard for Structured
Text''.

TIME:                3:30-4:30 p.m.

ROOM:                DC 1331

ABSTRACT

SQL  standard has gone through many stages and is still
growing  and  improving. SQL/MM is an ongoing effort to
develop  an  SQL  standard for multi-media applications
using the emerging SQL3 capabilities.

In  the existing SQL/MM document, a Full Text data type
is  used  to  construct text and to search patterns. It
has  very  limited  power  in  text  processing.  A new
standard  was  proposed,  by means of a Structured Text
data type, to extend the facilities. This seminar is to
introduce and describe the contents and the development
process of the proposed standard for structured text.

==================================================================

DEPARTMENT OF COMPUTER SCIENCE
UNIVERSITY OF WATERLOO
SEMINAR ACTIVITIES

MASTER'S ESSAY PRESENTATION

                    -Tuesday, January 30, 1996

Geck-Chan  Hong,graduate  student,  Dept.  Comp.  Sci.,
Univ.  Waterloo, will speak on ``Computation of Grobner
Bases by Change of Ordering''.

TIME:                4:00-5:00 p.m.

ROOM:                DC 13004

ABSTRACT

The  classical  method  of solving systems of algebraic
equations  is  by  resultants.  In the mid 1960s, a new
method  known as the Grobner Basis method was proposed.
The  method transforms the system into some ``reduced''
form called a Grobner basis which can be solved by back
substitution.  The main problem with this method is the
complexity  of  the  computation,  in  particular,  the
computation  of  the  Grobner  basis  in  lexicographic
order.

One approach is to calculate the Grobner basis in total
degree  order  and  convert  the  basis to the basis in
lexicographic  order.  In general, it is much faster to
compute  the  basis  in  total  degree  rather  than in
lexicographic term ordering.

This  presentation  will  discuss  two basis conversion
algorithms:   the  FGLM  method  and  the  Grobner Walk
method.   The first one is easy to understand but it is
limited  to  systems  that  have  a  finite  number  of
solutions.   The  Grobner Walk is more difficult but it
has   no   restrictions.   Experiments  show  that  the
conversion  methods  perform  better  than  the  direct
method  most  of the time.  The acceleration factor can
go up significantly.


10. Lab Cleanup (until 1:30 or 5 minutes)