NA Digest Sunday, January 15, 1989 Volume 89 : Issue 2
Today's Editor: Cleve Moler
From: Stephen Nash <SNASH%GMUVAX.BITNET@Forsythe.Stanford.EDU>
Date: Mon, 9 Jan 89 10:16 EST
Subject: Errata Sheet for Kahaner, Moler and Nash
An errata sheet has been prepared for the book NUMERICAL METHODS AND
SOFTWARE by D. Kahaner, C. Moler, and S. Nash. A copy can be obtained
by sending a message to Stephen Nash at NA.NASH on NA-NET, or by
writing to either
David Kahaner Stephen Nash
Technology Bldg., Room A151 ORAS Dept.
NIST George Mason University
Gaithersburg, MD 20899 Fairfax, VA 22030
Please include your mailing address, since the errata sheet will be sent
in paper rather than electronic form.
The errata sheet is constantly being updated. If you find additional errors,
or have suggestions for later printings of the book, please let one of us
From: Nancy Nichols <NANCY%NCSUMATH.BITNET@Forsythe.Stanford.EDU>
Date: Mon, 09 Jan 89 15:22:34 EST
Subject: Change of Address for Nancy Nichols
To friends and colleagues: I am once again at North Carilina State University
for this semester (until about May 6). My e-mail address is
email@example.com or the usual firstname.lastname@example.org
My office address and phone are
Department of Mathematics
North Carolina State University
Raleigh NC27695 - 8205
(919)-737-3100 (or 737-3796 for mssgs)
Home phone is (919) 782-4757
Hope you all have a verrrry Happppy New Year. Nancy
From: Wei Pai Tang <email@example.com>
Date: Mon, 9 Jan 89 15:56:32 EST
Subject: Unsymmetric Lanczos
I am developing a new tridiagonalization algorithm for unsymmetric
matrices and would like to compare with Lanczos algorithm. My experience
was: when n > 40, Lanczos provides no good information. I am not very
sure if there is some better software available out there in NA-Net.
I would very much appreciate it if you might give me a pointer or
send me a better version of unsymmetric Lanczos.
My email address is
Thank you in advance!
From: Adrian Barry <firstname.lastname@example.org>
Date: 9 Jan 89 14:47:12 GMT
Subject: Finite Differences Applied to Boundary Value Problems
Has anyone in this hitherto neglected area got an algorithm for solving
boundary value DE's, using the finite difference method, combined with
Richardson's method ? Please ?!
An answer combining both, or either one would be appreciated.
From: Mike Riley <email@example.com>
Date: 10 Jan 89 06:08:15 GMT
Subject: Info/code wanted on Patankar's SIMPLER model
Has anybody had any experience implementing Patankar's finite
difference SIMPLER model for solving elliptic pde's?
I'm trying to write code to calculate flow and temperature
variables in an enclosure. I'm assuming constant density
and steady state. Boundary conditions include no slip for the
walls, pressure and temperature may be either Dirichlet or
Von Neumann. I'm using primitive variables as I might later
want to go to 3-D.
Any information/code would be greatly appreciated. Please mail
replies to Mike Riley at:
Thanks in advance,
Colorado State University, Fort Collins
From: Michelle Jones <SIAM@wharton.upenn.edu>
Date: Mon, 9 Jan 89 13:33 EST
Subject: SIAM Upcoming Conferences
SIAM Conference on Domain Decomposition Methods
March 20-22, 1989
J.W. Marriott Hotel, Houston, TX
Ken Kennedy, Rice University; William Symes, Rice University; Jacques Periaux,
Avions Marcel Dassault/Breguet Aviation, France; Mary F. Wheeler, University
of Houston; Pierre-Louis Lions, Universite de Paris; Yuri Kuznetsov, USSR
Academy of Sciences; Alfio Quarteroni; Catholic University of Brescia, Italy,
and Istituto di Analisi Numerica del Consiglio Nazionale delle Rescherche;
Gerard Meurant, Centre d'Etudes de Limeil-Valenton; Robert M. Ferencz,
Lawrence Livermore National Laboratory; and James H. Bramble, Cornell
Third SIAM Conference on Optimization
April 3-5, 1989
57 Park Plaza Hotel, Boston, MA
Kurt Anstreicher, Yale University; Jorge Nocedal, Northwestern University;
Richard A. Tapia, Rice University; Philippe L. Toint, Faculte Universitaire
Notre Dame de la Paix, Namur, Belgium; Dimitri P. Bertsekas, Massachusetts
Institute of Technology; Clovis Gonzaga, Federal University of Rio de Janeiro,
Brazil; James B. Orlin, Massachusetts Institute of Technology; Robert B.
Schnabel, University of Colorado, Boulder; and Philip E. Gill, Stanford
SIAM Conference on Control in the 90's: Achievements, Opportunities and
May 17-19, 1989
Cathedral Hill Hotel, San Francisco, CA
John A. Burns, Virginia Polytechnic Institute and State University; Wendell H.
Fleming, Brown University; Alberto Isidori, Universita degli Studi di Roma "La
Sapienza"; P.R. Kumar, University of Illinois, Urbana; Alan J. Laub,
University of California, Santa Barbara; Jacques Louis-Lions, College de
France and Centre National d'Etudes Spatiales, France; Andrzej Manitius,
George Mason University; Steven I. Marcus, University of Texas, Austin; Jan C.
Willems, University of Groningen, The Netherlands.
SIAM Annual Meeting
July 17-21, 1989
Sheraton Harbor Island Hotel, San Diego, CA
Jerry L. Bona, Pennsylvania State University; Tony F. Chan, University of
California, Los Angeles; James W. Demmel, Courant Institute of Mathematical
Sciences; James P. Keener, University of Utah; Hanoch Lev-Ari, Stanford
University; Thomas A. Manteuffel, Los Alamos National Laboratory and
University of Colorado, Denver; Robert L. Parker, University of California,
San Diego; Philip G. Saffman, California Institute of Technology; Danny C.
Sorensen, Argonne National Laboratory; and Jeffrey Speiser, Naval Ocean
Systems Center, San Diego.
SIAM Conference on Mathematical and Computational
Issues in Geophysical Fluid and Solid Mechanics
September 25-28, 1989
Stouffer Greenway Plaza Hotel, Houston, TX
Alain Bamberger, Institut Francais du Petrol, France; Michael M. Carroll, Rice
University; James Dieterich, U.S. Geological Survey; Jim Douglas, Jr., Purdue
University; Bjorn Engquist, University of California, Los Angeles; Paul C.
Fife, University of Utah; James M. Hyman, University of Arizona and Los Alamos
National Laboratory; Barbara L. Keyfitz, University of Houston; Andrew J.
Majda, Princeton University; Peter Ortoleva, Indiana University, Bloomington;
George Pinder, University of Vermont; Luc Tartar, Carnegie-Mellon University;
Mary F. Wheeler, University of Houston; and Benjamin S. White, Exxon Research
and Engineering Company.
SIAM Conference on Geometric Design
November 6-10, 1989
Sheraton Mission Palm Hotel, Tempe, AZ
Alan H. Barr, California Institute of Technology; Philip J. Davis, Brown
University; Rida Farouki, IBM-T.J. Watson Research Center; John A. Gregory,
Brunel University; Dave Gossard, Massachusetts Institute of Technology; Cleve
Moler, Ardent Computer Corporation; John Rice, Purdue University; Larry
Schumaker, Vanderbilt University; Tom Sederberg, Brigham Young University;
Peter Wilson, General Electric Corporation; Michael Wozny, Rensselaer
For further information on registration or to obtain abstracts for
contributing papers or minisymposia, please contact:
SIAM Conference Coordinator
Suite 1400, Architects Building
117 S. 17th Street
Philadelphia, PA 19103-5052
E-Mail Address: SIAM@WHARTON.UPENN.EDU
From: Elizabeth Seaman <firstname.lastname@example.org>
Date: 11 Jan 89 07:58:10 GMT
Subject: Non-linear Unconstrained Optimization in Control
I am looking for references/pointers to the adaptive control
literature concerning the problem of search strategies in
"parameter space" for control problems.
This can also be thought of as belonging to the non-linear
unconstrained optimization realm.
In particular, I am interested in "diffusion" type approaches,
methods for controlling step size, methods for finding search
directions, and feedback paradigms.
Thanks for your help... e-mail to address below; I will post
a summary if there is interest.
Elizabeth C. Seaman
University of California, San Diego
From: Jonathan M. Smith <email@example.com>
Date: Wed, 11 Jan 89 11:14:48 est
Subject: Jenkins-Traub Algorithm for Polynomial Zeros
Does anyone have a typed-in, working version of ACM algorithm #419,
for zeros of a complex polynomial (a.k.a. Jenkins-Traub). Fortran is fine,
although C would be better for my purposes.
From: Youcef Saad <firstname.lastname@example.org>
Date: Thu, 12 Jan 89 13:54:59 PST
Subject: New Book Series on Scientific Computing
Manchester University Press announces a new series in computer
"ALGORITHMS AND ARCHITECTURES FOR ADVANCED SCIENTIFIC COMPUTING"
This new series will focus on recent advances in
scientific computing, with emphasis on new algorithmic methods
and computer architectures. It will reflect the rapid developments
in advanced scientific computing as a whole, including the aspects
that are purely algorithmic and those related to the
impact of supercomputer technology. Some of the topics where
books and monographs are actively sought are the following:
1) Parallel and vector numerical and non-numerical algorithms ;
2) Supercomputer applications;
3) New algorithms and methods in scientific computing;
4) Modern architectures for scientific computing.
The series will not include conference proceedings. Rather it will
consist of structured books written by an individual or a small
expert group. It is planned to co-publish the series in the US.
Dr. Yves Robert Dr. Youcef Saad
Laboratoire LIP-IMAG RIACS MS 230-5
Ecole Normale Superieure de Lyon NASA Ames res. Center
46 allee d'Italie Moffett Field, CA 94035
69364 Lyon Cedex 07 Lyon FRANCE
email@example.com (uucp) firstname.lastname@example.org
. Tony F. Chan, UCLA
. George Cybenko, University of Illinois
. Bill Gropp, Yale
. Marina Chen, Yale
. Michel Cosnard, Ecole Normale Sup., Lyon (France).
. Prof. D. J. Evans, Loughborough, UK
. Piero Sguazzero, IBM ECSEC Rome
If you are interested in publishing a book in the series
or would like further information write to the series editors.
From: Roger Grimes <email@example.com>
Date: Fri, 13 Jan 89 14:16:05 pst
Subject: Positions at Boeing Computer Services
Applied Mathematics Group
Boeing Computer Services
The Applied Mathematics Group in Boeing Computer Services, located in
Seattle, Washington anticipates openings in early 1989 for well
qualified individuals in
Numerical Linear Algebra
We are looking for entry level or experienced PhDs with either a
dissertation or post-graduate experience in one or more of the above
fields and an interest in applying this background to challenging real
Our group consists of about 35 mathematicians performing consulting
and research work for other parts of The Boeing Company, for
commercial customers, and for governmental agencies. We require
individuals capable of working independently in a dynamic,
multidisciplinary environment on complex algorithm, analysis, and
application software problems.
Qualified applicants should send a resume to:
Dr. James L. Philips
Manager, Applied Mathematics
Boeing Computer Services
P.O. Box 24346, MS 7L-21
Seattle, WA 98124-0346
Inquiries via electronic mail can be made to either John Lewis
(na.lewis) or Roger Grimes (na.grimes).
Boeing is an Equal Opportunity Employer
From: Arthur Wouk <wouk@BRL.MIL>
Date: Fri, 13 Jan 89 14:10:35 EST
Subject: Position in Computation at ARO
I will be retiring at the end of March, and the official announcement
of a search for my replacement follows. Please note that this is a
position at the Ph.D. level for someone with active research
experience in areas related to scientific computation and a wide range
of interests. For further details about the technical content of the
position, feel free to contact Dr. Jagdish Chandra, Director,
Mathematical Sciences Division, ARO, at 919-549-0641. Details of how
to obtain the forms required for applying for the position are listed
Arthur Wouk, ARO
U.S. ARMY RESEARCH OFFICE
P.O. BOX 12211
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27709
POSITION: Mathematician, GS-1520-14 or Computer Scientist, GS-1550-14
SALARY: $48,592 - $63,172 per annum
LOCATION: Mathematical Sciences Division, U.S. Army Research Office
Research Triangle Park, North Carolina
OPENING DATE: 17 January 1989
CLOSING DATE: 16 February 1989
AREA OF CONSIDERATION: All Sources. Applications will be accepted from
current career or career-conditional employees, reinstatement eligibles and
non-status applicants eligible for appointment from an appropriate Office
of Personnel Management register. Consideration will also be given to
individuals eligible under special authorities (e.g., physically
handicapped, 30% or more disabled veterans).
ALL QUALIFIED APPLICANTS WILL RECEIVE CONSIDERATION FOR THE ABOVE POSITION
WITHOUT DISCRIMINATION FOR ANY NON-MERIT REASON SUCH AS RACE, COLOR,
RELIGION, SEX, AGE, NATIONAL ORIGIN, POLITICS, MARITAL STATUS, PHYSICAL
HANDICAP WHICH DOES NOT INTERFERE WITH ACCOMPLISHMENT OF JOB, OR
MEMBERSHIP/NON-MEMBERSHIP IN AN EMPLOYEE ORGANIZATION.
DUTIES: Serves as Program Manager of the Computational Mathematics Program
which includes Numerical Analysis, Optimization, Discrete Mathematics and
Theoretical Computer Science. Research projects emphasize the development
and analysis of numerical methods for the solution of nonlinear equations
arising in such areas as solid and fluid mechanics, aerodynamics and
chemical kinetics, as well as the development of the underlying
technologies in numerical linear algebra, nonlinear algebraic systems and
optimization. Other research projects are directed toward new technologies
in high performance computing relating to scientific computing and
mathematical software and the computational aspects of robots, including
new mathematical methods in computational geometry and control logics for
robot manipulators. Evaluates and correlates the requirements, scope and
objectives of the overall Army research program in computational
mathematics and/or theoretical computer science. Evaluates and
disseminates the results of the research programs of the Army and also
those of other Government and private agencies (domestic and foreign) which
may be of significant interest to the Army. Initiates and exercises
surveillance over research contracts and/or grants. The actual research is
conducted in institutions of higher learning, private research
organizations, laboratories, and other Government agencies. Makes
significant decisions as to the acceptability of proposals to perform
research projects, taking into consideration such factors as the competence
of research investigators, relevancy and type of research proposed and
necessity for such research in terms of military and fiscal considerations.
Reviews progress and technical reports on research in progress, making
major decisions as to their significance and whether they warrant continued
support. Maintains personal contact with all scientists involved in the
program, overseeing their performance, aiding in the solution of their
problems and bringing Army needs to their attention. Pursues research
and/or advanced study within or outside of locally available institutions
on projects of personal interest and whose content is relevant to the
mission of the Army.
MINIMUM OPM QUALIFICATION REQUIREMENTS: Basic Requirements: Candidates
must show successful completion of a full 4-year or longer curriculum in an
accredited college or university leading to a bachelor's (or higher)
degree. For GS-1520, course of study must have included 24 semester hours
in mathematics. For GS-1550, course of study must have included 30
semester hours in a combination of mathematics, statistics, and computer
science that provided in-depth knowledge of the (1) theoretical foundations
and practical applications of computer science, including digital computer
system architecture and system software organization, the representation
and transformation of information structures, and the theoretical models
for such representations and transformations; and (2) essential
mathematical and statistical techniques. Professional Experience: In
addition to the basic requirement, candidates must have three years of
professional experience in computational mathematics and/or theoretical
computer science. This experience must have equipped the candidate with
the knowledge and ability to perform fully the work of the position.
Candidates may have equivalent combinations of professional experience and
graduate education as authorized in OPM Handbook X-118, GS-1520 and
GS-1550. Quality of Experience: At least one year of the required
experience must have been at a level of difficulty comparable to that of
the next lower grade level in the Federal service.
SELECTIVE PLACEMENT FACTOR: Ability to perform fundamental research in
mathematics or computer science.
TIME-IN-GRADE REQUIREMENT: Candidates for advancement to a position at the
GS-14 level must have completed at least one year of service in positions
no more than one grade lower than the position to be filled.
SPECIAL REQUIREMENT: Available to travel up to 30% of the time.
SECURITY REQUIREMENT: Selection for or retention in this position is
contingent upon satisfactory results of a security investigation and
ability to obtain appropriate security clearance. Offer of position may be
withdrawn or employee may be terminated if above conditions of employment
are not met.
CRITERIA FOR DETERMINING HIGHLY QUALIFIED: Applicants will be evaluated as
to degree of possession of Knowledge, Skills and Abilities (KSA) listed
1. Ability to plan, direct, and guide a research program in computational
mathematics and/or theoretical computer science.
2. Ability to evaluate research proposals for technical merit.
3. Advanced knowledge at doctorate level or equivalent scientific studies
and research in computational mathematics and/or theoretical computer
science and evidence that technical competence has been maintained.
4. Ability to interact with other scientists, professional groups and
EVALUATION METHODS: BASIC ELIGIBILITY will be determined by review of the
applicant's SF 171 against: (1) Minimum qualifications in accordance with
the USOPM Handbook, X-118, Mathematics Series, GS-1520 or Computer Science
Series, GS-1550, and (2) Any selective placement factors. HIGHLY QUALIFIED
determinations will be based upon evaluation of the candidate's background
as compared to the KSA's by application of the following rating
instruments: experience, training and education, supervisory appraisals
and related awards. BEST QUALIFIED candidates will be derived by further
evaluation and ranking of the highly qualified group. Ranking is required
when there are more than 10 basically eligible candidates.
HOW TO APPLY: Applicants who have status with the Federal Government
should forward one copy of the following documents to: U.S. Army Research
Office, ATTN: SLCIS-CP-AR, P.O. Box 12211, Research Triangle Park, NC
27709-2211. Documentation must be received by closing date. (1) A
completed Standard Form 171 (Personal Qualifications Statement). Item 1 of
the SF 171 should identify "Mathematician" or "Computer Scientist" and ARO
Announcement No. 89-5; (2) ARO Supplemental Applicant Statement; (3) ARO
Supervisory Appraisal of Job Related Knowledge, Skills and Abilities; (4)
Latest Supervisory Performance Rating; (5) OPM Form 1386; and (6) Current
SF 50, Notification of Personnel Action.
Non-status applicants should contact the ARO Support Branch, U.S. Army
Research Office (numbers listed below) to obtain appropriate forms to be
placed on the Office of Personnel Management register for Mathematician and
related positions in the southeastern United States.
ADDITIONAL INFORMATION: To obtain further information about the position,
contact ARO Support Branch, AUTOVON 935-3331 or Commercial (919) 549-0641,
RELOCATION SERVICES ARE AVAILABLE FOR THIS POSITION FOR CURRENT FEDERAL
USE OF POSTAGE-PAID GOVERNMENT AGENCY ENVELOPES TO FILE JOB APPLICATIONS IS
A VIOLATION OF FEDERAL LAW AND REGULATIONS. APPLICATIONS RECEIVED IN
POSTAGE-PAID GOVERNMENT ENVELOPES WILL NOT BE CONSIDERED.
End of NA Digest