NA Digest Sunday, December 24, 1995 Volume 95 : Issue 51

Today's Editor:
Cleve Moler
The MathWorks, Inc.

Submissions for NA Digest:

Mail to

Information about NA-NET:

Mail to

URL for the World Wide Web: -------------------------------------------------------

From: Jerzy Wasniewski <>
Date: Thu, 21 Dec 1995 14:04:01 +0100 (MET)
Subject: PARA96 & Merry Christmas

August 18-21, 1996

                                    + +
   A Merry Christmas and            +O+             =============
                                   + * +             |  . |  . |
     a Happy New Year              +* O+             | .  | .  |
                                  +   O +            |  . | .  |
          1 9 9 6                 + @   +            | .  | .  |
                                 +   @ * +           |----+----|
                                 + @  O  +           |   .| .. |
                                +   *  O  +          |  . |  . |
                                + OX  O  #+          | .  |  . |
                               +  O  *#*   +        =============
                               +O o   *  @ +
                              +     %   *   +
                              +O *    * O & +
                          ------    | |   -------
                          |#-#-|  ------- |#-#-#|
                          ------          -------
Best Regards,


From: Peter Deuflhard <deuflhard@ZIB-Berlin.DE>
Date: Thu, 21 Dec 1995 14:40:04 +0100
Subject: Konrad Zuse

It is extremely sad for me to inform the community about the death of
Konrad Zuse. He died on Dec. 18 at the age of 85 in his home in Huenfeld
near Fulda.
For those of you, who do not yet know his name: on an international level,
Konrad Zuse was the creator of the first really working fully automated,
program-controlled and freely programmable computer using binary floating-point
arithmetic. It was operational in 1941. He built this machine in the living
room of his parents here in Berlin. Fortunately, as he used to state, the Nazi
regime did not recognize the importance of his invention. After the Second
World War Zuse started to design a rather future oriented programming language
called "Plankalkuel" (plan calculus), which in fact shows certain construction
principles of modern object oriented languages. He started a firm, the ZUSE KG,
which, however, was later bought by Siemens, but not recognized as the jewel
that it really was (Siemens transformed it into a production site, so that all
know-how people left.) Konrad Zuse is the patron of the ZIB, the
Konrad-Zuse-Zentrum fuer Informationstechnik Berlin. As president of this
Center I got to know Zuse personally quite well. I really admired his
extremely abstract views, a high rank mathematician as well as a high rank
engineer! A universal genius of the kind that is rare worldwide. The same
degree of abstraction is also visible in his paintings, which were his hobby
throughout his life and the main occupation during his late years. Zuse rather
late enjoyed international recognition, only recently the History of Computer
Science put him into the right place. The Washington Museum still does not
(yet?) show him in the gallery of the main inventors of the field. Want to
know more about him?

Just look up URL: .

Do not hesitate to write me on the subject!

Peter Deuflhard
Konrad-Zuse-Zentrum Berlin.


From: G. W. Stewart <>
Date: Thu, 21 Dec 1995 11:33:16 -0500
Subject: Draft of New Book Available Via FTP

I am currently writing a multivolume Survey of Matrix Algorithms. The
present Volume I is entitled Basic Decompositions. Three chapters of
this volume can be obtained by anonymous ftp from in pub/survey

The first two chapters contain introductory material from mathematics
and computer science and the third chapter is on Gaussian elimination.
A fourth chapter on the QR decomposition and a fifth on rank
determination will complete the volume. For more information see the

I am distributing the book in the hope that it will be helpful to
others and that others will be willing to help me with their comments
and corrections. Please feel free to make copies for your personal
use. However, if you want to make copies to distribute to a class,
please ask my permission (it will generally be forthcoming).

Pete Stewart


From: Elizabeth Jessup <>
Date: Thu, 21 Dec 1995 15:38:08 -0700
Subject: New Textbook in High-Performance Scientific Computing

It's finally done! The following textbook is supposed to be
available from MIT Press for next semester's courses:

An Introduction to High-Performance Scientific Computing
by Lloyd D. Fosdick, Elizabeth R. Jessup, Carolyn J. C. Schauble,
and Gitta Domik

The authors begin with a survey of scientific computing and then
provide a review of background (numerical analysis, IEEE arithmetic,
Unix, Fortran) and tools (elements of MATLAB, IDL, AVS). Next, full
coverage is given to scientific visualization and to the architectures
(scientific workstations and vector and parallel supercomputers) and
performance evaluation needed to solve large-scale problems. The
concluding section on applications includes three problems (molecular
dynamics, advection, and computerized tomography) that illustrate the
challenge of solving problems on a variety of computer architectures as
well as the suitability of a particular architecture to solving a
particular problem.

The ISBN number is 0-262-06181-3, and the purchase price is $50
(in the U.S.).

The book, some lab materials, and syllabi are also available by
anonymous ftp from in the directory /pub/cs/HPSC.
(Beware that you will use up a LOT of paper if you decide to grab
and print the whole thing.)



From: Michele Benzi <>
Date: Fri, 22 Dec 1995 15:42:34 +0100
Subject: Change of Address for Michele Benzi

Dear colleagues,

Beginning January 2 I will be at CERFACS in Toulouse,
on a leave from the University of Bologna.
My address will be :

Michele Benzi
42 Ave. G. Coriolis
31057 Toulouse Cedex

My current e-mail address will also work.

Best wishes in the new year,

Michele Benzi


From: Martin Gutknecht <>
Date: Fri, 22 Dec 1995 16:14:13 +0100 (MET)
Subject: Reorganisation of Swiss Computing Centers

Reorganization of CSCS and IPS in Switzerland

By January 1, 1996, the Centro Svizzero di Calcolo Scientifico (CSCS) at
Manno near Lugano and the Interdisciplinary Project Center for Supercomputing
(IPS) at ETH Zurich will merge into a single organization that runs under
the acronym CSCS/SCSC; here, SCSC stands for the modified English name
`Swiss Center for Scientific Computing'. All the current members of the IPS
will keep their Zurich offices, but get new mail and email addresses:

ETH-Zentrum, RZ
CH-8092 Zurich

This includes, in particular, the following numerical analysts:

Armin Friedli
Gabor G. Groh
Martin H. Gutknecht
Urs von Matt
Wesley P. Petersen
Klaus J. Ressel
Eric de Sturler

Phone and fax numbers remain unchanged. The addresses of the Manno site,
where some of us will take over additional responsibilities, are

Via Cantonale
CH-6928 Manno

Along with the merger there goes a reorganization of the former CSCS.
The new CSCS/SCSC will have only two sections, `Services' (operating
the computing resources, which include a NEC SX-3 and a NEC SX-4, and
providing machine specific support) and `Scientific Projects and Support'
(SPS). The new organization is headed by a new director Jean-Pierre Therre
(formerly with Cray Research (Suisse)), along with Djordje Maric (Director
of Services) and Martin Gutknecht (Scientific Director), who are responsible
of the respective sections.

While the IPS will loose its name, it will survive as an enlarged new
section of the new CSCS/SCSC, and we will continue our research and
consulting work in scientific computing. The new SPS section will have
the following groups (and group leaders):

- Advanced Numerical Algorithms (ANA) - M. Gutknecht
- Computational Chemistry (CC) - H.P. L\"uthi
- Data Analysis/Solid State Physics (DA/SSP) - D. W\"urtz
- Computational Particle Physics (CPP) - P. de Forcrand
- Software Technology (ST) - K. Decker
- Interdisciplinary and Industrial Applications (IIA) - N.N.
- Scientific Visualization (SV) - R. Peikert

Martin H. Gutknecht
ETH-Zentrum, RZ
CH-8092 Zurich


From: Arthur Sherman <>
Date: Mon, 18 Dec 1995 12:07:48 -0500 (EST)
Subject: C Code for Gear Algorithm

Does anyone have or know of a C version of the Gear algorithm of
comparable quality to the fortran codes LSODE, DDRIV, etc.? Has anyone
tried f2c on those codes?

Arthur Sherman

National Institutes of Health
NIDDK/Mathematical Research Branch
BSA Building, Suite 350
BETHESDA MD 20892-2690
phone: (301) 496-4325
fax: (301) 402-0535


From: Stephanie Rault <>
Date: Fri, 22 Dec 1995 14:48:32 +0100
Subject: Satellite Trajectory Code Sought

I work on satellite trajectory with perturbations and i would like
to have programs in FORTRAN or formal calculus which give an
analytic solution.
Any reference would be helpful.
Thank you for your help.

Stephanie RAULT.
(33) 99 84 75 05
Institut de recherche en informatique et systemes aleatoires. Rennes. FRANCE.


From: Ivan Slapnicar <>
Date: Tue, 19 Dec 1995 17:32:13 +0000 (GMT)
Subject: Workshop in Croatia on Numerical Linear Algebria

- first announcement -
International Workshop on Accurate Eigensolving and Applications
Split, Croatia, July 11-17, 1996.


Jesse Barlow, The Pennsylvania State University
Ivan Slapnicar, University of Split (local manager)
Kresimir Veselic, Fernuniversitaet Hagen

Tentative speakers:

Beresford Parlett, University of California at Berkeley
James Demmel, University of California at Berkeley
Ilse Ipsen, North Carolina State University
Hongyuan Zha, The Pennsylvania State University
Zlatko Drmac, University of Colorado at Boulder
Jesse Barlow, The Pennsylvania State University
Ivan Slapnicar, University of Split
Kresimir Veselic, Fernuniversitaet Hagen

The main theme is the accuracy in the EVD or SVD computation or even
in solving linear systems. We will stress this to be a workshop and
not overload the program with too many speakers. We rather prefer to
give more time per speaker. We will therefore have invited lectures
(one or two) in the morning, while the afternoon will be reserved for
contributed lectures, including those from junior people, as well as
for less formal discussions and working groups.

Possible general themes for discussion are:
(i) how to turn recent results on accurate algorithms
systematically into a good software,
(ii) how to present them in a monograph,
(iii) how to let them influence the teaching of numerical linear
(iv) real world engineering problems.

In order to keep the local costs as low as possible, we expect to have
a number of single and double bedrooms in a comfortable student dormitory
located by the sea at a walking distance from the city centre.

Split is a picturesque city with population around 200,000 which in
this decade celebrates 1700 years of urban settlement. It is located
on the Croatian Adriatic coast, the beauty of which is proverbial with
its hundreds of islands and still quite unpolluted nature.
Neither Split nor its neighbourhood carry traces of the (we may now hope)
definitely passed war in Croatia.

Split is accessible by train (9 hours trip from Zagreb), by ship
(one night by ferry from Rijeka, Pescara or Ancona), by car (6 hours
from Zagreb, 8 hours from Trieste), and by plane. Split airport has
daily connections to all world destinations. There are direct flights
from Dusseldorf, Frankfurt, and Rome on some days. All other
connections are via Zagreb by Croatian Airlines.
INFORMATION FOR YOUR TRAVEL AGENT: the Croatia Airlines timetable is
accessed through the AMADEUS Distributive System.

WWW sites:
Split and the University:

For further information please contact

Kresimir Veselic
Fernuniversitaet Hagen
LG Mathematische Physik
P.O. Box 940
D-58084 Hagen


From: Tony Skjellum <tony@Aurora.CS.MsState.Edu>
Date: Tue, 19 Dec 1995 23:52:00 -0600
Subject: Object-Oriented Numerics Conference

The Fourth Annual Object-Oriented Numerics Conference
Mississippi State University
March 27-29, 1996

Scientific application software continues to grow in complexity due to
increasing functionality, evolution of codes, the use of parallelism,
and other factors. As a result, more and more time is spent writing
and maintaining computer code, and less time is spent on the
application itself. Object technology offers a way to break this
cycle. Structuring applications as assemblages of software objects,
rather than as monolithic programs, may pave the way for more
efficient and complex applications by making software reusable,
distributable, and modular.

The annual OONSCI conference promotes the use of object technology for
constructing numerical applications. OONSCI provides a forum for
computer scientists and scientific programmers, in both academia and
industry, to learn about issues related to object-oriented programming
and discuss their experiences.


Examples of appropriate topics include:
* Experiences with object technology
* Use of non-programming language specific objects - for
example, CORBA or Visual Basic controls.
* Object-oriented languages and environments
* Object-oriented design
* Reusable software objects
* Performance issues related to object-oriented programming

Application domain specific papers should focus on the
object-oriented techniques used. Sample application domains of
interest to OONSCI are as follows:
* Image processing
* Computer graphics
* PDE solvers
* Engineering


Submissions should be prepared for an audience with diverse
backgrounds who come from many disciplines, rather than specialists
in a particular discipline. General knowledge of object-oriented
terminology can be assumed.

Proposals are being solicited for:
1) Papers
2) Survey papers
3) Tutorials
4) Panel Discussions
5) Posters


December 15, 1995 [extended to January 15, 1996]
* Tutorials

January 15, 1996 [Abstracts only are due]
* Papers
* Survey papers
* Panels
* Posters

For submission guidelines see


Complete and up-to-the-minute information about the conference can be
found on the Web at

or contact

Anthony Skjellum Internet:
Department of Computer Science Phone: (601) 325-4524
Mississippi State University Phone: (601) 325-8435
Mississippi State, MS 39762 Fax: (601) 325-8997


From: Adhemar Bultheel <>
Date: Wed, 20 Dec 1995 15:10:22 +0100 (MET)
Subject: Congress on Computational and Applied Mathematics

Seventh International Congress on

Katholieke Universiteit Leuven, Belgium
July 21-26, 1996


The Congress will concentrate on the analysis of computational techniques
for solving real scientific problems.

There will be sessions on:

-Parallel algorithms
-Numerical conformal mapping
-Constructive techniques for solving ordinary and partial differential
-Computational complex analysis
-Numerical quadrature and integral equations
-Numerical software
-Computational techniques in operations research and statistics
-Mathematical techniques for financial and actuarial sciences


Prof. F. Broeckx
Prof. R. Piessens
Prof. M. Goovaerts
Prof. L. Wuytack

Invited speakers:

Prof. H. Brunner (Canada)
Prof. M.E.H. Ismail (USA)
Prof. F. Marcellan (Spain)
Prof. M. Nakao (Japan)
Prof. J. Nedoma (Szech Republic)
Dr. W. Sweldens (Belgium)
Prof. P. Toint (Belgium)

Short Communications:

Short Communications (20 minutes duration) will be accepted for presentation.
Participants who would like to present a paper should submit a title and a
short abstract (at most 1 page) not later than March,1, 1996 to

Prof. M.J. Goovaerts
huis Eygen Heerd
Minderbroederstraat 5
B-3000 Leuven
tel/fax (32) 16 29 53 46
e-mail :

This address should also be used if more information is required.


From: Glenn Luecke <>
Date: Wed, 20 Dec 95 13:57:59 CST
Subject: Supercomputing on IBM Systems

SUPER! 96 - A Conference for HPC on IBM Systems
Focus - The RISC System/6000 SP
Iowa State University
Ames, Iowa
April 21 -24, 1996

SUPER!, a group of users and technical support staff involved in high
performance parallel and clustered computing on IBM systems, announces
the 1996 conference. The penetration of the SP2 computer into a
variety of roles - academic and commerical; scientific, technical and
data management - provides a focus for this year's conference.

Researchers, computing center management and technical support staff
from all SP sites are invited to attend, as well as those anticipating
the arrival of an SP. Conference activities include the following:

o Plenary talks by distinguished speakers in the field.

o Technical sessions on systems management, emerging uses, computational
applications and new products.

o Panel discussion by SP sites on SP system administration.

o Open Forum with IBM product executives - open mike Q&A for the audience.

o Tutorial on SP Network Tuning.

o An SP User Group BOF.

o An SP User Gallery of SP sites.

o Variety of informal discussion opportunities, including a special
dinner excursion.

Registration forms and updated information are available from the
SUPER! Home Page:
and by email to


From: Grigory Gadiyak <>
Date: Thu, 21 Dec 1995 12:18:34 +0600
Subject: Conference in Russia on Simulation of Devices

MAY 13-17, 1996
In Honour of P.L.Chebyshev's



- Simulation of semiconductor devices

- Simulation of processes in thin insulator films in strong
electric fields

- Simulation of technologies ( including simulation of crystall
growth in microgravity, simulation of nuclear technologies and
nuclear power plants)

- Numerical methods for simulation devices and technologies

All the correspondence should be sent to the
V.Ginkin or G.Gadiyak.
Addresses for correspondence:
Dr. V. Ginkin,
Bondarenko Sq.1.,Obninsk, Kaluga Region,249020,
Telex: 412509 URAN SU, Fax: (095)8833112, e-mail:
Prof.G.Gadiyak Ac.Lavrentjev Ave. 6,
Novosibirsk, 630090, Russia


From: Ligia Ribeiro <>
Date: Thu, 21 Dec 95 16:33:17 +0000
Subject: Meeting in Portugal on Vector and Parallel Processing


PROCESSING (Systems and Applications)

Faculdade de Engenharia da Universidade do Porto, Porto, Portugal
1996, September 25--27

The Meeting is interdisciplinary in nature, bringing together people from
Science, Engineering and Industry to explore some of the many challenges
and promises of vector and parallel processing. The first edition of this
Conference was held in Porto in September 1993, and was attended by more
than 100 participants from 15 countries. The event aims at disseminating
present knowledge on the topic and at providing a forum for presentation
and discussion of basic research and applications in this area. The focus
of the scientific section of the Meeting will be key invited lectures to
describe the concepts behind vector/parallel processing and to highlight
some of the major trends in the field. After these main talks, contributed
papers of approximately 20 minutes duration will be presented.

Papers may address a broad range of research fields of current interest.
A list of possible topics includes (but is not limited to) the following:

* Architectures, operating systems, environments, software tools and languages
* Numerical and symbolical algorithms
* Applications in Science and Engineering
(e.g. computational fluid dynamics, reservoir modelling, etc.)
* Industrial and commercial systems and applications
(e.g. database based systems, traffic flow modelling, optimization)
* Signal processing and both image processing and synthesis

Papers on systems and applications to Science and Engineering will be

* Extended abstracts due: 15 March 1996
* Notification of acceptance: 10 May 1996
* Final papers due: 30 August 1996
* Meeting in Porto, Portugal: 25, 26 and 27 September 1996

If you are interested on receiving information on this Conference,
please register on the mailing-list sending your name and address to

Attention of Dr. Jose Laginha Palma
Faculdade de Engenharia, DEMEGI-Seccao de Fluidos e Calor
Rua dos Bragas
4099 Porto Codex, Portugal

Tel.: International+2+2001746
Fax: International+2+312476
WWW address:

Dr. Jose M. Laginha M. Palma
Faculdade de Engenharia
Departamento de Engenharia Mecanica e Gestao Industrial
Seccao de Fluidos e Calor
Rua dos Bragas
4099 Porto Codex, Portugal
FAX: International+2+312476


From: Robert van de Geijn <>
Date: Tue, 19 Dec 1995 11:51:04 -0600
Subject: Position at University of Texas, Austin

The Department of Computer Sciences of the University of Texas at
Austin invites applications for tenure-track positions at all levels,
particularly at the assistant professor level. Of interest to the
department are candidates whose research accomplishments are in
experimental systems that would broaden and complement the research
interests of our faculty in architecture, compilers, computational
mathematics, databases, graphics, networking, operating systems, and
robotics. Candidates must hold or be making satisfactory progress
towards a Ph.D. or equivalent degree in computer science or a related
area, with a reasonable expectation of completion by August 31,
1996. Offers of employment are contingent upon completion of the
Ph.D. degree requirements by that date. Successful candidates are
expected to pursue an active research program, perform both graduate
and undergraduate teaching, and supervise graduate students.

The Department is ranked among the top ten computer science
departments in the country. It has 40 faculty members across all
areas of computer science, and participates in the university's
Computational and Applied Mathematics interdisciplinary

Women and minority candidates are especially encouraged to apply. The
University of Texas is an Equal Opportunity/Affirmative Action

For further details see


From: Marek Behr <>
Date: Tue, 19 Dec 95 13:19:24 CST
Subject: Positions at University of Minnesota Research Center

The Army High Performance Computing Research Center at the University of
Minnesota is funded by the US Army to establish programs for collaborative
research involving high performance computing, between University and Army

Position(s) at the Army Research Laboratory Aberdeen Proving Ground, MD,
require research experience in Material Science and Advanced Materials.
Position(s) at the Corps of Engineers Waterways Experiment Station, Vicksburg,
MS, require research experience in Computational Fluid Mechanics or
Computational Solid Mechanics. Position(s) at the Tank Automotive Research,
Development and Engineering Center, Warren, MI, require research experience
in Computational Solid Mechanics or Multi-Body Dynamics. In addition, a Ph.D.
in engineering, applied mathematics, physical sciences, computer science, or
related fields is required. (Applicants who hold US citizenship at time of
application will be given preference.)

Salary range is $35,000 to $75,000 DOQ. Additional information is available
on the www ( or by email to

The University of Minnesota is an equal opportunity educator and employer

Anita Anderson
Tel: 612-626-8095 Fax: 612-626-1596 Email:


From: Iain Duff <>
Date: Fri, 22 Dec 95 15:16:57 GMT
Subject: Postdoctoral Position at Rutherford Appleton Laboratory


A Postdoctoral Research position is available at the Rutherford Appleton
Laboratory (URL to work in the
Numerical Analysis Group (URL
The post is supported by a Long Term Research (LTR) Project funded by the
European Union. The Project is charged with developing new parallel
algorithms for the direct solution of sparse systems of equations and for
their preconditioning. These algorithms will be implemented as portable
prototypes and will be integrated into a public-domain Library. The
prototype codes will be extensively tested on a range of industrial

The work can commence immediately. Normally the position is for one year
with possible extensions to a second and even third year. Applicants should
be comfortable at programming in Fortran and preferably a familiarity with
message passing concepts. MPI will be principally used in the Project. A
background in numerical analysis and knowledge of numerical linear algebra is

Interested people should send an email message to Jennifer Scott
(, preferably including a short CV and contact
details for three people willing to provide a reference.

Applications can also be sent by regular mail to:

Dr Jennifer Scott
Rutherford Appleton laboratory
Chilton, Didcot

[Note that this LTR Project is the same as the one in which CERFACS is
involved and that there are two positions, one in each laboratory]


From: Iain Duff <>
Date: Fri, 22 Dec 95 15:17:26 GMT
Subject: Postdoctoral Position at CERFACS


A Postdoctoral Research position is available at CERFACS in Toulouse France
( to work in the Parallel Algorithms Project
( The post is supported by a Long Term
Research (LTR) Project funded by the European Union. The Project is charged
with developing new parallel algorithms for the direct solution of sparse
systems of equations and for their preconditioning. These algorithms will be
implemented as portable prototypes and will be integrated into a public-
domain Library. The prototype codes will be extensively tested on a range of
industrial applications.

The work can commence immediately. Normally the position is for one year
with possible extensions to a second and even third year. Applicants should
be comfortable at programming in Fortran and preferably a familiarity with
message passing concepts. MPI will be principally used in the Project. A
background in numerical analysis and knowledge of numerical linear algebra is

Interested people should send an email message to Chiara Puglisi
(, preferably including a short CV and contact
details for three people willing to provide a reference.

Applications can also be sent by regular mail to:

Dr Chiara Puglisi
Parallel Algorithms Project
42 Ave G Coriolis
31057 Toulouse Cedex
Fax: +33-61-19-30-00

[Note that this LTR Project is the same as the one in which the Rutherford
Appleton Laboratory is involved and that there are two positions, one in each


From: IEEE CS&E <>
Date: Mon, 18 Dec 95 17:25:25 est
Subject: Contents, IEEE Computational Science & Engineering

Vol. 2, No. 4, Winter 1995.

Editor-in-Chief through 12/31/95:
Ahmed Sameh, U. of Minnesota,
Editor-in-Chief as of 1/1/96:
George Cybenko, Dartmouth Coll.,
Associate EIC: Francis Sullivan, IDA Ctr. for Computing Sciences


Reinventing Electromagnetics: Emerging Applications for FD-TD Computation
Allen Taflove 24

Computational Science and the Future of Computing Research
John R. Rice 35

Theme Editor's Introduction: Computational Inverse Problems in
Christopher R. Johnson 42

A New Approach to Detecting Leukemia: Using Computational
David Colton and Peter Monk 46

Issues in Electrical Impedance Imaging
Margaret Cheney and David Isaacson 53

MRI-Guided Optical Tomography: Prospects and Computation for a New
Imaging Method
Randall L. Barbour, Harry L. Graber, Jenghwa Chang, San-Lian S.
Barbour, Ping C. Koo, and Raphael Aronson 63


Editor-in-Chief 1
About the Cover 4
Visualizing Geophysical Data: Teasing Meaning from Models
CSE Education 6
The Four A's of CSE Education
Awards Recognize Excellence and Innovation in CSE Education
Site Report 10
UVA Center for Computational Structures Technology
CSE at Work 15
Crunching Numbers, Shaping Steel
CSE in Europe 19
Europort's Parallelized Codes Yield Results
About the Articles 22
Product News 78
In the News 82
Thin-film lubricants may damage disk drives
Protein structures calculated quickly
Supercomputer looking for oil
Modeling ceramics may improve yields
Geomagnetic field reversals simulated
Distributed climate simulation
Smart guitars
Book News & Reviews 85
Studies in Computational Science: Parallel Programming Paradigms, by Per
Brinch Hansen; reviewed by Eric F. Van de Velde
Scientific Visualization: Advances and Challenges, edited by Lawrence J.
Rosenblum et al.; reviewed by Andrew J. Hanson
Literature: Random Access 89
Conferences & Workshops 90
First Int'l Conf. on Complementarity Problems
11th ACM Symp. on Computational Geometry
High-Performance Computational Chemistry Tutorial and Workshop
1995 IEEE Antennas and Propagation Society Int'l Symp.
1995 SIAM Annual Mtg.
Calendar & Call for Papers 93
1995 Annual Index 94
Editor's Perspective 96
Margaret H. Wright on Medical Applications and Optimization


From: Marilyn Radcliff <>
Date: Wed, 20 Dec 1995 09:39:12 -0500 (EST)
Subject: Contents, J. Approximation Theory

Table of Contents: J. Approx. Theory, Volume 83, Number 1, October 1995

Mourad E. H. Ismail and David R. Masson
Generalized orthogonality and continued fractions

A. Kro\'o and J. Szabados
Weighted polynomial approximation on the real line

Lefan Zhong and Lai-Yi Zhu
The Marcinkiewicz-Zygmund inequality on a smooth simple arc

R. C. Gayle
Inverse rational $L^1$ approximation

Dietrich Braess
Asymptotics for the approximation of wave functions by exponential sums

Dorothee Haroske
Approximation numbers in some weighted function spaces

Table of Contents: J. Approx. Theory, Volume 83, Number 2, November 1995

E. Belinskii
Functional classes with "dominated" mixed derivative and the

Kirill A. Kopotun
Coconvex polynomial approximation of twice differentiable functions

Dang Vu Giang and Ferenc M\'oricz
Strong approximation by Fourier transforms and Fourier series in

Dimitar K. Dimitrov
Markov inequalities for weight functions of Chebyshev type

S. L. Lee and Rohaisan Osman
Asymptotic formulas for convolution operators with spline kernels

Fathi B. Saidi
On the smoothness of the metric projection and its applications to
proximinality in $L^p(S, X)$

Rumen Kozarev
Reconstruction of functions on the basis of sequences of linear

Ian H. Sloan
Polynomial interpolation and hyperinterpolation over general regions

M. A. Bokhari
Interpolation mixed with $l_2$-approximation

M. Alfaro, F. Marcell\'an, M. L. Rezola, and A. Ronveaux
Sobolev--type orthogonal polynomials: the nondiagonal case

Table of Contents: J. Approx. Theory, Volume 83, Number 3, December 1995

Henrik Laurberg Pedersen
Stieltjes moment problems and the Friedrichs extension of a
positive--definite operator

K. A. Driver
Simultaneous rational approximants for a pair of functions with smooth
Maclaurin series coefficients

Paul Sablonniere
Positive Bernstein-Sheffer operators

Ulrich Schmid
On the approximation of positive functions by power series

M. I. Ganzburg and E. R. Liflyand
Estimates of best approximation and Fourier transforms in integral

R. K. Kovacheva
Zeros of Pad\'e error functions for functions with smooth Maclaurin

Leonid Golinskii, Paul Nevai, and Walter van Assche
Perturbation of orthogonal polynomials on an arc of the unit circle

Manfred Sommer and Hans Strauss
Interpolation by uni- and multivariate generalized splines

Author index for volume 83


End of NA Digest