From Thu Dec 23 20:40:56 2004 Date: Thu, 23 Dec 2004 20:40:56 -0600 (CST) From: Postmaster Subject: Message from mail server Content-Length: 94 Mime-Version: 1.0 Status: RO X-IMAP: 1103856056 38 Delete. This is a system message. --END+PSEUDO-- From owner-reliable_computing [at] interval [dot] louisiana.edu Fri Nov 1 08:28:08 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gA1ES7i17219 for reliable_computing-outgoing; Fri, 1 Nov 2002 08:28:07 -0600 (CST) Received: (from rbk5287@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gA1ES1t17213 for reliable_computing [at] interval [dot] louisiana.edu; Fri, 1 Nov 2002 08:28:01 -0600 (CST) Received: from lcyoung.math.wisc.edu (lcyoung.math.wisc.edu [144.92.166.90]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gA11Hj514737 for ; Thu, 31 Oct 2002 19:17:46 -0600 (CST) Received: from ultra7.math.wisc.edu (ultra7.math.wisc.edu [144.92.166.188]) by lcyoung.math.wisc.edu (8.11.6/8.11.6) with ESMTP id gA11GRX19728; Thu, 31 Oct 2002 19:16:27 -0600 (CST) Date: Thu, 31 Oct 2002 19:16:27 -0600 (CST) From: Hans Schneider To: NETS -- at-net , E-LETTER , Pradeep Misra , Shaun Fallat , "na.digest" , ipnet-digest [at] math [dot] msu.edu, wim@bell-labs.com, hjt [at] eos [dot] ncsu.edu, SMBnet [at] smb [dot] org, vkm [at] eedsp [dot] gatech.edu, reliable_computing [at] interval [dot] louisiana.edu Subject: LAA contents Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-MailScanner: Found to be clean Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: RO X-Status: $$$$ X-UID: 0000000001 Dear Net Organizer: Please circulate the attached LAA contents over your net. Thanks hans ********************************************************************* Hans Schneider Office: Home: Mathematics Department 910 S. Midvale Blvd. Van Vleck Hall Madison, WI 53711 USA University of Wisconsin 608-271-7252 480 Lincoln Drive Madison, WI 53706-1313 USA Email: hans [at] math [dot] wisc.edu Office Phone: 608-262-1402 WWW: http://www.math.wisc.edu/~hans Math Dept Phone: 608-263-3054 Math Dept Fax: 608-263-8891 ********************************************************************* Journal: Linear Algebra and its Applications ISSN : 0024-3795 Volume : 357 Issue : 1-3 Date : 15-Dec-2002 For more information about this journal visit: http://www.elsevier.com/locate/jnlnr/07738 ______________________________________________________ Table of Contents: Editorial board pp ii-iii Symmetric conference matrices and locally largest regular crosspolytopes in cubes A. Packer pp 1-13 Graded polynomial identities of matrices Y. Bahturin, V. Drensky pp 15-34 Normal matrices with a dominant eigenvalue and an eigenvector with no zero entries R.A. Horn pp 35-44 Products of commutators of transvections over local rings B. Zheng, H. You pp 45-57 On reduction of elements of the full matrix superalgebra to a block-diagonal form by conjugation I.M. Trishin pp 59-82 Stability of the feasible set for linear inequality systems: A carrier index set approach M.A. Lopez, M.J. Canovas, J. Parra pp 83-105 Spectrum and commutativity preserving mappings on triangular matrices T. Petek pp 107-122 Additivity of Jordan maps on standard operator algebras F. Lu pp 123-131 Inertially arbitrary (2r-1)-diagonal sign patterns Z. Miao, J. Li pp 133-141 Stability theory for linear dissipative Hamiltonian systems S.-J. Chern pp 143-162 A PLU-factorization of rectangular matrices by the Neville elimination M. Gasso, J.R. Torregrosa pp 163-171 Meromorphic interpolation in several variables G. Popescu pp 173-196 Linear transformations between matrix spaces that map one rank specific set into another C.-K. Li, L. Rodman, P. Semrl pp 197-208 When is HyperlatT=Hyperlatf(T) in finite dimension? G.T. Prajitura pp 209-215 Global monotone convergence of Newton iteration for a nonlinear eigen-problem Y.S. Choi, I. Koltracht, P.J. McKenna, N. Savytska pp 217-228 Small transitive families of subspaces in finite dimensions M.S. Lambrou, W.E. Longstaff pp 229-245 Factorial Stirling matrix and related combinatorial sequences G.-S. Cheon, J.-S. Kim pp 247-258 Inversion of polynomial matrices via state-space J.C. Basilio pp 259-271 Patterns, linesums, and symmetry E.E. Eischen, C.R. Johnson, K. Lange, D.P. Stanford pp 273-289 An algorithm for a result on minimal polynomials S.D. Agashe pp 291-293 Numerical range of Aluthge transform of operator P.Y. Wu pp 295-298 A note on the minimal nonnegative solution of a nonsymmetric algebraic Riccati equation C.-H. Guo pp 299-302 Volume and variance in the linear statistical model I.C. Araujo, M.P. de Oliveira pp 303-306 Call for papers: special issue in honor of Heydar Radjavi pp 307-308 Author index pp 309-310 ________________________________________ For calls for papers for current LAA special issues, see http://www.math.wisc.edu/~hans/speciss.html . From owner-reliable_computing [at] interval [dot] louisiana.edu Fri Nov 1 11:12:47 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gA1HClo17710 for reliable_computing-outgoing; Fri, 1 Nov 2002 11:12:47 -0600 (CST) Received: from goedel.louisiana.edu (rbk5287 [at] goedel [dot] louisiana.edu [130.70.132.143]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gA1HCf517706 for ; Fri, 1 Nov 2002 11:12:42 -0600 (CST) Received: (from rbk5287@localhost) by goedel.louisiana.edu (8.11.3/8.11.3/ull-interval-math-mailhost_1.1) id gA1HCe816515 for reliable_computing [at] interval [dot] louisiana.edu; Fri, 1 Nov 2002 11:12:40 -0600 (CST) Received: from lcyoung.math.wisc.edu (lcyoung.math.wisc.edu [144.92.166.90]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gA11Hj514737 for ; Thu, 31 Oct 2002 19:17:46 -0600 (CST) Received: from ultra7.math.wisc.edu (ultra7.math.wisc.edu [144.92.166.188]) by lcyoung.math.wisc.edu (8.11.6/8.11.6) with ESMTP id gA11GRX19728; Thu, 31 Oct 2002 19:16:27 -0600 (CST) Date: Thu, 31 Oct 2002 19:16:27 -0600 (CST) From: Hans Schneider To: NETS -- at-net , E-LETTER , Pradeep Misra , Shaun Fallat , "na.digest" , ipnet-digest [at] math [dot] msu.edu, wim@bell-labs.com, hjt [at] eos [dot] ncsu.edu, SMBnet [at] smb [dot] org, vkm [at] eedsp [dot] gatech.edu, reliable_computing [at] interval [dot] louisiana.edu Subject: LAA contents Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-MailScanner: Found to be clean Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000002 Dear Net Organizer: Please circulate the attached LAA contents over your net. Thanks hans ********************************************************************* Hans Schneider Office: Home: Mathematics Department 910 S. Midvale Blvd. Van Vleck Hall Madison, WI 53711 USA University of Wisconsin 608-271-7252 480 Lincoln Drive Madison, WI 53706-1313 USA Email: hans [at] math [dot] wisc.edu Office Phone: 608-262-1402 WWW: http://www.math.wisc.edu/~hans Math Dept Phone: 608-263-3054 Math Dept Fax: 608-263-8891 ********************************************************************* Journal: Linear Algebra and its Applications ISSN : 0024-3795 Volume : 357 Issue : 1-3 Date : 15-Dec-2002 For more information about this journal visit: http://www.elsevier.com/locate/jnlnr/07738 ______________________________________________________ Table of Contents: Editorial board pp ii-iii Symmetric conference matrices and locally largest regular crosspolytopes in cubes A. Packer pp 1-13 Graded polynomial identities of matrices Y. Bahturin, V. Drensky pp 15-34 Normal matrices with a dominant eigenvalue and an eigenvector with no zero entries R.A. Horn pp 35-44 Products of commutators of transvections over local rings B. Zheng, H. You pp 45-57 On reduction of elements of the full matrix superalgebra to a block-diagonal form by conjugation I.M. Trishin pp 59-82 Stability of the feasible set for linear inequality systems: A carrier index set approach M.A. Lopez, M.J. Canovas, J. Parra pp 83-105 Spectrum and commutativity preserving mappings on triangular matrices T. Petek pp 107-122 Additivity of Jordan maps on standard operator algebras F. Lu pp 123-131 Inertially arbitrary (2r-1)-diagonal sign patterns Z. Miao, J. Li pp 133-141 Stability theory for linear dissipative Hamiltonian systems S.-J. Chern pp 143-162 A PLU-factorization of rectangular matrices by the Neville elimination M. Gasso, J.R. Torregrosa pp 163-171 Meromorphic interpolation in several variables G. Popescu pp 173-196 Linear transformations between matrix spaces that map one rank specific set into another C.-K. Li, L. Rodman, P. Semrl pp 197-208 When is HyperlatT=Hyperlatf(T) in finite dimension? G.T. Prajitura pp 209-215 Global monotone convergence of Newton iteration for a nonlinear eigen-problem Y.S. Choi, I. Koltracht, P.J. McKenna, N. Savytska pp 217-228 Small transitive families of subspaces in finite dimensions M.S. Lambrou, W.E. Longstaff pp 229-245 Factorial Stirling matrix and related combinatorial sequences G.-S. Cheon, J.-S. Kim pp 247-258 Inversion of polynomial matrices via state-space J.C. Basilio pp 259-271 Patterns, linesums, and symmetry E.E. Eischen, C.R. Johnson, K. Lange, D.P. Stanford pp 273-289 An algorithm for a result on minimal polynomials S.D. Agashe pp 291-293 Numerical range of Aluthge transform of operator P.Y. Wu pp 295-298 A note on the minimal nonnegative solution of a nonsymmetric algebraic Riccati equation C.-H. Guo pp 299-302 Volume and variance in the linear statistical model I.C. Araujo, M.P. de Oliveira pp 303-306 Call for papers: special issue in honor of Heydar Radjavi pp 307-308 Author index pp 309-310 ________________________________________ For calls for papers for current LAA special issues, see http://www.math.wisc.edu/~hans/speciss.html . From owner-reliable_computing [at] interval [dot] louisiana.edu Sat Nov 2 09:43:44 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gA2Fhha20517 for reliable_computing-outgoing; Sat, 2 Nov 2002 09:43:43 -0600 (CST) Received: from clmboh1-smtp5.columbus.rr.com (clmboh1-smtp5.columbus.rr.com [65.24.0.116]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gA2Fhb520513 for ; Sat, 2 Nov 2002 09:43:38 -0600 (CST) Received: from oemcomputer (dhcp065-024-174-102.columbus.rr.com [65.24.174.102]) by clmboh1-smtp5.columbus.rr.com (8.11.2/8.11.2) with SMTP id gA2Fhab09216 for ; Sat, 2 Nov 2002 10:43:36 -0500 (EST) Message-ID: <000c01c28285$adcf4b20$66ae1841 [at] columbus [dot] rr.com> From: "Ramon Moore" To: "interval" Subject: Sun interval compilers Date: Sat, 2 Nov 2002 10:36:57 -0500 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0009_01C2825B.C4A556C0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000003 This is a multi-part message in MIME format. ------=_NextPart_000_0009_01C2825B.C4A556C0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Dear Colleagues, I would like to hear from any or all who have had successes in = mathematics, science, engineering, business or industry using Sun's = interval compilers: 1.. Sun ONE Studio 7, Compiler Collection Fortran 95 Interval = Arithmetic Support 2.. Sun ONE Studio 7, Compiler Collection C++ Interval Arithmetic = Support Sincerely, Ramon Moore ------=_NextPart_000_0009_01C2825B.C4A556C0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
 Dear=20 Colleagues,
 
I would like to hear from any or all who have = had=20 successes in mathematics, science, engineering, business or=20 industry using Sun's interval compilers:
  1. Sun = ONE=20 Studio 7, Compiler Collection Fortran 95 Interval Arithmetic = Support
  2. Sun = ONE=20 Studio 7, Compiler Collection C++ Interval Arithmetic=20 Support
Sincerely,
 
Ramon Moore
 ------=_NextPart_000_0009_01C2825B.C4A556C0-- From owner-reliable_computing [at] interval [dot] louisiana.edu Sun Nov 3 09:54:27 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gA3FsRg23704 for reliable_computing-outgoing; Sun, 3 Nov 2002 09:54:27 -0600 (CST) Received: (from rbk5287@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gA3FsNE23699 for reliable_computing [at] interval [dot] louisiana.edu; Sun, 3 Nov 2002 09:54:23 -0600 (CST) Received: from lcyoung.math.wisc.edu (lcyoung.math.wisc.edu [144.92.166.90]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gA11Hj514737 for ; Thu, 31 Oct 2002 19:17:46 -0600 (CST) Received: from ultra7.math.wisc.edu (ultra7.math.wisc.edu [144.92.166.188]) by lcyoung.math.wisc.edu (8.11.6/8.11.6) with ESMTP id gA11GRX19728; Thu, 31 Oct 2002 19:16:27 -0600 (CST) Date: Thu, 31 Oct 2002 19:16:27 -0600 (CST) From: Hans Schneider To: NETS -- at-net , E-LETTER , Pradeep Misra , Shaun Fallat , "na.digest" , ipnet-digest [at] math [dot] msu.edu, wim@bell-labs.com, hjt [at] eos [dot] ncsu.edu, SMBnet [at] smb [dot] org, vkm [at] eedsp [dot] gatech.edu, reliable_computing [at] interval [dot] louisiana.edu Subject: LAA contents Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-MailScanner: Found to be clean Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000004 Dear Net Organizer: Please circulate the attached LAA contents over your net. Thanks hans ********************************************************************* Hans Schneider Office: Home: Mathematics Department 910 S. Midvale Blvd. Van Vleck Hall Madison, WI 53711 USA University of Wisconsin 608-271-7252 480 Lincoln Drive Madison, WI 53706-1313 USA Email: hans [at] math [dot] wisc.edu Office Phone: 608-262-1402 WWW: http://www.math.wisc.edu/~hans Math Dept Phone: 608-263-3054 Math Dept Fax: 608-263-8891 ********************************************************************* Journal: Linear Algebra and its Applications ISSN : 0024-3795 Volume : 357 Issue : 1-3 Date : 15-Dec-2002 For more information about this journal visit: http://www.elsevier.com/locate/jnlnr/07738 ______________________________________________________ Table of Contents: Editorial board pp ii-iii Symmetric conference matrices and locally largest regular crosspolytopes in cubes A. Packer pp 1-13 Graded polynomial identities of matrices Y. Bahturin, V. Drensky pp 15-34 Normal matrices with a dominant eigenvalue and an eigenvector with no zero entries R.A. Horn pp 35-44 Products of commutators of transvections over local rings B. Zheng, H. You pp 45-57 On reduction of elements of the full matrix superalgebra to a block-diagonal form by conjugation I.M. Trishin pp 59-82 Stability of the feasible set for linear inequality systems: A carrier index set approach M.A. Lopez, M.J. Canovas, J. Parra pp 83-105 Spectrum and commutativity preserving mappings on triangular matrices T. Petek pp 107-122 Additivity of Jordan maps on standard operator algebras F. Lu pp 123-131 Inertially arbitrary (2r-1)-diagonal sign patterns Z. Miao, J. Li pp 133-141 Stability theory for linear dissipative Hamiltonian systems S.-J. Chern pp 143-162 A PLU-factorization of rectangular matrices by the Neville elimination M. Gasso, J.R. Torregrosa pp 163-171 Meromorphic interpolation in several variables G. Popescu pp 173-196 Linear transformations between matrix spaces that map one rank specific set into another C.-K. Li, L. Rodman, P. Semrl pp 197-208 When is HyperlatT=Hyperlatf(T) in finite dimension? G.T. Prajitura pp 209-215 Global monotone convergence of Newton iteration for a nonlinear eigen-problem Y.S. Choi, I. Koltracht, P.J. McKenna, N. Savytska pp 217-228 Small transitive families of subspaces in finite dimensions M.S. Lambrou, W.E. Longstaff pp 229-245 Factorial Stirling matrix and related combinatorial sequences G.-S. Cheon, J.-S. Kim pp 247-258 Inversion of polynomial matrices via state-space J.C. Basilio pp 259-271 Patterns, linesums, and symmetry E.E. Eischen, C.R. Johnson, K. Lange, D.P. Stanford pp 273-289 An algorithm for a result on minimal polynomials S.D. Agashe pp 291-293 Numerical range of Aluthge transform of operator P.Y. Wu pp 295-298 A note on the minimal nonnegative solution of a nonsymmetric algebraic Riccati equation C.-H. Guo pp 299-302 Volume and variance in the linear statistical model I.C. Araujo, M.P. de Oliveira pp 303-306 Call for papers: special issue in honor of Heydar Radjavi pp 307-308 Author index pp 309-310 ________________________________________ For calls for papers for current LAA special issues, see http://www.math.wisc.edu/~hans/speciss.html . From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 4 09:44:12 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gA4FiBe26668 for reliable_computing-outgoing; Mon, 4 Nov 2002 09:44:12 -0600 (CST) Received: (from rbk5287@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gA4Fi8126663 for reliable_computing [at] interval [dot] louisiana.edu; Mon, 4 Nov 2002 09:44:08 -0600 (CST) Received: from lcyoung.math.wisc.edu (lcyoung.math.wisc.edu [144.92.166.90]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gA1Itt518021 for ; Fri, 1 Nov 2002 12:55:55 -0600 (CST) Received: from ultra8.math.wisc.edu (ultra8.math.wisc.edu [144.92.166.178]) by lcyoung.math.wisc.edu (8.11.6/8.11.6) with ESMTP id gA1IssX18958; Fri, 1 Nov 2002 12:54:54 -0600 (CST) Date: Fri, 1 Nov 2002 12:54:53 -0600 (CST) From: Hans Schneider To: NETS -- at-net , E-LETTER , Pradeep Misra , ipnet-digest [at] math [dot] msu.edu, wim@bell-labs.com, hjt [at] eos [dot] ncsu.edu, SMBnet [at] smb [dot] org, vkm [at] eedsp [dot] gatech.edu, reliable_computing [at] interval [dot] louisiana.edu, Richard Brualdi Subject: :LAA special issue Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=X-UNKNOWN X-MailScanner: Found to be clean Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from QUOTED-PRINTABLE to 8bit by interval.louisiana.edu id gA1Itu518022 Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000005 Dear Net Organizer: Please circulate the attached LAA announcment over your net. Thanks hans ********************************************************************* Hans Schneider Office: Home: Mathematics Department 910 S. Midvale Blvd. Van Vleck Hall Madison, WI 53711 USA University of Wisconsin 608-271-7252 480 Lincoln Drive Madison, WI 53706-1313 USA Email: hans [at] math [dot] wisc.edu Office Phone: 608-262-1402 WWW: http://www.math.wisc.edu/~hans Math Dept Phone: 608-263-3054 Math Dept Fax: 608-263-8891 ********************************************************************* LINEAR ALGEBRA AND ITS APPLICATIONS Special issue on Determinants and the Legacy of Sir Thomas Muir Thomas Muir was born in 1844 in Scotland, was educated at the University of Glasgow, and died in 1934 in South Africa. His monumental work "The Theory of Determinants in the Historical Order of Development" in 5 volumes was published from 1890 (volume 1) to 1930 (volume 5). It covers the history of determinants from its foundation by Leibniz (1693) and Cayley (1841) to 1920. A sixth volume was being prepared when Muir died in 1934. Determinants arise not only in linear algebra but in many other parts of mathematics and science, such as combinatorial enumeration, graph theory, representation theory, symmetric functions, statistics, number theory, interpolation and approximation, tilings, special function theory, statistical mechanics, and theoretical computer science. Entries of the associated matrices can vary from just 0's and 1's (or 0's, 1's and -1's) to multivariable polynomials to special functions to general functions. Matrices whose determinants are to be evaluated can be unstructured or highly structured (e.g. Laplace, Vandermonde, Hankel, Fredholm, Toeplitz). For this special issue, we seek papers that, to name a few possibilities, advance the theory of determinants, provide special formulas for determinants, use determinants crucially in the context of solving a problem in another field, and give new application of determinants. In editing this special issue we seek to honor the legacy of Muir as well as to showcase the central role of determinants in mathematics. All papers submitted must meet the publication standards of Linear Algebra and its Applications and will be refereed in the usual way. They should be submitted to one of the special editors of this issue listed below by November 30, 2003. Wayne Barrett Samad Hedayat Department of Mathematics Department of Mathematics, Brigham Young University Statistics & Computer Science Provo, UT 84602, USA 322 SEO, 851 S. Morgan St. wayne [at] math [dot] byu.edu University of Illinois at Chicago Chicago, IL 60607-7045, USA hedayat [at] uic [dot] edu Christian Krattenthaler Raphael Loewy Institut Girard Desargues Department of Mathematics Université Claude Bernard Lyon-I Technion - I.I.T. Bâtiment Braconnier Haifa 32000, ISRAEL 21 Avenue Claude Bernard loewy [at] techunix [dot] technion.ac.il F-69622 Villeurbanne Cedex, FRANCE kratt [at] euler [dot] univ-lyon1.fr From owner-reliable_computing [at] interval [dot] louisiana.edu Tue Nov 12 08:30:34 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gACEUX812719 for reliable_computing-outgoing; Tue, 12 Nov 2002 08:30:33 -0600 (CST) Received: (from rbk5287@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gACEUSU12714 for reliable_computing [at] interval [dot] louisiana.edu; Tue, 12 Nov 2002 08:30:28 -0600 (CST) Received: from imeil.udg.es (imeil.udg.es [130.206.45.97]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gABFRMi09987 for ; Mon, 11 Nov 2002 09:27:22 -0600 (CST) Received: from imeil.udg.es (localhost.localdomain [127.0.0.1]) by imeil.udg.es (8.11.6/out/otb) with ESMTP id gABMPXB18915 for ; Mon, 11 Nov 2002 16:25:33 -0600 Received: from eia.udg.es (silver.udg.es [130.206.129.16]) by imeil.udg.es (8.11.6/in/otb) with ESMTP id gABMPW318863; Mon, 11 Nov 2002 16:25:32 -0600 Received: from PCEIA000 (pceia000.udg.es [130.206.129.66]) by eia.udg.es (8.8.5/8.8.5) with ESMTP id QAA22565; Mon, 11 Nov 2002 16:28:34 GMT From: "Josep Vehi" To: , , , Subject: New action group on "interval Methods for Control" Date: Mon, 11 Nov 2002 16:26:48 +0100 Organization: Universitat de Girona Message-ID: <006301c28996$c98ce8f0$4281ce82@PCEIA000> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPart_000_0064_01C2899F.2B5150F0" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.2627 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Importance: Normal Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: RO X-Status: $$$$ X-UID: 0000000006 This is a multi-part message in MIME format. ------=_NextPart_000_0064_01C2899F.2B5150F0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Please, circulate this message among the persons who may be interested. =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Dear all, A new working group on "Interval Methods for Control" has been started. The group is part the IEEE Control Systems Society technical committee on Computer Aided Control System Design.=20 The group will be devoted to cover research and software development activities in the field of Interval Methods and their applications to Control. A very informative home page for this Action Group has been launch at http://eia.udg.es/ieee_interval/ , with links to active groups, current projects, available software and people working in this field. The goal of this site is to provide a convenient entry point into the world of interval methods and software for control. It provides links giving access to the up-to-date and most important information resources related to the topic: research projects that are underway, research groups and individuals who are active in the field, selected papers and software. It also provides the latest news about conferences, workshops, special issues and the like.=20 The current home page is still a very preliminary and incomplete one. Comments, news, suggestions and contributions are very welcome. Thanks in advance. Josep Veh=ED MICELab (http://eia.udg.es/mice) University of Girona. Campus Montilivi. Edifici P4. 17071 Girona, Spain=20 Tel: +34 972 418 495 Fax: +34 972 418 976 ------=_NextPart_000_0064_01C2899F.2B5150F0 Content-Type: text/x-vcard; name="Josep Vehi (vehi [at] eia [dot] udg.es).vcf" Content-Transfer-Encoding: quoted-printable Content-Disposition: attachment; filename="Josep Vehi (vehi [at] eia [dot] udg.es).vcf" BEGIN:VCARD VERSION:2.1 N:Vehi;Josep FN:Josep Vehi (vehi [at] eia [dot] udg.es) ORG:Universitat de Girona;D.E.I.A. TEL;WORK;VOICE:+34 972 418 495 TEL;WORK;FAX:+34 972 418 098 ADR;WORK;ENCODING=3DQUOTED-PRINTABLE:;;Campus Montilivi. Edifici = P-IV=3D0D=3D0A17071 Girona LABEL;WORK;ENCODING=3DQUOTED-PRINTABLE:Campus Montilivi. Edifici = P-IV=3D0D=3D0A17071 Girona ADR;POSTAL;ENCODING=3DQUOTED-PRINTABLE:;;Campus Montilivi. Edifici = P-IV=3D0D=3D0A17071 Girona LABEL;POSTAL;ENCODING=3DQUOTED-PRINTABLE:Campus Montilivi. Edifici = P-IV=3D0D=3D0A17071 Girona EMAIL;PREF;INTERNET:vehi [at] eia [dot] udg.es REV:20021015T080114Z END:VCARD ------=_NextPart_000_0064_01C2899F.2B5150F0-- From owner-reliable_computing [at] interval [dot] louisiana.edu Tue Nov 12 13:01:01 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gACJ10G13360 for reliable_computing-outgoing; Tue, 12 Nov 2002 13:01:00 -0600 (CST) Received: from nwkea-mail-2.sun.com (nwkea-mail-2.sun.com [192.18.42.14]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gACJ0qi13356 for ; Tue, 12 Nov 2002 13:00:56 -0600 (CST) Received: from engmail1mpk.Eng.Sun.COM ([129.146.1.45]) by nwkea-mail-2.sun.com (8.9.3+Sun/8.9.3) with ESMTP id LAA24176 for ; Tue, 12 Nov 2002 11:00:33 -0800 (PST) Received: from phys-mpkmaila (phys-mpkmaila.Eng.Sun.COM [129.146.18.131]) by engmail1mpk.Eng.Sun.COM (8.12.2+Sun/8.12.2/ENSMAIL,v2.2) with ESMTP id gACJ0XMq011834 for ; Tue, 12 Nov 2002 11:00:33 -0800 (PST) Received: from gww (gww.Eng.Sun.COM [129.146.78.116]) by mpkmail.eng.sun.com (iPlanet Messaging Server 5.2 Patch 1 (built Apr 2 2002)) with SMTP id <0H5H00EJZ8SVZR [at] mpkmail [dot] eng.sun.com> for reliable_computing [at] interval [dot] louisiana.edu; Tue, 12 Nov 2002 11:00:32 -0800 (PST) Date: Tue, 12 Nov 2002 11:00:32 -0800 (PST) From: William Walster Subject: Square Brackets in Standard Fortran To: reliable_computing [at] interval [dot] louisiana.edu Reply-to: William Walster Message-id: <0H5H00EK08SVZR [at] mpkmail [dot] eng.sun.com> MIME-version: 1.0 X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.4.2 SunOS 5.8 sun4u sparc Content-type: TEXT/plain; charset=us-ascii Content-transfer-encoding: 7BIT Content-MD5: 3jmc5/pAzgzAjFxDqgr4SA== Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: RO X-Status: $$$$ X-UID: 0000000007 Fellow Intervalers, If you believe it will be helpful to have square brackets available for intervals in Standard Fortran, now is the time to send email to the InterNational Committee for Information Technology Standards (incits). More information can be found at: www.ncits.org/public_review_process.html The deadline is Nov 18. Best regards, Bill ------------- Begin Forwarded Message ------------- Date: Tue, 12 Nov 2002 10:56:32 -0800 (PST) From: William Walster Subject: Square Brackets To: ddonovan [at] itic [dot] org, psa [at] ansi [dot] org Cc: Bill.Walster [at] Eng [dot] Sun.COM MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-MD5: 07nDvjTWjdW0JRyiAtIwLw== I am writing to request that the proposed use of square brackets "[.]" in the Fortran 2000 standard be reconsidered. It is difficult for me to believe that the proposed use of square brackets has higher priority than keeping them available to support interval data types. G. William (Bill) Walster, Ph.D. Interval Technology Engineering Manager Sun Microsystems, Inc. 16 Network Circle, MS UMPK16-304 Menlo Park, CA 94025 (650) 786-9004 Direct (650) 786-9551 Fax bill.walster [at] eng [dot] sun.com ------------- End Forwarded Message ------------- From owner-reliable_computing [at] interval [dot] louisiana.edu Tue Nov 12 17:09:43 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gACN9hk13888 for reliable_computing-outgoing; Tue, 12 Nov 2002 17:09:43 -0600 (CST) Received: from nwkea-mail-2.sun.com (nwkea-mail-2.sun.com [192.18.42.14]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gACN9bi13883 for ; Tue, 12 Nov 2002 17:09:38 -0600 (CST) Received: from engmail1mpk.Eng.Sun.COM ([129.146.1.45]) by nwkea-mail-2.sun.com (8.9.3+Sun/8.9.3) with ESMTP id PAA06731 for ; Tue, 12 Nov 2002 15:09:32 -0800 (PST) Received: from phys-mpkmaila (phys-mpkmaila.Eng.Sun.COM [129.146.18.131]) by engmail1mpk.Eng.Sun.COM (8.12.2+Sun/8.12.2/ENSMAIL,v2.2) with ESMTP id gACN9VMq028600 for ; Tue, 12 Nov 2002 15:09:31 -0800 (PST) Received: from gww (gww.Eng.Sun.COM [129.146.78.116]) by mpkmail.eng.sun.com (iPlanet Messaging Server 5.2 Patch 1 (built Apr 2 2002)) with SMTP id <0H5H00C44KBUXW [at] mpkmail [dot] eng.sun.com> for reliable_computing [at] interval [dot] louisiana.edu; Tue, 12 Nov 2002 15:09:30 -0800 (PST) Date: Tue, 12 Nov 2002 15:09:31 -0800 (PST) From: William Walster Subject: The URL is www.ncits.org/public_review_process.htm To: reliable_computing [at] interval [dot] louisiana.edu, Bill.Walster [at] Eng [dot] Sun.COM Reply-to: William Walster Message-id: <0H5H00C45KBUXW [at] mpkmail [dot] eng.sun.com> MIME-version: 1.0 X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.4.2 SunOS 5.8 sun4u sparc Content-type: TEXT/plain; charset=us-ascii Content-transfer-encoding: 7BIT Content-MD5: bWybm95BIgfuJoAwiJWzfg== Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000008 not ".html". Sorry for the mistake. Thanks to Arnold Neumaier for pointing this out. Bill >Date: Tue, 12 Nov 2002 11:00:32 -0800 (PST) >From: William Walster >Subject: Square Brackets in Standard Fortran >To: reliable_computing [at] interval [dot] louisiana.edu >MIME-version: 1.0 >Content-transfer-encoding: 7BIT >Content-MD5: 3jmc5/pAzgzAjFxDqgr4SA== > > >Fellow Intervalers, > >If you believe it will be helpful to have square brackets available for >intervals in Standard Fortran, now is the time to send email to the >InterNational Committee for Information Technology Standards (incits). > >More information can be found at: > > www.ncits.org/public_review_process.html > >The deadline is Nov 18. > >Best regards, > >Bill > >------------- Begin Forwarded Message ------------- > >Date: Tue, 12 Nov 2002 10:56:32 -0800 (PST) >From: William Walster >Subject: Square Brackets >To: ddonovan [at] itic [dot] org, psa [at] ansi [dot] org >Cc: Bill.Walster [at] Eng [dot] Sun.COM >MIME-version: 1.0 >Content-transfer-encoding: 7BIT >Content-MD5: 07nDvjTWjdW0JRyiAtIwLw== > > >I am writing to request that the proposed use of square brackets "[.]" >in the Fortran 2000 standard be reconsidered. It is difficult for me >to believe that the proposed use of square brackets has higher priority >than keeping them available to support interval data types. > > > G. William (Bill) Walster, Ph.D. > Interval Technology Engineering Manager > Sun Microsystems, Inc. > 16 Network Circle, MS UMPK16-304 > Menlo Park, CA 94025 > (650) 786-9004 Direct > (650) 786-9551 Fax > > bill.walster [at] eng [dot] sun.com > > > >------------- End Forwarded Message ------------- > > From owner-reliable_computing [at] interval [dot] louisiana.edu Wed Nov 13 09:25:46 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gADFPke16112 for reliable_computing-outgoing; Wed, 13 Nov 2002 09:25:46 -0600 (CST) Received: from marnier.ucs.louisiana.edu (root [at] marnier [dot] ucs.louisiana.edu [130.70.132.233]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gADFPfi16108 for ; Wed, 13 Nov 2002 09:25:41 -0600 (CST) Received: from liberty (liberty.louisiana.edu [130.70.46.171]) by marnier.ucs.louisiana.edu (8.11.3/8.11.3/ull-ucs-mx-host_1.6) with SMTP id gADFPeF28045 for ; Wed, 13 Nov 2002 09:25:40 -0600 (CST) Message-Id: <2.2.32.20021113152709.016c1c10 [at] pop [dot] louisiana.edu> X-Sender: rbk5287 [at] pop [dot] louisiana.edu X-Mailer: Windows Eudora Pro Version 2.2 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 13 Nov 2002 09:27:09 -0600 To: reliable_computing [at] interval [dot] louisiana.edu From: "R. Baker Kearfott" Subject: Re: square brackets in Fortran Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000009 It looks like I needed to stay on the committee!!! :-( I encourage EVERYONE to voice their opinion. Sincerely, Baker At 03:09 PM 11/12/2002 -0800, William Walster wrote: > >not ".html". > >Sorry for the mistake. > >Thanks to Arnold Neumaier for pointing this out. > >Bill > > >Date: Tue, 12 Nov 2002 11:00:32 -0800 (PST) > >From: William Walster > >Subject: Square Brackets in Standard Fortran > >To: reliable_computing [at] interval [dot] louisiana.edu > >MIME-version: 1.0 > >Content-transfer-encoding: 7BIT > >Content-MD5: 3jmc5/pAzgzAjFxDqgr4SA== > > > > > >Fellow Intervalers, > > > >If you believe it will be helpful to have square brackets available for > >intervals in Standard Fortran, now is the time to send email to the > >InterNational Committee for Information Technology Standards (incits). > > > >More information can be found at: > > > > www.ncits.org/public_review_process.html > > > >The deadline is Nov 18. > > > >Best regards, > > > >Bill > > > >------------- Begin Forwarded Message ------------- > > > >Date: Tue, 12 Nov 2002 10:56:32 -0800 (PST) > >From: William Walster > >Subject: Square Brackets > >To: ddonovan [at] itic [dot] org, psa [at] ansi [dot] org > >Cc: Bill.Walster [at] Eng [dot] Sun.COM > >MIME-version: 1.0 > >Content-transfer-encoding: 7BIT > >Content-MD5: 07nDvjTWjdW0JRyiAtIwLw== > > > > > >I am writing to request that the proposed use of square brackets "[.]" > >in the Fortran 2000 standard be reconsidered. It is difficult for me > >to believe that the proposed use of square brackets has higher priority > >than keeping them available to support interval data types. > > > > > > G. William (Bill) Walster, Ph.D. > > Interval Technology Engineering Manager > > Sun Microsystems, Inc. > > 16 Network Circle, MS UMPK16-304 > > Menlo Park, CA 94025 > > (650) 786-9004 Direct > > (650) 786-9551 Fax > > > > bill.walster [at] eng [dot] sun.com > > > > > > > >------------- End Forwarded Message ------------- > > > > > > --------------------------------------------------------------- R. Baker Kearfott, rbk [at] louisiana [dot] edu (337) 482-5346 (fax) (337) 482-5270 (work) (337) 981-9744 (home) URL: http://interval.louisiana.edu/kearfott.html Department of Mathematics, University of Louisiana at Lafayette Box 4-1010, Lafayette, LA 70504-1010, USA --------------------------------------------------------------- From owner-reliable_computing [at] interval [dot] louisiana.edu Wed Nov 13 11:49:38 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gADHnbE16485 for reliable_computing-outgoing; Wed, 13 Nov 2002 11:49:37 -0600 (CST) Received: from clmboh1-smtp5.columbus.rr.com (clmboh1-smtp5.columbus.rr.com [65.24.0.116]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gADHnWi16481 for ; Wed, 13 Nov 2002 11:49:32 -0600 (CST) Received: from oemcomputer (dhcp065-024-174-102.columbus.rr.com [65.24.174.102]) by clmboh1-smtp5.columbus.rr.com (8.11.2/8.11.2) with SMTP id gADHhsO04604; Wed, 13 Nov 2002 12:44:00 -0500 (EST) Message-ID: <004401c28b3b$fdc9fe40$66ae1841 [at] columbus [dot] rr.com> From: "Ramon Moore" To: "Josep Vehi" , , , References: <006301c28996$c98ce8f0$4281ce82@PCEIA000> Subject: Re: New action group on "interval Methods for Control" Date: Wed, 13 Nov 2002 12:42:03 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000010 Dear Josep and all, This is an excellent project ! One suggestion I have is to add, if possible, a link to an updatable list of electronically available publications in the area your group is working on: "Interval Methods for Control". It would be helpful also, if you could point out a few introductory articles for those who are not yet expert in that area, but want to learn. best wishes, Ramon Moore ----- Original Message ----- From: "Josep Vehi" To: ; ; ; Sent: Monday, November 11, 2002 10:26 AM Subject: New action group on "interval Methods for Control" Please, circulate this message among the persons who may be interested. ======================================================================== Dear all, A new working group on "Interval Methods for Control" has been started. The group is part the IEEE Control Systems Society technical committee on Computer Aided Control System Design. The group will be devoted to cover research and software development activities in the field of Interval Methods and their applications to Control. A very informative home page for this Action Group has been launch at http://eia.udg.es/ieee_interval/ , with links to active groups, current projects, available software and people working in this field. The goal of this site is to provide a convenient entry point into the world of interval methods and software for control. It provides links giving access to the up-to-date and most important information resources related to the topic: research projects that are underway, research groups and individuals who are active in the field, selected papers and software. It also provides the latest news about conferences, workshops, special issues and the like. The current home page is still a very preliminary and incomplete one. Comments, news, suggestions and contributions are very welcome. Thanks in advance. Josep Vehí MICELab (http://eia.udg.es/mice) University of Girona. Campus Montilivi. Edifici P4. 17071 Girona, Spain Tel: +34 972 418 495 Fax: +34 972 418 976 From owner-reliable_computing [at] interval [dot] louisiana.edu Thu Nov 14 17:57:21 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAENvLq19948 for reliable_computing-outgoing; Thu, 14 Nov 2002 17:57:21 -0600 (CST) Received: from imf38bis.bellsouth.net (mail121.mail.bellsouth.net [205.152.58.61]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAENvFi19944 for ; Thu, 14 Nov 2002 17:57:15 -0600 (CST) Received: from u8174 ([65.83.166.143]) by imf38bis.bellsouth.net (InterMail vM.5.01.04.19 201-253-122-122-119-20020516) with SMTP id <20021114235856.FWPZ14859.imf38bis.bellsouth.net@u8174> for ; Thu, 14 Nov 2002 18:58:56 -0500 Message-Id: <2.2.32.20021114235713.009fe024 [at] pop [dot] louisiana.edu> X-Sender: rbk5287 [at] pop [dot] louisiana.edu X-Mailer: Windows Eudora Pro Version 2.2 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Thu, 14 Nov 2002 17:57:13 -0600 To: reliable_computing [at] interval [dot] louisiana.edu From: "R. Baker Kearfott" Subject: Re: Fortran's use of "[" and "]" other than for intervals Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000011 Colleagues, I am addressing the issue Bill brought up a couple of days ago: > > >Date: Tue, 12 Nov 2002 11:00:32 -0800 (PST) > > >From: William Walster > > >Subject: Square Brackets in Standard Fortran > > >To: reliable_computing [at] interval [dot] louisiana.edu > > > > > > > > >Fellow Intervalers, > > > > > >If you believe it will be helpful to have square brackets available for > > >intervals in Standard Fortran, now is the time to send email to the > > >InterNational Committee for Information Technology Standards (incits). > > > > > >More information can be found at: > > > > > > www.ncits.org/public_review_process.html (The preceeding URL is incorrect; see below.) > > > > > >The deadline is Nov 18. > > > > > >Best regards, > > > > > >Bill > It is interesting that square brackets are introduced at this point, since, in the past, the committee had been VERY conservative about introducing new symbols. In any case, it seems to me that the use for array constructors is nowhere near as close to "accepted practice" as is the use of "[" and "]" for intervals. I suggest we say as much in public comment. The draft standard can be found at http://std.dkuug.dk/jtc1/sc22/open/n3501.pdf The relevant section of 4.8, pages 63ff. It would be easy for the Fortran committee to find alternate symbols for array constructors, but more awkward for us. Let's say as much (if, of course, you agree with me). You can register your public comment by going to http://www.ncits.org/prsweb.htm then going to the part with Committee: J3 Approval of a Standard. CD 1539-1 You can send your comment by clicking on ddonovan [at] itic [dot] org There is also a link where you can see previous peoples' comments. I suggest that, when we request reservation of "[" and "]", we suggest alternatives that would be reasonable for array constructors. Best regards, Baker --------------------------------------------------------------- R. Baker Kearfott, rbk [at] louisiana [dot] edu (337) 482-5346 (fax) (337) 482-5270 (work) (337) 981-9744 (home) URL: http://interval.louisiana.edu/kearfott.html Department of Mathematics, University of Louisiana at Lafayette Box 4-1010, Lafayette, LA 70504-1010, USA --------------------------------------------------------------- From owner-reliable_computing [at] interval [dot] louisiana.edu Fri Nov 15 01:57:49 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAF7vmu20949 for reliable_computing-outgoing; Fri, 15 Nov 2002 01:57:48 -0600 (CST) Received: from mailbox.univie.ac.at (mailbox.univie.ac.at [131.130.1.27]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAF7vei20945 for ; Fri, 15 Nov 2002 01:57:41 -0600 (CST) Received: from univie.ac.at (hektor.mat.univie.ac.at [131.130.16.21]) by mailbox.univie.ac.at (8.12.2/8.12.2) with ESMTP id gAF7vLj8113316; Fri, 15 Nov 2002 08:57:24 +0100 Message-ID: <3DD4A8E1.16FF52F7 [at] univie [dot] ac.at> Date: Fri, 15 Nov 2002 08:57:21 +0100 From: Arnold Neumaier Organization: University of Vienna X-Mailer: Mozilla 4.79 [en] (X11; U; Linux 2.4.18-10 i686) X-Accept-Language: en, de MIME-Version: 1.0 To: ddonovan [at] itic [dot] org, psa [at] ansi [dot] org CC: bill.walster [at] eng [dot] sun.com Subject: Square Brackets in Standard Fortran Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000012 I would like to support Bill Walster's request that square brackets are not used to denote subsidiary issues, conflicting with the important (and mathematically established) usage to denote intervals. As a major coordinator of activities in the field of global optimization, where interval techniques play a crucial role, and where more and more applied problems (e.g., FEM calculations in structural engineering) have to consider the effects of bounded data uncertainties (specified by intervals), it would be a disservice to the community to disable the natural format of specifying intervals by the proposed standard. Yours sincerely, Arnold Neumaier Professor of Computational Mathematics University of Vienna, Austria http://www.mat.univie.ac.at/~neum http://www.mat.univie.ac.at/coconut > >I am writing to request that the proposed use of square brackets "[.]" >in the Fortran 2000 standard be reconsidered. It is difficult for me >to believe that the proposed use of square brackets has higher priority >than keeping them available to support interval data types. > > > G. William (Bill) Walster, Ph.D. > Interval Technology Engineering Manager > Sun Microsystems, Inc. > 16 Network Circle, MS UMPK16-304 > Menlo Park, CA 94025 > (650) 786-9004 Direct > (650) 786-9551 Fax > > bill.walster [at] eng [dot] sun.com > From owner-reliable_computing [at] interval [dot] louisiana.edu Fri Nov 15 18:04:20 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAG04KC22983 for reliable_computing-outgoing; Fri, 15 Nov 2002 18:04:20 -0600 (CST) Received: from mailgate.urz.tu-dresden.de (mailgate.urz.tu-dresden.de [141.30.66.154]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAG04Ci22979 for ; Fri, 15 Nov 2002 18:04:13 -0600 (CST) Received: from [141.30.70.195] (helo=NBTS02.math.tu-dresden.de) by mailgate.urz.tu-dresden.de with esmtp (exim-3.32) for id 18CqRS-0000vY-00; Sat, 16 Nov 2002 01:04:06 +0100 Received: from NWRW09.math.tu-dresden.de (NWRW09 [141.30.70.9]) by NBTS02.math.tu-dresden.de (8.9.3+Sun/8.9.1) with ESMTP id BAA15823; Sat, 16 Nov 2002 01:04:05 +0100 (MET) From: Wolfgang Walter Received: (from wwalter@localhost) by NWRW09.math.tu-dresden.de (8.10.2+Sun/8.9.1) id gAG044b04430; Sat, 16 Nov 2002 01:04:04 +0100 (MET) Message-Id: <200211160004.gAG044b04430 [at] NWRW09 [dot] math.tu-dresden.de> Subject: Re: Fortran's use of "[" and "]" other than for intervals To: reliable_computing [at] interval [dot] louisiana.edu Date: Sat, 16 Nov 2002 01:04:04 +0100 (MET) Cc: wwalter [at] math [dot] tu-dresden.de In-Reply-To: <2.2.32.20021114235713.009fe024 [at] pop [dot] louisiana.edu> from "R. Baker Kearfott" at Nov 14, 2002 05:57:13 PM X-Mailer: ELM [version 2.5 PL2] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000013 Dear Friends of Intervals, I fully agree with Bill Walster's request and with the additional comments made by Baker Kearfott and Arnold Neumaier. I would like to stress that there is absolutely NO technical reason why Fortran needs yet another purely syntactic variant of an existing, well established and perfectly good notation (there are too many ways to say the same things already) --- especially since the proposed syntax is really a purely LEXICAL variant of a notation that was only introduced by the Fortran 90 standard (the beginning of the "modern" era for Fortran). In any case, just two simple global change commands in any editor will suffice to perform this local, context-free transformation: '[' -> '(/' and ']' -> '/)' --- if you are willing to accept that these changes will also be made in character literals and in comments. Here are some additional arguments against an alternative notation for array constructors, and against square brackets in particular: 1. Most syntactic variants were introduced into Fortran in order to improve the readability, reliability and security of Fortran code, not just the notation of an old (F66 or F77) feature/concept. In fact, they often led to important generalizations or other substantial enhancements of the functionality of the language. Noteworthy examples of this are ends of DO loops (and other block-structured syntax elements), attributes in type declarations, kind parameters, etc. . In the present case I cannot discern an enhancement of the existing language in any respect. Also, the proposed use of square brackets is highly unvonventional in mathematics and unusual in general programming languages (specialized mathematical systems such as Maple are an exception). 2. The Fortran committees have repeatedly rejected new ways of saying the same thing, especially if it was only a different spelling. The long discussions about CONSTANT as an alternative spelling for the PARAMETER attribute spring to mind. It was rejected in the late phases of defining Fortran 90, and I believe it was discussed and rejected again for Fortran 95. 3. Some of the most popular languages around, among them C, C++, and Java, use CURLY BRACES { } to enclose the array elements in an array notation, and that is not the only use of braces in these languages. In these and many other languages since Algol 60, square brackets [ ] are used for array dimensioning and indexing, but NOT for array construction. Pascal also uses square brackets to construct sets, but not arrays (Pascal does not provide array constructors). Modula-2 does not provide array constructors either, but switches to braces for set construction, having discontinued the use of braces for comments. Square brackets are used for subrange types and for array indexing in Modula-2. Interestingly enough, Ada folows the example of Algol 68 and uses normal parentheses ( ) for array constructors. However, this is not an option in Fortran since it would result in ambiguities. 4. Array constructors were only introduced into Fortran through the Fortran 90 standard, and the decision then was that Fortran (or computers or the world) were not ready to use previously unused characters for this purpose. What has changed since the late 80's? Not much, I would say, except that the percentage of systems that do not have square brackets and curly braces in their character set has dwindled even further. Is that a reason to use these characters now? Not necessarily, but their use should certainly be considered. If Fortran really wants to, after almost 50 years of existence, start using these "new" characters, then it should do so very carefully and very deliberately, and with the consciousness that [ ] and { } are probably the ONLY bracketing characters besides ( ) there will ever be (for all practical purposes). Does Fortran really want to waste one of these two pairs of symbols on a redundant notation, thus precluding much more interesting uses, e.g. in parallel and high performance computing, for intervals, etc. ? I do not see Fortran ready to take this historic step at this time. Before making such an irreversible decision, I would like to see a thorough investigation and discussion of the potential uses of these "precious" symbols. With best regards, Wolfgang Walter Baker Kearfott wrote: > > It is interesting that square brackets are introduced at this point, > since, in the past, the committee had been VERY conservative about > introducing new symbols. In any case, it seems to me that the > use for array constructors is nowhere near as close to "accepted > practice" as is the use of "[" and "]" for intervals. I suggest we > say as much in public comment. > > The draft standard can be found at > > http://std.dkuug.dk/jtc1/sc22/open/n3501.pdf > > The relevant section of 4.8, pages 63ff. It would be easy for the > Fortran committee to find alternate symbols for array constructors, > but more awkward for us. Let's say as much (if, of course, you agree > with me). > > [ ... parts deleted ... ] > > I suggest that, when we request reservation of "[" and "]", we > suggest alternatives that would be reasonable for array constructors. > > Best regards, > > Baker ======================================================================= Wolfgang V. Walter Technische Universitaet Dresden Institut f. Wissenschaftliches Rechnen Fachrichtung Mathematik D - 01062 Dresden Germany Tel: +49 - 351 / 463-33996 or -34266 (secr.) FAX: +49 - 351 / 463-37096 e-mail: wwalter [at] math [dot] tu-dresden.de From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 05:05:37 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIB5aW00631 for reliable_computing-outgoing; Mon, 18 Nov 2002 05:05:36 -0600 (CST) Received: from interferon.mpi-sb.mpg.de (mail [at] mail [dot] mpi-sb.mpg.de [139.19.1.1]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAIB5Vi00626 for ; Mon, 18 Nov 2002 05:05:31 -0600 (CST) Received: from amavis by interferon.mpi-sb.mpg.de with scanned-ok (Exim 3.12 #1 (Debian)) id 18Djic-00083Y-00 for ; Mon, 18 Nov 2002 12:05:30 +0100 Received: from data.mpi-sb.mpg.de ([::ffff:139.19.90.238] helo=mpi-sb.mpg.de) by interferon.mpi-sb.mpg.de with esmtp (Exim 3.12 #1 (Debian)) id 18DjiZ-00083P-00 for ; Mon, 18 Nov 2002 12:05:27 +0100 Received: from contact.mpi-sb.mpg.de (contact [139.19.2.26]) by mpi-sb.mpg.de (8.9.3/8.9.3) with ESMTP id MAA26881; Mon, 18 Nov 2002 12:05:27 +0100 (MET) Received: (from ratschan@localhost) by contact.mpi-sb.mpg.de (8.11.6+Sun/8.9.3) id gAIB5Rg29319; Mon, 18 Nov 2002 12:05:27 +0100 (MET) Date: Mon, 18 Nov 2002 12:05:27 +0100 (MET) Message-Id: <200211181105.gAIB5Rg29319 [at] contact [dot] mpi-sb.mpg.de> From: Stefan Ratschan To: reliable_computing [at] interval [dot] louisiana.edu Subject: Solutions of systems of equations X-Virus-Scanned: by AMaViS perl-11 Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000014 Dear Colleagues, one of the most basic tasks of our field is to find solutions of non-linear systems of equations. However, I was very surprised that, having checked several textbooks, I could not find a single algorithm for which it is proven that it does the following: Input: A system of non-linear equations F=0, and a box B Output: Yes, if F=0 contains a zero in B that does not vanish under perturbations, no, if F=0 does not contain zeros in B. Otherwise (there are some zeros in B, but all of them vanish under perturbations), the algorithm does not terminate. Of course the zeros might be arbitrarily ill-conditioned. However, this just requires the use of arbitrarily high floating-point precision. The algorithms that I found seem to verify solutions just sometimes, where "sometimes" is very often not explicitely defined, or an internal algorithmic condition that does not mean anything to a user. However, it might be that I overlooked something. So can anybody provide pointers to algorithms that solve the above problem or that might be useful? Best regards, Stefan Ratschan ----------------------------------- Stefan Ratschan WWW: http://www.mpi-sb.mpg.de/~ratschan Max-Planck-Institut fuer Informatik e-mail: stefan.ratschan@mpi-sb.mpg.de Stuhlsatzenhausenweg 85 phone: +49 681 9325 229 66123 Saarbruecken fax: +49 681 9325 299 Germany From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 06:59:23 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAICxN100967 for reliable_computing-outgoing; Mon, 18 Nov 2002 06:59:23 -0600 (CST) Received: from imf06bis.bellsouth.net (mail006.mail.bellsouth.net [205.152.58.26]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAICxIi00963 for ; Mon, 18 Nov 2002 06:59:18 -0600 (CST) Received: from u8174 ([68.155.33.221]) by imf06bis.bellsouth.net (InterMail vM.5.01.04.19 201-253-122-122-119-20020516) with SMTP id <20021118130059.ECHF10021.imf06bis.bellsouth.net@u8174>; Mon, 18 Nov 2002 08:00:59 -0500 Message-Id: <2.2.32.20021118125849.009de4ec [at] pop [dot] louisiana.edu> X-Sender: rbk5287 [at] pop [dot] louisiana.edu X-Mailer: Windows Eudora Pro Version 2.2 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 18 Nov 2002 06:58:49 -0600 To: Stefan Ratschan , reliable_computing [at] interval [dot] louisiana.edu From: "R. Baker Kearfott" Subject: Re: Solutions of systems of equations Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000015 Stefan, If I understand your question, I think it reduces to a fundamental problem with finite machines, namely, "is it possible to determine whether a particular quantity is equal to zero or not?" In particular, there are some problems for which a unique, non-singular solution exists within B, but for which either the solution is extremely close to the boundary of B or the Jacobi matrix is extremely ill-conditioned at B. No matter what finite-precision arithmetic system is used, there are such problems, for which the accuracy of the computations is not able to distinguish whether the zero is within the box or not. Thinking of the computer as a telescope, you can think of it as not being powerful enough to resolve the differences. A bit of theory has been developed concerning this situation. (Can someone help me with references?) Thus, any algorithm that finds a zero must have three outputs: 1. Yes, there is a unique zero; 2. No, there is no zero; 3. There MIGHT be a unique zero. Better algorithms return with the third alternative less often (for a given machine precision), but they must still have that alternative. Of course, symbolic and exact methods avoid the problems with finite precision arithmetic. However, they are not general either, and thus they fail to give definitive answers in other ways. Colleagues, please correct me if I have made any inaccurate statements. Best regards, Baker At 12:05 PM 11/18/2002 +0100, Stefan Ratschan wrote: > >Dear Colleagues, > >one of the most basic tasks of our field is to find solutions of >non-linear systems of equations. However, I was very surprised that, >having checked several textbooks, I could not find a single algorithm >for which it is proven that it does the following: > >Input: > A system of non-linear equations F=0, and > a box B >Output: > Yes, if F=0 contains a zero in B that does not vanish under perturbations, > no, if F=0 does not contain zeros in B. > >Otherwise (there are some zeros in B, but all of them vanish under >perturbations), the algorithm does not terminate. > >Of course the zeros might be arbitrarily ill-conditioned. However, >this just requires the use of arbitrarily high floating-point >precision. > >The algorithms that I found seem to verify solutions just sometimes, >where "sometimes" is very often not explicitely defined, or an >internal algorithmic condition that does not mean anything to a user. > >However, it might be that I overlooked something. > >So can anybody provide pointers to algorithms that solve the above >problem or that might be useful? > >Best regards, > > Stefan Ratschan > > >----------------------------------- >Stefan Ratschan WWW: http://www.mpi-sb.mpg.de/~ratschan >Max-Planck-Institut fuer Informatik e-mail: stefan.ratschan@mpi-sb.mpg.de >Stuhlsatzenhausenweg 85 phone: +49 681 9325 229 >66123 Saarbruecken fax: +49 681 9325 299 >Germany > > --------------------------------------------------------------- R. Baker Kearfott, rbk [at] louisiana [dot] edu (337) 482-5346 (fax) (337) 482-5270 (work) (337) 981-9744 (home) URL: http://interval.louisiana.edu/kearfott.html Department of Mathematics, University of Louisiana at Lafayette Box 4-1010, Lafayette, LA 70504-1010, USA --------------------------------------------------------------- From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 07:11:13 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIDBCv01104 for reliable_computing-outgoing; Mon, 18 Nov 2002 07:11:12 -0600 (CST) Received: from clmboh1-smtp5.columbus.rr.com (clmboh1-smtp5.columbus.rr.com [65.24.0.116]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAIDB7i01100 for ; Mon, 18 Nov 2002 07:11:07 -0600 (CST) Received: from oemcomputer (dhcp065-024-174-102.columbus.rr.com [65.24.174.102]) by clmboh1-smtp5.columbus.rr.com (8.11.2/8.11.2) with SMTP id gAIDArO18002; Mon, 18 Nov 2002 08:10:54 -0500 (EST) Message-ID: <002d01c28f03$e8a07960$66ae1841 [at] columbus [dot] rr.com> From: "Ramon Moore" To: "Stefan Ratschan" , , "R. Baker Kearfott" References: <2.2.32.20021118125849.009de4ec [at] pop [dot] louisiana.edu> Subject: Re: Solutions of systems of equations Date: Mon, 18 Nov 2002 08:10:44 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4133.2400 X-MIMEOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000016 See Oliver Aberth's book "Precise Numerical Methods Using C++", Academic Press 1998, where there are proofs of what is possible and what is not possible with finite machines. Ramon Moore ----- Original Message ----- From: "R. Baker Kearfott" To: "Stefan Ratschan" ; Sent: Monday, November 18, 2002 7:58 AM Subject: Re: Solutions of systems of equations > Stefan, > > If I understand your question, I think it reduces to a > fundamental problem with finite machines, namely, "is it possible > to determine whether a particular quantity is equal to zero > or not?" > > In particular, there are some problems for which a unique, non-singular > solution exists within B, but for which either the solution is > extremely close to the boundary of B or the Jacobi matrix is > extremely ill-conditioned at B. No matter what finite-precision > arithmetic system is used, there are such problems, for which > the accuracy of the computations is not able to distinguish whether > the zero is within the box or not. Thinking of the computer as > a telescope, you can think of it as not being powerful enough to > resolve the differences. > > A bit of theory has been developed concerning this situation. (Can > someone help me with references?) > > Thus, any algorithm that finds a zero must have three outputs: > > 1. Yes, there is a unique zero; > > 2. No, there is no zero; > > 3. There MIGHT be a unique zero. > > Better algorithms return with the third alternative less often > (for a given machine precision), but they must still have > that alternative. > > Of course, symbolic and exact methods avoid the problems with > finite precision arithmetic. However, they are not general > either, and thus they fail to give definitive answers in other > ways. > > Colleagues, please correct me if I have made any inaccurate statements. > > Best regards, > > Baker > > > At 12:05 PM 11/18/2002 +0100, Stefan Ratschan wrote: > > > >Dear Colleagues, > > > >one of the most basic tasks of our field is to find solutions of > >non-linear systems of equations. However, I was very surprised that, > >having checked several textbooks, I could not find a single algorithm > >for which it is proven that it does the following: > > > >Input: > > A system of non-linear equations F=0, and > > a box B > >Output: > > Yes, if F=0 contains a zero in B that does not vanish under perturbations, > > no, if F=0 does not contain zeros in B. > > > >Otherwise (there are some zeros in B, but all of them vanish under > >perturbations), the algorithm does not terminate. > > > >Of course the zeros might be arbitrarily ill-conditioned. However, > >this just requires the use of arbitrarily high floating-point > >precision. > > > >The algorithms that I found seem to verify solutions just sometimes, > >where "sometimes" is very often not explicitely defined, or an > >internal algorithmic condition that does not mean anything to a user. > > > >However, it might be that I overlooked something. > > > >So can anybody provide pointers to algorithms that solve the above > >problem or that might be useful? > > > >Best regards, > > > > Stefan Ratschan > > > > > >----------------------------------- > >Stefan Ratschan WWW: http://www.mpi-sb.mpg.de/~ratschan > >Max-Planck-Institut fuer Informatik e-mail: stefan.ratschan@mpi-sb.mpg.de > >Stuhlsatzenhausenweg 85 phone: +49 681 9325 229 > >66123 Saarbruecken fax: +49 681 9325 299 > >Germany > > > > > --------------------------------------------------------------- > R. Baker Kearfott, rbk [at] louisiana [dot] edu (337) 482-5346 (fax) > (337) 482-5270 (work) (337) 981-9744 (home) > URL: http://interval.louisiana.edu/kearfott.html > Department of Mathematics, University of Louisiana at Lafayette > Box 4-1010, Lafayette, LA 70504-1010, USA > --------------------------------------------------------------- > From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 07:19:03 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIDJ3c01228 for reliable_computing-outgoing; Mon, 18 Nov 2002 07:19:03 -0600 (CST) Received: from interferon.mpi-sb.mpg.de (mail [at] mail [dot] mpi-sb.mpg.de [139.19.1.1]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAIDIvi01224 for ; Mon, 18 Nov 2002 07:18:57 -0600 (CST) Received: from amavis by interferon.mpi-sb.mpg.de with scanned-ok (Exim 3.12 #1 (Debian)) id 18Dlnk-0005Os-00 for ; Mon, 18 Nov 2002 14:18:56 +0100 Received: from data.mpi-sb.mpg.de ([::ffff:139.19.90.238] helo=mpi-sb.mpg.de) by interferon.mpi-sb.mpg.de with esmtp (Exim 3.12 #1 (Debian)) id 18Dlni-0005Of-00; Mon, 18 Nov 2002 14:18:54 +0100 Received: from contact.mpi-sb.mpg.de (contact [139.19.2.26]) by mpi-sb.mpg.de (8.9.3/8.9.3) with ESMTP id OAA22220; Mon, 18 Nov 2002 14:18:53 +0100 (MET) Received: (from ratschan@localhost) by contact.mpi-sb.mpg.de (8.11.6+Sun/8.9.3) id gAIDIrR12963; Mon, 18 Nov 2002 14:18:53 +0100 (MET) Date: Mon, 18 Nov 2002 14:18:53 +0100 (MET) Message-Id: <200211181318.gAIDIrR12963 [at] contact [dot] mpi-sb.mpg.de> From: Stefan Ratschan To: reliable_computing [at] interval [dot] louisiana.edu CC: rbk [at] louisiana [dot] edu In-reply-to: <2.2.32.20021118125849.009de4ec [at] pop [dot] louisiana.edu> (rbk [at] louisiana [dot] edu) Subject: Re: Solutions of systems of equations References: <2.2.32.20021118125849.009de4ec [at] pop [dot] louisiana.edu> X-Virus-Scanned: by AMaViS perl-11 Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000017 Thanks, Baker, for taking the time to write such a fundamental answer. I think it will be very helpful for clarifying the situation. > there are some problems for which a unique, non-singular > solution exists within B, but for which either the solution is > extremely close to the boundary of B or the Jacobi matrix is > extremely ill-conditioned at B. That is true. However, "extremely close to the boundary" does not mean "on the boundary" and "extremely ill-conditioned" does not mean "ill-posed". > No matter what finite-precision > arithmetic system is used, there are such problems, for which > the accuracy of the computations is not able to distinguish whether > the zero is within the box or not. For extremely ill-conditioned, but still well-posed problems: Yes, if we fix the precision beforehand. But NO, if we are able to add precision during computation. > 1. Yes, there is a unique zero; > 2. No, there is no zero; > 3. There MIGHT be a unique zero. I agree (we might replace possibility 3 with non-termination). The challenge is to find algorithms for which 3 occurs only when really theoretically necessary and to prove this. My question was, whether the reliable computing community can provide this. Best regards, Stefan From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 08:27:52 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIERp101574 for reliable_computing-outgoing; Mon, 18 Nov 2002 08:27:51 -0600 (CST) Received: (from rbk5287@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIERlh01569 for reliable_computing [at] interval [dot] louisiana.edu; Mon, 18 Nov 2002 08:27:47 -0600 (CST) Received: from keen.esi.ac.at (keen.esi.ac.at [193.170.117.2]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAIDX2i01395 for ; Mon, 18 Nov 2002 07:33:03 -0600 (CST) Received: (from herman@localhost) by keen.esi.ac.at (8.11.6/8.11.6) id gAIDWsK26715; Mon, 18 Nov 2002 14:32:54 +0100 Date: Mon, 18 Nov 2002 14:32:54 +0100 From: Hermann Schichl To: "R. Baker Kearfott" Cc: Stefan Ratschan , reliable_computing [at] interval [dot] louisiana.edu Subject: Re: Solutions of systems of equations Message-ID: <20021118143254.A24038 [at] keen [dot] esi.ac.at> References: <2.2.32.20021118125849.009de4ec [at] pop [dot] louisiana.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: <2.2.32.20021118125849.009de4ec [at] pop [dot] louisiana.edu>; from rbk [at] louisiana [dot] edu on Mon, Nov 18, 2002 at 06:58:49AM -0600 Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000018 Hello, I hope I did understand everything correctly, but this information might help. On Mon, Nov 18, 2002 at 06:58:49AM -0600, R. Baker Kearfott wrote: > Stefan, > >> ... things deleted ... > > Thus, any algorithm that finds a zero must have three outputs: > > 1. Yes, there is a unique zero; > > 2. No, there is no zero; > > 3. There MIGHT be a unique zero. > > Better algorithms return with the third alternative less often > (for a given machine precision), but they must still have > that alternative. > Yes. There does not exist an algorithm, which solves all problems without having alternative 3. Such an algorithm would be, I think, equivalent to a 'halting tester', an algorithm which can be fed with programs and input and decides whether the program halts on the given input or not. Such an algorithm does not exist; this leads to a contradiction. Is there anything known about F except that it maps $\R^n\to\R^n$? Anyway, there are diophantine equations with one parameter $k$ such that it is dependend on the exact axioms for the natural numbers (e.g. Peano vs. construction via Zermelo-Fraenkel) whether the $k$-th equation has finitely many or infinitely many solutions. So there cannot be a finite state machine deciding which of both is true. * Algorithmic Information Theory, Gregory J. Chaitin, Cambridge University Press, 1987. * Randomness in Arithmetic, Gregory J. Chaitin, Scientific American 259, No. 1 (July 1988), pp. 80-85 Things like this extend to $\Q^n$ hence $\R^n$ perfectly well. > > Of course, symbolic and exact methods avoid the problems with > finite precision arithmetic. However, they are not general > either, and thus they fail to give definitive answers in other > ways. > So, even symbolic methods or arbitrarily high precision do not help. > > Colleagues, please correct me if I have made any inaccurate statements. > > Best regards, > > Baker > Best regards, Hermann > > At 12:05 PM 11/18/2002 +0100, Stefan Ratschan wrote: > > > >Dear Colleagues, > > > >one of the most basic tasks of our field is to find solutions of > >non-linear systems of equations. However, I was very surprised that, > >having checked several textbooks, I could not find a single algorithm > >for which it is proven that it does the following: > > > >Input: > > A system of non-linear equations F=0, and > > a box B > >Output: > > Yes, if F=0 contains a zero in B that does not vanish under perturbations, > > no, if F=0 does not contain zeros in B. > > > >Otherwise (there are some zeros in B, but all of them vanish under > >perturbations), the algorithm does not terminate. > > > >Of course the zeros might be arbitrarily ill-conditioned. However, > >this just requires the use of arbitrarily high floating-point > >precision. > > > >The algorithms that I found seem to verify solutions just sometimes, > >where "sometimes" is very often not explicitely defined, or an > >internal algorithmic condition that does not mean anything to a user. > > > >However, it might be that I overlooked something. > > > >So can anybody provide pointers to algorithms that solve the above > >problem or that might be useful? > > > >Best regards, > > > > Stefan Ratschan > > > > > >----------------------------------- > >Stefan Ratschan WWW: http://www.mpi-sb.mpg.de/~ratschan > >Max-Planck-Institut fuer Informatik e-mail: stefan.ratschan@mpi-sb.mpg.de > >Stuhlsatzenhausenweg 85 phone: +49 681 9325 229 > >66123 Saarbruecken fax: +49 681 9325 299 > >Germany > > > > > --------------------------------------------------------------- > R. Baker Kearfott, rbk [at] louisiana [dot] edu (337) 482-5346 (fax) > (337) 482-5270 (work) (337) 981-9744 (home) > URL: http://interval.louisiana.edu/kearfott.html > Department of Mathematics, University of Louisiana at Lafayette > Box 4-1010, Lafayette, LA 70504-1010, USA > --------------------------------------------------------------- From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 08:40:25 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIEeOb01744 for reliable_computing-outgoing; Mon, 18 Nov 2002 08:40:24 -0600 (CST) Received: from marnier.ucs.louisiana.edu (root [at] marnier [dot] ucs.louisiana.edu [130.70.132.233]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAIEeJi01740 for ; Mon, 18 Nov 2002 08:40:19 -0600 (CST) Received: from liberty (liberty.louisiana.edu [130.70.46.171]) by marnier.ucs.louisiana.edu (8.11.3/8.11.3/ull-ucs-mx-host_1.6) with SMTP id gAIEeFH05112; Mon, 18 Nov 2002 08:40:15 -0600 (CST) Message-Id: <2.2.32.20021118144157.0169eee4 [at] pop [dot] louisiana.edu> X-Sender: rbk5287 [at] pop [dot] louisiana.edu X-Mailer: Windows Eudora Pro Version 2.2 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 18 Nov 2002 08:41:57 -0600 To: Stefan Ratschan , reliable_computing [at] interval [dot] louisiana.edu From: "R. Baker Kearfott" Subject: Re: Solutions of systems of equations Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000019 Stefan et al, Please see my interpolated comment. At 02:18 PM 11/18/2002 +0100, Stefan Ratschan wrote: > >Thanks, Baker, for taking the time to write such a fundamental >answer. I think it will be very helpful for clarifying the situation. > >> there are some problems for which a unique, non-singular >> solution exists within B, but for which either the solution is >> extremely close to the boundary of B or the Jacobi matrix is >> extremely ill-conditioned at B. > >That is true. However, "extremely close to the boundary" does not mean >"on the boundary" and "extremely ill-conditioned" does not mean >"ill-posed". > >> No matter what finite-precision >> arithmetic system is used, there are such problems, for which >> the accuracy of the computations is not able to distinguish whether >> the zero is within the box or not. > >For extremely ill-conditioned, but still well-posed problems: Yes, if >we fix the precision beforehand. But NO, if we are able to add >precision during computation. > Yes, that's true. If we have a bound on the precision, then there are problems that cannot be determined. However, even if we have no bound on the precision, there are still problems that we cannot determine, if we have a bound on the execution time. Is that not true? Best regards, Baker >> 1. Yes, there is a unique zero; >> 2. No, there is no zero; >> 3. There MIGHT be a unique zero. > >I agree (we might replace possibility 3 with non-termination). >The challenge is to find algorithms for which 3 occurs only when >really theoretically necessary and to prove this. My question was, >whether the reliable computing community can provide this. > >Best regards, > > Stefan > > --------------------------------------------------------------- R. Baker Kearfott, rbk [at] louisiana [dot] edu (337) 482-5346 (fax) (337) 482-5270 (work) (337) 981-9744 (home) URL: http://interval.louisiana.edu/kearfott.html Department of Mathematics, University of Louisiana at Lafayette Box 4-1010, Lafayette, LA 70504-1010, USA --------------------------------------------------------------- From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 09:12:19 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIFCIJ01984 for reliable_computing-outgoing; Mon, 18 Nov 2002 09:12:18 -0600 (CST) Received: from interferon.mpi-sb.mpg.de (mail [at] mail [dot] mpi-sb.mpg.de [139.19.1.1]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAIFCDi01979 for ; Mon, 18 Nov 2002 09:12:13 -0600 (CST) Received: from amavis by interferon.mpi-sb.mpg.de with scanned-ok (Exim 3.12 #1 (Debian)) id 18DnZL-0003gd-00 for ; Mon, 18 Nov 2002 16:12:11 +0100 Received: from data.mpi-sb.mpg.de ([::ffff:139.19.90.238] helo=mpi-sb.mpg.de) by interferon.mpi-sb.mpg.de with esmtp (Exim 3.12 #1 (Debian)) id 18DnZJ-0003gN-00; Mon, 18 Nov 2002 16:12:09 +0100 Received: from contact.mpi-sb.mpg.de (contact [139.19.2.26]) by mpi-sb.mpg.de (8.9.3/8.9.3) with ESMTP id QAA20211; Mon, 18 Nov 2002 16:12:09 +0100 (MET) Received: (from ratschan@localhost) by contact.mpi-sb.mpg.de (8.11.6+Sun/8.9.3) id gAIFC8t25396; Mon, 18 Nov 2002 16:12:08 +0100 (MET) Date: Mon, 18 Nov 2002 16:12:08 +0100 (MET) Message-Id: <200211181512.gAIFC8t25396 [at] contact [dot] mpi-sb.mpg.de> From: Stefan Ratschan To: rbk [at] louisiana [dot] edu, reliable_computing [at] interval [dot] louisiana.edu In-reply-to: <2.2.32.20021118144157.0169eee4 [at] pop [dot] louisiana.edu> (rbk [at] louisiana [dot] edu) Subject: Re: Solutions of systems of equations References: <2.2.32.20021118144157.0169eee4 [at] pop [dot] louisiana.edu> X-Virus-Scanned: by AMaViS perl-11 Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000020 >>> No matter what finite-precision >>> arithmetic system is used, there are such problems, for which >>> the accuracy of the computations is not able to distinguish whether >>> the zero is within the box or not. >> >>For extremely ill-conditioned, but still well-posed problems: Yes, if >>we fix the precision beforehand. But NO, if we are able to add >>precision during computation. > >Yes, that's true. If we have a bound on the precision, then there >are problems that cannot be determined. However, even if we have >no bound on the precision, there are still problems that we cannot >determine, if we have a bound on the execution time. Is that not >true? Again, a similar argument applies: If we bound the run-time BEFOREHAND, then we are restricted to well-conditioned problems. However, if we are allowed to add run-time dynamically (i.e., an algorithm with no condition-independent bound on the run-time) then one should be able to solve ALL well-posed (but possibly extremely ill-conditioned) root-finding problems. Of course, in practice, one hopes to do everything with bounded (i.e., machine) precision, in a small amount of time. But the ability of an algorithm to solve EVERY well-posed (but possibly extremely ill-conditioned) problem would be a very important property for * re-using such an algorithm as a sub-algorithm, * reasoning about the properties of the resulting algorithms, and * giving people guidance about which algorithm to use for which problems. HS> Is there anything known about F except that it maps $\R^n\to\R^n$? Let's say it is given by a finite symbolic expression involving the usual continuous function symbols (+, ., sin, cos, exp). Stefan From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 09:13:57 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIFDvb02083 for reliable_computing-outgoing; Mon, 18 Nov 2002 09:13:57 -0600 (CST) Received: from marnier.ucs.louisiana.edu (root [at] marnier [dot] ucs.louisiana.edu [130.70.132.233]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAIFDmi02075 for ; Mon, 18 Nov 2002 09:13:49 -0600 (CST) Received: from liberty (liberty.louisiana.edu [130.70.46.171]) by marnier.ucs.louisiana.edu (8.11.3/8.11.3/ull-ucs-mx-host_1.6) with SMTP id gAIFDmH06710 for ; Mon, 18 Nov 2002 09:13:48 -0600 (CST) Message-Id: <2.2.32.20021118151529.016b8258 [at] pop [dot] louisiana.edu> X-Sender: rbk5287 [at] pop [dot] louisiana.edu X-Mailer: Windows Eudora Pro Version 2.2 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 18 Nov 2002 09:15:29 -0600 To: reliable_computing [at] interval [dot] louisiana.edu From: "R. Baker Kearfott" Subject: Re: Solutions of systems of equations Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000021 Stefan et al, I'd like to add that, although I still believe there are theoretical limits on a given actual machine, the utility of verification algorithms can certainly be extended with variable precision. On the other hand, it can also be argued that, for real-world problems, the data itself is given to a limited, fixed precision (and should be so represented as intervals). In this case, variable precision is of more limited use. Best regards, Baker >Stefan et al, > >Please see my interpolated comment. > >At 02:18 PM 11/18/2002 +0100, Stefan Ratschan wrote: >> >>Thanks, Baker, for taking the time to write such a fundamental >>answer. I think it will be very helpful for clarifying the situation. >> >>> there are some problems for which a unique, non-singular >>> solution exists within B, but for which either the solution is >>> extremely close to the boundary of B or the Jacobi matrix is >>> extremely ill-conditioned at B. >> >>That is true. However, "extremely close to the boundary" does not mean >>"on the boundary" and "extremely ill-conditioned" does not mean >>"ill-posed". >> >>> No matter what finite-precision >>> arithmetic system is used, there are such problems, for which >>> the accuracy of the computations is not able to distinguish whether >>> the zero is within the box or not. >> >>For extremely ill-conditioned, but still well-posed problems: Yes, if >>we fix the precision beforehand. But NO, if we are able to add >>precision during computation. >> > >Yes, that's true. If we have a bound on the precision, then there >are problems that cannot be determined. However, even if we have >no bound on the precision, there are still problems that we cannot >determine, if we have a bound on the execution time. Is that not >true? > >Best regards, > >Baker > >>> 1. Yes, there is a unique zero; >>> 2. No, there is no zero; >>> 3. There MIGHT be a unique zero. >> >>I agree (we might replace possibility 3 with non-termination). >>The challenge is to find algorithms for which 3 occurs only when >>really theoretically necessary and to prove this. My question was, >>whether the reliable computing community can provide this. >> >>Best regards, >> >> Stefan >> >> > --------------------------------------------------------------- R. Baker Kearfott, rbk [at] louisiana [dot] edu (337) 482-5346 (fax) (337) 482-5270 (work) (337) 981-9744 (home) URL: http://interval.louisiana.edu/kearfott.html Department of Mathematics, University of Louisiana at Lafayette Box 4-1010, Lafayette, LA 70504-1010, USA --------------------------------------------------------------- From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 09:59:09 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIFx8g02358 for reliable_computing-outgoing; Mon, 18 Nov 2002 09:59:08 -0600 (CST) Received: from mercury.uwe.ac.uk (mercury.uwe.ac.uk [164.11.132.23]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAIFx0i02351 for ; Mon, 18 Nov 2002 09:59:00 -0600 (CST) Received: from cornet (cornet.cems.uwe.ac.uk [164.11.243.231]) by mercury.uwe.ac.uk (2.0.4/SMS 2.0.4-devel) with SMTP id PAA18221; Mon, 18 Nov 2002 15:58:57 GMT Message-ID: <00f401c28f1c$800eb920$e7f30ba4 [at] csm [dot] uwe.ac.uk> Reply-To: "Simon Langley" From: "Simon Langley" To: Cc: "Stefan Ratschan" References: <2.2.32.20021118125849.009de4ec [at] pop [dot] louisiana.edu> <200211181318.gAIDIrR12963 [at] contact [dot] mpi-sb.mpg.de> Subject: Re: Solutions of systems of equations Date: Mon, 18 Nov 2002 15:38:34 -0000 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4522.1200 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4522.1200 Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000022 How general is your set of equations? For example, if it's multivariate polynomials and the solution set is finite there is a lower bound for how close roots can be (cf Canny: "Complexity of Robot Motion Planning" for example). In theory that could set limits for how small your boxes need to be until a box always contains either 1 (possibly repeated) root or 0 roots. Aberths book is a good introduction but there are some later papers by him on using topological degree to solve the the sort of problem you are describing. I don't have the references to hand I'm afraid. Simon Langley ~~~~~~~~~~~~~~~ CS/CEMS/UNIWOE ----- Original Message ----- From: "Stefan Ratschan" To: Cc: Sent: Monday, November 18, 2002 1:18 PM Subject: Re: Solutions of systems of equations > > Thanks, Baker, for taking the time to write such a fundamental > answer. I think it will be very helpful for clarifying the situation. > > > there are some problems for which a unique, non-singular > > solution exists within B, but for which either the solution is > > extremely close to the boundary of B or the Jacobi matrix is > > extremely ill-conditioned at B. > > That is true. However, "extremely close to the boundary" does not mean > "on the boundary" and "extremely ill-conditioned" does not mean > "ill-posed". > > > No matter what finite-precision > > arithmetic system is used, there are such problems, for which > > the accuracy of the computations is not able to distinguish whether > > the zero is within the box or not. > > For extremely ill-conditioned, but still well-posed problems: Yes, if > we fix the precision beforehand. But NO, if we are able to add > precision during computation. > > > 1. Yes, there is a unique zero; > > 2. No, there is no zero; > > 3. There MIGHT be a unique zero. > > I agree (we might replace possibility 3 with non-termination). > The challenge is to find algorithms for which 3 occurs only when > really theoretically necessary and to prove this. My question was, > whether the reliable computing community can provide this. > > Best regards, > > Stefan > From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 09:59:14 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIFxDk02367 for reliable_computing-outgoing; Mon, 18 Nov 2002 09:59:13 -0600 (CST) Received: from mercury.uwe.ac.uk (mercury.uwe.ac.uk [164.11.132.23]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAIFx0i02350 for ; Mon, 18 Nov 2002 09:59:00 -0600 (CST) Received: from cornet (cornet.cems.uwe.ac.uk [164.11.243.231]) by mercury.uwe.ac.uk (2.0.4/SMS 2.0.4-devel) with SMTP id PAA18218 for ; Mon, 18 Nov 2002 15:58:57 GMT Message-ID: <00f301c28f1c$7ff341e0$e7f30ba4 [at] csm [dot] uwe.ac.uk> Reply-To: "Simon Langley" From: "Simon Langley" To: References: <2.2.32.20021118144157.0169eee4 [at] pop [dot] louisiana.edu> Subject: Re: Solutions of systems of equations Date: Mon, 18 Nov 2002 15:08:20 -0000 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4522.1200 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4522.1200 Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000023 > Yes, that's true. If we have a bound on the precision, then there > are problems that cannot be determined. However, even if we have > no bound on the precision, there are still problems that we cannot > determine, if we have a bound on the execution time. Is that not > true? True, but perhaps unduly pessimistic. The problem is solvable over algebraic numbers, and (probably) is still solvable if you add exp() and log() (then the 'proof' depends on Schanual's conjecture). If it's true with exp() and log() you can add elementary functions. It's not true (using the halting problem argument) if arbitrary computable functions are allowed. Simon Langley ~~~~~~~~~~~~~~ CS/CEMS/UNIWOE ----- Original Message ----- From: "R. Baker Kearfott" To: "Stefan Ratschan" ; <> Sent: Monday, November 18, 2002 2:41 PM Subject: Re: Solutions of systems of equations > Stefan et al, > > Please see my interpolated comment. > > At 02:18 PM 11/18/2002 +0100, Stefan Ratschan wrote: > > > >Thanks, Baker, for taking the time to write such a fundamental > >answer. I think it will be very helpful for clarifying the situation. > > > >> there are some problems for which a unique, non-singular > >> solution exists within B, but for which either the solution is > >> extremely close to the boundary of B or the Jacobi matrix is > >> extremely ill-conditioned at B. > > > >That is true. However, "extremely close to the boundary" does not mean > >"on the boundary" and "extremely ill-conditioned" does not mean > >"ill-posed". > > > >> No matter what finite-precision > >> arithmetic system is used, there are such problems, for which > >> the accuracy of the computations is not able to distinguish whether > >> the zero is within the box or not. > > > >For extremely ill-conditioned, but still well-posed problems: Yes, if > >we fix the precision beforehand. But NO, if we are able to add > >precision during computation. > > > > Yes, that's true. If we have a bound on the precision, then there > are problems that cannot be determined. However, even if we have > no bound on the precision, there are still problems that we cannot > determine, if we have a bound on the execution time. Is that not > true? > > Best regards, > > Baker > > >> 1. Yes, there is a unique zero; > >> 2. No, there is no zero; > >> 3. There MIGHT be a unique zero. > > > >I agree (we might replace possibility 3 with non-termination). > >The challenge is to find algorithms for which 3 occurs only when > >really theoretically necessary and to prove this. My question was, > >whether the reliable computing community can provide this. > > > >Best regards, > > > > Stefan > > > > > --------------------------------------------------------------- > R. Baker Kearfott, rbk [at] louisiana [dot] edu (337) 482-5346 (fax) > (337) 482-5270 (work) (337) 981-9744 (home) > URL: http://interval.louisiana.edu/kearfott.html > Department of Mathematics, University of Louisiana at Lafayette > Box 4-1010, Lafayette, LA 70504-1010, USA > --------------------------------------------------------------- > > From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 12:18:54 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIIIrI02877 for reliable_computing-outgoing; Mon, 18 Nov 2002 12:18:53 -0600 (CST) Received: from wrath.cs.utah.edu (wrath.cs.utah.edu [155.99.198.100]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAIIImi02873 for ; Mon, 18 Nov 2002 12:18:48 -0600 (CST) Received: from faith.cs.utah.edu (faith.cs.utah.edu [155.99.198.108]) by wrath.cs.utah.edu (8.11.6/8.11.6) with ESMTP id gAIIIf912867; Mon, 18 Nov 2002 11:18:41 -0700 (MST) Received: from localhost (sikorski@localhost) by faith.cs.utah.edu (8.11.6+Sun/8.11.1) with ESMTP id gAIIIeZ20890; Mon, 18 Nov 2002 11:18:41 -0700 (MST) Date: Mon, 18 Nov 2002 11:18:40 -0700 (MST) From: Kris Sikorski To: Stefan Ratschan cc: reliable_computing [at] interval [dot] louisiana.edu Subject: Re: Solutions of systems of equations In-Reply-To: <200211181105.gAIB5Rg29319 [at] contact [dot] mpi-sb.mpg.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000024 Stefan: you need to specify the class of functions for which such algorithm is supposed to work. For general Lipschitz functions we can only solve such problems under the residual error criterion (the problem with absolute error criterion has infinite worst case complexity) Please check my book: Optimal solution of Nonlinear Equations , Oxford Press, 2001 (for related results). chris --------------------------------------- Christopher Sikorski, Ph.D. Professor Director of the CES Program (www.ces.utah.edu) School of Computing 50 S Central Campus Drive Rm 3190 University of Utah Salt Lake City, UT 84112-9205 e-mail: sikorski [at] cs [dot] utah.edu Tel: 801-5818579 Fax: 801-5815843 www.cs.utah.edu/~sikorski --------------------------------------- On Mon, 18 Nov 2002, Stefan Ratschan wrote: > > Dear Colleagues, > > one of the most basic tasks of our field is to find solutions of > non-linear systems of equations. However, I was very surprised that, > having checked several textbooks, I could not find a single algorithm > for which it is proven that it does the following: > > Input: > A system of non-linear equations F=0, and > a box B > Output: > Yes, if F=0 contains a zero in B that does not vanish under perturbations, > no, if F=0 does not contain zeros in B. > > Otherwise (there are some zeros in B, but all of them vanish under > perturbations), the algorithm does not terminate. > > Of course the zeros might be arbitrarily ill-conditioned. However, > this just requires the use of arbitrarily high floating-point > precision. > > The algorithms that I found seem to verify solutions just sometimes, > where "sometimes" is very often not explicitely defined, or an > internal algorithmic condition that does not mean anything to a user. > > However, it might be that I overlooked something. > > So can anybody provide pointers to algorithms that solve the above > problem or that might be useful? > > Best regards, > > Stefan Ratschan > > > ----------------------------------- > Stefan Ratschan WWW: http://www.mpi-sb.mpg.de/~ratschan > Max-Planck-Institut fuer Informatik e-mail: stefan.ratschan@mpi-sb.mpg.de > Stuhlsatzenhausenweg 85 phone: +49 681 9325 229 > 66123 Saarbruecken fax: +49 681 9325 299 > Germany > From owner-reliable_computing [at] interval [dot] louisiana.edu Mon Nov 18 13:32:02 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIJW1E03182 for reliable_computing-outgoing; Mon, 18 Nov 2002 13:32:01 -0600 (CST) Received: (from rbk5287@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAIJVvQ03176 for reliable_computing [at] interval [dot] louisiana.edu; Mon, 18 Nov 2002 13:31:57 -0600 (CST) Received: from kelly.bath.ac.uk (IDENT:lvo4hhfnibjh2mwks8oe [at] kelly [dot] bath.ac.uk [138.38.32.20]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAIIe8i03079 for ; Mon, 18 Nov 2002 12:40:09 -0600 (CST) Received: from mmdf by kelly.bath.ac.uk with pleasure id 18DqoV-0004Px-01 for reliable_computing [at] interval [dot] louisiana.edu; Mon, 18 Nov 2002 18:40:03 +0000 Received: from mary.bath.ac.uk ([138.38.32.14] ident=mmdf) by kelly.bath.ac.uk with smtp id 18DqoT-0003UQ-01; Mon, 18 Nov 2002 18:40:01 +0000 Date: Mon, 18 Nov 2002 18:39:58 +0000 (GMT) From: D Richardson To: Stefan Ratschan cc: reliable_computing [at] interval [dot] louisiana.edu, rbk [at] louisiana [dot] edu Subject: Re: Solutions of systems of equations In-Reply-To: <200211181318.gAIDIrR12963 [at] contact [dot] mpi-sb.mpg.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000025 Assuming we have n equations in n unknowns, then we can use recursive subdivision and interval arithmetic, provided we have the possibility of unlimited precision computation, and we are willing to replace possibility 3) with 3*) There might be a solution with singular Jacobian. Some exact computation results are also known. Suppose the box is in complex n space, and the functions are polynomials in x1,..., xn and exp(x1), ... ,exp(xn) with integral coefficients. Then there is an algorithm to solve this problem, which always terminates if the Schanuel conjuecture is true. Schanuel conjecture: if z1,..., zn are complex numbers linearly independent over the rationals, then there are at least n algebraically independent numbers in {z1,..., zn, exp(z1),..., exp(zn)}. However the algorithm is not very practical at present. Some progress has been made in the attempt to decide the fundamental problem referred to by Baker, ie.e is a constant =0. Some references can be found on my web site: www.bath.ac.uk/~masdr or look up uniformity conjecture in Google. Daniel Richardson On Mon, 18 Nov 2002, Stefan Ratschan wrote: > > Thanks, Baker, for taking the time to write such a fundamental > answer. I think it will be very helpful for clarifying the situation. > > > there are some problems for which a unique, non-singular > > solution exists within B, but for which either the solution is > > extremely close to the boundary of B or the Jacobi matrix is > > extremely ill-conditioned at B. > > That is true. However, "extremely close to the boundary" does not mean > "on the boundary" and "extremely ill-conditioned" does not mean > "ill-posed". > > > No matter what finite-precision > > arithmetic system is used, there are such problems, for which > > the accuracy of the computations is not able to distinguish whether > > the zero is within the box or not. > > For extremely ill-conditioned, but still well-posed problems: Yes, if > we fix the precision beforehand. But NO, if we are able to add > precision during computation. > > > 1. Yes, there is a unique zero; > > 2. No, there is no zero; > > 3. There MIGHT be a unique zero. > > I agree (we might replace possibility 3 with non-termination). > The challenge is to find algorithms for which 3 occurs only when > really theoretically necessary and to prove this. My question was, > whether the reliable computing community can provide this. > > Best regards, > > Stefan > From owner-reliable_computing [at] interval [dot] louisiana.edu Thu Nov 21 11:07:11 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gALH7Ar11571 for reliable_computing-outgoing; Thu, 21 Nov 2002 11:07:10 -0600 (CST) Received: from cs.utep.edu (mail.cs.utep.edu [129.108.5.3]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gALH73i11567 for ; Thu, 21 Nov 2002 11:07:04 -0600 (CST) Received: from aragorn (aragorn [129.108.5.35]) by cs.utep.edu (8.11.3/8.11.3) with SMTP id gALH6sT03871 for ; Thu, 21 Nov 2002 10:06:55 -0700 (MST) Message-Id: <200211211706.gALH6sT03871 [at] cs [dot] utep.edu> Date: Thu, 21 Nov 2002 10:06:55 -0700 (MST) From: Vladik Kreinovich Reply-To: Vladik Kreinovich To: reliable_computing [at] interval [dot] louisiana.edu MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Content-MD5: GCsXdzenzmEFDuYTl0gF+w== X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.4 SunOS 5.8 sun4u sparc Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000026 Dear Friends, I just wanted to let you all know that this year, Sun Microsystems and Bill Walster again organized a Special Interest Group on Interval Arithmetic at the High Performance Computing Consortium Meeting, a meeting that was held in conjunction with SuperComputing'2002 conference in Baltimore, Maryland. SIG was followed by a tutorial on intervals given by Bill Walster and by his toreless promotion of intervals at SC'2002 meeting itself. Thanks a lot to Bill for his hard work and to Sun Microsystems for their tireless promotion of intervals! Vladik From owner-reliable_computing [at] interval [dot] louisiana.edu Sat Nov 23 01:38:51 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAN7coQ16020 for reliable_computing-outgoing; Sat, 23 Nov 2002 01:38:50 -0600 (CST) Received: from mail.comset.net (mail.comset.net [213.172.0.3]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAN7cgi16016 for ; Sat, 23 Nov 2002 01:38:43 -0600 (CST) Received: from 3-061.dialup.comset.net ([213.172.3.61] helo=e0gumi46) by mail.comset.net with smtp (Exim 3.33 #1) id 18FUun-0002vO-00 for reliable_computing [at] interval [dot] louisiana.edu; Sat, 23 Nov 2002 10:41:21 +0300 Message-ID: <009e01c292c2$ca33d360$710bfea9 [at] wplus [dot] net> From: "Vyacheslav Nesterov" To: "RC mailing list" Subject: Reliable Computing, vol. 9, issue 1, 2003 Date: Sat, 23 Nov 2002 10:19:57 +0300 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_005E_01C292D9.DF78F540" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000027 This is a multi-part message in MIME format. ------=_NextPart_000_005E_01C292D9.DF78F540 Content-Type: text/plain; charset="koi8-r" Content-Transfer-Encoding: quoted-printable Reliable Computing=20 Volume 9, issue 1, 2003 Mathematical Research Diagrammatic Analysis of Interval Linear Equations: =20 Part I: Basic Notions and the One-Dimensional Case Zenon Kulpa=20 1-20 Controlling the Wrapping Effect in the Solution=20 of ODEs for Asteroids Jens Hoefkens, Martin Berz, Kyoko Makino 21-41 =20 Review Taylor Forms - Use and Limits Arnold Neumaier 43-79 =20 Information COCOS'02: A Workshop on Global Constrained Optimization and Constraint Satisfaction R. Baker Kearfott 81-87 ------=_NextPart_000_005E_01C292D9.DF78F540 Content-Type: text/html; charset="koi8-r" Content-Transfer-Encoding: quoted-printable
           = Reliable=20 Computing
         Volume 9, = issue=20 1, 2003
 
           = Mathematical=20 Research
 
Diagrammatic Analysis of Interval = Linear=20 Equations:          &nb= sp;=20
Part I: Basic Notions and the One-Dimensional Case
Zenon Kulpa=20
1-20
 
Controlling the Wrapping Effect = in the=20 Solution
of ODEs for Asteroids
Jens Hoefkens, Martin Berz, Kyoko=20 Makino
21-41
         = =20
           &nb= sp;     =20 Review
 
Taylor Forms - Use and = Limits
Arnold=20 Neumaier
43-79
       =20
           &nb= sp;    =20 Information
 
COCOS'02: A Workshop on Global = Constrained=20 Optimization and Constraint
Satisfaction
R. Baker=20 Kearfott
81-87
------=_NextPart_000_005E_01C292D9.DF78F540-- From owner-reliable_computing [at] interval [dot] louisiana.edu Sat Nov 23 16:31:39 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gANMVdh17889 for reliable_computing-outgoing; Sat, 23 Nov 2002 16:31:39 -0600 (CST) Received: from cs.utep.edu (mail.cs.utep.edu [129.108.5.3]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gANMVWi17885 for ; Sat, 23 Nov 2002 16:31:33 -0600 (CST) Received: from aragorn (aragorn [129.108.5.35]) by cs.utep.edu (8.11.3/8.11.3) with SMTP id gANMVS227634 for ; Sat, 23 Nov 2002 15:31:28 -0700 (MST) Message-Id: <200211232231.gANMVS227634 [at] cs [dot] utep.edu> Date: Sat, 23 Nov 2002 15:31:28 -0700 (MST) From: Vladik Kreinovich Reply-To: Vladik Kreinovich Subject: from NA Digest: interval computations mentioned To: reliable_computing [at] interval [dot] louisiana.edu MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Content-MD5: eho8ETzNgTwu2VjJ9Pxa1w== X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.4 SunOS 5.8 sun4u sparc Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000028 From: Theodore Simos Date: Mon, 18 Nov 2002 03:00:47 +0200 Subject: Conference in Cambridge on Numerical Analysis & Computational Math FIRST ANNOUNCEMENT AND CALL FOR PAPERS International Conference on NUMERICAL ANALYSIS & COMPUTATIONAL MATHEMATICS (NACoM-2003) May 23-24-25-26, 2003 Anglia Polytechnic University (APU), Cambridge, UK http://www.apu.ac.uk/appsci/maths/NACoM-2003/ The multidisciplinary field of Numerical Analysis and Computational Mathematics is long suffering from major delays in the publication of original papers. In most cases it takes at least 2 to 3 years for a paper to be accepted and the situation seems to get worse by the year. The NACoM-2003 will address this problem, as follows: 1) This first time the NACoM Conference will serve as the medium for the official LAUNCH of a new WILEY JOURNAL on "APPLIED NUMERICAL ANALYSIS AND COMPUTATIONAL MATHEMATICS (ANACM)" (Editor-in-Chief T.E. SIMOS). Generally, in ANACM it is expected that the time between a paper submission and its acceptance for publication to be around 6 months. ANACM will be registered with the ISI CITATION INDEX without delay. 2) The NACoM-2003 Conference is expected to attract very high quality papers and a refereed selection of them will be published in the two special issues of the new journal. 3) This will be an excellent opportunity to MEET IN PERSON members of the ANACM EDITORIAL BOARD. Furthermore, high-ranking representatives from Wiley-Europe will be present and will address the participants. MAIN TOPICS Numerical ODEs, Numerical PDEs (inc. BVPs), Differential Algebraic Equations, Numerical Stochastic Differential Equations, Approximation, Numerical Linear Algebra, Numerical Integral Equations, Error Analysis and Interval Analysis, Difference Equations and Recurrence Relations, Numerical problems in Dynamical Systems, Applications to the Sciences (Computational Physics-Chemistry-Engineering etc.), Scientific Computing and Algorithms, Numerical methods in Fourier analysis. GENERAL CHAIR & ORGANISER Dr Georgios Psihoyios, Anglia Pol. University, Cambridge, UK VICE-CHAIRS Prof. Theodore E. Simos, University of Peloponnisos, Greece Boz Kempski, Anglia Pol. University, Cambridge, UK PLENARY SPEAKERS (to be finalised) Prof. J.R. Cash, Imperial College, London, UK. Prof. A. Cuyt, University of Antwerp, Belgium. Prof. B. Fischer, Medical University of Luebeck, Germany. Prof. M. Hochbruck, University of Duesseldorf, Germany. Prof. W.F. Mitchell, National Institute of Standards & Technology, USA. Prof. G. Vanden Berghe, University of Gent, Belgium. Prof. G.A. Watson, University of Dundee, UK CALL FOR PAPERS & EXPRESSION OF INTEREST Submission of original papers is invited for the NACoM-2003 Conference. Please initially submit an ABSTRACT OF UP TO FOUR A4 PAGES directly to the General Chair (g.y.psihoyios [at] apu [dot] ac.uk) by February 28, 2003. Please send a standard Latex article style abstract without page numbers. For further instructions please refer to the WEBSITE. Only electronic submission of abstracts will be accepted. DEADLINES Submission of Abstract (on or before): February 28, 2003 Submission of Mini-symposia proposals: February 28, 2003 Abstracts refereed selection by: March 20, 2003 Submission of full-length Papers: To be announced From owner-reliable_computing [at] interval [dot] louisiana.edu Sun Nov 24 10:00:45 2002 Received: (from daemon@localhost) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) id gAOG0ia20351 for reliable_computing-outgoing; Sun, 24 Nov 2002 10:00:44 -0600 (CST) Received: from cs.utep.edu (mail.cs.utep.edu [129.108.5.3]) by interval.louisiana.edu (8.11.3/8.11.3/ull-interval-math-majordomo-1.3) with ESMTP id gAOG0di20347 for ; Sun, 24 Nov 2002 10:00:39 -0600 (CST) Received: from aragorn (aragorn [129.108.5.35]) by cs.utep.edu (8.11.3/8.11.3) with SMTP id gAOG0ZB03295 for ; Sun, 24 Nov 2002 09:00:36 -0700 (MST) Message-Id: <200211241600.gAOG0ZB03295 [at] cs [dot] utep.edu> Date: Sun, 24 Nov 2002 09:00:34 -0700 (MST) From: Vladik Kreinovich Reply-To: Vladik Kreinovich To: reliable_computing [at] interval [dot] louisiana.edu MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Content-MD5: EL3Ys88t1j7GVVsEiR58zg== X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.4 SunOS 5.8 sun4u sparc Sender: owner-reliable_computing [at] interval [dot] louisiana.edu Precedence: bulk Status: O X-Status: $$$$ X-UID: 0000000029 Dear Friends, This is the 2003 conference about which I mentioned that the 2004 oen can be back-toback with Denmark meeting. Vladik *********************************************************************** XXXI Summer School "Advanced Problems in Mechanics" June 23-July 2, 2003, St. Petersburg, Russia First Announcement A P M 2 0 0 3 http://www.eng.abdn.ac.uk/~apm/2003 To receive our announcements, please send a void e-mail with the subject "subscribe" to apm [at] eng [dot] abdn.ac.uk We will be grateful if you spread this announcement among your colleagues GENERAL INFORMATION The Summer school Advanced Problems in Mechanics 2003 is organized by the Institute for Problems in Mechanical Engineering of the Russian Academy of Sciences (IPME RAS) under the patronage of the Russian Academy of Sciences and with information support of Gesellschaft fuer Angewandte Mathematik und Mechanik (GAMM). The main purpose of the meeting is to gather specialists from different branches of mechanics to provide a platform for cross-fertilisation of ideas. HISTORY OF THE SCHOOL The first Summer School was organized by Ya.G. Panovko and his colleagues in 1971. In the early years the main focus of the School was on nonlinear oscillations of mechanical systems with a finite number of degrees of freedom. The School specialized in this way because at that time in Russia (USSR) there were held regular National Meetings on Theoretical and Applied Mechanics, and also there were many conferences on mechanics with a more particular specialization. After 1985 many conferences and schools on mechanics in Russia were terminated due to financial problems. In 1994 the Institute for Problems in Mechanical Engineering of the Russian Academy of Sciences restarted the Summer School. The traditional name of "Summer School" has been kept, but the topics covered by the School have been much widened, and the School has been transformed into an international conference. The topics of the School cover now all fields of mechanics and associated into interdisciplinary problems. SCIENTIFIC COMMITTEE (changes are possible) * A.K. Abramian (IPME RAS, St. Petersburg, Russia) * V.V. Beletsky (Space Research Institute, Moscow, Russia) * I.I. Blekhman (IPME RAS, Mechanobr Research Institute, St. Petersburg, Russia) * M. Cartmell (University of Glasgow, UK) * Y. Fu (University of Keele, UK) * S.N. Gavrilov (IPME RAS, St. Petersburg, Russia) * I.G. Goryacheva (Institute for Problems in Mechanics RAS, Moscow, Russia) * E.F. Grekova (IPME RAS, St. Petersburg, Russia) * D. Harris (UMIST, Manchester, UK) * A.B. Freidin (IPME RAS, St. Petersburg, Russia) * D.A. Indeitsev (IPME RAS, St. Petersburg, Russia) -- Co-Chairman * E.A. Ivanova (IPME RAS, St. Petersburg State Technical University, Russia) * A.M. Krivtsov (IPME RAS, St. Petersburg State Technical University, Russia) * G.A. Leonov (IPME RAS, St. Petersburg State University, Russia) * E.V. Lomakin (Moscow State Lomonosov University, Russia) * G.A. Maugin (University of P. et M. Curie, Paris, France) * N.F. Morozov (IPME RAS, St. Petersburg State University, Russia) * R.W. Ogden (University of Glasgow, UK) * V.A. Palmov (St. Petersburg State Technical University, Russia) -- Co-Chairman * M. Wiercigroch (University of Aberdeen, UK) * P.A. Zhilin (IPME RAS, St. Petersburg State Technical University, Russia) * F. Ziegler (Technical University of Wien, Austria) LOCAL ORGANIZING COMMITTEE * A.D. Firsova, IPME RAS * S.N. Gavrilov, IPME RAS * E.F. Grekova, IPME RAS - Scientific Secretary * A.D. Sergeyev, IPME RAS * E.V. Serogo, IPME RAS * E.V. Shishkina, IPME RAS SCIENTIFIC PROGRAM Presentations devoted to fundamental aspects of mechanics, or spreading the field of applications of mechanics, are invited. We are particularly keen to receive contributions that SHOW NEW EFFECTS AND PHENOMENA OR DEVELOP NEW MATHEMATICAL MODELS. The topics of the School cover all fields of mechanics, including, but NOT RESTRICTED, to * mechanics of generalized continua (polar and micromorphic continua, mixtures, porous media, electromagnetic continua, grains, powders etc) * solids and structures * phase transitions * nanostructures and thin films * wave motion * nonlinear dynamics, chaos and vibration * dynamics of rigid bodies and multibody dynamics * fluid and gas * computational mechanics * mechanical and civil engineering applications Four different forms of presentations are offered, namely, plenary lectures (1 hour), presentations at minisymposia (30 minutes), short communications (20 minutes), and posters. The working language is English. For the attention of Russian participants: the Russian language as an exception can be used only in posters, but even for posters English is strongly recommended. Regrettably we can not provide simultaneous translation, and due to the international nature of the School all the oral presentations must be in English. Attention: each participant may only give ONE oral presentation. The number of posters for each participant is not limited. MINISYMPOSIA AND CHAIRS (preliminary information) * Instabilities, Waves, and Phase Transitions in Nonlinear Elasticity R.W. Ogden (Glasgow, UK), A.B. Freidin (St. Petersburg, Russia), and Y. Fu (Keele, UK) * Nonlinear Dynamics of Engineering Systems E.E. Pavlovskaia, M. Wiercigroch (Aberdeen, UK) * Nonlinear Strain Waves in Complex Deformable Solids G.A. Maugin (Paris, France), A.V. Porubov, A.M. Samsonov (St.Petersburg, Russia) * Problems in the Mechanics of Granular Materials and Systems D. Harris (Manchester, UK) * Recent Advances in Random Vibrations D. Iourtchenko (Miami, USA) * Vibrational Mechanics and Synchronization I.I. Blekhman (St. Petersburg, Russia) Speakers include (to be completed): Prof. E.L. Aero (St.Petersburg, Russia) Acad. V.V. Beletsky (Moscow, Russia) http://www.iki.rssi.ru/eng/people/beletsky_res.htm Prof. I.I. Blekhman (St. Petersburg, Russia) http://www.ipme.ru/ipme/labs/vm/index.htm Prof. G. Capriz (Pisa, Italy) - to be confirmed Dr. D. Harris (Manchester, UK) http://www.ma.umist.ac.uk/dh/index.htm Prof. Y. Fu (Keele, UK) http://www.keele.ac.uk/depts/ma/people/yf/ Acad. I.G. Goryacheva (Moscow, Russia) http://hp.iitp.ru/win/38/3883.htm Prof. D.A. Indeitsev (St. Petersburg, Russia) http://www.ipme.ru/ipme/labs/he/ind.htm Prof. E.V. Lomakin (Moscow, Russia) Prof. G.A. Maugin (Paris, France) http://www.lmm.jussieu.fr/MEMBRES/MAUGIN/directeur.html Acad. N.F. Morozov (St. Petersburg, Russia) http://www.math.spbu.ru/user/morozov/ Prof. R.W. Ogden (Glasgow, UK) http://www.maths.gla.ac.uk/~rwo/ Prof. V.A. Palmov (St. Petersburg, Russia) http://www.unilib.neva.ru/stu/face_13.html Prof. M. Wiercigroch (Aberdeen, UK) http://wwwcad.eng.abdn.ac.uk/~eng373 Prof. P. Zhilin (St. Petersburg, Russia) ORGANIZATIONAL DETAILS AND IMPORTANT DATES Details concerning abstract submission and other organizational details are available at our website http://www.eng.abdn.ac.uk/~apm/2003 DATES TO REMEMBER: * Application for financial support for young scientists from Russia (abstract and registration form are required): January 15, 2003. * Abstract submission: February 1, 2003. * Submission of a visa form (required to issue the invitation to Russia): February 1, 2003. * Notification of acceptance: March 15, 2003. * Paper submission (for plenary lectures, minisymposia and invited papers): June 23, 2003. * Conference: June 23-July 2, 2003. REGISTRATION FEE The conference fee is: * for foreign participants giving a presentation: $250 * for foreign participants not giving a presentation: $190 * for foreign students: $190 The fee covers invitation costs, postage, book of abstracts, proceedings, social program (excursion, get-together party), and organizational costs. * The fee for foreign