From chemistry-request@server.ccl.net Wed Jan 31 04:12:19 2001 Received: from stark.udg.es (stark.udg.es [130.206.128.61]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0V9CJw30036 for ; Wed, 31 Jan 2001 04:12:19 -0500 Received: from iqc.udg.es (woodstock.udg.es [130.206.128.118]) by stark.udg.es (980427.SGI.8.8.8/980728.SGI.AUTOCF) via ESMTP id FAA29025 for ; Wed, 31 Jan 2001 05:18:00 -0800 (PST) Message-ID: <3A77E746.E98445B7@iqc.udg.es> Date: Wed, 31 Jan 2001 10:21:59 +0000 From: Pedro Salvador X-Mailer: Mozilla 4.7 [en] (Win95; I) X-Accept-Language: en MIME-Version: 1.0 CC: chemistry@ccl.net Subject: Re: CCL:BBSE corrections References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sean, Which energy are you talking about? The energy of the complex or the interaction energy? In any case, the energy of the complex should increase after correction (less negative) so that the interaction energy decreases. Of course,the correction depends on the geometry were you apply it but, by definition, is not possible that the interaction energy is larger(energy of the complex is more negative) after correction. If it happens it can be due to: a) If you're using the counterpoise method, the state, spin, etc.. of the monomer and the corresponding monomer + ghost orbital calculation are different. b) There is simply a bug in the program (I remember having lots of problems with Gaussian and pseudopotentials...) Hope it helps Pedro -- -------------------------------------------------------------------- Pedro Salvador Sedano | Ph.D. Student | I'll stare the sundown Institut de Quimica Computacional | until my eyes go blind... Universitat de Girona | I won't change direction Campus Montilivi 17071 | and I won't change my mind. Girona Spain | e-mail: perico@stark.udg.es | WWW: http://stark.udg.es/~perico | Telf: +34 972 418358 | -------------------------------------------------------------------- From chemistry-request@server.ccl.net Wed Jan 31 04:45:05 2001 Received: from vytautas.vdu.lt (mail@vytautas.vdu.lt [193.219.38.33]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0V9hQw30265 for ; Wed, 31 Jan 2001 04:43:29 -0500 Received: from vaidila.vdu.lt ([193.219.38.52] ident=root) by vytautas.vdu.lt with esmtp for chemistry@ccl.net id 14Ntko-0005KO-00; Wed, 31 Jan 2001 11:40:42 +0200 Received: from vaidila.vdu.lt (vejas.vdu.lt [193.219.38.82]) by vaidila.vdu.lt (8.9.3/8.9.3) with ESMTP id LAA00034 for ; Wed, 31 Jan 2001 11:40:42 +0200 (GMT) Message-ID: <3A774221.BCD64127@vaidila.vdu.lt> Date: Tue, 30 Jan 2001 23:37:22 +0100 From: "art'" X-Mailer: Mozilla 4.7 [en] (Win98; I) X-Accept-Language: en MIME-Version: 1.0 CC: CCL Subject: CCL: BBSE corrections- how to ? References: Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 7bit Dear CCL'ers, Where I could get routine of BSSE correction ? How I could make those corrections ? I am user of Gaussian, if it makes any difference. best wishes to all Arturas Z From chemistry-request@server.ccl.net Wed Jan 31 04:54:35 2001 Received: from century.fen.bilkent.edu.tr (century.fen.bilkent.edu.tr [139.179.97.100]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0V9sWw30353 for ; Wed, 31 Jan 2001 04:54:33 -0500 Received: from pinar (pinar [139.179.97.99]) by century.fen.bilkent.edu.tr (8.9.1/8.9.1) with ESMTP id LAA04800 for ; Wed, 31 Jan 2001 11:40:19 +0200 (EET) Date: Wed, 31 Jan 2001 11:46:29 +0200 (EET) From: Ulrike Salzner X-Sender: salzner@pinar To: ccl Subject: programming language Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII We are a chemistry department and our first year students take a computer science cours learning JAVA. They told me that they don't think it is of any use to them. I would like to collect opinions which programming language is most useful for chemists. Let's assume they are learning only one. Thanks, Ulrike Salzner =================================================================== Dr. Ulrike Salzner Associate Professor Department of Chemistry Tel.: (312) 290-2122 Bilkent University Fax.: (312) 266-5097 06533 Bilkent, Ankara e-mail: salzner@fen.bilkent.edu.tr Turkey ==================================================================== From chemistry-request@server.ccl.net Wed Jan 31 10:12:09 2001 Received: from lx00.csrsrc.mi.cnr.it (IDENT:root@lx00.csrsrc.mi.cnr.it [155.253.3.10]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VFC8w00624 for ; Wed, 31 Jan 2001 10:12:08 -0500 Received: from csrsrc.mi.cnr.it (IDENT:alex@pitpc7.csrsrc.mi.cnr.it [155.253.3.246]) by lx00.csrsrc.mi.cnr.it (8.9.3/8.8.7) with ESMTP id QAA07516 for ; Wed, 31 Jan 2001 16:12:08 +0100 Sender: alex@csrsrc.mi.cnr.it Message-ID: <3A7822C2.EC86D603@csrsrc.mi.cnr.it> Date: Wed, 31 Jan 2001 15:35:46 +0100 From: Alessandro X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.14-5.0 i586) X-Accept-Language: en MIME-Version: 1.0 To: chemistry@ccl.net Subject: CCL:Modelling of DNA-Protein interaction Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear cclers, I am looking for some good reviews about DNA-Protein interaction, both on modelling and calculations. Despite some experiences on protein studies, I have no idea on DNA problems... so pratical suggestion on this topics are gratefully accepted, too. This is a "first look" search, just for getting an idea of the state of the art, so every good information could be useful. Thanks a lot. Ale From chemistry-request@server.ccl.net Wed Jan 31 10:13:20 2001 Received: from imc21.ex.nus.edu.sg (imc21.ex.nus.edu.sg [137.132.14.62]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VFDJw00641 for ; Wed, 31 Jan 2001 10:13:19 -0500 Received: by imc21.ex.nus.edu.sg with Internet Mail Service (5.5.2650.21) id ; Wed, 31 Jan 2001 23:13:19 +0800 Message-ID: <415039BB7DE8D011BC4600805F311E16066BB622@exs25.ex.nus.edu.sg> From: Liu Xianghui To: "'Serguei Patchkovskii'" Cc: "'chemistry@ccl.net'" Subject: Annihilation Date: Wed, 31 Jan 2001 23:12:20 +0800 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Content-Type: text/plain; charset="iso-8859-1" Dear all I also have questions about this.... :-) That are some methods for spin correction for B3LYP. The famous one is by K.Yamaguchi. "A spin correction procedure for unrestricted HF and MP wavefunctions for singlet diradicals and polyradicals."(Chemical physics letters V149 537-542.) Someone used it for spin correction for B3LYP. for example, E.Goldstein, K.N.Houk used it in "Density Fuctional Theory prediction of Relative energies and istope effects for the concerted pathway and stepwise pathway mechanism of the Diels-alder reaction of Butadiene and ethylene" (J.AM.Chem.Soc. V118.No.25,1996. ) However, On page 6041, They said " Initial values for the diradical stationary points(5-7) range from 0.56 to 0.98 and become 0.07 and 0.23 after spin annihilation. " I don't how this 0.07 and 0.23 come. For Gaussian output, there are S2, S2-1,S2A, what are they correspondingly? Thanks. Liu xianghui -----Original Message----- From: Serguei Patchkovskii [mailto:patchkov@ucalgary.ca] Sent: Tuesday, January 30, 2001 7:13 AM To: hu eric Cc: chemistry@ccl.net Subject: CCL:Annihilation On Mon, 29 Jan 2001, hu eric wrote: > In DFT singlet or triplet diradical g98 output, there's a S**2 after > annihilation. What is annihilation for? Does that mean I should use the new > S**2 value as the reference? Thanks! You should keep in mind that the "" expectation value, printed by Gaussian at the end of DFT calculations is not particularly meaningful in any case. This value is computed for the non-interacting Kohn-Sham reference wavefunction, and has no relation to the of the real, interacting system. The Gaussians' "" may be useful as a qualitative indication of the degree of spin-contamination, but that's about it. /Serge.P --- Home page: http://www.cobalt.chem.ucalgary.ca/ps/ -= This is automatically added to each message by mailing script =- CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net From chemistry-request@server.ccl.net Wed Jan 31 10:18:53 2001 Received: from postoffice.mail.cornell.edu (postoffice.mail.cornell.edu [132.236.56.7]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VFIrw00821 for ; Wed, 31 Jan 2001 10:18:53 -0500 Received: from cornell.edu (carbo2.foodsci.cornell.edu [132.236.147.61]) by postoffice.mail.cornell.edu (8.9.3/8.9.3) with ESMTP id KAA13770 for ; Wed, 31 Jan 2001 10:18:49 -0500 (EST) Sender: dwood@cornell.edu Message-ID: <3A782B72.2A6E0CE9@cornell.edu> Date: Wed, 31 Jan 2001 10:12:51 -0500 From: "Dr. Richard Wood" Organization: Cornell University X-Mailer: Mozilla 4.7C-SGI [en] (X11; I; IRIX 6.5 IP32) X-Accept-Language: en MIME-Version: 1.0 To: chemistry@ccl.net Subject: Re: CCL:programming language References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Ulrike Salzner wrote: > We are a chemistry department and our first year students take a computer > science cours learning JAVA. They told me that they don't think it is of > any use to them. > > I would like to collect opinions which programming language is most useful > for chemists. Let's assume they are learning only one. Dear Dr. Salzner Being rather advanced in age myself, I would think that it would be beneficial for one to know Fortran, which was what I learned when I was in college. However, as time has gone by, I have seen that most things are written in C or C++, so I would think that would be the most useful language for one to learn, in my opinion. It certainly is the one I've regretted not learning. Richard -- Richard L. Wood, Ph. D. Physical/Computational Chemist Post-doctoral Associate Cornell University, Ithaca, NY 14853 From chemistry-request@server.ccl.net Wed Jan 31 07:14:32 2001 Received: from martin.u.arizona.edu (IDENT:mcafiero@martin.U.Arizona.EDU [128.196.137.11]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VCEVw32011 for ; Wed, 31 Jan 2001 07:14:31 -0500 Received: from localhost (mcafiero@localhost) by martin.u.arizona.edu (8.8.8/8.8.8) with ESMTP id FAA52710; Wed, 31 Jan 2001 05:14:15 -0700 Date: Wed, 31 Jan 2001 05:14:15 -0700 (MST) From: Mauricio L Cafiero To: Ulrike Salzner cc: ccl Subject: Re: CCL:programming language In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII hello: It really depends. For the most global usefulness, C++ would be best. It is powerful, extremely widely used, and capable of scientific programming. Java is also widely used, and there are scientific and mathematical libraries being developed for Java. Also, it is similar to C++, so that is an added benefit. As far as theoretical chemistry is concerned, I think Fortran is by far the best choice. It has many mathematical libraries, and is designed for scientific computing. Since many chemistry undergraduates do not go on to become theorists, then perhaps C++ or Java would be best, as with a background in one language, any other language is picked up very easily. I hope this addresses your question. Mauricio Cafiero Doctoral Candidate: Theoretical and Computational Quantum Chemistry Dept. of Chemistry University of Arizona Tucson, AZ From chemistry-request@server.ccl.net Wed Jan 31 07:52:33 2001 Received: from mail-c.bcc.ac.uk (mail-c.bcc.ac.uk [144.82.100.23]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VCqWw32124 for ; Wed, 31 Jan 2001 07:52:32 -0500 Received: from socrates-a.ucl.ac.uk by mail-c.bcc.ac.uk with SMTP (Mailer); Wed, 31 Jan 2001 12:51:25 +0000 From: uccatvm Message-Id: <10896.200101311252@socrates-a.ucl.ac.uk> Subject: Re: CCL:BBSE corrections- how to ? To: chemistry@ccl.net (CCL) Date: Wed, 31 Jan 2001 12:51:46 +0000 (GMT) In-Reply-To: <3A774221.BCD64127@vaidila.vdu.lt> from "art'" at Jan 30, 2001 11:37:22 PM X-Mailer: ELM [version 2.5 PL3] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi Arturas, > > Where I could get routine of BSSE correction ? How I could make those > corrections ? > I am user of Gaussian, if it makes any difference. With Gaussian, the keyword Massage has to used for counterpoise corrections. Have a look at the Gaussian manual: http://www.gaussian.com/00000452.htm. There is an example there as well. However, don't believe the sentence about counterpoise being only a crude estimate, have a look at the following papers to learn more about BSSE and Counterpoise: - F.B. van Duijneveldt, J.G.C.M. van Duijneveldt-van de Rijdt, and J.H. van Lenthe, Chem. Rev. 94, 1873, 1994 (on counterpoise theory) - S. Simon, M. Duran, J.J. Dannenberg, J. Chem. Phys. 105, 1996 (on geometry optimizations on BSSE-corrected potential energy surfaces). - F.B. van Duijneveldt, "Basis Superposition Error" in "Molecular Interactions" S. Scheiner (ed.), Wiley (1997) - T. van Mourik, A.K. Wilson, K.A. Peterson, D.E. Woon, T.H. Dunning, Jr., Adv. Quant. Chem. 31, 105, 1999 (showing BSSE effects on interaction energies, distances, and frequencies). HTH, Tanja -- ==================================================================== Tanja van Mourik Royal Society University Research Fellow Chemistry Department University College London phone: +44 (0)20-7679-4663 20 Gordon Street e-mail: work: T.vanMourik@ucl.ac.uk London WC1H 0AJ, UK home: tanja@netcomuk.co.uk http://www.chem.ucl.ac.uk/people/vanmourik/index.html ==================================================================== From chemistry-request@server.ccl.net Wed Jan 31 08:32:22 2001 Received: from narnia4.rutgers.edu (narnia4.rutgers.edu [165.230.180.169]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VDWMw32358 for ; Wed, 31 Jan 2001 08:32:22 -0500 Received: from eden.rutgers.edu (molly.envsci.rutgers.edu [165.230.5.148]) by narnia4.rutgers.edu (8.8.8/8.8.8) with ESMTP id IAA28282; Wed, 31 Jan 2001 08:32:08 -0500 (EST) Message-ID: <3A78159E.D29BCCEF@eden.rutgers.edu> Date: Wed, 31 Jan 2001 08:39:42 -0500 From: "Mary O'Connor" X-Mailer: Mozilla 4.08 [en] (Win98; I) MIME-Version: 1.0 To: Ulrike Salzner , chemistry@ccl.net Subject: Re: CCL:programming language References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Ulrike Salzner wrote: > We are a chemistry department and our first year students take a computer > science cours learning JAVA. They told me that they don't think it is of > any use to them. > > I would like to collect opinions which programming language is most useful > for chemists. Let's assume they are learning only one. > > Thanks, > Ulrike Salzner > > =================================================================== > > Dr. Ulrike Salzner > Associate Professor > Department of Chemistry Tel.: (312) 290-2122 > Bilkent University Fax.: (312) 266-5097 > 06533 Bilkent, Ankara e-mail: salzner@fen.bilkent.edu.tr > Turkey > > ==================================================================== > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net I think JAVA is of most use to people working on webpages (I'm no expert). My own inquiries into which languages to learn have led me to the conclusion that the basis of scientific programming is still FORTRAN, being gradually overtaken by C++ Mary O'Connor Doctoral Candidate Rutgers University New Brunswick,NJ USA From chemistry-request@server.ccl.net Wed Jan 31 09:23:44 2001 Received: from mail2.organik.uni-erlangen.de (lesath.chemie.uni-erlangen.de [131.188.127.212]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VENhw00361 for ; Wed, 31 Jan 2001 09:23:43 -0500 Received: from chemie.uni-erlangen.de (xugu [131.188.127.224]) by mail2.organik.uni-erlangen.de (8.8.8/8.1.4-FAU) with ESMTP id PAA18737; Wed, 31 Jan 2001 15:23:43 +0100 (MET) Sender: Frank.Oellien@CCC.Chemie.Uni-Erlangen.DE Message-ID: <3A781FEF.845F6A8F@chemie.uni-erlangen.de> Date: Wed, 31 Jan 2001 15:23:43 +0100 From: Frank Oellien Organization: Computer-Chemie-Centrum - http://www2.ccc.uni-erlangen.de/ChemVis/ X-Mailer: Mozilla 4.7C-SGI [en] (X11; I; IRIX 6.5 IP32) X-Accept-Language: en MIME-Version: 1.0 To: chemistry@ccl.net, chminf-l@listserv.indiana.edu Subject: molecular surfaces online Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear Computational Chemists, the ChemVis project is pleased to announce a new online service for calculation and visualization of molecular surfaces and electrostatic potentials - MolSurf. http://www2.chemie.uni-erlangen.de/services/molsurf/ MolSurf is able to calculate and visualize molecular surfaces as VRML scenes. Our service supports three methods of input chemical structure data: Molecular editor, SMILES String, direct loading of structure files (with surface information). The most easiest way to create a new structure is to use the JME Molecular Editor (courtesy of Peter Ertl, Novartis). Users who are familiar with the SMILES format can also input the SMILES String by hand. The third input possibility is useful for people with structure files on their disk. They can select their structure file by clicking the button in the main frame. In the moment we only support the VAMP file format. The computation of 3D coordinates is done by the CORINA program (Prof. Gasteiger). The computation of the surface information is done by the VAMP program (Dr. Clark). Please send us other file formats that contain surface information. This helps us to enlarge the number of file formats that can be read and visulized by our service. Please inform us, when problems occur. With regards Frank Oellien -- Computer-Chemistry-Center and Institute of Organic Chemistry University of Erlangen-Nuernberg Naegelsbachstrasse 25, D-91052 Erlangen (Germany) Tel (+49)-(0)9131-85-26579 Fax (+49)-(0)9131-85-26566 email: Frank.Oellien@chemie.uni-erlangen.de WWW : http://www2.ccc.uni-erlangen.de/oellien/ http://www2.ccc.uni-erlangen.de/ChemVis/ From chemistry-request@server.ccl.net Wed Jan 31 09:33:15 2001 Received: from asnmail.asc.edu (asnmail.asc.edu [129.66.12.7]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VEXEw00386 for ; Wed, 31 Jan 2001 09:33:14 -0500 Received: from localhost (asnmaz01@localhost) by asnmail.asc.edu (8.9.3/8.9.3) with ESMTP id IAA06271; Wed, 31 Jan 2001 08:32:58 -0600 Date: Wed, 31 Jan 2001 08:32:58 -0600 (CST) From: "Mark A. Zottola" To: Ulrike Salzner cc: Subject: CCL:programming language In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear Ulrike, While I think that any programming language is a good addition to anyone's skill set, I believe that FORTRAN has unique advantages over other programming languages. Before I explain myself, let me state up front that I am prejudiced. I have programmed in FORTRAN for over 20 years and teach courses in FORTRAN and code optimization in FORTRAN. FORTRAN is the pre-eminent scientific programming language. As the an analyst for the Alabama Supercomputer Center, I can tell you that *all* the codes we have are either 100% FORTRAN or 98+% written in FORTRAN. I do not see this trend changing. While there are religous wars between C and FORTRAN concerning which is the better coding language, the fact remains that FORTRAN remains the language of choice for scientific programming. Therefore, if a student wanted to examine a piece of code, be it AMBER, Gaussian or GAMESS (or any of a multitude of other programs), a student would be well served having FORTRAN under her/his belt. Witness the recent postings by Jeff Saxe on trying to determine the use of the variable mv in the CURVES code as a good example for the need for FORTRAN. Knowing C/C++, JAVA or some other language and *not* FORTRAN, ferreting out the answer to that question would have been significantly more difficult. If one wanted to modify a piece of code to augment its cuurrent capabilities, chances are those changes would have to be done in FORTRAN. A third reason for learning FORTRAN is its readability. C/C++, JAVA can be very powerful languages due to their pointer constructs. However, this power also tends to make code somewhat obtuse and difficult to follow. I ahve found that FORTRAN code is much more readable. In my classes I have found that even after three concentrated classes in FORTRAN, most of my students can follow some very high level codes. I can not say the same for people who are learning other languages. Tying into the ideas of the above paragraph, readability has a direct consequence in terms of debugging, modifying and understanding code. The final reason for learning FORTRAN is that it is eminently optimizable, as compared to other languages. I do not wish to get into the computer science of this issue. Rather let me state as a fact which can be validated in the computer science literature that it is easier to optimize FORTRAN code. I do not want to get into a holy war between those who believe FORTRAN is faster as opposed to those who believe C is faster. That flame war is truly a code-dependent issue. However, for the general case, it is easier and more straightforward to optimize FORTRAN code. As computers get faster and faster, it is important to be able to make codes run faster. It is inherently easier to get FORTRAN codes to do such. I hope you find these few thoughts useful, Mark Zottola Ph.D. Alababa Research and Education Network 119 Rust Research Center University of Alabama at Birmingham 1801 University Boulevard VOICE:(205) 934-3893 Birmingham, AL 35294 EMAIL:asnmaz01@asc.edu On Wed, 31 Jan 2001, Ulrike Salzner wrote: > > We are a chemistry department and our first year students take a computer > science cours learning JAVA. They told me that they don't think it is of > any use to them. > > I would like to collect opinions which programming language is most useful > for chemists. Let's assume they are learning only one. > > Thanks, > Ulrike Salzner > > =================================================================== > > Dr. Ulrike Salzner > Associate Professor > Department of Chemistry Tel.: (312) 290-2122 > Bilkent University Fax.: (312) 266-5097 > 06533 Bilkent, Ankara e-mail: salzner@fen.bilkent.edu.tr > Turkey > > ==================================================================== > > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > From chemistry-request@server.ccl.net Wed Jan 31 10:10:50 2001 Received: from mail2.zrz.tu-berlin.de (mail2.zrz.TU-Berlin.DE [130.149.4.14]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VFAow00601 for ; Wed, 31 Jan 2001 10:10:50 -0500 Received: from mailszrz.zrz.tu-berlin.de ([130.149.4.11]) by mail2.zrz.tu-berlin.de with esmtp (exim-3.22) id 14NyuI-00053L-00; Wed, 31 Jan 2001 16:10:50 +0100 Received: from natrium.chem.tu-berlin.de ([130.149.42.210]) by mailszrz.zrz.tu-berlin.de with esmtp (exim-3.22) id 14NyuH-0000bC-00; Wed, 31 Jan 2001 16:10:49 +0100 Mime-Version: 1.0 X-Sender: chwu0531@mailszrz.zrz.tu-berlin.de Message-Id: In-Reply-To: References: Date: Wed, 31 Jan 2001 16:10:49 +0100 To: Ulrike Salzner From: Christoph van =?iso-8859-1?Q?W=FCllen?= Subject: Re: CCL:programming language Cc: Chemistry@ccl.net Content-Type: text/plain; charset="us-ascii" ; format="flowed" >I would like to collect opinions which programming language is most useful >for chemists. Let's assume they are learning only one. > FORTRAN. If you don't like it, learn C (or C++, but this makes no difference for beginners). -- +---------------------------------+-------------------------------------+ | Prof. Christoph van Wullen | Tele-Phone (+49) (0)30 314 27870 | | Technische Universitat Sekr. C3 | Tele-Fax (+49) (0)30 314 23727 | | Strasse des 17. Juni 135 | eMail | | D-10623 Berlin, Germany | Christoph.vanWullen@TU-Berlin.De | +---------------------------------+-------------------------------------+ From chemistry-request@server.ccl.net Wed Jan 31 10:15:12 2001 Received: from hotmail.com (f185.law7.hotmail.com [216.33.237.185]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VFFCw00710 for ; Wed, 31 Jan 2001 10:15:12 -0500 Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC; Wed, 31 Jan 2001 07:13:13 -0800 Received: from 130.91.65.29 by lw7fd.law7.hotmail.msn.com with HTTP; Wed, 31 Jan 2001 15:13:13 GMT X-Originating-IP: [130.91.65.29] From: "JAYASRI DASSARMA" To: chemistry@ccl.net Subject: Looking for any free builder software Date: Wed, 31 Jan 2001 10:13:13 -0500 Mime-Version: 1.0 Content-Type: text/plain; format=flowed Message-ID: X-OriginalArrivalTime: 31 Jan 2001 15:13:13.0346 (UTC) FILETIME=[54635620:01C08B98] Hi everyone, I am looking for any graphical freeware available to build molecules (similar to what Builder module does in Insight), which runs on Windows or Linux/Unix platforms. Any help will be appreciated. Thanks in advance, Jayasri _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com From chemistry-request@server.ccl.net Wed Jan 31 13:03:47 2001 Received: from fischer.pharm.uu.nl (fischer.pharm.uu.nl [131.211.56.46]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VI3lw03456 for ; Wed, 31 Jan 2001 13:03:47 -0500 Received: from fischer (localhost [127.0.0.1]) by fischer.pharm.uu.nl (SGI-8.9.3/8.9.3) with ESMTP id TAA23509 for ; Wed, 31 Jan 2001 19:03:47 +0100 (MET) Message-Id: <200101311803.TAA23509@fischer.pharm.uu.nl> To: chemistry@ccl.net Reply-To: Malcolm Gillies Subject: Re: CCL:programming language In-reply-to: Your message of "Wed, 31 Jan 2001 08:39:42 MET." <3A78159E.D29BCCEF@eden.rutgers.edu> Date: Wed, 31 Jan 2001 19:03:47 +0100 From: Malcolm Gillies I think it's difficult to give a good general answer to this one. If you are considering doing any 'serious' programming in the future (e.g. writing programs of more than a thousand lines or so), then it's a great help to get a firm grasp of the fundamentals, and a waste of time to use anything other than modern tools and techniques. Java or C++ would be good (conservative) choices in that case. Picking up C or Fortran once you have a grounding in either of those is not much work at all. If the intention is just to hack lightly on existing code written in Fortran, then perhaps it's reasonable to take a shortcut and start there. For the situations somewhere in between, for instance if you're interested in the kind of day-to-day programming involved in scripting, file format conversion, building simple web interfaces etc., I would recommend Python (or at a pinch, Perl) as a good place to start. It's an object-oriented language with clear syntax, and plenty of libraries for useful everyday tasks. As an interpreted language, it's easy to experiment and get instant feedback. You also have the advantage of a user community which is very actively developing libraries and applications with direct relevance to computational chemistry and bioinformatics. I should add the disclaimer that I have studied computer science, and that my own history has gone something like Basic (1980) - C (1985) - Assembler - Pascal (1990) - [AWK,Bourne shell,etc] - Common Lisp - Standard ML - Perl (1995) - Python (1997) - Splus - Fortran 77 (1999). Learning C was definitely the turning point, though picking up functional programming concepts (mostly through learning Common Lisp) broadened my horizons, as did the concepts of object-oriented programming (first in CLOS, an extension to Common Lisp). cheers, Malcolm -- Malcolm Gillies PhD student, computational medicinal chemistry Dept Medicinal Chemistry, Faculty of Pharmacy, Utrecht University, The Netherlands From chemistry-request@server.ccl.net Wed Jan 31 13:11:02 2001 Received: from newmail.DKFZ-Heidelberg.DE (newmail.inet.dkfz-heidelberg.de [193.174.53.39]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VIB2w03507 for ; Wed, 31 Jan 2001 13:11:02 -0500 Received: from DKFZ-heidelberg.de (vishnu.inet.dkfz-heidelberg.de [193.174.62.115]) by newmail.DKFZ-Heidelberg.DE (8.9.3/8.9.3/DKFZ-8.9.3) with ESMTP id TAA04474 for ; Wed, 31 Jan 2001 19:09:02 +0100 (MET) Message-ID: <3A7854BF.A7CC0D21@DKFZ-heidelberg.de> Date: Wed, 31 Jan 2001 19:09:03 +0100 From: Chaitanya Athale Organization: DKFZ-Heidelberg X-Mailer: Mozilla 4.7 [en] (WinNT; I) X-Accept-Language: en MIME-Version: 1.0 To: chem computing Subject: Persistence Length of DNA Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, a question concerning an approach to make multiple measurements on a single fibre of a polymer like DNA to obtain a persistence length. There is a method of using a wondow and "scanning the lenth of the fibre. I would be grateful if somebody can direct me to some literature. I have looked and havent encountered it yet. The input is an image of the polymer. Cheers, Chaitanya. -- Chaitanya Athale PhD student at: DKFZ Heidelberg, Im Neuenheimer Feld 280, 69120, Heidelberg. From chemistry-request@server.ccl.net Wed Jan 31 10:40:19 2001 Received: from iitkgp.iitkgp.ernet.in ([202.141.127.2]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VFeFw01324 for ; Wed, 31 Jan 2001 10:40:16 -0500 Received: from cts.iitkgp.ernet.in (IDENT:root@cts.iitkgp.ernet.in [144.16.192.220]) by iitkgp.iitkgp.ernet.in (8.9.3/8.9.3) with ESMTP id VAA25012; Wed, 31 Jan 2001 21:08:27 -0500 (GMT) Received: from localhost (rajarshi@localhost) by cts.iitkgp.ernet.in (8.9.3/8.9.3) with ESMTP id UAA02872; Wed, 31 Jan 2001 20:56:55 +0530 X-Authentication-Warning: cts.iitkgp.ernet.in: rajarshi owned process doing -bs Date: Wed, 31 Jan 2001 20:56:54 +0530 (IST) From: Rajarshi Guha X-X-Sender: To: Ulrike Salzner cc: ccl Subject: Re: CCL:programming language In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Wed, 31 Jan 2001, Ulrike Salzner wrote: > science cours learning JAVA. They told me that they don't think it is of > any use to them. > > I would like to collect opinions which programming language is most useful > for chemists. Let's assume they are learning only one. Well I think it depends a lot on what you are programming - if you're doing numerical codes / simulations etc then C (or Fortran) is well suited. If you want to quickly generate a GUI for simple visualization then Java is a pretty good option. Other languages like Python/Perl etc are more oriented towards text processing - so it would be a good tool for say translating file formats. Another factor which affects choice of language is the learning aspect - if people are already grounded in C then i think Java should be no problem. For that matter Perl is also syntactically close to C. Personally I would say C is the best choice - with a proper choice of libraries both numerical and graphical (I talk from a Unix point of view) applications can be put together. At the same time I also admit Perl would be a language to keep handy - for rapid prototyping. C pointers still mess up my day ;-) ------------------------------------------------------------------------ Rajarshi Guha | Last yeer I kudn't spel Engineer. Dept Of Chemistry, | Now I are won. IIT Kharagpur. | | email: rajarshi@presidency.com | web : www.psynet.net/jijog | ------------------------------------------------------------------------- From chemistry-request@server.ccl.net Wed Jan 31 12:59:47 2001 Received: from mail.csusb.edu (mail.csusb.edu [139.182.2.16]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VHxlw03375 for ; Wed, 31 Jan 2001 12:59:47 -0500 Received: from localhost (kcousins@localhost) by mail.csusb.edu (8.11.1/8.11.1) with SMTP id f0VHx3H09085; Wed, 31 Jan 2001 09:59:03 -0800 (PST) Date: Wed, 31 Jan 2001 09:59:03 -0800 (PST) From: Kimberley Cousins To: "Mary O'Connor" cc: Ulrike Salzner , chemistry@ccl.net Subject: Re: CCL:programming language In-Reply-To: <3A78159E.D29BCCEF@eden.rutgers.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Which "language" you teach to chem majors depends on what you expect them to do: if they are likely to do scientific programmming, I concur that C++ is a good choice. If you want them to learn to think logically, while providing them with tools they can really use, you might consider an "applied" programming environment instead, like LabView for instrument simulation or control, or perhaps a Mathematical programming like Mathematica, Maple, or MathCad. These tools teach a sort of programming skill while providing tools for use in other areas of chemistry. Kimberley Cousins Associate Professor of Chemistry California State University, San Bernardino 5500 University Parkway San Bernardino, CA 92407 (909)880-5391 kcousins@csusb.edu http://chem.csusb.edu/~kcousins On Wed, 31 Jan 2001, Mary O'Connor wrote: > > > Ulrike Salzner wrote: > > > We are a chemistry department and our first year students take a computer > > science cours learning JAVA. They told me that they don't think it is of > > any use to them. > > > > I would like to collect opinions which programming language is most useful > > for chemists. Let's assume they are learning only one. > > > > Thanks, > > Ulrike Salzner > > > > =================================================================== > > > > Dr. Ulrike Salzner > > Associate Professor > > Department of Chemistry Tel.: (312) 290-2122 > > Bilkent University Fax.: (312) 266-5097 > > 06533 Bilkent, Ankara e-mail: salzner@fen.bilkent.edu.tr > > Turkey > > > > ==================================================================== > > > > I think JAVA is of most use to people working on webpages (I'm no expert). My > own inquiries into which languages to learn have led me to the conclusion that > the basis of scientific programming is still FORTRAN, being gradually overtaken > by C++ > > Mary O'Connor > Doctoral Candidate > Rutgers University > New Brunswick,NJ > USA > > > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > > From chemistry-request@server.ccl.net Wed Jan 31 13:16:30 2001 Received: from toronto.acdlabs.com (toronto.acdlabs.com [207.176.233.2]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VIGUw03598 for ; Wed, 31 Jan 2001 13:16:30 -0500 Received: from gavin (gavin.acdlabs.com [207.176.233.80]) by toronto.acdlabs.com (8.9.0/8.9.0) with SMTP id NAA09367; Wed, 31 Jan 2001 13:16:12 -0500 (EST) From: "Gavin Shear" To: "Ulrike Salzner" , "ccl" Subject: RE: programming language Date: Wed, 31 Jan 2001 13:23:16 -0500 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 In-Reply-To: Dear Dr. Salzner, I would like to offer some examples of using JAVA in chemistry applications. Please see: http://www.acdlabs.com/products/java/ You could download the structure drawing applet from here: http://www.acdlabs.com/download/sda.html Another language to learn could be ChemBasic.. a language similar to basic that incorporates custom Chemical functions into our free ChemSketch software. We have many examples of ChemBasic (including all instructions) available at http://www.acdlabs.com/download/ In my first year of Chemistry, our programming was with Fortran77.. I have not programmed Fortran since then. Visual Basic might have been more useful, although the programming fundamentals learned with Fortran did carry over to other areas of programming. C++ would have been (and I think still is) rather challenging for me at that time :-) Many thanks, Gavin Shear > -----Original Message----- > From: Computational Chemistry List [mailto:chemistry-request@ccl.net]On > Behalf Of Ulrike Salzner > Sent: 31 January 2001 04:46 > To: ccl > Subject: CCL:programming language > > > > We are a chemistry department and our first year students take a computer > science cours learning JAVA. They told me that they don't think it is of > any use to them. > > I would like to collect opinions which programming language is most useful > for chemists. Let's assume they are learning only one. > > Thanks, > Ulrike Salzner > > =================================================================== > > Dr. Ulrike Salzner > Associate Professor > Department of Chemistry Tel.: (312) 290-2122 > Bilkent University Fax.: (312) 266-5097 > 06533 Bilkent, Ankara e-mail: salzner@fen.bilkent.edu.tr > Turkey > > ==================================================================== > > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- > To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: > gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: > jkl@ccl.net > > > > > > From chemistry-request@server.ccl.net Wed Jan 31 11:13:29 2001 Received: from neon.chem.ucla.edu (neon.chem.ucla.edu [128.97.35.251]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VGDTw01755 for ; Wed, 31 Jan 2001 11:13:29 -0500 Received: from localhost (renxiao@localhost) by neon.chem.ucla.edu (8.8.0/8.8.0) with SMTP id IAA21891 for ; Wed, 31 Jan 2001 08:13:28 -0800 (PST) X-Authentication-Warning: neon.chem.ucla.edu: renxiao owned process doing -bs Date: Wed, 31 Jan 2001 08:13:28 -0800 (PST) From: Renxiao Wang To: Comp Chem List Subject: Summary: solvation algorithm Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear CCLers, Last week I posted a query looking for algorithm/program for solvating molecules with explicit waters. Thanks for all the kind answers and here comes the summary. This is my original post: For some reasons we need to solvate organic molecules or even proteins with explicit water molecules. Will anybody recommend a reliable algorithm for doing this? It will be ideal if it is able to reproduce the 'real' Please note that we prefer an algorithm rather than a certain program. It is simply because we are writing programs in our approach. We certainly know about some programs, such as AMBER and CHARMm, which can place waters around a given molecule. But we are a little bit skeptical for what they generate on the screen. And these are the answers we got: ------------------------------------------------------------------------------- > From krauss@carb.nist.gov Thu Jan 25 19:34:57 2001 Fragment potentials for waters in GAMESS will represent the first shell well if you know how to distribute them. Prior sampling with MD simulation with care in selection of snapshots. Gas-phase cluster can also be easily optimized. ------------------------------------------------------------------------------- > From lensink@sun3.oulu.fi Thu Jan 25 23:40:58 2001 Since you're aware of amber and charmm, I'm sure you're aware of gromacs as well (http://md.chem.rug.nl/~gmx/online2.1.html). You should use its program genbox, with which you can add a layer of water molecules around any given structure. Alternatively, you could generate a box of solvent around your molecule, and perform energy minimization and possibly position restrained molecular dynamics (PR on your molecule, thus let the waters relax around it). ------------------------------------------------------------------------------- > From spoel@xray.bmc.uu.se Fri Jan 26 01:02:09 2001 If this makes you skeptical, then try quantum mechanics or X-ray crystallography instead. Writing a program in "your own approach" will certainly not improve the result unless you use a better description than molecular mechanics (i.e. quantum chemistry). ------------------------------------------------------------------------------- > From vitorge@azurite.cea.fr Fri Jan 26 01:04:39 2001 I have started to study aqueous mineral species (usually ions). I first performed usual ab initio studies, up to the begining of the second hydratation spheres, however ab initio caculations are virtually impossible for second sphere (geometry optimisation cannot be reliable) Next step would be to parametrise a model with the ab initio result (and actually validate it with ab initio results for other similar molecules or clusters) because usual ab initio calculations cannot presently model liquid solvent, neither actually temperature corrections (despite they produce some). The only difficulty is in finding a reliable model for your solvent (I cannot help you for that): then you would use the model for molecular dynamics. All this is a classical approach. ------------------------------------------------------------------------------- > From psenet@ruca.ua.ac.be Fri Jan 26 02:48:31 2001 I am interested by the same topic and I would be interested to know the answers. Could you forward me a resume? Myself, I intend to use Monte Carlo simulated annealing with potentials fitted on Quantum Chemical calculations ------------------------------------------------------------------------------- > From dshobe@sud-chemieinc.com Fri Jan 26 07:07:28 2001 I haven't "played" with it yet, but I just got a program called Agoa that is supposed to do just that: (http://www.dqf.ufpe.br/~zaldini/agoa.html). You might want to talk to the developer of Agoa...he may even wish to collaborate! ------------------------------------------------------------------------------- > From dgregory@msi.com Fri Jan 26 08:26:18 2001 I thought I'd send you a short note, just wondering what problems you may have encountered with adding waters using CHARMm. The standard way CHARMm researchers do this, is to first 'solvate' around a protein, using something like an 8A or 10A shell around all the surface residues, and then do some equiliibration dynamics to allow the waters to 'adjust' to the protein environment. The waters used to make the beginning 'shell', have been equilibrated (as bulk, just by themselves) using periodic boundary conditions, and adjusting the water characteristics to reflect experimental water values such as density and the O-O radial-dist.function. Boxes of this 'equilibrated' water are freely available, and we as well as our customers, have used them with no noticable problems. What kind of situations have you encountered to make you "skeptical " ? ------------------------------------------------------------------------------- > From gadres@pn2.vsnl.net.in Fri Jan 26 18:31:32 2001 We have developed a model, named as EPIC, for this purpose. You may look up our papers in J.Phys. Chem. and Ind. J. Chem. in 1998-2000 for this purpose. Currently we are extending it for larger number of water molecules. Of course, what contitutes the "first layer" or "first shell" is rather difficult to define. We have been able to use EPIC structure for ab initio studies. The agreement for energies is reasobaly good. For example, we have studied thyamine..n H2O with n upto 20, formaimide with upto 20 etc. Please contact us for details if interested. Thanks...Shridhar Gadre ------------------------------------------------------------------------------- > From zaldini@npd.ufpe.br Sat Jan 27 06:32:28 2001 I believe that you have not see my message at CCL, anuncing my AGOA program, that I judge that may be usefull to you. Please, visit the HP at: http://www.dqf.ufpe.br/~zaldini/agoa.html. AGOA does not do simulation, but it purpose hydration shells from QM - ESP. If you get interested, please, contact me. ------------------------------------------------------------------------------ Best wishes, Renxiao Wang From chemistry-request@server.ccl.net Wed Jan 31 14:04:23 2001 Received: from SEMOVM.SEMO.EDU (semovm.semo.edu [150.201.1.1]) by server.ccl.net (8.11.0/8.11.0) with SMTP id f0VJ4Mw04523 for ; Wed, 31 Jan 2001 14:04:22 -0500 Received: from tony [150.201.3.191] by SEMOVM.SEMO.EDU (IBM VM SMTP Level 310) via TCP with SMTP ; Wed, 31 Jan 2001 13:01:30 CST Message-Id: <3.0.5.32.20010131130558.0087eda0@semovm.semo.edu> X-Sender: ajduben@semovm.semo.edu X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) Date: Wed, 31 Jan 2001 13:05:58 -0600 To: chemistry@ccl.net From: "Anthony J. Duben" Subject: Re: CCL:Programming language Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" I hope that this discussion does not become a religious war, but I believe that any modern language would be quite suitable -- C, FORTRAN, C++, JAVA -- are all good candidates. You are not training chemists to maintain legacy code (much of which is in FORTRAN, plenty is in C, and some more in C++). You want the students to learn how to think algorithmically. JAVA is a good candidate because of its object-orientation is more thoroughly part of it than C++. By the time the students graduate and maybe finish grad school, whatever language they learn now will have gone through another 3 editions. So, I would urge the students to get on with it and really work hard at it. The hardest computer language to learn is the first one in which they need to learn how to think in algorithms. The next one to learn is a lot easier. Anthony J. Duben (ajduben@semovm.semo.edu) ******************************************************** Anthony J. Duben Professor and Chairman Computer Science Dept., MS 5950 Southeast Missouri State University 1 University Plaza Cape Girardeau MO 63701-4799 phone: 573-651-2194 fax: 573-651-2791 e-mail: ajduben@semovm.semo.edu or c867buc@semovm.semo.edu "Education is not the filling of a pail but the lighting of a fire." -- William Butler Yeats "Amateurs talk strategy; professionals talk logistics." -- Gen. Omar N. Bradley From chemistry-request@server.ccl.net Wed Jan 31 15:08:31 2001 Received: from mallard.duc.auburn.edu (mallard.duc.auburn.edu [131.204.2.11]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VK8Vw05632 for ; Wed, 31 Jan 2001 15:08:31 -0500 Received: from localhost (newhoir@localhost) by mallard.duc.auburn.edu (8.8.8+Sun/8.7.2) with SMTP id OAA15065; Wed, 31 Jan 2001 14:08:01 -0600 (CST) X-Authentication-Warning: mallard.duc.auburn.edu: newhoir owned process doing -bs Date: Wed, 31 Jan 2001 14:08:01 -0600 (CST) From: Irene Newhouse X-Sender: newhoir@mallard To: "Mark A. Zottola" cc: Ulrike Salzner , chemistry@ccl.net Subject: Re: CCL:programming language In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Since a poll question was asked, I wish to be polled in favor of FORTRAN as well, for scientific uses. *IS* there an example yet of a C/C++ application that runs faster then the FORTRAN? Speed is not an advantage I've heard computer scientists cite in the religious wars. The advantages I've seen cited are, to my jaundiced view, disguises for "programming this way results in code easier for professors/TAs to grade", not performance advantages. Someone once said "if we had a teraflop machine with petabytes of memory today, the comp. chemists would be happy for about a week. Then they'd start asking 'why can't we do this?'" When the problems you're interested in push the limits of the available machines, performance outranks most other language criteria. When you're interested in doing something with a machine, you want to be able to reuse what's out there that's close & not have to start from scratch. For chemists, what's out there is FORTRAN. If it ain't broken, don't fix it... Irene From chemistry-request@server.ccl.net Wed Jan 31 15:03:55 2001 Received: from copland.udel.edu (copland.udel.edu [128.175.13.92]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VK3tw05456 for ; Wed, 31 Jan 2001 15:03:55 -0500 Received: from localhost (giju@localhost) by copland.udel.edu (8.9.3/8.9.3) with ESMTP id PAA06070 for ; Wed, 31 Jan 2001 15:03:56 -0500 (EST) Date: Wed, 31 Jan 2001 15:03:53 -0500 (EST) From: Giju Thomas Kalathingal To: Subject: conformational minima Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear CCLers, I have a question about molecular rotations connecting conformational minima. Literature discussions are usually based on single group rotations (e.g., CH3 group in ethane, CH2 in ethene, etc.). Is it possible to have rotational isomers connected by simultaneous rotations of more than one group (e.g., A with two rotating groups isomerises to rotomer B)? Is there any examples? Thanks, Giju Giju T. Kalathingal Department of Chemistry and Biochemistry University of Delaware Newark, Delaware 19716, USA. From chemistry-request@server.ccl.net Wed Jan 31 17:29:27 2001 Received: from dns2.seanet.com ([199.181.164.2]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VMTRw07450 for ; Wed, 31 Jan 2001 17:29:27 -0500 Received: from C1353359A (dialup-63.214.14.209.Seattle1.Level3.net [63.214.14.209]) by dns2.seanet.com (8.9.3/8.9.0) with SMTP id OAA15346; Wed, 31 Jan 2001 14:29:10 -0800 (PST) Message-ID: <005c01c08bd4$9151aaf0$d10ed63f@sttln1.wa.home.com> From: "Mark A. Thompson" To: "JAYASRI DASSARMA" , References: Subject: Re: CCL:Looking for any free builder software Date: Wed, 31 Jan 2001 14:24:23 -0800 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 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Try ArgusLab 2.0 http://www.planaria-software.com It runs on Windows. Mark Thompson Factotum, Planaria-Software Seattle, WA. ----- Original Message ----- From: "JAYASRI DASSARMA" To: Sent: Wednesday, January 31, 2001 7:13 AM Subject: CCL:Looking for any free builder software > Hi everyone, > > I am looking for any graphical freeware available to build molecules > (similar to what Builder module does in Insight), which runs on Windows or > Linux/Unix platforms. Any help will be appreciated. > > Thanks in advance, > > Jayasri > > _________________________________________________________________ > Get your FREE download of MSN Explorer at http://explorer.msn.com > > > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > From chemistry-request@server.ccl.net Wed Jan 31 12:06:11 2001 Received: from mail-c.bcc.ac.uk (mail-c.bcc.ac.uk [144.82.100.23]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VH6Aw02424 for ; Wed, 31 Jan 2001 12:06:10 -0500 Received: from socrates-a.ucl.ac.uk by mail-c.bcc.ac.uk with SMTP (Mailer); Wed, 31 Jan 2001 17:05:03 +0000 From: uccatvm Message-Id: <21727.200101311705@socrates-a.ucl.ac.uk> Subject: Re: CCL:Modelling of DNA-Protein interaction To: chemistry@ccl.net (CCL) Date: Wed, 31 Jan 2001 17:05:56 +0000 (GMT) In-Reply-To: <3A7822C2.EC86D603@csrsrc.mi.cnr.it> from "Alessandro" at Jan 31, 2001 03:35:46 PM X-Mailer: ELM [version 2.5 PL3] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi Alessandro, > > I am looking for some good reviews about DNA-Protein interaction, both > on modelling and calculations. > Despite some experiences on protein studies, I have no idea on DNA > problems... so pratical suggestion on this topics are gratefully > accepted, too. > > This is a "first look" search, just for getting an idea of the state of > the art, so every good information could be useful. The following two papers contain some information on DNA-protein complexes: "Hydration of DNA: take 2", H.M. Berman, Current Opin. Struct. Biol. 4, 345-350 (1994) "Weakly Bound Clusters of Biological Interest", C. Desfrancois, S. Carles and P.J. Schermann, Chem. Rev. 100, 3943-3962 (2000) (paragraph 4.3) Best wishes, Tanja -- ==================================================================== Tanja van Mourik Royal Society University Research Fellow Chemistry Department University College London phone: +44 (0)20-7679-4663 20 Gordon Street e-mail: work: T.vanMourik@ucl.ac.uk London WC1H 0AJ, UK home: tanja@netcomuk.co.uk http://www.chem.ucl.ac.uk/people/vanmourik/index.html ==================================================================== From chemistry-request@server.ccl.net Wed Jan 31 14:48:25 2001 Received: from mallard.duc.auburn.edu (mallard.duc.auburn.edu [131.204.2.11]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VJmPw05144 for ; Wed, 31 Jan 2001 14:48:25 -0500 Received: from localhost (newhoir@localhost) by mallard.duc.auburn.edu (8.8.8+Sun/8.7.2) with SMTP id NAA03542; Wed, 31 Jan 2001 13:48:12 -0600 (CST) X-Authentication-Warning: mallard.duc.auburn.edu: newhoir owned process doing -bs Date: Wed, 31 Jan 2001 13:48:12 -0600 (CST) From: Irene Newhouse X-Sender: newhoir@mallard To: Mauricio L Cafiero cc: Ulrike Salzner , ccl Subject: Re: CCL:programming language In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I do agree that FORTRAN is most useful to scientists & unlikely to be replaced any time soon. I got support I didnt' expect from a colleague with a Ph. D. in computer science who specialized in comp. chemistry programming when he explained to me that in C/C++, arithmetic instructions are implemented far less efficiently than in FORTRAN. This is because C was originally written to create operating systems & math is an after-thought [as should be obvious from: if you need to do math with C, you must 'include math.h'] While FORTRAN was written specifically to do computations. However, computer scientists have a completely different view of what computers are for, and many departments no longer even teach FORTRAN. [My sister once asked a math prof a question about the physcial consequences of curl in electricy & magnetism & he said "I don't do applications". That's rather how computer scientists view comp. chemistry]. Unfortunately, when students learn C/C++ from a CS dept., they also pick up the attitude & then someone has to pound it out of their heads, because it's counterproductive for chemists. Irene Newhouse From chemistry-request@server.ccl.net Wed Jan 31 14:40:47 2001 Received: from postoffice.mail.cornell.edu (postoffice.mail.cornell.edu [132.236.56.7]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VJelw05092 for ; Wed, 31 Jan 2001 14:40:47 -0500 Received: from cornell.edu (phobos.mbg.cornell.edu [128.84.203.151]) by postoffice.mail.cornell.edu (8.9.3/8.9.3) with ESMTP id OAA08212; Wed, 31 Jan 2001 14:40:32 -0500 (EST) Sender: richard@cornell.edu Message-ID: <3A786B15.46B95749@cornell.edu> Date: Wed, 31 Jan 2001 14:44:21 -0500 From: Richard Gillilan Reply-To: reg8@cornell.edu X-Mailer: Mozilla 4.75 [en] (X11; U; Linux 2.2.16-3 i686) X-Accept-Language: en MIME-Version: 1.0 To: Ulrike Salzner , chemistry@ccl.net Subject: Re: CCL:programming language References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Ulrike Salzner wrote: > We are a chemistry department and our first year students take a computer > science cours learning JAVA. They told me that they don't think it is of > any use to them. > > I would like to collect opinions which programming language is most useful > for chemists. Let's assume they are learning only one. Having been in a computer-science environment for many years and then having moved into an experimental lab environment, I have a couple different perspectives to offer. JAVA is very popular among students, CS and non-CS, however there are some drawbacks. JAVA is very slow, even in compiled form and it is my understanding that industry has largely failed to adopt it for large projects because of its lack of support for templates and other reasons. In practice, we don't use it at all, although it may be a valuable skill and a good choice for a first language for the following reasons: 1. It is a very good introduction to object-oriented program design. 2. It is a valuable skill for web design, which most students have to do at some point in their career. You might alternatively consider Python or Perl. Both are used probably more than JAVA for scientific applications. In genomics and various kinds of informatics. FORTRAN is very easy to learn, very fast and leads to unmaintainable, unreadable, unreusable, interdependent code that tends to get lumped into a single file with no dynamic memory allocation (do I sound biased? ;). The software is usually further obfuscated by layers of shell scripts, or worse, layers of invented scripting language. Yes some people do write good FORTRAN code, unfortunately most learn by example. FORTRAN will be around for a long time to come, so you'll probably need to learn it ... it won't take long. C/C++ are the real standards in the CS world. C++ is harder to learn and has considerable potential for misuse. It's very hard to find students with good C++ skills. I write all my new code in C++ and will never go back to FORTRAN. It is nearly as fast as FORTRAN, and may infact exceed it in some special applications. (There was a FORTRAN vs C++ thread on this list some time ago). I HIGHLY recommend this book if you can find a copy: Enough Rope to Shoot Yourself in the Foot : Rules for C and C++ Programming by Allen I. Holub The emerging software design concepts like "Design Patterns" and "Refactoring" are not on most people's radar yet. They may or not not become popular ... the big problem I see is that, like C++, they take additional time to learn and will most likely not filter into the scientific programing world for a long time. Richard Gillilan MacCHESS and Molecular Biology and Genetics Cornell University From chemistry-request@server.ccl.net Wed Jan 31 15:36:55 2001 Received: from havana.ks.uiuc.edu (havana.ks.uiuc.edu [130.126.120.73]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VKatw06166 for ; Wed, 31 Jan 2001 15:36:55 -0500 Received: from verdun.ks.uiuc.edu (verdun.ks.uiuc.edu [130.126.120.129]) by havana.ks.uiuc.edu (8.9.1/8.9.1) with ESMTP id OAA18813; Wed, 31 Jan 2001 14:36:56 -0600 (CST) Received: from localhost (jim@localhost) by verdun.ks.uiuc.edu (8.9.3+Sun/8.9.1) with ESMTP id OAA11391; Wed, 31 Jan 2001 14:36:56 -0600 (CST) X-Authentication-Warning: verdun.ks.uiuc.edu: jim owned process doing -bs Date: Wed, 31 Jan 2001 14:36:55 -0600 (CST) From: Jim Phillips To: "Mark A. Zottola" cc: Ulrike Salzner , chemistry@ccl.net Subject: Re: CCL:programming language In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, I would say FORTRAN is the default scientific programming language because plain vanilla FORTRAN compiles just about everywhere. Of course, with a language like Java it will run the same everywhere as well whereas FORTRAN leaves behaviors like aliasing undefined. FORTRAN also teaches bad programming practices like fixed size arrays, global data, and all caps. As for performance, most students don't know about -O and their code just runs "fast enough". For them a better algorithm or interface matters much more. Start with a scripting language like Python, or just use Matlab and Mathematica. If you care about performance, learn C and use something like SWIG to speed up the part of the code that matters. If you really care about performance, learn C because it is an excellent model of what the compiler is really doing and you can handle any data structure you want. If you still need FORTRAN at that point, go for it, but only after you learn what the "restrict" keyword is for in C. -Jim From chemistry-request@server.ccl.net Wed Jan 31 15:45:11 2001 Received: from jam.ca (ns.worldsfinest.net [207.236.154.51] (may be forged)) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VKjAw06224 for ; Wed, 31 Jan 2001 15:45:11 -0500 Received: from localhost.localdomain (firewall.chemcomp.com [209.226.194.162]) by jam.ca (8.9.3/8.9.3) with ESMTP id PAA18297; Wed, 31 Jan 2001 15:45:04 -0500 Received: from taz (cotter.chemcomp.com [192.1.1.154]) by localhost.localdomain (8.9.3/8.9.3) with ESMTP id PAA28838; Wed, 31 Jan 2001 15:38:05 -0500 Message-ID: <01d601c08bc7$372813a0$9a0101c0@chemcomp.com> From: "Bill Hayden" To: "Mary O'Connor" , "Ulrike Salzner" , References: <3A78159E.D29BCCEF@eden.rutgers.edu> Subject: Re: CCL:programming language Date: Wed, 31 Jan 2001 15:46:44 -0500 Organization: Chemical Computing Group 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 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Please check out our web site at: http://www.chemcomp.com We have a built-in high level programing language, SVL - Scientific Vector Language which is "chemistry aware". We use it to for developing our built-in applications in our Life Science software, MOE - the Molecular Operating Environment. There are a number of academic sites world-wide that use it as a teaching aid. Thanks, Bill Bill Hayden Vice President, Sales & Marketing CHEMICAL COMPUTING GROUP hayden@chemcomp.com phone - (514) 393 1055 ext. 27 fax - (514) 874 9538 http://www.chemcomp.com ----- Original Message ----- From: Mary O'Connor To: Ulrike Salzner ; Sent: Wednesday, January 31, 2001 8:39 AM Subject: CCL:programming language > > > Ulrike Salzner wrote: > > > We are a chemistry department and our first year students take a computer > > science cours learning JAVA. They told me that they don't think it is of > > any use to them. > > > > I would like to collect opinions which programming language is most useful > > for chemists. Let's assume they are learning only one. > > > > Thanks, > > Ulrike Salzner > > > > =================================================================== > > > > Dr. Ulrike Salzner > > Associate Professor > > Department of Chemistry Tel.: (312) 290-2122 > > Bilkent University Fax.: (312) 266-5097 > > 06533 Bilkent, Ankara e-mail: salzner@fen.bilkent.edu.tr > > Turkey > > > > ==================================================================== > > > > I think JAVA is of most use to people working on webpages (I'm no expert). My > own inquiries into which languages to learn have led me to the conclusion that > the basis of scientific programming is still FORTRAN, being gradually overtaken > by C++ > > Mary O'Connor > Doctoral Candidate > Rutgers University > New Brunswick,NJ > USA > > > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > From chemistry-request@server.ccl.net Wed Jan 31 17:43:47 2001 Received: from darwin-hme0.sepracor.com (mail.sepracor.com [209.48.19.169] (may be forged)) by server.ccl.net (8.11.0/8.11.0) with SMTP id f0VMhlw07611 for ; Wed, 31 Jan 2001 17:43:47 -0500 Received: from mailhost.sepracor.com by darwin-hme0.sepracor.com via smtpd (for server.ccl.net [192.153.40.39]) with SMTP; 31 Jan 2001 23:43:34 UT Received: from getmail.sepracor.com by lovelace-qfe0 via smtpd (for server.ccl.net [192.153.40.39]) with SMTP; 31 Jan 2001 22:43:49 UT Received: by getmail.sepracor.com with Internet Mail Service (5.5.2650.21) id ; Wed, 31 Jan 2001 17:43:30 -0500 Message-ID: From: Gerry Tanoury To: "'chemistry@ccl.net'" Subject: Programming language Date: Wed, 31 Jan 2001 17:43:29 -0500 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Content-Type: text/plain; charset="iso-8859-1" Being relatively new to computational chemistry, and having read the recent posts about FORTRAN as the language for scientific computing, could somebody please suggest a book (or books) that I could use to teach myself FORTRAN. Kind regards, Gerald J. Tanoury gtanoury@sepracor.com From chemistry-request@server.ccl.net Wed Jan 31 18:11:46 2001 Received: from emb22.resnet.cornell.edu (emb22.resnet.cornell.edu [128.253.200.106]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VNBkw07846 for ; Wed, 31 Jan 2001 18:11:46 -0500 Received: from pobox.com (128.253.86.74) by emb22.resnet.cornell.edu with ESMTP (Eudora Internet Mail Server 1.3.1); Wed, 31 Jan 2001 18:13:22 -0500 Message-ID: <3A789BA5.8BE7736F@pobox.com> Date: Wed, 31 Jan 2001 18:11:33 -0500 From: Eric Bennett X-Mailer: Mozilla 4.75 [en] (X11; U; OSF1 V4.0 alpha) X-Accept-Language: en MIME-Version: 1.0 To: chemistry@ccl.net Subject: Re: CCL:programming language References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Irene Newhouse wrote: > *IS* there an example yet of a C/C++ application that runs faster then the > FORTRAN? Speed is not an advantage I've heard computer scientists cite in > the religious wars. I've not heard of FORTRAN being faster either. That should be more of a compiler issue. A good C compiler should produce faster code than a mediocre FORTRAN compiler. The FORTRAN advantage is that it's *easier* to write scientific number-crunching code because of things like built-in support for complex numbers, which C does not have. But C is a more flexible language overall, so it's no surprise computer scientists would prefer it for general purposes and for teaching programming. I learned numerous versions of BASIC, then some assembly, then Pascal, then C. Then I picked up a FORTRAN 77 book and the general structure of the language made me want to gag, so I put the book down and never picked it up again. I use C today. But a friend of mine who is in astrophysics and used to be more of a C zealot than I ever was quickly stopped making fun of FORTRAN when he started writing programs to do calculations. He thought FORTRAN was that much better for what he was doing. But, instead of the personal anecdotes, why not look at what the people who are writing major chemical/biophysical software packages use? There are numerically intensive scientific apps written under both FORTRAN and C/C++, so that's good evidence that either choice is appropriate, but > from what I can tell the majority still seem to be FORTRAN. And many are still under very active development, so FORTRAN is hardly being consigned to the dustbin. Even though I don't like FORTRAN myself I'd have to give it a slight edge for chemistry applications. But either language would be more appropriate than Java for a chemist who is only going to learn one language. How many numerically intensive Java programs are there in widespread use? I'm not aware of any. >When the problems you're interested >in push the limits of the available machines, performance outranks most >other language criteria. Hmmm, then you should have suggested assembly language. :*) Performance only ranks highly in a language comparison when the two languages have very *different* speeds. If the language is compiled and there are highly optimized compilers are available, as for C/C++ and FORTRAN, then this shouldn't be a major factor. -- Eric Bennett Cornell University, Chemistry & Chemical Biology From chemistry-request@server.ccl.net Wed Jan 31 18:48:59 2001 Received: from fulcrum.physbio.mssm.edu (fulcrum.physbio.mssm.edu [146.203.4.14]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VNmww08346 for ; Wed, 31 Jan 2001 18:48:58 -0500 Received: from mezei (helo=localhost) by fulcrum.physbio.mssm.edu with local-esmtp (Exim 3.11 #1) id 14O6zk-008C3k-00 for CHEMISTRY@ccl.net; Wed, 31 Jan 2001 18:49:00 -0500 Date: Wed, 31 Jan 2001 18:49:00 -0500 From: Mihaly Mezei X-Sender: mezei@fulcrum.physbio.mssm.edu To: Computationl Chemistry List Subject: Programming language Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: Mihaly Mezei We had plenty of postings, so one more should not hurt too much :) . First, let's remenber that the origial posting was asking for suggestions that applies to all chemistry majors. Since theorist are not an overwhelming majority, the issue of Fortran vs C/C++/Java is not too relevant to the original posting. For the general chemistry (or science) student, I would tend to agree with the suggestion of Prof. Kimberley Cousins: Matlab or Mathematica, or Maple and the like. These all include the basic programming elements (including graphics in most cases!) without the more computer-specific hassle (like format statements). On the Fortan vs object oriented language issue I would only venture to suggest that it helps to have a foundation of the basic programming elements before getting into OO stuff - once you are writing a 'method', it still has to be a straight program. Mihaly Mezei, Dept. of Physiology & Biophysics, Mount Sinai School of Medicine Voice: (212) 241-2186 Fax: (212) 860-3369| WWW: http:// adsr13.mssm.edu/domains/dept/facultyInfo.epl?objname=physbio&user=mezeim01 From chemistry-request@server.ccl.net Wed Jan 31 19:10:37 2001 Received: from chemsun.chem.umn.edu (chemsun.chem.umn.edu [128.101.157.2]) by server.ccl.net (8.11.0/8.11.0) with SMTP id f110Aaw08462 for ; Wed, 31 Jan 2001 19:10:36 -0500 Received: (qmail 43113 invoked by uid 1120); 31 Jan 2001 18:10:38 -0600 Received: from localhost (sendmail-bs@127.0.0.1) by localhost with SMTP; 31 Jan 2001 18:10:38 -0600 Date: Wed, 31 Jan 2001 18:10:38 -0600 (CST) From: Lakshmi Devi Kesavan To: chemistry@ccl.net Subject: CCL: self-learning fortran In-Reply-To: <3.0.5.32.20010131130558.0087eda0@semovm.semo.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello all, While ccl is discussing about programming languages, I would like to get more information on self-learning fortran. Please recommend 1. web tutorials; 2. good easy-to-follow books; or any other resources. Thanks in advance, Lakshmi ************************* Lakshmi S Devi Kesavan Graduate Student 1012, 27th Ave SE, Department of Chemistry Apt # A 207, Pleasant Street S.E, Minneapolis, University of Minnesota MN-55414 Minneapolis, MN 55455-0431 Off # : 612-625-5325 ************************* From chemistry-request@server.ccl.net Wed Jan 31 19:01:08 2001 Received: from havana.ks.uiuc.edu (havana.ks.uiuc.edu [130.126.120.73]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f11018w08402 for ; Wed, 31 Jan 2001 19:01:08 -0500 Received: from geneseo.ks.uiuc.edu (geneseo.ks.uiuc.edu [130.126.120.127]) by havana.ks.uiuc.edu (8.9.1/8.9.1) with ESMTP id SAA20817; Wed, 31 Jan 2001 18:01:09 -0600 (CST) Received: (from johns@localhost) by geneseo.ks.uiuc.edu (8.9.3+Sun/8.9.1) id SAA00613; Wed, 31 Jan 2001 18:01:09 -0600 (CST) Date: Wed, 31 Jan 2001 18:01:09 -0600 From: John Stone To: Irene Newhouse Cc: "Mark A. Zottola" , Ulrike Salzner , chemistry@ccl.net Subject: Re: CCL:programming language Message-ID: <20010131180109.B541@geneseo.ks.uiuc.edu> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: ; from newhoir@mail.auburn.edu on Wed, Jan 31, 2001 at 02:08:01PM -0600 Hi, Someone forwarded this note to me, and I can't resist interjecting a few thoughts for people to keep in mind.... On Wed, Jan 31, 2001 at 02:08:01PM -0600, Irene Newhouse wrote: > Since a poll question was asked, I wish to be polled in favor of FORTRAN > as well, for scientific uses. > > *IS* there an example yet of a C/C++ application that runs faster then the > FORTRAN? Sure, there are tons of programs that are entirely impractical to write in Fortran, which is why they are written in other languages. Programs requiring sophisticated data structures are often difficult to write in Fortran. Granted, you can express the same algorithm many different ways, but some algorithms map particularly well to certain data structures, and these data structures often map well to something other than simple arrays. This is not to say that there's no place for Fortran, but there are many applications where Fortran is simply an ill-suited language. The same goes for C, C++, Java, etc... None of these languages is master of all tasks, they each have advantages and disadvantages. As far as more directly addressing your question about whether or not there are C/C++ applications that run faster than Fortran versions of the same applications, I'm sure there are many. Most graphics programs are written in C/C++ due to the system interfaces being implemented in C, but the other side of that coin is that much of the processing performed by graphics software tends to benefit from the use of complex and dynamic data structures. Granted, if you use Fortran 90 you can probably do many of these same things now, but most "Fortran" codes out there are still old-style Fortran 77 type stuff. There are lots of things that I'd have a hard time imagining someone using Fortran to implement such as 3-D graphics software (not entirely Fortran's fault), GUIs, complex text or other parsing software, systems programs like web servers, network daemons, etc. Sure, you could use Fortran to implement these things if there were appropriate Fortran system interfaces, but would certainly be a painful task... > Speed is not an advantage I've heard computer scientists cite in > the religious wars. The advantages I've seen cited are, to my jaundiced > view, disguises for "programming this way results in code easier for > professors/TAs to grade", not performance advantages. That's an awefully academia-centric view... In the real-world, people use whatever language does the job best _for them_. Some people use whatever language they already know just because it is much more convenient than having to learn yet another one. If development time is their primary constraint, then this is a perfectly reasonable choice. If performance is your main criteria, then you should use the language which yields peak performance on your algorithm, with the type of data structures that you use. There's still a lot more to this issue than just the language itself. In truth, many languages are _fine_ for most tasks, but the compilers and system libraries available on the computers in the field are sorely lacking for one of the "less popular" languages. For example, I could give lots of reasons why Pascal and Modula are nicer languages than C or Fortran, but that doesn't change the fact that I'll have one _heck_ of a time trying to get compilers for all of the different machines that I might want to run my code on, nevertheless machines and compilers than yield good performance on the final executable code. So, given these considerations, you can easily see why nobody writes heavy networking, graphics, or GUI code in Fortran.... Its not that it cannot be done, but because its much more trouble than it is worth. There have been lots of graphics programs written in C and Fortran, and I'd have to say that to cite these as a specific example, C is probably the better overall language for that type of work, especially now.... I'd hate to contemplate implementing something like the computational core of Renderman with its shader language (compiled/interpreted) in Fortran... You could probably do it, but it would be a nightmare. But, by the same token, I don't think I'd want to use C for implementing algorithms with lots of complex math or linear algebra, provided that I didn't need sophisticated data structures, if performance were my main criteria. Fortran is the obvious choice there... > Someone once said "if we had a teraflop machine with petabytes of memory > today, the comp. chemists would be happy for about a week. Then they'd > start asking 'why can't we do this?'" When the problems you're interested > in push the limits of the available machines, performance outranks most > other language criteria. Well, the truth is that _FUNDING_ is the fundamental constraint. You can do about anything you want, including building custom hardware, if you have the funding. The real constraint is that we all have to do our research within the funding that is available, so we are forced to balance our costs between the "human costs" of developing high performance software and continuing to maintain it, versus the hardware costs of buying a faster machine, or building custom hardware just for our particular simulation. If you want a fascinating example of a field where they DID build special hardware, take a look at the "GRAPE" project by the astrophysics guys... http://www.abcnews.go.com/sections/tech/CuttingEdge/cuttingedge000602.html http://www.supercomp.org/sc95/proceedings/721_JMAK/SC95.HTM If performance was truly the limiting factor, we'd all build our own special hardware like these guys have, and like the genomics people have recently started to. Actually, this is another good example of truly hard-core performance: http://www.timelogic.com/ > When you're interested in doing something with a > machine, you want to be able to reuse what's out there that's close & not > have to start from scratch. For chemists, what's out there is FORTRAN. > If it ain't broken, don't fix it... I agree entirely with what you say here. This is why people don't throw away legacy code that still performs perfectly well, and why there are new college graduates taking jobs maintaining COBOL, etc. Again, I'm not trying to say that Fortran is bad, but that it isn't _ALWAYS_ the best language, not by convenience, and not by performance either. Best to be open minded and evaluate which language is best for each task rather than assuming any one is better than the rest.... John Stone johns@ks.uiuc.edu (not speaking for my employer) -- NIH Resource for Macromolecular Modeling and Bioinformatics Beckman Institute for Advanced Science and Technology University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801 Email: johns@ks.uiuc.edu Phone: 217-244-3349 WWW: http://www.ks.uiuc.edu/~johns/ Fax: 217-244-6078 From chemistry-request@server.ccl.net Wed Jan 31 17:27:14 2001 Received: from ccl.net (atlantis.ccl.net [192.148.249.4]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0VMRDw07422 for ; Wed, 31 Jan 2001 17:27:13 -0500 Received: from gothos.ccl.net (gothos.ccl.net [192.148.249.204]) by ccl.net (8.9.3/8.9.3/OSC 2.0) with ESMTP id RAA20185; Wed, 31 Jan 2001 17:26:54 -0500 (EST) Date: Wed, 31 Jan 2001 17:26:54 -0500 From: Scott Brozell To: Ulrike Salzner cc: ccl Subject: Re: CCL:programming language In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, For first year undergraduate chemistry students I think Java is a good choice; Basic also would be fine. My reasoning is that the simple Basic or the omnipresent Java will be most useful to undergraduate students because what little programming any of them will do is likely to involve the web or to be so simple that any procedural language will suffice. For first year graduate chemistry students C++ or Fortran 95 would be my first choices followed by Java, C, or Fortran 77. C++, Fortran 95, and Java offer support for more programming paradigms than the older languages and imo encourage sound design and coding practices. But I think most any procedural language will suffice for even graduate chemistry students, most of whom won't do much programming. Another type of language that has not been mentioned yet is the symbolic math language, such as, Maple or Mathematica. These can be quite useful for graduate level work. Obviously, for theoreticians and computational scientists knowledge of Fortran is essential. However, C++ and C are very popular among graduate students based on enrollment in C++ and C programming courses offered by the Ohio SuperComputer Center. Those languages are being used in the sciences: http://oonumerics.org/blitz/ http://www-fp.mcs.anl.gov/petsc/ http://www.acl.lanl.gov/pooma/ http://math.nist.gov/sparselib/ etc. Scott Brozell, Ph.D. in Chemical Physics Science and Technology Support Ohio SuperComputer Center On Wed, 31 Jan 2001, Ulrike Salzner wrote: > > We are a chemistry department and our first year students take a computer > science cours learning JAVA. They told me that they don't think it is of > any use to them. > > I would like to collect opinions which programming language is most useful > for chemists. Let's assume they are learning only one. > > Thanks, > Ulrike Salzner > > =================================================================== > > Dr. Ulrike Salzner > Associate Professor > Department of Chemistry Tel.: (312) 290-2122 > Bilkent University Fax.: (312) 266-5097 > 06533 Bilkent, Ankara e-mail: salzner@fen.bilkent.edu.tr > Turkey From chemistry-request@server.ccl.net Wed Jan 31 19:43:50 2001 Received: from ccl.net (atlantis.ccl.net [192.148.249.4]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f110hnw08698 for ; Wed, 31 Jan 2001 19:43:49 -0500 Received: from sgi04-e.std.COM (sgi04-e.std.com [199.172.62.134]) by ccl.net (8.9.3/8.9.3/OSC 2.0) with ESMTP id TAA21902 for ; Wed, 31 Jan 2001 19:43:48 -0500 (EST) Received: from world.std.com (world-f.std.com [199.172.62.5]) by sgi04-e.std.COM (8.9.3/8.9.3) with ESMTP id SAA15624 for ; Wed, 31 Jan 2001 18:33:15 -0500 (EST) Received: (from jle@localhost) by world.std.com (8.9.3/8.9.3) id TAA14294 for chemistry@ccl.net; Wed, 31 Jan 2001 19:43:46 -0500 (EST) Date: Wed, 31 Jan 2001 19:43:46 -0500 (EST) From: Joe M Leonard Message-Id: <200102010043.TAA14294@world.std.com> To: chemistry@ccl.net Subject: CCL:programming language Folks, I've tried to avoid emailing a reply to this, as it's basically a flame war with no real resolution - but I like wearing Nomex underwear... Learning Fortran alone is career-limiting, period. There are too many restrictions - no memory allocation, limited system calls, poor string handling, etc. "Good code" written in many languages will outperform "bad code" written in Fortran. Algebraic/formulaic code looks pretty much the same in any language... If one has to maintain dusty-deck codes, then yeah, Fortran is required. However, knowing a second language lets one do the CPU-intensive stuff in Fortran (if one desires) and the "interacts with world" stuff in a more-suited language. Select the tool for the task. Rather than concentrating on a language, concentrate on how programs are designed, constructed and maintained. A small bit of C lets one, for example, ship a single binary which is capable of sizing itself - rather than forcing a set of binaries for differing problem-size memory requirements. Finally, Fortran has never been loved by those who write compilers (except for some for-purchase versions). Therefore, if you're limiting yourself to Open Source tools, writing in Fortran will slow you down as nobody's taken the time to add the necessary optimizing stuff to g77. Times change - I've written miles of stuff in various languages, and (repeating myself, sorry) the trick is to write code which solves the problem and can be maintained. Don't teach "Fortran Programming", teach "Programming". If you want to teach numerically intensive programming, use whatever language you want, and avoid the language features which do not positively contribute to performance... Also consider the effect of debugging and testing tools - graduate student time is cheap, but it can be hard on the grad student. Joe jle@world.std.com