From owner-chemistry@ccl.net Tue Oct 25 04:12:01 2005
From: "Ding Xunlei dxl^^^ustc.edu.cn" <owner-chemistry|a|server.ccl.net>
To: CCL
Subject: CCL: calculation on C60 using DMol3
Message-Id: <-29719-051024233121-27481-61ExEcgEI6Yy2M6+alOZKA|a|server.ccl.net>
X-Original-From: "Ding Xunlei" <dxl[a]ustc.edu.cn>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain;
charset="gb2312"
Date: Tue, 25 Oct 2005 11:31:02 +0800
Mime-Version: 1.0
Sent to CCL by: "Ding Xunlei" [dxl%%ustc.edu.cn]
Dear CCLs,
I get an error when I do an energy calculation on C60 with DMol3, using dnd basis set and set "symmetry on"( C60 has Ih symmetry). The message is below:
"Point group symmetry ih symmetry orbital prototypes generated (SYMDEC)
Too many coefficients procof **** in T1U.1
Message: DMol3 job failed
Error: DMol3 exiting"
If set "symmetry off" or use dn basis set, the calculation is successful.
Then how can I do the calculation with symmetry on and dnd basis set?
Thank you!
Yours sincerely,
Ding Xunlei
2005-10-25
______________________________________________
From owner-chemistry@ccl.net Tue Oct 25 05:03:00 2005
From: "Van Dam, HJJ \(Huub\) h.j.j.vandam|*|dl.ac.uk" <owner-chemistry],[server.ccl.net>
To: CCL
Subject: CCL: W:Calculation of Point Group
Message-Id: <-29720-051025050013-9190-eSc6C2yCCSce8pLLLmWjCA],[server.ccl.net>
X-Original-From: "Van Dam, HJJ \(Huub\)" <h.j.j.vandam]=[dl.ac.uk>
Content-Class: urn:content-classes:message
Content-Transfer-Encoding: 8bit
Content-Type: text/plain;
charset="us-ascii"
Date: Tue, 25 Oct 2005 09:24:57 +0100
MIME-Version: 1.0
Sent to CCL by: "Van Dam, HJJ \(Huub\)" [h.j.j.vandam===dl.ac.uk]
Hi,
Are you running this from the windows command prompt or are you double
clicking it?
Huub
-----Original Message-----
> From: owner-chemistry*_*ccl.net [mailto:owner-chemistry*_*ccl.net]
Sent: 24 October 2005 19:34
To: Vandam, Huub
Subject: CCL: W:Calculation of Point Group
Hi Serguei,
i went to the mentioned website and downloaded the symmetry.c file. i am
using Bloodshed Dev C++ IDE, it can handle C files too. i build the file
and execute but all it does is give a black command prompt box, just
like we the one we get in WinXp. i don;t understand the mistake i am
committing.
further i thought probably the zip files would be the updated version
but same thing happened. could you plz tell what is mistake i am making.
thanks=
From owner-chemistry@ccl.net Tue Oct 25 06:07:01 2005
From: "David Cornil cornildavid_._yahoo.fr" <owner-chemistry:-:server.ccl.net>
To: CCL
Subject: CCL: W:MAKPOL?!
Message-Id: <-29721-051025060616-4666-tj+775NnS7/RvPy7U9iqNg:-:server.ccl.net>
X-Original-From: "David Cornil" <cornildavid]~[yahoo.fr>
Sent to CCL by: "David Cornil" [cornildavid,;,yahoo.fr]
Hello all,
We did calculation on stacks with different molecules to calculate the ratio between the total dipole moment of the stack by the number of molecules. But we can not increase the number of molecules because of the atoms number limitation in the Mopac package.
Our purpose is to do a AM1 calculation on a stack composed by an "infinite" number of molecules. This can be done with the help of periodic conditions. We heard about the possibility of doing this with the MAKPOL program included in the Mopac package.
The trick in this program is to define a cell unit and to define the directions (the vector(s)) in wich the cell is repeating.
Execute MAKPOL just gives us a repetition of the unit cell but does not give a result of an AM1 calculation.
Execute a AM1 calculation with the following keywords, atom coordinates and unit cell vectors
mers=(3) AM1 1SCF GRADIENTS
C 0.000000 0 0.000000 0 0.000000 0 0 0 0
C 1.388330 1 0.000000 0 0.000000 0 1 0 0
C 1.406220 1 119.267000 1 0.000000 0 2 1 0
C 1.404060 1 120.616510 1 -0.012470 1 3 2 1
C 1.390640 1 119.257700 1 0.006220 1 4 3 2
C 1.404270 1 120.757950 1 0.011770 1 1 2 3
H 1.102180 1 119.631660 1 -179.989580 1 1 2 3
H 1.103920 1 120.469820 1 179.991700 1 2 1 3
H 1.103860 1 120.330350 1 -179.994720 1 4 3 2
H 1.101150 1 119.006770 1 -179.999380 1 5 4 3
S 1.686630 1 116.640520 1 179.994790 1 6 1 2
H 1.322650 1 100.218760 1 179.985000 1 11 6 1
N 1.483870 1 119.687200 1 179.987820 1 3 2 1
O 1.202210 1 118.993250 1 0.043290 1 13 3 2
O 1.202430 1 118.960780 1 -179.955960 1 13 3 2
XX 4.000000 0 90.000000 0 90.000000 0 1 2 3
TV 4.000000 1 90.000000 1 90.000000 1 1 2 3
gives the dipole moment of the molecule included in our unit cell but when we change mers=(3) by mers=(100), the dipole moment does not change !!!!!
Maybe we miss something, is there anybody can help us :
- to use MAKPOL in a right way
- to understand the problem with Mopac calculation here above
- to find maybe an other way to do periodic condition with AM1 Hamiltonian
Thank you in advance
Cornil David
Centre de recherche Materia Nova
Service de chimie des matriaux nouveaux
Mail : cornildavid]*[yahoo.fr
From owner-chemistry@ccl.net Tue Oct 25 06:42:02 2005
From: "Noel O Boyle no228||cam.ac.uk" <owner-chemistry*_*server.ccl.net>
To: CCL
Subject: CCL: Summary: Where can you publish articles on software?
Message-Id: <-29722-051025062544-17808-6EcaXVDyVUTrvgKN6rN2Tg*_*server.ccl.net>
X-Original-From: "Noel O'Boyle" <no228.:.cam.ac.uk>
Content-Transfer-Encoding: 8bit
Content-Type: text/plain; charset=ISO-8859-1
Date: Tue, 25 Oct 2005 11:25:32 +0100
Mime-Version: 1.0
Sent to CCL by: "Noel O'Boyle" [no228,;,cam.ac.uk]
Dear CCLers,
My original question was:
"Where can one publish articles on programs that are of general interest
to computational chemists, but which do not contain new algorithms?"
Thanks to all those that contributed to the very lively discussion that
followed. There appear to be two journals in which it may be possible to
publish: Computer Physics Communications, and Journal of Computational
Chemistry.
Off-topic a number of issues were raised, which make for interesting
reading (see below). I don't want to prolong this topic, but few people
addressed the effect on science of not encouraging chemists to share
code (which is one of the side-effects of not encouraging publication).
On this list, people often write in looking for free code - how long
will people continue to share code if they cannot get any academic
recognition for the time they spend on it?
Here is the original email:
======================================================================
Dear CCLers,
Where can one publish articles on programs that are of general interest
to computational chemists, but which do not contain new algorithms?
For example, I have a program that parses the output of the logfiles of
certain computational chemistry programs, shows the progress of the
calculation graphically and can convolute some spectra. I would say that
this is of general interest (judging by the number of downloads), but I
cannot figure out whether any journals would accept an article
describing the software (which is freely available, and open source).
Regards,
Noel O'Boyle.
(http://gausssum.sf.net is the program in question)
======================================================================
Replies:
======================================================================
Sent to CCL by: David van der Spoel [spoel_._xray.bmc.uu.se]
On Tue, 2005-10-11 at 09:45 +0100, Noel O Boyle no228 ~~ cam.ac.uk
wrote:
> Sent to CCL by: "Noel O'Boyle" [no228]_[cam.ac.uk]
> Dear CCLers,
>
> Where can one publish articles on programs that are of general
interest
> to computational chemists, but which do not contain new algorithms?
Comp. Phys. Comm. does that.
>
> For example, I have a program that parses the output of the logfiles
of
> certain computational chemistry programs, shows the progress of the
> calculation graphically and can convolute some spectra. I would say
that
> this is of general interest (judging by the number of downloads), but
I
> cannot figure out whether any journals would accept an article
> describing the software (which is freely available, and open source).
>
> Regards,
> Noel O'Boyle.
> (http://gausssum.sf.net is the program in question)>
>
--
David.
________________________________________________________________________
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596, 75124 Uppsala, Sweden
phone: 46 18 471 4205 fax: 46 18 511 755
spoel#,#xray.bmc.uu.se spoel#,#gromacs.org
http://xray.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
======================================================================
Sent to CCL by: Sergio Emanuel Galembeck [segalemb%%usp.br]
Noel,
I suggest Journal of Computational Chemistry.
Best regards,
Sergio
Sergio Emanuel Galembeck
Assistant Professor in Physical Chemistry
Departamento de Quimica
Faculdade de Filosofia, Ciencias e Letras de Ribeirao Preto
Universidade de Sao Paulo
Av Bandeirantes, 3900
Ribeirao Preto, SP
Brasil
======================================================================
Sent to CCL by: Jonathan Brecher [jsb],[cambridgesoft.com]
At 4:45 AM -0400 10/11/05, Noel O Boyle no228 ~~ cam.ac.uk wrote:
>Where can one publish articles on programs that are of general interest
>to computational chemists, but which do not contain new algorithms?
>
>For example, I have a program that parses the output of the logfiles of
>certain computational chemistry programs, shows the progress of the
>calculation graphically and can convolute some spectra. I would say
that
>this is of general interest (judging by the number of downloads), but I
>cannot figure out whether any journals would accept an article
>describing the software (which is freely available, and open source).
JCIM (the former JCICS) is pretty flexible about accepting papers
that describe new software. But *please* don't waste everyone's time
if you're simply looking for publicity. Find something
scientifically interesting to say, or don't say anything.
I once reviewed a paper that spent most of a page describing the
behavior of a program's About Box. That paper didn't exactly get
high marks from me...
Jonathan Brecher
jsb/a\cambridgesoft.com
======================================================================
Sent to CCL by: "Konrad Hinsen" [khinsen(-)cea.fr]
On Oct 11, 2005, at 10:45, Noel O Boyle no228 ~~ cam.ac.uk wrote:
> Where can one publish articles on programs that are of general
interest
> to computational chemists, but which do not contain new algorithms?
The Journal of Computational Chemistry has a section for articles
describing programs, but I don't
know what their criteria for acceptance are.
There is also Computer Physics Communications, which has some
chemistry-oriented contributions in
spite of "physics" in the title.
---------------------------------------------------------------------
Konrad Hinsen
Laboratoire Lon Brillouin, CEA Saclay,
91191 Gif-sur-Yvette Cedex, France
Tel.: +33-1 69 08 79 25
Fax: +33-1 69 08 82 61
E-Mail: khinsen+/-cea.fr
======================================================================
Sent to CCL by: Michel Petitjean [ptitjean . itodys.jussieu.fr]
To: chemistry^^ccl.net
Subject: CCL: Re: Where can you publish articles on software?
This programme is surely useful, but the editors of research journals
would probably not accept it for publication. If this kind of article
is acceptable, I guess that many people would produce tons of technical
programmes and have hundreds of papers in their CV, then apply for
some full Professor position without having done any scientific
research.
Publishing programmes for computational chemistry (or other) means
(a) creating an original algorithm, and (b) programming it, and (c)
running it on suitable data of interest.
Regrettably, numerous molecular modeling papers published on the basis
of step (c) alone bear the same value that papers involving (a)+(b)+(c).
Michel Petitjean, Email:
petitjean^^itodys.jussieu.fr
ITODYS (CNRS, UMR 7086) ptitjean^^ccr.jussieu.fr
1 rue Guy de la Brosse Phone: +33 (0)1 44 27 48 57
75005 Paris, France. FAX : +33 (0)1 44 27 68 14
http://petitjeanmichel.free.fr/itoweb.petitjean.freeware.html
======================================================================
Sent to CCL by: "Noel O'Boyle" [no228:+:cam.ac.uk]
On Tue, 2005-10-11 at 16:23 +0200, Michel Petitjean
ptitjean-.-itodys.jussieu.fr wrote:
> Sent to CCL by: Michel Petitjean [ptitjean . itodys.jussieu.fr]
> To: chemistry^^ccl.net
> Subject: CCL: Re: Where can you publish articles on software?
>
> This programme is surely useful, but the editors of research journals
> would probably not accept it for publication. If this kind of article
> is acceptable, I guess that many people would produce tons of
technical
> programmes
This is exactly what I think *should* be encouraged! If we look at
bioinformatics, the latest issue of "BMC Bioinformatics" has 7
software articles (this journal has an impact factor of 5.42); the
latest issue of "Bioinformatics" has 9 'Application Notes' (this journal
has an impact factor of 5.74). There is a real interest in useful
software that can make it easier to do science.
> and have hundreds of papers in their CV, then apply for
> some full Professor position without having done any scientific
research.
I don't think this has happened in bioinformatics, but I understand your
point. However, I think that the benefits to the community outweigh this
possibility.
> Publishing programmes for computational chemistry (or other) means
> (a) creating an original algorithm, and (b) programming it, and (c)
> running it on suitable data of interest.
>
> Regrettably, numerous molecular modeling papers published on the basis
> of step (c) alone bear the same value that papers involving
(a)+(b)+(c).
So where is the encouragement for me to write software that can help
other people? To deal with bug reports, feature requests, installation
problems. (You are probably familiar with this also, with your software)
In addition, science should involve building on someone else's work, not
reinventing the wheel. Wouldn't it be better if authors made their code
available so that others could use it? Otherwise every person who wants
to do a particular type of charge analysis, for example, has to read the
algorithm, write the code, and get their result. During my PhD I found
this to be the case for Hirshfeld Population Analysis. I spent a week
and a half reinventing the wheel (and not very well).
This is why I would like to be able to publish an article on a program.
Regards,
Noel
======================================================================
Sent to CCL by: "Perry E. Metzger" [perry++piermont.com]
"Noel O'Boyle" writes:
>>
>> This programme is surely useful, but the editors of research journals
>> would probably not accept it for publication. If this kind of article
>> is acceptable, I guess that many people would produce tons of
technical
>> programmes
>
> This is exactly what I think *should* be encouraged!
Yup...
>> Publishing programmes for computational chemistry (or other) means
>> (a) creating an original algorithm, and (b) programming it, and (c)
>> running it on suitable data of interest.
>>
>> Regrettably, numerous molecular modeling papers published on the
basis
>> of step (c) alone bear the same value that papers involving
(a)+(b)+(c).
> So where is the encouragement for me to write software that can help
> other people? To deal with bug reports, feature requests, installation
> problems. (You are probably familiar with this also, with your
software)
>
> In addition, science should involve building on someone else's work,
not
> reinventing the wheel. Wouldn't it be better if authors made their
code
> available so that others could use it? Otherwise every person who
wants
> to do a particular type of charge analysis, for example, has to read
the
> algorithm, write the code, and get their result. During my PhD I found
> this to be the case for Hirshfeld Population Analysis. I spent a week
> and a half reinventing the wheel (and not very well).
>
> This is why I would like to be able to publish an article on a
program.
Everything you say here makes sense. Science cannot progress on the
attitude that tool building is an unimportant activity (it is not) or
if we discourage tool building, or if we treat our tools as though
sharing them with others would somehow be negative instead of an
unmitigated positive thing.
One great thing about open source (and I mean real open source, not
the pseudo-open source that frequently is found in the Computational
Chemistry world) is that it can be built on and everyone can share in
the benefits of the improvements, just as we all share in the body of
ideas that is science and try to contribute back so that the knowledge
base increases.
(When I say "pseudo-open", I mean the typical "you can use this for
free but only if you send us a letter in triplicate, and don't dare
alter it or publish the altered code". What is interesting about open
source is *not* the price of the code -- it is the freedom to take
parts of it, or improve it, or rearrange it to fix new problems.)
Perry
======================================================================
Sent to CCL by: "Andy Holder" [holdera%umkc.edu]
In reply to:
> This programme is surely useful, but the editors of research journals
> would probably not accept it for publication. If this kind of article
> is acceptable, I guess that many people would produce tons of
technical
> programmes
>>This is exactly what I think *should* be encouraged! If we look at
>>bioinformatics, the latest issue of "BMC Bioinformatics" has 7
>>software articles (this journal has an impact factor of 5.42); the
>>latest issue of "Bioinformatics" has 9 'Application Notes' (this
journal
>>has an impact factor of 5.74). There is a real interest in useful
>>software that can make it easier to do science.
I am the editor of the Journal of Molecular Graphics and Modelling, and
we specifically EXCLUDE program announcements/descriptions from our
pages. Programs are not original research, but may be implementations
of original research.
Being a former member of the Dewar group, they certainly published a
large number of papers on semiempirical methods, but never published any
article on AMPAC itself. These programs were even run on just a few
examples of "suitable data of interest"! The algorithm is perhaos
worthy of publication, and the program can certainly be mentioned, but
is not in and of itself worthy of the valuable space in a research
journal.
Comparison to various other journals with high impact factors is
anecdotal at best. The data from JMGM indicate that our IF went DOWN
when we included these item and has been steadily increasing since we
eliminated them. But that could also have been for a number of other
reasons.
My 0.02.
Regards, Andy Holder
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
JJJJJJJJ MMM MMM GGG MMM MMM | ANDREW J. HOLDER
JJ MM MM GG GG MM MM | Editor
JJ MMM MMM GG MMM MMM |J. of Molclr Grphcs & Modelling
JJ MM M MM GG GGGG MM M MM | Dept. of Chemistry
JJ JJ MM M MM GG GG MM M MM | Univ. of Missouri-Kansas City
JJ GGG | Kansas City, MO 64110
| holdera-$-umkc.edu
Published by Elsevier Science | (816)235-2293 * (816)235-6543F
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
======================================================================
Sent to CCL by: jle [jle-.-theworld.com]
On Oct 11, 2005, at 11:47 AM, Noel O Boyle no228^-^cam.ac.uk wrote:
>
> In addition, science should involve building on someone else's work,
> not
> reinventing the wheel. Wouldn't it be better if authors made their
code
> available so that others could use it? Otherwise every person who
wants
> to do a particular type of charge analysis, for example, has to read
> the
> algorithm, write the code, and get their result. During my PhD I found
> this to be the case for Hirshfeld Population Analysis. I spent a week
> and a half reinventing the wheel (and not very well).
>
> This is why I would like to be able to publish an article on a
program.
>
Well, if what they want is done by a commercially available program,
they can purchase and use it without concern (assuming most such
programs, particularly in the QM field) are validated and tested well.
Otherwise, they can make the code available under the suitable open-
source license or as a binary distribution on a website and post
pointers
on places such as CCL. We're still quite a small community, so if
something's
good and available, it'll turn up, or be Google-able.
Writing the code's actually the smallest about of work. Debugging it,
validating it, maintaining it and possibly porting it are usually way
more
work :-).
Joe Leonard
jle]![theworld.com
======================================================================
Sent to CCL by: "Noel O'Boyle" [no228===cam.ac.uk]
On Tue, 2005-10-11 at 19:52 -0400, jle jle^^theworld.com wrote:
> Sent to CCL by: jle [jle-.-theworld.com]
> > This is why I would like to be able to publish an article on a
program.
> >
> Well, if what they want is done by a commercially available program,
> they can purchase and use it without concern (assuming most such
> programs, particularly in the QM field) are validated and tested well.
Except that they are restricted to the features available within that
program. For example, I used a well-known commercially-available package
during my PhD - I was very happy with its performance in general, but
when I wanted to do Morokuma decomposition I found that it wasn't
implemented. It *is* implemented in another freely-available package,
but that package wasn't able to converge the SCF. So what we have are
several different programs (some free and open-source, and some
commercially-available) which implement different subsets of useful
algorithms, and which are not interoperable.
> Otherwise, they can make the code available under the suitable open-
> source license or as a binary distribution on a website and post
> pointers
> on places such as CCL. We're still quite a small community, so if
> something's
> good and available, it'll turn up, or be Google-able.
My point is that this doesn't happen - authors do *not* make the code
available, and so other scientists have to write it themselves (and not
all computational chemists can program).
> Writing the code's actually the smallest about of work. Debugging it,
> validating it, maintaining it and possibly porting it are usually way
> more
> work :-).
This is indeed true, and is part of the reason code is not released -
people aren't willing to support it, because of the work involved.
Regards,
Noel
======================================================================
Sent to CCL by: "Warren DeLano" [warren(~)delsci.com]
> Sent to CCL by: "Andy Holder" [holdera%umkc.edu]
Ironically, JMGM editor Andy Holder offers compelling reasons for why
developer-scientists should shift their attention away from
closed-access journals like JMGM and toward peer-reviewed
open-access/open-source venues that better serve the overall needs of
the field.
> I am the editor of the Journal of Molecular Graphics and Modelling,
> and we specifically EXCLUDE program announcements/descriptions from
> our pages. Programs are not original research, but may be
> implementations of original research.
This fails to acknowledge the critical enabling role of software in
computational chemistry and conveys a shockingly dismissive view of the
original research aspects of computer programs! To put it as politely
as I can, I most vehemently disagree. Such comments insult and demean
the creative efforts of those who might have otherwise considered
publishing papers in JMGM.
> The data from JMGM indicate that our IF went DOWN
> when we included these item and has been steadily increasing since we
> eliminated them.
Effective running software is at least as impactful as algorithmic
descriptions, if not more so. (We live in real buildings, not
architectural plans; we travel by plane, not via designs of planes).
"Impact Factor" (I.F.) in terms of follow-on citations has little
relevance as a measure of impact for scientific software and especially
so once usage becomes widespread (e.g. ChemDraw, RasMol, Excel, etc.).
Speaking from personal experience, recognizable PyMOL images are now
routinely found in Science, Nature, C&EN, and many other journals, but
such usage is rarely cited. So what is PyMOL's impact factor? Zero! It
has never been officially "published" per se, even though its source
code has been published continuously on the internet since early 2000.
> The algorithm is perhaps worthy of
> publication, and the program can certainly be mentioned, but is not in
> and of itself worthy of the valuable space in a research journal.
Such comments reflect little "editorial" sympathy for the
publish-or-perish plight faced by academic scientists who develop
practical algorithms in the form of working software programs instead of
abstract algorithms in the form of non-working descriptive publications.
Without such publications, the impact of developer-scientists is
invisible to tenure review committees and funding agencies. That is
unfair to many scientists and damaging to the field as a whole since
better research software is very much needed. Developer-scientists need
to be appropriately recognized and rewarded through career advancement,
and thus it is not by accident that so many talented scientific software
developers have chosen the private sector over the academic world. The
system of academic credit is largely broken with respect to creation of
quality research software.
While JMCM may indeed have a fine "editorial" policy on this matter, it
is justified for the wrong reasons.
Programs cannot truly be "published" in traditional journals because
they are too complex to describe precisely in such venues. A standalone
print article does not deliver reproducibility and verifiability with
respect to research software. Thus, journal articles describing
software do not meet fundamental requirements of the scientific method.
The only way to publish software in a scientifically robust manner is to
share source code, and that means publishing via the internet in an
open-access/open-source fashion. Anything short of that amounts to
issuing unproven claims based on limited empirical tests regarding what
a given program allegedly does. What is that called outside of science?
Advertising! And as such, I agree that it does not belong in a
scientific journal. Either you publish software with source code and
stand behind it, or you are blowing smoke and quite *literally* hiding
something -- no matter how noble your intent.
> Being a former member of the Dewar group, they certainly published a
> large number of papers on semiempirical methods, but never published
> any article on AMPAC itself.
Such comments confirm the notion that research software developers are
not recognized for the critical enabling contributions they make, and
reveal just how incapable the old closed-access, print-journal
publication system is of meeting current needs.
So, go live fellow research software developers, go live! Bypass an
inadequate and obsolete system, and instead pursue internet-based
open-access/open-source publishing of your work. If you lead, the world
will follow.
http://www.plos.org , http://www.doaj.org ,
http://pubchem.ncbi.nlm.nih.gov , http://www.blueobelisk.org ,
http://www.chmoogle.com , http://www.sf.net , etc.
Cheers,
Warren
--
Warren L. DeLano, Ph.D.
Principal Scientist
. DeLano Scientific LLC
. 400 Oyster Point Blvd., Suite 213
. South San Francisco, CA 94080 USA
. Biz:(650)-872-0942 Tech:(650)-872-0834
. Fax:(650)-872-0273 Cell:(650)-346-1154
. mailto:warren%x%delsci.com
======================================================================
Sent to CCL by: John McKelvey [jmmckel]-[attglobal.net]
Folks,
Am I wrong on this: Didn't Tom Ziegler, or someone, publish a nice
article on ADF a while ago? Gave a lot of insight to a very credible
code!!!
Cheers,
John McKelvey
======================================================================
Sent to CCL by Marcel Swart:
Yes, see:
Chemistry with ADF
G. te Velde, F.M. Bickelhaupt, E.J. Baerends, C. Fonseca Guerra, S.J.A.
van Gisbergen, J.G. Snijders, T. Ziegler
Journal of Computational Chemistry, Vol. 22, No. 9, 931-967 (2001)
But that was on the scientific aspects of ADF, not merely the technical
part.
======================================================================
Sent to CCL by: Per-Ola Norrby [pon-$-kemi.dtu.dk]
Dear all,
I would recommend finding an interesting application for your
software, publish the application, and in the same publication, add
the description of the program as supporting information. This will
open up a large number of possible journals, you can still use the
reference as a required one when using your software, and if you
cannot find an interesting, chemistry-related application, it
probably isn't worth publishing in a chemistry journal anyway.
/Per-Ola
--
Per-Ola Norrby, Assoc. Professor, http://organisk.kemi.dtu.dk/PON/
Technical University of Denmark, Department of Chemistry
Building 201, Kemitorvet, DK-2800 Kgs. Lyngby, Denmark
Email: pon]=[kemi.dtu.dk tel +45-45252123, fax +45-45933968
======================================================================
Sent to CCL by: Michel Petitjean [ptitjean[-]itodys.jussieu.fr]
To: chemistry;;ccl.net
Subject: CCL: Re(2): Where can you publish articles on software?
The programming step is often the longest and the most difficult step
(includes intensive validation tests, not to be confused with the
"tests on suitable data of scientific interest").
Most scientists have each morning excellent ideas about new concepts,
algorithms, or methods, but after having tried to programme the stuff,
the ideas may appear to be very difficult to be realized. Worse: anyone
remembers the chief telling to the young scientist: please just do it,
the idea is important, not the technical task. Then the young
collaborator
spends much time and fails, but the chief admits the incompetence of
the collaborator rather than the misconception of the idea.
Clearly, I mean here that the programming step has much value, and that
a good algorithm bears his full value AFTER being implemented in a
programme.
Nevertheless, I share the views of JMGM editor Andy Holder.
(Note: I produce myself softwares and frewares, and I am the
editor in chief of an open access journal),
You can spend weeks and months to produce complicated programmes
of general interest, it does not mean that it should be published
in a scientific journal. The task may be done by an undergraduate
student in computer sciences, or by a team of experienced software
developpers: we do not care. If the utility programme has value,
you can do one of the following (not exclusive):
- Add a section or an appendix in the manuscript to be submitted
- Provide the programme as "supplementary material" to the editor
- Sell the programme (documentation may mention your contribution)
- Put it in open access (documentation may mention your contribution)
- Find a job and earn money in a company producing scientific softwares
But please not:
- Publishing a paper containing only programming stuff.
Even the computer scientists do not publish that, and consider that
the programming step is strictly technical.
If you need to produce a new text editor incorporating utility functions
for computational chemists (e.g. advancerd handling molecular
graphics displays), would you try to publish it ?
If you need this product, you can ask to somebody which is insensitive
to the publish-or-perish requirement. If you have to do the task
yourself, good luck !
The real problem is that using programmes to get results is better
recognized that producing algorithms and programmes, although this
latter scientific task is often longer and more difficult
(furthermore, it needs pluridisciplinarity).
Progress will be made when software will have the same value than
patents (but please NO software patent).
Michel Petitjean Email:
petitjean;;itodys.jussieu.fr
Editor-in-Chief of Entropy entropy;;mdpi.org
ITODYS (CNRS, UMR 7086) ptitjean;;ccr.jussieu.fr
1 rue Guy de la Brosse Phone: +33 (0)1 44 27 48 57
75005 Paris, France. FAX : +33 (0)1 44 27 68 14
http://www.mdpi.net/entropy http://www.mdpi.org/entropy
http://petitjeanmichel.free.fr/itoweb.petitjean.freeware.html
======================================================================
Sent to CCL by: "Konrad Hinsen" [khinsen]^[cea.fr]
On 12.10.2005, at 11:42, Noel O Boyle no228*cam.ac.uk wrote:
>> Well, if what they want is done by a commercially available program,
>> they can purchase and use it without concern (assuming most such
>> programs, particularly in the QM field) are validated and tested
well.
>
> Except that they are restricted to the features available within that
> program. For example, I used a well-known commercially-available
package
Plus they are not extensible.
I think one needs to make a difference between software that is
intimately related to one's research
and software that fulfills generic roles that different readily
available programs can occupy. For
example, I need a lot of control over the software that runs my MD
simulations, but I can live very
well with the text editors and C compilers that the outside world
provides me.
For anyone doing research in computational science, lack of control over
their essential software
implies a sever limitation to scientific creativity. I know many
scientists who do not calculate the
quantities that they would like to look at but the quantities that their
programs allow them to
calculate, even when those quantities are not really adequate. This is
one reason why I consider the
current state of the art in scientific computing (not just in chemistry)
unsatisfactory. Given that it is
not feasible for every scientist to write his own complex software, but
that everyone should be able
to extend and modify it, code quality, in particular readability and
modularity, needs to acquire a
much higher status than it has at the moment, and the education of
computational scientists needs
to include such elements as well as modern development techniques.
> This is indeed true, and is part of the reason code is not released -
> people aren't willing to support it, because of the work involved.
That is indeed a significant problem which I am facing myself as well.
But ultimately it is also a sign
of the lack of recognition for the utility of scientific software
development. In established and
recognized aspects of research, support staff is considered an evident
necessity. Hardly any lab
chemist needs to argue for the necessity of having lab technicians
(obtaining a position in a specific
case if of course another matter). If software development were equally
recognized, we would have
programmer and support staff for big projects.
--
-------------------------------------------------------------------------------
Konrad Hinsen
Laboratoire Leon Brillouin (CEA-CNRS), CEA Saclay,
91191 Gif-sur-Yvette Cedex, France
Tel.: +33-1 69 08 79 25
Fax: +33-1 69 08 82 61
E-Mail: khinsen#%#cea.fr
-------------------------------------------------------------------------------
======================================================================
Sent to CCL by: "Konrad Hinsen" [khinsen!=!cea.fr]
On Oct 13, 2005, at 11:29, Michel Petitjean ptitjean*|*itodys.jussieu.fr
wrote:
> But please not:
> - Publishing a paper containing only programming stuff.
> Even the computer scientists do not publish that, and consider that
> the programming step is strictly technical.
The same can be said for other aspects of research: as long as some work
involves nothing but the
straightforward application of known techniques, it should not be
published in scientific journals.
With occasional exceptions, that's what happens, thanks to the vigilance
of editors and referees.
However, a paper is not automatically strictly technical because its
topic is a piece of software. If
the design or the implementation of software involves new ideas that are
of interest to the
community, then it does in my opinion deserve publication in a
scientific journal. In the case of
complex applications, even the considerations that led the authors adopt
one algorithm over
another one can be sufficient to deserve publication, just as we see
papers that compare different
force fields.
Here are a couple of papers on computational chemistry software that
illustrate my point (pretty
much an arbitrary selection from my paper collection):
J. Comp. Chem. 18, 1848 (1997)
J. Comp. Phys. 151, 283 (1999)
J. Comp. Chem. 21, 79 (2000)
J. Appl. Crystallography 37, 174 (2003)
J. Comp. Chem. 24, 657 (2003)
J. Comp. Chem. 26, 252 (2004)
J. Comp. Chem. 26, 1647 (2005)
None of them describes a straightforward implementation of known
algorithms.
I would wish that journal editors not refuse software-related papers as
a matter of principle, but
that they impose high standards for their evaluation.
--
---------------------------------------------------------------------
Konrad Hinsen
Laboratoire Lon Brillouin, CEA Saclay,
91191 Gif-sur-Yvette Cedex, France
Tel.: +33-1 69 08 79 25
Fax: +33-1 69 08 82 61
E-Mail: khinsen.:.cea.fr
---------------------------------------------------------------------
======================================================================
Sent to CCL by: "Ivanciuc, Ovidiu I." [oiivanci]_[UTMB.EDU]
>>>You can spend weeks and months to produce complicated
>>>programmes of general interest, it does not mean that
>>>it should be published in a scientific journal.
This misconception might explain why chemistry is well
behind biology in terms of number of free software,
Web servers, and databases.
Software and database notes are regularly published in
high-impact biology journals: Bioinformatics (Impact Factor=5.742),
BMC Bioinformatics (IF=5.42), Nucleic Acids Research (IF=7.26).
Each year, Nucleic Acids Research has a special issue for
databases (see 2005 Database Issue,
http://nar.oxfordjournals.org/content/vol33/suppl_1/index.dtl)
and another one for Web servers (see Web Server Issue 2005,
http://nar.oxfordjournals.org/content/vol33/suppl_2/index.dtl).
Updates for databases (GenBank, PDB, SwissProt) or major
bioinformatics software are regularly published, and the rule
is to cite the most recent publication when using a database
or a software (such as alignment tools, Web servers).
The recent trend in computational biology is to move in
areas traditionally belonging to computational chemistry.
An example: NIH started PubChem, a database of organic compounds
(http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pcsubstance).
Very instructive was the answer from the American Chemical Society:
ACS asked the Congress to cut the funding for PubChem, because
it is a public project that competes with a private one (CAS).
If the chemical community wants useful and free software, then it should
accept papers on software.
For the moment, the best advice is to find a potential
application to a biological problem and publish the
software in Bioinformatics or other similar journals.
Regards,
Ovidiu
======================================================================
Sent to CCL by: "TJ O'Donnell" [tjo^-^acm.org]
Well, I've been staying out of this fray, but having just started to
market a closed-source product, this issue has been on my mind a lot
over the past year.
Warren writes:
> The only way to publish software in a scientifically robust manner is
to
> share source code, and that means publishing via the internet in an
> open-access/open-source fashion. Anything short of that amounts to
> issuing unproven claims based on limited empirical tests regarding
what
> a given program allegedly does. What is that called outside of
science?
> Advertising! And as such, I agree that it does not belong in a
> scientific journal. Either you publish software with source code and
> stand behind it, or you are blowing smoke and quite *literally* hiding
> something -- no matter how noble your intent.
This is a rather extreme statement, that needs to be tempered, IMHO.
Hiding source code is NOT tantamount to deception, as is implied above.
Results of a program are, perhaps, the only ~scientifically~ relevant
aspect of
a program; the only useful effect of it; and the only thing that must be
subject
to scientific scrutiny. Results MUST be carefully considered, reviewed
and
verified. If they are in error, one must consider the assumptions,
theory,
algorithms and, of course, the code behind those erroneous results.
Do instrument designers publish their schematics (I honestly don't
know).
Does Varian, et. al.? Does that make those instruments tantamount to
scientific fraud? Absolutely not! Why? Because the RESULTS (spectra,
etc.)
are subject to intense scientific scrutiny.
Let me just finish by saying that I subject my own programs to intense
scientific scrutiny and welcome such from my users. It becomes my
responsibility to correct errors, as any experimentalist does.
For those who may be interested, my new work is at:
http://www.gnova.com/
Sincerely,
TJ O'Donnell, Ph.D.
gNova Scientific Software
http://www.gnova.com/
======================================================================
Sent to CCL by: "Warren DeLano" [warren*|*delsci.com]
> From: TJ O'Donnell [mailto:tjo*acm.org]
>
> Warren writes:
> > The only way to publish software in a scientifically robust
> manner is
> > to share source code, and that means publishing via the
> internet in an
> > open-access/open-source fashion. Anything short of that amounts to
> > issuing unproven claims based on limited empirical tests regarding
> > what a given program allegedly does. What is that called
> outside of science?
> > Advertising! And as such, I agree that it does not belong in a
> > scientific journal. Either you publish software with
> source code and
> > stand behind it, or you are blowing smoke and quite
> *literally* hiding
> > something -- no matter how noble your intent.
> This is a rather extreme statement, that needs to be tempered, IMHO.
> Hiding source code is NOT tantamount to deception, as is
> implied above.
No -- let me clarify -- I do not imply that closed-source is tantamount
to deception. It is simply non-disclosure -- a willful holding back of
pertinent helpful information. It is tantamount to saying "trust me" --
I have correctly applied chemistry, physics, math, and computer science
to create a working solution to your problem.
Thorough testing of closed-source code can of course lay an empirical
foundation for extending such trust, and testing is equally necessary
with open-source code. But testing alone is not the same as disclosing
an implementation that can itself be subjected to direct intellectual
scrutiny.
While there are valid personal, economic, political, legal, practical,
and insitutional reasons for not disclosing source code, I challenge
anyone to come up with a compelling scientific reason for why source
code should not be disclosed -- when possible -- to enable
understanding, reproduction, verification, and extension of
computational advances.
Is there ever a legitimate *purely scientific* reason for settling with
empirical evidence alone (just test results) when mathematical proof is
itself attainable (via inspection of source code)? I cannot think of
any.
Or are we all agreed that making source code available is the
*scientific* ideal to which we should all aspire?
If so, then when we do not make source available, we should certainly
have some compelling non-scientific reason for holding it back, and as
honest scientists, we must realize that doing so will have the effect of
limiting the value and impact of our work -- at least from a scientific
standpoint. Intellectual advances are either shared or lost, and
software implementations are no exception to this.
Cheers,
Warren
PS. A trivial concrete illustration:
I write some function called "add_two_numbers" and share it will my
colleagues. And for the millions of pairs of numbers they test it on,
it returns the sum. So much so good. But without source, no one can
even be sure that there isn't some untested pair of numbers for which it
fails.
Now I share the source:
def add_two_numbers(a,b):
return a+b
> From that point on, everyone can sleep well knowing for certain that
my
function will always return the sum of two numbers, since the
implementation is public and exact. Science can progress because Warren
has re-implemented addition. Hooray!
On the other hand, if the source was reveald to be:
def add_two_numbers(a,b):
if (a == 1823723) and (b == 8374723):
return 6
else:
return a+b
Then everyone would immediately know for certain that my function is
flawed and needs to be fixed. Could testing have found this? Assuming
that the inputs are 32-bits wide, then there are 2^64 combinations of
possible inputs with only one flaw. 2^64 is far too many inputs to
practially examine, so this flaw would almost certainly never have been
found through testing.
Thus, even in the simplest function, it is possible to introduce a flaw
that cannot be found through rigorous regression testing. Admittedly
this example is contrived, but it isn't hard to see how testing can
easily miss subtle problems that could well be identified through
critical analysis of source code.
--
Warren L. DeLano, Ph.D.
Principal Scientist
. DeLano Scientific LLC
. 400 Oyster Point Blvd., Suite 213
. South San Francisco, CA 94080 USA
. Biz:(650)-872-0942 Tech:(650)-872-0834
. Fax:(650)-872-0273 Cell:(650)-346-1154
. mailto:warren*delsci.com
======================================================================
Sent to CCL by: Cory Pye [cpye[-]crux.smu.ca]
On Fri, 14 Oct 2005, Warren DeLano warren],[delsci.com wrote:
> No -- let me clarify -- I do not imply that closed-source is
tantamount
> to deception. It is simply non-disclosure -- a willful holding back
of
> pertinent helpful information. It is tantamount to saying "trust me"
--
> I have correctly applied chemistry, physics, math, and computer
science
> to create a working solution to your problem.
Every time somebody publishes a paper, there is a matter of trust
inherent
between the readers, journal, and author. This trust entails, for
example,
the author declaring "I did not fabricate my data", "I have not already
published this work elsewhere", and the editor declaring "This
manuscript has
undergone a rigorous peer-review process", "I did not let any personal
connection with the author influence my views", etc. There is a certain
amount
of trust inherent in scientific publishing. A (complicated) program is
the
author's "baby", and it is up to the author as to whether he or she
wishes to
make it publically available. I don't think that the trust argument is
applicable here, as it is an inevitable fact.
One difficulty with making a program publically available is it can
mutate into
a non-viable form by an inexperienced programmer, and if the mutant
happens to
be widely circulated, then a lot of the blame ends up on the original
programmer, who has essentially given up his "baby" for adoption,
instead of
the modifier. One can easily waste weeks of time addressing irate
"customers"
because of someone else's goof-up.
>
> Thorough testing of closed-source code can of course lay an empirical
> foundation for extending such trust, and testing is equally necessary
> with open-source code. But testing alone is not the same as
disclosing
> an implementation that can itself be subjected to direct intellectual
> scrutiny.
>
> While there are valid personal, economic, political, legal, practical,
> and insitutional reasons for not disclosing source code, I challenge
> anyone to come up with a compelling scientific reason for why source
> code should not be disclosed -- when possible -- to enable
> understanding, reproduction, verification, and extension of
> computational advances.
Suppose a junior faculty member publishes a paper and publicly releases
some
code with his first graduate student. The idea becomes so popular that a
senior
researcher or company takes that open-source code, with
acknowledgements, and
incorporates it into a popular commercial program, with some
modifications and
writes 2 or 3 papers describing the advances. Science is advanced. Lots
of
people use it, but only quote the paper in the manual of the senior
researcher.
Now suppose that the advances were supposed to be the rest of the
graduate
student's Ph. D. work. This student cannot re-publish this work because
the
other ideas have been published already by the borrower. The junior
faculty,
for lack of sufficient publications, loses his grant, and is denied
tenure.
The student's thesis, being mostly unpublishable, is not accepted.
Has science advanced at the expense of the careers of the two
individuals?
Are computational chemists a species known for eating their young? :-)
Had the junior faculty member not disclosed his source, it would have
been
more difficult for this hypothetical travesty to happen.
The sad reality is that things like this can and do happen and a little
paranoia goes a long way. It would have been far better if the company
had
approached the junior faculty and come to an agreement.
In order to be able to do your own science, pragmatics such as having
tenure,
a grant, etc. have to be looked at first. It is a lot less stressful
(and the
temptation to cut corners less i.e. thorough testing) to implement a new
code,
building on some older code, when you don't have someone breathing down
your
neck trying to outdo you. Writing and debugging code is a lot of effort,
and
you want to be rewarded for your efforts, either through publications,
citations, or financial remuneration.
I would like to say for the record that my experience coding the COSMO
routine
within the ADF package (97-99) has been absolutely wonderful, as I have
found
SCM to be very cooperative in pointing out my errors and vice versa. I
am also
grateful to the many users who have been patient when they hit the
occasional
snag, esp. Heiko Jacobsen and Michael Atanasov. Through
dialogue the resulting code was much better. I will certainly celebrate
with
a bottle of good scotch when I hit 100 citations on the TCA paper
describing
our implementation either late his year or early next year.
>
> Is there ever a legitimate *purely scientific* reason for settling
with
> empirical evidence alone (just test results) when mathematical proof
is
> itself attainable (via inspection of source code)? I cannot think of
> any.
>
Source code is "not" mathematical proof, just ask anyone who had to use
a buggy
Fortran compiler, say when the runtime "check array bounds" debugging
option,
didn't work with a character array.
> Or are we all agreed that making source code available is the
> *scientific* ideal to which we should all aspire?
>
> If so, then when we do not make source available, we should certainly
> have some compelling non-scientific reason for holding it back, and as
> honest scientists, we must realize that doing so will have the effect
of
> limiting the value and impact of our work -- at least from a
scientific
> standpoint. Intellectual advances are either shared or lost, and
> software implementations are no exception to this.
>
> Cheers,
> Warren
>
> [stuff deleted]
************* ! Dr. Cory C. Pye
***************** ! Associate Professor
*** ** ** ** ! Theoretical and Computational Chemistry
** * **** ! Department of Chemistry, Saint Mary's University
** * * ! 923 Robie Street, Halifax, NS B3H 3C3
** * * ! cpye .. crux.stmarys.ca
http://apwww.stmarys.ca/~cpye
*** * * ** ! Ph: (902)-420-5654 FAX:(902)-496-8104
***************** !
************* ! Les Hartree-Focks (Apologies to Montreal Canadien
Fans)
======================================================================
Sent to CCL by: "Warren DeLano" [warren() delsci.com]
Dear Cory,
Thank you for that response, which includes excellent discussion of the
"valid personal, economic, political, legal, practical, and insitutional
reasons for not disclosing source code" to which I referred. Releasing
source code can very well make it harder for the involved parties to do
*more science*, due to distraction, giving up control, lack of proper
credit, loss of funding, risk of tenure, and so forth.
But also understand that by not releasing source code, you absolutely
deny the rest of the scientific community the opportunity of doing *more
science* using your source code. Perhaps your TCA paper would now have
1,000 citations had the code been open from the start. Who can say? It
depends.
Regardless, both sides of the above argument are pragmatic in nature and
do not address the question I again pose:
If we solely consider the standards of disclosure, verifiability, and
reproducibility inherent in the scientific method, is it not the case
that sharing source code comes much closer to meeting that ideal than
does not sharing source code?
And if so, then as good scientists, shouldn't we seek opportunities to
advance science through open source scientific code, whenever possible,
while fully taking into account pragmatic concerns like those you
described?
Cheers,
Warren
PS. Aside...
> Source code is "not" mathematical proof, just ask anyone who
> had to use a buggy Fortran compiler, say when the runtime
> "check array bounds" debugging option, didn't work with a
> character array.
Source code is a symbolic representation of an information process that
has a precise mathematical meaning, subject to the nature of the
implementation language. One can indeed prove mathematical equivalence
or non-equivalence between an abstract mathematical expression and the
mathematical meaning of a given source code implementation.
For example, take the expression "F=ma", and the source code
implementation "def F(m,a): return m*a". In Python, this
implementation is proven to be correct to the precision limits of the
double-precision numeric representation. In contrast "def F(m,a):
return m+a" is proven to be non-equivalent.
In other words, having a broken "+" key on a calculator does not
invalidate Addition any more than having a broken Fortran compiler
invalidates the mathematical validity of the Fortran program you
compile.
--
Warren L. DeLano, Ph.D.
Principal Scientist
. DeLano Scientific LLC
. 400 Oyster Point Blvd., Suite 213
. South San Francisco, CA 94080 USA
. Biz:(650)-872-0942 Tech:(650)-872-0834
. Fax:(650)-872-0273 Cell:(650)-346-1154
. mailto:warren]![delsci.com
======================================================================
Sent to CCL by: John McKelvey [jmmckel : attglobal.net]
Corey Pye commented:
"One difficulty with making a program publically available is it can
mutate into
a non-viable form by an inexperienced programmer, and if the mutant
happens to
be widely circulated, then a lot of the blame ends up on the original
programmer, who has essentially given up his "baby" for adoption,
instead of
the modifier. One can easily waste weeks of time addressing irate
"customers"
because of someone else's goof-up."
I believe that this is exactly what happened to one version of Gaussian
when someone hacked a d-orbital capability and got it wrong ...
John McKelvey
======================================================================
Sent to CCL by: "Perry E. Metzger" [perry*|*piermont.com]
"Cory Pye" writes:
> One difficulty with making a program publically available is it can
> mutate into a non-viable form by an inexperienced programmer, and if
> the mutant happens to be widely circulated, then a lot of the blame
> ends up on the original programmer, who has essentially given up his
> "baby" for adoption, instead of the modifier. One can easily waste
> weeks of time addressing irate "customers" because of someone else's
> goof-up.
I've been writing open source software for something like 15 years
now. I've never once had this happen to me or to anyone I know -- not
once.
> Suppose a junior faculty member publishes a paper and publicly
> releases some code with his first graduate student. The idea becomes
> so popular that a senior researcher or company takes that
> open-source code, with acknowledgements, and incorporates it into a
> popular commercial program, with some modifications and writes 2 or
> 3 papers describing the advances. Science is advanced. Lots of
> people use it, but only quote the paper in the manual of the senior
> researcher.
>
> Now suppose that the advances were supposed to be the rest of the
graduate
> student's Ph. D. work. This student cannot re-publish this work
because the
> other ideas have been published already by the borrower.
Well, suppose you publish some research and, even without the use of
your code, other people make advancements that you wanted to make
first just by looking at your work and saying "hey, this is a cool
thing to try next". Tough luck for you. That's how science works. Of
course, if you aren't silly, you don't mind and it doesn't hurt
you. Einstein didn't come up with spacetime -- Minkowski did, looking
at Einstein's 1905 papers. Did Einstein then winge "you did that
before I could"? No. Did Poincare spend all his time complaining that
Einstein was influenced by him and published first? No.
As it happens, I've had code I've written incorporated into commercial
products that have made other people money and I haven't gotten a
nickel. I've had code I've written taken by others, improved and
re-released as open source, too. I have no regrets at all and continue
to write open source software. My goal when I release something is for
it to be used, you see. If no one uses it, it's as useless as writing
a symphony no one will ever hear or painting a canvas no one will ever
see. Who wants that?
> Has science advanced at the expense of the careers of the two
individuals?
I'm going to be so bold as to say that science doesn't exist to
advance people's careers, and that good science isn't done very often
by people so obsessed with careerism that they withhold information in
the hope of gaining an edge over scientific competitors. I'll go
further and say that people obsessed with their careers to this extent
don't actually have good ones, either on the level of enjoying them or
on the level of getting ahead. Not so paradoxically, doing good work
is more important than relentless scheming and gamesmanship. That
doesn't mean you should ignore personal advancement, but it does mean
that spending all your time being a backbiting bitter schemer doesn't
really do as much good as having good ideas and getting them out.
If you want to live in a world filled with smart people who withhold
information from each other and spend their whole time being
cuthroats, become a hedge fund manager instead of a chemist. Most
chemists are smart enough to do that for a living, and it pays
literally orders of magnitude (no exaggeration) better than
chemistry. If your goal is not to live in that sort of world -- if the
reason you are in science is because you love science (and if you are
in science for any other reason you're probably foolish -- the money
sucks and the recognition sucks compared to almost anything else a
smart person can do for a living), then feel happy that others have
built on your work and build further on it or do something else
interesting. The whole point of science is for people to stand on each
other's shoulders, not on each other's feet.
Long ago in my computer science career I was one of the co-inventors
of a computer security protocol that is currently in use by a large
fraction of the planet. (It is called "IPSEC" for those that know
anything about such things.) Very quickly after I helped create it,
other people became far more expert on it than I was and advanced it
far past where we started. Am I bitter? Of course not. If you're bitter
about stuff like that you're in the wrong field of endeavor. I'm
HAPPY. The thing I helped bring into the world grew up and had a life
past me.
> The sad reality is that things like this can and do happen and a
little
> paranoia goes a long way.
So you claim, but I really think you're presenting a very twisted view
of the world. If you really live your life that way, you're going to
be bitter all the time instead of productive.
>> Is there ever a legitimate *purely scientific* reason for settling
with
>> empirical evidence alone (just test results) when mathematical proof
is
>> itself attainable (via inspection of source code)? I cannot think of
>> any.
>
> Source code is "not" mathematical proof,
No, but it is just as valuable to show your code to someone as to say
what your equipment was, and IMHO just as necessary.
By the way, I'll point out that the bioinformatics guys seem to be
doing all of this right. They don't seem to spend all their time being
paranoid that someone out there might actually USE something they
built as intended -- they seem happy when it happens. They've got big
public databases filled with useful information, lots of tools anyone
can use, they code in high level languages, they aren't insanely
secretive, etc. I think they're a good model to follow.
Perry
======================================================================
Sent to CCL by: Cory Pye [cpye:_:crux.smu.ca]
On Sat, 15 Oct 2005, Perry E. Metzger perry[]piermont.com wrote:
>
> Sent to CCL by: "Perry E. Metzger" [perry*|*piermont.com]
>
[stuff deleted]
>
> Well, suppose you publish some research and, even without the use of
> your code, other people make advancements that you wanted to make
> first just by looking at your work and saying "hey, this is a cool
> thing to try next". Tough luck for you. That's how science works. Of
> course, if you aren't silly, you don't mind and it doesn't hurt
> you. Einstein didn't come up with spacetime -- Minkowski did, looking
> at Einstein's 1905 papers. Did Einstein then winge "you did that
> before I could"? No. Did Poincare spend all his time complaining that
> Einstein was influenced by him and published first? No.
>
Sure, this can happen. When you publish a paper, you allow your ideas to
be
peer-reviewed, then adapted by others to suit their needs. If I had
plans to
publish a followup article, I would probably not release the source
until after
it was published. I am certainly not obligated to HELP others scoop my
ideas by
giving them my source.
One shouldn't underestimate the educational benefit of having to
re-implement
code. Re-writing code gives you a much greater appreciation and
understanding
of the original author's work.
In some ways, this is similar to drug companies having a 5 or 10 year
patent
protection from generics. Perhaps a similar thing, but shorter time (2-3
years), could be a useful compromise.
In my case, the original idea was Klamt's. I took a buggy and limited
implementation in ADF 2.0, reincorporated it into 2.3, then
reincorporated it
into ADF98 development version, which was a major rewrite because of the
change
in the integration details. At that point I gave the code to ADF so that
they could maintain it, and wrote up the paper describing the finer
details of
the implementation. I encourage folks when they use ADF COSMO to cite
both my
paper, and the original paper of Klamt. There are numerous versions of
COSMO
now, as implemented in Gaussian 92 (Truong), Gaussian 98 (Barone), ADF
(me),
DMol, MNDO/d, PARAGAUSS, GAMESS-US, etc. The details of the
implementation are
described in the papers. If I had kept it totally to myself, only in ADF
2.3,
it would now be obsolete. By giving it to SCM and the affiliated
developers,
it has a life of its own. Lots of people use it and suggest
improvements, even
though it isn't open source. I'm quite happy to see people using and
acknowledging it, I don't want money for it.
[stuff deleted]
>
> I'm going to be so bold as to say that science doesn't exist to
> advance people's careers, and that good science isn't done very often
> by people so obsessed with careerism that they withhold information in
> the hope of gaining an edge over scientific competitors. I'll go
> further and say that people obsessed with their careers to this extent
> don't actually have good ones, either on the level of enjoying them or
> on the level of getting ahead. Not so paradoxically, doing good work
> is more important than relentless scheming and gamesmanship. That
> doesn't mean you should ignore personal advancement, but it does mean
> that spending all your time being a backbiting bitter schemer doesn't
> really do as much good as having good ideas and getting them out.
This is true. However, if the hypothetical professor in my previous
e-mail
doesn't get tenure and his student doesn't get his Ph. D., then the
short-term
advancement in science is more than nullified by the fact that these two
individuals now cannot make any further contributions in academia. Who
would
re-hire a person denied tenure or without a Ph. D.? They are consigned
to a
sessional appointment or lecturer position with no possibility of
advancement.
I feel that the long-term scientific contributions are traded for
short-term
gain.
[stuff deleted]
> > The sad reality is that things like this can and do happen and a
little
> > paranoia goes a long way.
>
> So you claim, but I really think you're presenting a very twisted view
> of the world. If you really live your life that way, you're going to
> be bitter all the time instead of productive.
I know of people who go to conferences to hear talks about unpublished
work by
some junior guy or person at a small institution, essentially stealing
their
ideas, then put their army of 20 postdocs on the same problem in an
attempt to
beat them to the press. I have also heard of senior faculty rejecting
grant
proposals from junior guys, and submitting a very similar proposal in
the next
round. If push comes to shove, who will be believed, the senior or
junior guy?
Perhaps one's paranoia can be switched off once one has the letter from
the
university president congratulating you on one's tenure and promotion to
associate :-)
I propose a new reality TV show called "Tenure-Track". It would be a
cross
between "Survivor" and "Cops".
[more stuff deleted]
************* ! Dr. Cory C. Pye
***************** ! Associate Professor
*** ** ** ** ! Theoretical and Computational Chemistry
** * **** ! Department of Chemistry, Saint Mary's University
** * * ! 923 Robie Street, Halifax, NS B3H 3C3
** * * ! cpye%a%crux.stmarys.ca
http://apwww.stmarys.ca/~cpye
*** * * ** ! Ph: (902)-420-5654 FAX:(902)-496-8104
***************** !
************* ! Les Hartree-Focks (Apologies to Montreal Canadien
Fans)
======================================================================
Sent to CCL by: Steve Bowlus [chezbowlus_+_goldrush.com]
$0.02 from a (grateful) enduser (who can't spell "C") of everyone
else's
code:
At one time, the CompChem community had its own vehicle for this
purpose. It was (and still is) called the "Quantum Chemistry Program
Exchange." For nominal cost (basically then media costs), academics
and
industry could get access to cutting edge programming. As QCPE matured
in the mid- to late '80's, there was published (again, at modest cost)
the quarterly "QCPE Bulletin," in which contributors could (as my
memory
serves me) publish exactly the kind of information being debated in
this
thread. Descriptions of new contributions were made in every issue by
the director/editor (Richard Counts/Peggy Edwards). A uniform
mechanism
was provided for citation of contributions.
I would suggest it is time to re-examine the utility of such a
repository, and revive it along the lines envisioned by Counts. I
don't
know how IU currently supports QCPE (AFAIK, its capabilities are
considerably curtailed); Jan's experience in obtaining funding for CCL
does not bode well for the venture. But perhaps someone knowledgeable
in these areas might inquire.
Cheers,
Steve
======================================================================
Sent to CCL by: "Noel O'Boyle" [no228{}cam.ac.uk]
On Mon, 2005-10-17 at 06:57 -0700, Steve Bowlus
chezbowlus[*]goldrush.com wrote:
> Sent to CCL by: Steve Bowlus [chezbowlus_+_goldrush.com]
> $0.02 from a (grateful) enduser (who can't spell "C") of everyone
else's
> code:
>
> At one time, the CompChem community had its own vehicle for this
> purpose. It was (and still is) called the "Quantum Chemistry Program
> Exchange." For nominal cost (basically then media costs), academics
and
> industry could get access to cutting edge programming.
Things have improved in this regard since the 80s. The facilities
provided by open-source repositories such as SourceForge
(http://www.sf.net) are light-years beyond simple
dump-it-and-leave-it-there type repositories. For example, SourceForge
provides, for each program, a web site, a CVS repository (this is for
coordinated the work of several developers), mailing lists, forums, bug
trackers and feature requests. Whether one uses these or not is up to
the individual developer(s). And it's cheaper than the QCPE - it's free
for users *and* developers!
> As QCPE matured
> in the mid- to late '80's, there was published (again, at modest
cost)
> the quarterly "QCPE Bulletin," in which contributors could (as my
memory
> serves me) publish exactly the kind of information being debated in
this
> thread. Descriptions of new contributions were made in every issue
by
> the director/editor (Richard Counts/Peggy Edwards). A uniform
mechanism
> was provided for citation of contributions.
Anything that provides a mechanism for developers to tell users about
their software, as well as providing an encouragement for people to
share code openly, sounds good to me. I would prefer though a more
formal publication procedure with a peer review process, examples of
which have been mentioned in both Physics and Bioinformatics. Dropping
the Q in QCPE might also be a good idea; and it would encourage people
to share code in the wider chemistry community.
Regards,
Noel
======================================================================
Emails that did not appear on the list:
======================================================================
Dear Noel,
Computer Physics Communications (published by Elsevier).
We published a code there recently.
Best regards,
Per-Olof �strand
======================================================================
Hi Noel,
Publishing an article concerning software could be difficult.
However, an article describing the application of a program to
solve particular chemical problem (compounds, reactions, mechanism,
etc.) is much easier to publish.
So, find an application for your programs and show on the example
how it works. Solve some chemically important problem using G03
and your program. The results can be easily published and you may
include the description of key-features of your program in this
publication.
Best regards,
Valentin.
Zelinsky Institute of Organic Chemistry,
Moscow.
======================================================================
Dr Noel O'Boyle:
I saw you message in the CCL list. If your programs are related with
biological research, then please take a look at www.bioinformation.net.
Its an open source forum for bioinformatics and related work. We shall
be happy to consider your manuscript describing the programs if they
have biological applications.
Sincerely,
Sandeep Kumar
Member, Editorial Board, Bioinformation.net
-------------------------------------------------------------------
Dr. Sandeep Kumar,
Associate Research Scientist,
Johns Hopkins University Department of Biology,
106 Mudd Hall, 3400 N. Charles Street,
Baltimore, MD 21218, USA.
Phone: 410-516-8433,
======================================================================
On Oct 11, 2005, at 10:45, Noel O Boyle no228 ~~ cam.ac.uk wrote:
> Where can one publish articles on programs that are of general
interest
> to computational chemists, but which do not contain new algorithms?
There's a list of places to publish comp chem at
http://membership.acs.org/C/COMP/resources.html#Anchor-Publishing-47857
, perhaps some of them might be appropriate.
Lisa
--
-------------------------------------------------------------------------------------------------------------------------------
Lisa M. Balbes, Ph.D. lisa],[balbes.com http://www.balbes.com/
Balbes Consultants Scientific Writing/Editing since 1992
American Chemical Society - St Louis Section Treasurer, COMP division
WebMaven
ACS/Career Services Presenter and Career Consultant
Chair, Washington University APAP, St Louis South Committee
Consultants and Independent Contractors SIG Manager, St Louis Society
for Technical Communication
Advancement Coordinator, Boy Scout Troop 352
======================================================================
Hello Noel!
A good place for this is Computer Physics Communications,
published by Elsevier.
kind regards,
Scotty
http://www.chem.ubc.ca/personnel/faculty/scotty/index.shtml
======================================================================
> In addition, science should involve building on someone else's
> work, not
> reinventing the wheel. Wouldn't it be better if authors made their
> code
> available so that others could use it?
Yes! And of course that's the point of open source. :-) It make sense
in the context of science, where building on old work and peer review
is expected. I was just asked to give a talk on Open Babel at a
nanoscience workshop, so maybe things are changing. Everyone seemed
happy about the ideas of programs like GaussSum and Open Babel.
I hope you find a place to publish an article -- people seem to be
unhappy about citing programs that don't have publications. Except
Gaussian itself... I guess that's OK to cite even though it's not
published anywhere.
Cheers and best regards,
-Geoff
--
-Dr. Geoffrey Hutchison <grh25],[cornell.edu>
Cornell University, Department of Chemistry and Chemical Biology
Abru�a Group http://abruna.chem.cornell.edu/
======================================================================
Hi Noel,
Let's consider MOLDEN as example. I always cite the reference
suggested by the author of MOLDEN, when I use this program.
Even if that reference does not provide all the information
about the particular MOLDEN option, still it is very useful
to cite it. The readers may find general information about the
program features and the link to website, where complete
manual is given.
So, if you suggest a reference (a few references) for the
users of your software, most likely these references will be
cited.
Writing software seems very easy from the users' point of view
(those who never did it) and it is really hard task from the
other side. You should be able to use some benefits in terms of
citation associated with your program.
As you can see from the CCL replies most of the community
is against publishing articles about programs. One of the
editors told me: "we don't publish software manuals, we
publish chemistry."
Best regards,
Valentin.
> Dear Valentin,
> Thanks for your answer.
>
> I'm familiar with this method of publication, and it's true that I've
> used my program to do some science. However, it's not really a valid
way
> of spreading information about the program's features. For example, it
> does not make a lot of sense to reference a paper about ruthenium
> chemistry (in my case) just because you used my software. Also, the
> reviewers will rightly say that a long description of features of
> gausssum is not appropriate to the topic of the paper.
>
> If I cannot find any other way to do it, I will follow your advice,
but
> I regard it as a workaround for the current situation in chemistry,
> rather than the ideal solution.
>
> Thanks again,
>
> Noel
======================================================================
From owner-chemistry@ccl.net Tue Oct 25 07:17:00 2005
From: "Vlad Cojocaru Vlad.Cojocaru---eml-r.villa-bosch.de" <owner-chemistry(_)server.ccl.net>
To: CCL
Subject: CCL: Gamess problem: error message missing
Message-Id: <-29723-051025064316-26200-D93sDSvgnmaRqVcRXGbfVA(_)server.ccl.net>
X-Original-From: Vlad Cojocaru <Vlad.Cojocaru]^[eml-r.villa-bosch.de>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=us-ascii
Date: Tue, 25 Oct 2005 12:42:50 +0200
MIME-Version: 1.0
Sent to CCL by: Vlad Cojocaru [Vlad.Cojocaru^eml-r.villa-bosch.de]
Dear CCL ers,
Please except my appologies .. In my previous email about the gamess
program I forgot to give you the error message that I get. Here is the
error message that I get when trying to run gamess on single CPU under
Debian Linux:
"Running gamess.Jun202002R1.x on localhost as compute process 0
** Address Error **
ddikick: Trapped SIGCHLD: Unexpected termination of a child process."
Of course the gamess.Jun202002R1.x is my executable
Sorry for the missing info!!
Cheers
vlad
--
Dr. Vlad Cojocaru
EML Research gGmbH
Molecular and Cellular Modeling Group
Villa Bosch
Schloss-Wolfsbrunnenweg 33
69118 Heidelberg
Germany
Phone: +49-6221-533266
Fax: +49-6221-533298
e-mail: Vlad.Cojocaru||eml-r.villa-bosch.de
From owner-chemistry@ccl.net Tue Oct 25 09:04:06 2005
From: "kevin abbot kevanabbot*yahoo.com" <owner-chemistry#server.ccl.net>
To: CCL
Subject: CCL: W:bounded DNA-Protein Docking
Message-Id: <-29725-051025070427-7797-WZYrvqLc5brsdTDiFc9JTA#server.ccl.net>
X-Original-From: "kevin abbot" <kevanabbot!=!yahoo.com>
Sent to CCL by: "kevin abbot" [kevanabbot]^[yahoo.com]
Hi all,
I'm interested in docking DNA into a protein with a covalent bond among these two biopolymers. Is there some programs (possible free) that do this kind of docking?
regards
kevin
From owner-chemistry@ccl.net Tue Oct 25 08:30:03 2005
From: "Olegas Eicher-Lorka lorka[-]ktl.mii.lt" <owner-chemistry]|[server.ccl.net>
To: CCL
Subject: CCL: bigger single-core or smaller dual-core?
Message-Id: <-29724-051025080349-29365-Ph8lMWaIPUMF4XZgN1PYMA]|[server.ccl.net>
X-Original-From: "Olegas Eicher-Lorka" <lorka~~ktl.mii.lt>
Content-Type: multipart/alternative;
boundary="----=_NextPart_000_00E5_01C5D96F.05621B40"
Date: Tue, 25 Oct 2005 14:18:50 +0300
MIME-Version: 1.0
Sent to CCL by: "Olegas Eicher-Lorka" [lorka(_)ktl.mii.lt]
This is a multi-part message in MIME format.
------=_NextPart_000_00E5_01C5D96F.05621B40
Content-Type: text/plain;
charset="windows-1257"
Content-Transfer-Encoding: quoted-printable
Dear all:
Which is the better performance/price choice for computational chemistry =
(energy and frequency jobs) :
Two dual-core AMD Opteron 1.8GHz(265) or four single-core AMD Opteron =
2.4GHz(250) ?
Thanks to all:
Olegas
------=_NextPart_000_00E5_01C5D96F.05621B40
Content-Type: text/html;
charset="windows-1257"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Dwindows-1257">
<META content=3D"MSHTML 6.00.2900.2769" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT size=3D2></FONT> </DIV>
<DIV>Dear all:<BR></DIV>
<DIV>Which is the better performance/price choice for computational =
chemistry=20
(energy and frequency jobs) :</DIV>
<DIV>Two dual-core AMD Opteron 1.8GHz(265) or four single-core AMD =
Opteron=20
2.4GHz(250) ?</DIV>
<DIV> </DIV>
<DIV>Thanks to all:<BR><BR>Olegas</DIV></BODY></HTML>
------=_NextPart_000_00E5_01C5D96F.05621B40--
From owner-chemistry@ccl.net Tue Oct 25 09:39:01 2005
From: "Laurynas Riauba laurynas.Riauba]_[chf.vu.lt" <owner-chemistry^server.ccl.net>
To: CCL
Subject: CCL: Vibrational analysis
Message-Id: <-29726-051025090152-12082-+Ba3Dkx1C8WC5jJIHlha7w^server.ccl.net>
X-Original-From: Laurynas Riauba <laurynas.Riauba{:}chf.vu.lt>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Date: Tue, 25 Oct 2005 15:05:30 +0300
MIME-Version: 1.0
Sent to CCL by: Laurynas Riauba [laurynas.Riauba!=!chf.vu.lt]
Dear CCLers,
I'm doing vibrational analysis with some organic molecules and I have
got some strange results.
Calculations were performed with Gaussian 03 and calculated vibrational
frequencies in output file are in quite good agreement with experimental
data. But when I try to calculate these frequencies using only cartesian
force constants and cartesian coordinates, I get very different results.
Below are listed main steps of calculation:
0. Assume, that molecule has C1 symetry, has N atoms, 3N-6 vibrations
1. Converting *.chk to *.fch
2. Extraction of cartesian force constants from *.fch and converting
extracted data to a matrix of size 3N * 3N (*.fch lists only triangular
matrix elements). Denote this matrix FX
3. Extraction of cartesian coordinates of atoms from *.fch and scaling
extracted data with a value of 0.529177237.
4. Creation of B matrix. This step was done with BMat program from QCPE,
and input for that program i have used properly formated file with
cartesian coordinates from step3 and 3N-6 internal coordinates(indeed
these coordinates were generated using Prometheus program for windows).
Output from BMat program was formated to give B matrix of size
(3N-6)*(3N), where 3N is number of columns
5. Creation of diagonal M matrix with size 3N*3N and filled with atomic
weight values
6. Calculation of G matrix: G = B * M(-1) * B'
7. Calculation of force matrix F in internal coordinates representation:
F = B * FX * B'
8. Calculation of GF matrix: GF = G*F
9. Calculations of GF eigenvalues and eigenvectors.
10. It is stated, that eigenvalues of GF matrix (denote Lambda) are
connected with vibrational frequence Liambda(k) = 4 * sqr(PI) *
sqr(v(k)), where Liambda(k) is k-th eigenvalue and v(k) is frequency of
vibration. Frequency can be converted to wavenumbers by dividing it with
c, speed of light. And the last thing, units have to be converted.
After all these calculations I get frequencies, which are totally
different compared to frequencies from gaussian output.
Is this calculation scheme correct?
Thank you in advance
Laurynas Riauba
PhD student,
Faculty of Chemistry
Vilnius University,
Vilnius, Lithuania
From owner-chemistry@ccl.net Tue Oct 25 10:14:00 2005
From: "Ding Xunlei dxl#ustc.edu.cn" <owner-chemistry|-|server.ccl.net>
To: CCL
Subject: CCL: calculation on C60 using DMol3
Message-Id: <-29727-051025091732-16144-5ntI5CsinuADGjqzj/Ti/Q|-|server.ccl.net>
X-Original-From: "Ding Xunlei" <dxl|ustc.edu.cn>
Content-Transfer-Encoding: 8bit
Content-Type: text/plain;
charset="gb2312"
Date: Tue, 25 Oct 2005 21:17:21 +0800
Mime-Version: 1.0
Sent to CCL by: "Ding Xunlei" [dxl%ustc.edu.cn]
Dear All��
I used 1.45327 and 1.36701 (angstrom) for single and double bonds in C60, the calculation will have a error.
����I now find that using the bond lengths 1.452 and 1.399 for single and double bonds in C60, the calculation is successful.
I don't know why.
Thank you for your help!
Yours sincerely,
Ding Xunlei
2005-10-25
______________________________________________
======= 2005-10-25 11:31:02 You have written: =======
>
>Sent to CCL by: "Ding Xunlei" [dxl%%ustc.edu.cn]
>Dear CCLs,
>
>I get an error when I do an energy calculation on C60 with DMol3, using dnd basis set and set "symmetry on"( C60 has Ih symmetry). The message is below:
>"Point group symmetry ih symmetry orbital prototypes generated (SYMDEC)
>Too many coefficients procof **** in T1U.1
>Message: DMol3 job failed
>Error: DMol3 exiting"
>If set "symmetry off" or use dn basis set, the calculation is successful.
>Then how can I do the calculation with symmetry on and dnd basis set?
>
>Thank you!
>
>
>Yours sincerely,
>Ding Xunlei
>2005-10-25
>______________________________________________>
>
>
>.
= = = = = = = = = = = = = = = = = = = =
From owner-chemistry@ccl.net Tue Oct 25 10:49:01 2005
From: "Jane Ou janeou.:.che.rochester.edu" <owner-chemistry\a/server.ccl.net>
To: CCL
Subject: CCL: W:Transition Dipole Moments with Gaussain 03
Message-Id: <-29728-051025101116-24628-VRt7CYWKD936Sa29ycPsjA\a/server.ccl.net>
X-Original-From: "Jane Ou" <janeou:_:che.rochester.edu>
Sent to CCL by: "Jane Ou" [janeou=-=che.rochester.edu]
Dear Sirs:
What is the general accuracy of the computed transition dipole
moments using Gaussian 03 with TD B3LYP/6-31G(d) functionals?
Are there any data or literatures I can look into? Thank you.
Sincerely,
Jane
From owner-chemistry@ccl.net Tue Oct 25 11:24:00 2005
From: "Istvan Mayer mayer-#-chemres.hu" <owner-chemistry ~ server.ccl.net>
To: CCL
Subject: CCL: W:Bond order and virial-theorem based energy decomposition program
Message-Id: <-29729-051025110822-28192-TL7fmqAOlz/iGNJ0A+nVDw ~ server.ccl.net>
X-Original-From: "Istvan Mayer" <mayer#%#chemres.hu>
Sent to CCL by: "Istvan Mayer" [mayer-,-chemres.hu]
Dear CCL-ers,
this is to inform you that a new small program called
BO-VIR
has been prepared for calculating BOND ORDER and VALENCE INDICES, as
defined in I. Mayer, Chem. Phys. Lett. 97, 270, 1983 (closed shell), ibid.
117, 396, 1985 (open shell) etc. as well as ENERGY DECOMPOSITION based on
the VIRIAL THEOREM. It can be used at the RHF, UHF level, and is
applicable to large systems, too. The program uses my original bond
order routine of 1982/83 and some routines from the program APOST by I.
Mayer and A. Hamza (Budapest, 2000)
The program requires only a formatted checkpoint file calculated in a
Gaussian run (G92...G03). It is applicable for large systems (up to 2000
orbitals) and can easily be extended further.
The program may be freely downloaded from our website
http://occam.chemres.hu
where other programs of our group (other energy decomposition, fuzzy atom
analysis, prediction of primary mass spectrometric fragmentations etc.)
are available, too.
Istvan Mayer
------------------------------------------------------
Prof. Istvan Mayer
Chemical Research Center
Hungarian Academy of Sciences
H-1525 Budapest, P.O.Box 17, Hungary
Phone: (+361) 438-4141, 438-4123 (ext.295,107)
Fax: (+361) 325-7554, 325-7750
e-mail: mayer*at*chemres.hu
------------------------------------------------------
From owner-chemistry@ccl.net Tue Oct 25 11:59:00 2005
From: "Richard Wood rwoodphd%msn.com" <owner-chemistry_._server.ccl.net>
To: CCL
Subject: CCL: help needed
Message-Id: <-29730-051025102656-6277-5MUfiH69YKE+0hQ91eyKPQ_._server.ccl.net>
X-Original-From: "Richard Wood" <rwoodphd~~msn.com>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain;
format=flowed;
charset="iso-8859-1";
reply-type=response
Date: Tue, 25 Oct 2005 10:26:51 -0400
MIME-Version: 1.0
Sent to CCL by: "Richard Wood" [rwoodphd[A]msn.com]
Hi all,
Thanks to those whom have offered suggestions so far.
Last night I discovered a problem that I don't think is related to the large
velocities encountered when I begin an MD simulation in NAMD. The protein
that I was trying to do calculations on has two chains, A and B. There are
about 350 residues in each chain or do, and each is numbered consecutively
in each chain. That is, they are numbered 1, 2, 3 ... in Chain A, and 1, 2,
3 ... in Chain B as well. Thus, when I run them through my script in
psfgen, I lose the B chain residues! So when I look at the molecule, I have
half the protein, but the waters surround the B chain are there. I guess
the residues of Chain B need to be renumbered, but I don't know of an "easy"
way to do this, other than to do a "Replace" in a word processor.
As I said, this is a new problem, unrelated to the others, but one I am sure
will occur again and again.
Thank you for your help,
Richard
Richard L. Wood, Ph. D.
Computational Chemist
Cockeysville, MD 21030
rwoodphd(_)yahoo.com
----- Original Message -----
> From: "Richard Wood rwoodphd+*+msn.com" <owner-chemistry(_)ccl.net>
To: "Wood, Richard L. " <rwoodphd(_)msn.com>
Sent: Monday, October 24, 2005 10:29 AM
Subject: CCL: help needed
>
> Sent to CCL by: "Richard Wood" [rwoodphd]|[msn.com]
> Hi all,
>
> I have some questions, some may be simple and some may not be. So, please
> bear with me :)
>
> The first is this. Where can I find a free program that runs under
> Windows that will read in a pdb file and do an energy minimization so that
> I can run dynamics calculations. The pdb files have been "generated"
> using psfgen that comes with NAMD.
>
> The second question may not be so simple. How can I generate
> "Charmm-like" topologies and parameters for small molecules? Again, I
> wish to use NAMD to perform dynamics calculations on these small
> molecules.
>
> The third one may be the most difficult. I have a protein-ligand complex
> that I wish to do dynamics on. The hard part is that the ligand is also a
> peptide. How can I set up my peptide/system so that when I generate a psf
> file and pdb from psfgen, it doesn't think that the peptide is part of the
> larger protein?
>
> Again, any programs to do this need to be free as I am doing this all
> freelance.
>
> Thanks in advance,
> Richard
>
>
> Richard L. Wood, Ph. D.
> Computational Chemist
> Cockeysville, MD 21030
> rwoodphd(~)yahoo.comhttp://www.ccl.net/chemistry/sub_unsub.shtml>
>
>
>
From owner-chemistry@ccl.net Tue Oct 25 12:34:00 2005
From: "Jane-Jane Ou janeou * che.rochester.edu" <owner-chemistry%a%server.ccl.net>
To: CCL
Subject: CCL: Transition Dipole Moment with Gaussian
Message-Id: <-29731-051025105607-21764-+lGU/EsVZYGzPkayKqg/Ow%a%server.ccl.net>
X-Original-From: Jane-Jane Ou <janeou-#-che.rochester.edu>
Date: Tue, 25 Oct 2005 10:05:19 -0400 (EDT)
Sent to CCL by: Jane-Jane Ou [janeou(~)che.rochester.edu]
Dear Sirs:
What is the general accuracy of the computed transition dipole
moments using Gaussian 03 with TD B3LYP/6-31G(d) functionals?
Are there any data or literatures I can look into? Thank you.
Sincerely,
Jane
./
From owner-chemistry@ccl.net Tue Oct 25 13:37:01 2005
From: "Mima Staikova mima.staikova _ gmail.com" <owner-chemistry**server.ccl.net>
To: CCL
Subject: CCL: rigid scan in Gaussian
Message-Id: <-29732-051025132302-7266-WzPAl+q2ObipMAJqZsUfWA**server.ccl.net>
X-Original-From: Mima Staikova <mima.staikova^^gmail.com>
Content-Type: multipart/alternative;
boundary="----=_Part_15614_16162762.1130257036109"
Date: Tue, 25 Oct 2005 12:17:16 -0400
MIME-Version: 1.0
Sent to CCL by: Mima Staikova [mima.staikova!A!gmail.com]
------=_Part_15614_16162762.1130257036109
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Hi everyone,
I need to model a reaction path of hydrogen abstraction, mediated by water
molecule.
A hydrogen from the substrate is transferred to the water molecule and
simultaneously one of the water hydrogens is transferred to the acceptor. I
am interested in the energy along this pathway without reoptimizing the
substrate and the acceptor at every step along the reaction coordinate. The
reaction coordinate is a straight line.
The Scan keyword (rigid scan) is going to give me a potential surface, not =
a
potential curve, if I specify the coordinates of the two hydrogens, that ar=
e
moving simultaneously, calculating at every coordinate change of the first =
H
the all the changes for the second H.
How can I do the calculation in such a way that I have for each change in
the first H distance to the substrate just one change in the distance for
the second hydrogen, without having to resort to multiple single point
calculations?
Thank you.
Mima Staikova
------=_Part_15614_16162762.1130257036109
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Hi everyone,<br><br>I need to model a reaction path of hydrogen abstraction=
, mediated by water molecule.<br>A hydrogen from the substrate is tra=
nsferred to the water molecule and simultaneously one of the water h=
ydrogens is transferred to the acceptor. I am interested in the energy alon=
g this pathway without reoptimizing the substrate and the acceptor at every=
step along the reaction coordinate. The reaction coordinate is a straight =
line.
<br><br>The Scan keyword (rigid scan) is going to give me a potential surfa=
ce, not a potential curve, if I specify the coordinates of the two hydrogen=
s, that are moving simultaneously, calculating at every coordinate change o=
f the first H the all the changes for the second H.
<br><br>How can I do the calculation in such a way that I have for each cha=
nge in the first H distance to the substrate just one change in the distanc=
e for the second hydrogen, without having to resort to multiple single poin=
t calculations?
<br><br>Thank you.<br>Mima Staikova<br><br><br><br>
------=_Part_15614_16162762.1130257036109--
From owner-chemistry@ccl.net Tue Oct 25 14:14:01 2005
From: "Loan Huynh huynhkl2000(a)yahoo.ca" <owner-chemistry()server.ccl.net>
To: CCL
Subject: CCL: W:bounded DNA-Protein Docking
Message-Id: <-29733-051025140439-18099-nZVztnYoa0p2u+YAOYFeTg()server.ccl.net>
X-Original-From: Loan Huynh <huynhkl2000]![yahoo.ca>
Content-Transfer-Encoding: 8bit
Content-Type: multipart/alternative; boundary="0-910503302-1130259871=:86230"
Date: Tue, 25 Oct 2005 13:04:31 -0400 (EDT)
MIME-Version: 1.0
Sent to CCL by: Loan Huynh [huynhkl2000[-]yahoo.ca]
--0-910503302-1130259871=:86230
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Hi Kevin,
You may want to use Autodock. It's free but you need to sign a form in order to have it for free. I have the web site below.
http://www.scripps.edu/mb/olson/doc/autodock/
"kevin abbot kevanabbot*yahoo.com" <owner-chemistry_-_ccl.net> wrote:
Sent to CCL by: "kevin abbot" [kevanabbot]^[yahoo.com]
Hi all,
I'm interested in docking DNA into a protein with a covalent bond among these two biopolymers. Is there some programs (possible free) that do this kind of docking?
regards
kevinhttp://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txtLoan Huynh
M.S. Polymer and Materials Chemistry
University of Toronto
Work: 416-946-0040 or: 416-946-8469
Cell: 416-779-7173
***************************************************
---------------------------------
Find your next car at Yahoo! Canada Autos
--0-910503302-1130259871=:86230
Content-Type: text/html; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
<DIV>Hi Kevin,</DIV>
<DIV> </DIV>
<DIV>You may want to use Autodock. It's free but you need to sign a form in order to have it for free. I have the web site below.</DIV>
<DIV> </DIV>
<DIV><A href="http://www.scripps.edu/mb/olson/doc/autodock/">http://www.scripps.edu/mb/olson/doc/autodock/</A><BR><BR><B><I>"kevin abbot kevanabbot*yahoo.com" <owner-chemistry_-_ccl.net></I></B> wrote:
<BLOCKQUOTE class=replbq style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #1010ff 2px solid"><BR>Sent to CCL by: "kevin abbot" [kevanabbot]^[yahoo.com]<BR>Hi all,<BR>I'm interested in docking DNA into a protein with a covalent bond among these two biopolymers. Is there some programs (possible free) that do this kind of docking?<BR><BR>regards<BR><BR>kevin<BR><BR><BR><BR<BR<BR<BR<BR><BR<BR>http://www.ccl.net/cgi-bin/ccl/send_ccl_message<BR><BR<BR>http://www.ccl.net/cgi-bin/ccl/send_ccl_message<BR><BR<BR>http://www.ccl.net/chemistry/sub_unsub.shtml<BR><BR<BR><BR<BR><BR<BR>http://www.ccl.net/spammers.txt<BR><BR<BR><BR><BR><BR></BLOCKQUOTE></DIV><BR><BR>Loan Huynh <br>M.S. Polymer and Materials Chemistry<br>University of Toronto <br>Work: 416-946-0040 or: 416-946-8469 <br>Cell: 416-779-7173 <br>***************************************************<p>
<hr size=1>Find your next car at <a href="http://autos.yahoo.ca"><b>Yahoo! Canada Autos</b></a>
--0-910503302-1130259871=:86230--
From owner-chemistry@ccl.net Tue Oct 25 16:12:00 2005
From: "Pierre Mignon pmignon||vub.ac.be" <owner-chemistry(_)server.ccl.net>
To: CCL
Subject: CCL: electronic embedding in ONIOM
Message-Id: <-29734-051025152617-31696-9olITfLdW3zKA+Y29fZX5w(_)server.ccl.net>
X-Original-From: Pierre Mignon <pmignon=vub.ac.be>
Date: Tue, 25 Oct 2005 20:32:39 +0200 (CEST)
Sent to CCL by: Pierre Mignon [pmignon;;vub.ac.be]
Dear,
Is it possible to perform electronic embedding with an ONIOM(B3LYP/6-31G*:MNDO) calculation ? I read in the manual that it is possible for ONIOM(MO:MM) calculations, but for ONIOM(QM/QM) ?
Thank you in advance,
Pierre.
Mignon Pierre PhD
Free University of Brussel (VUB)
Dienst Algemene Chemie (ALGC)
Pleinlaan, 2
1050 Brussels
Belgium
Tel + 32 2 629 33 15
Fax + 32 2 629 33 17
e-mail pmignon]-[vub.ac.be
From owner-chemistry@ccl.net Tue Oct 25 16:57:01 2005
From: "Robert Brunner rbrunner[-]uiuc.edu" <owner-chemistry(!)server.ccl.net>
To: CCL
Subject: CCL: help needed
Message-Id: <-29735-051025165251-23928-zd9//CDZHavFyOT2thLfjA(!)server.ccl.net>
X-Original-From: Robert Brunner <rbrunner|-|uiuc.edu>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Date: Tue, 25 Oct 2005 14:59:12 -0500
Mime-Version: 1.0 (Apple Message framework v734)
Sent to CCL by: Robert Brunner [rbrunner!A!uiuc.edu]
Hello Richard,
This sounds like something you should be able to do with NAMD and
VMD, but I'm not all that knowledgeable about setting up simulations.
However, I would recommend that you try posting your question to the
namd-l mailing list:
<http://www.ks.uiuc.edu/Research/namd/mailing_list/>
Robert
On Oct 24, 2005, at 9:29 AM, Richard Wood rwoodphd+*+msn.com wrote:
>
> Sent to CCL by: "Richard Wood" [rwoodphd]|[msn.com]
> Hi all,
>
> I have some questions, some may be simple and some may not be. So,
> please bear with me :)
>
> The first is this. Where can I find a free program that runs under
> Windows that will read in a pdb file and do an energy minimization
> so that I can run dynamics calculations. The pdb files have been
> "generated" using psfgen that comes with NAMD.
>
> The second question may not be so simple. How can I generate
> "Charmm-like" topologies and parameters for small molecules?
> Again, I wish to use NAMD to perform dynamics calculations on these
> small molecules.
>
> The third one may be the most difficult. I have a protein-ligand
> complex that I wish to do dynamics on. The hard part is that the
> ligand is also a peptide. How can I set up my peptide/system so
> that when I generate a psf file and pdb from psfgen, it doesn't
> think that the peptide is part of the larger protein?
>
> Again, any programs to do this need to be free as I am doing this
> all freelance.
>
> Thanks in advance,
> Richard
>
>
> Richard L. Wood, Ph. D.
> Computational Chemist
> Cockeysville, MD 21030
> rwoodphd(~)yahoo.com
>
From owner-chemistry@ccl.net Tue Oct 25 17:31:01 2005
From: "Perry E. Metzger perry^piermont.com" <owner-chemistry]-[server.ccl.net>
To: CCL
Subject: CCL: help needed
Message-Id: <-29736-051025170128-28325-eswlZp+ibq/Bsw4umt0DaA]-[server.ccl.net>
X-Original-From: "Perry E. Metzger" <perry ~ piermont.com>
Content-Type: text/plain; charset=us-ascii
Date: Tue, 25 Oct 2005 17:01:17 -0400
MIME-Version: 1.0
Sent to CCL by: "Perry E. Metzger" [perry-#-piermont.com]
"Dr Seth Olsen s.olsen1(_)uq.edu.au" <owner-chemistry]~[ccl.net> writes:
> You may find your options are greater if you use something like Cygwin
> under windows. I couldn't get that technical about the description, but
> Cygwin essentially gives you a linux-imitating environment that runs
> under windows.
It is a reasonable emulation environment, though you have to recompile
anything you need to use, and the edges are a bit sharp. I use
it when I'm forced to set stuff up in Windows environments -- it is a
lot more productive than the usual Microsoft tools.
Microsoft also has a similar offering that they give away called
(inexplicably) "Services for Unix". I've heard they're getting rid of
it though.
Performance under Cygwin will not rival performance on a real Linux
box -- Windows has architectural issues that make it perform pretty
badly in general, and Cygwin is just a POSIX scrim over Windows.
> Cygwin is free. I think the site is www.cygwin.org or somesuch.
http://www.cygwin.com/ actually.
Perry
From owner-chemistry@ccl.net Tue Oct 25 18:09:00 2005
From: "Arindam Ganguly arindamganguly^gmail.com" <owner-chemistry|-|server.ccl.net>
To: CCL
Subject: CCL: W:Calculation of Point Group
Message-Id: <-29737-051025170748-3654-31K2VQzWKWecpn069QkwwA|-|server.ccl.net>
X-Original-From: Arindam Ganguly <arindamganguly++gmail.com>
Content-Type: multipart/alternative;
boundary="----=_Part_2169_27492544.1130264186570"
Date: Tue, 25 Oct 2005 13:16:26 -0500
MIME-Version: 1.0
Sent to CCL by: Arindam Ganguly [arindamganguly-.-gmail.com]
------=_Part_2169_27492544.1130264186570
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Hi Huub,
i am not doing either of them. the Bloodshed Dev C++ IDE has compile and ru=
n
commmand in it. so i am compile the C source code and executing it. it just
gives me a black box. just like what we get in windows command prompt,
except the fact that in windows commad prompt it mentions the path, but her=
e
nothing that sort of comes up. that's why i was confused as to where is the
mistake. i hope this helps, thanks,
Arindam
On 10/25/05, Van Dam, HJJ (Huub) h.j.j.vandam|*|dl.ac.uk <
owner-chemistry=-=ccl.net> wrote:
>
>
> Sent to CCL by: "Van Dam, HJJ \(Huub\)" [h.j.j.vandam=3D=3D=3Ddl.ac.uk]
> Hi,
>
> Are you running this from the windows command prompt or are you double
> clicking it?
>
> Huub
>
> -----Original Message-----
> > From: owner-chemistry#,#ccl.net [mailto:owner-chemistry#,#ccl.net]
> Sent: 24 October 2005 19:34
> To: Vandam, Huub
> Subject: CCL: W:Calculation of Point Group
>
> Hi Serguei,
> i went to the mentioned website and downloaded the symmetry.c file. i am
> using Bloodshed Dev C++ IDE, it can handle C files too. i build the file
> and execute but all it does is give a black command prompt box, just
> like we the one we get in WinXp. i don;t understand the mistake i am
> committing.
> further i thought probably the zip files would be the updated version
> but same thing happened. could you plz tell what is mistake i am making.
> thanks=3D
>
>
>
> -=3D This is automatically added to each message by the mailing script =
=3D->
>
>
>
------=_Part_2169_27492544.1130264186570
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Hi Huub,<br>
i am not doing either of them. the Bloodshed Dev C++ IDE has compile
and run commmand in it. so i am compile the C source code and executing
it. it just gives me a black box. just like what we get in windows
command prompt, except the fact that in windows commad prompt it
mentions the path, but here nothing that sort of comes up. that's why i
was confused as to where is the mistake. i hope this helps, thanks,<br>
<br>
Arindam<br><br><div><span class=3D"gmail_quote">On 10/25/05, <b class=3D"gm=
ail_sendername">Van Dam, HJJ (Huub) h.j.j.vandam|*|dl.ac.uk</b> <<a href=
=3D"mailto:owner-chemistry=-=ccl.net">owner-chemistry=-=ccl.net</a>> wrote:<=
/span>
<blockquote class=3D"gmail_quote" style=3D"border-left: 1px solid rgb(204, =
204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br>Sent to CCL b=
y: "Van Dam, HJJ \(Huub\)" [h.j.j.vandam=3D=3D=3Ddl.ac.uk]<br>Hi,=
<br><br>
Are you running this from the windows command prompt or are you double<br>c=
licking it?<br><br>Huub<br><br>-----Original Message-----<br>> From: own=
er-chemistry#,#ccl.net [mailto:<a href=3D"mailto:owner-chemistry#">owner-ch=
emistry#
</a>,#ccl.net]<br>Sent: 24 October 2005 19:34<br>To: Vandam, Huub<br>Subjec=
t: CCL: W:Calculation of Point Group<br><br>Hi Serguei,<br>i went to the me=
ntioned website and downloaded the symmetry.c file. i am<br>using Bloodshed=
Dev C++ IDE, it can handle C files too. i build the file
<br>and execute but all it does is give a black command prompt box, just<br=
>like we the one we get in WinXp. i don;t understand the mistake i am<br>co=
mmitting.<br>further i thought probably the zip files would be the updated =
version
<br>but same thing happened. could you plz tell what is mistake i am making=
.<br>thanks=3D<br><br><br><br>-=3D This is automatically added to each mess=
age by the mailing script =3D-<br>To recover the email address of the autho=
r of the message, please change
<br<br>l=
ook up the X-Original-From: line in the mail header.<br><br>E-mail to subsc=
ribers: <a href=3D"mailto:CHEMISTRY=-=ccl.net">CHEMISTRY=-=ccl.net</a> or use:
<br> <a href=3D"http://www.ccl.net/cgi-b=
in/ccl/send_ccl_message">http://www.ccl.net/cgi-bin/ccl/send_ccl_message</a=
><br><br>E-mail to administrators: <a href=3D"mailto:CHEMISTRY-REQUEST=-=ccl.=
net">CHEMISTRY-REQUEST=-=ccl.net
</a> or use<br> <a href=3D"http://www.cc=
l.net/cgi-bin/ccl/send_ccl_message">http://www.ccl.net/cgi-bin/ccl/send_ccl=
_message</a><br><br>Subscribe/Unsubscribe:<br>  =
; <a href=3D"http://www.ccl.net/chemistry/sub_unsub.shtml"</a><br><br>Before posting chec=
k wait time for next message at: <a href=3D"http://www.ccl.net">http://www.=
ccl.net</a><br><br>Job advertisements: <a href=3D"http://www.ccl.net/jobs">
http://www.ccl.net/jobs</a><br><br>If your mail bounces from CCL with 5.7.1=
error, check:<br> <a href=3D"http://www=
.ccl.net/spammers.txt">http://www.ccl.net/spammers.txt</a><br><br>-+-+-+-+-=
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
<br><br><br><br></blockquote></div><br>
------=_Part_2169_27492544.1130264186570--
From owner-chemistry@ccl.net Tue Oct 25 19:06:00 2005
From: "Alexander Martins Silva alex.msilva,uol.com.br" <owner-chemistry**server.ccl.net>
To: CCL
Subject: CCL: Gaussian parallel
Message-Id: <-29738-051025190350-31902-MbSl4M0Hmk2j8LN4H+jwIQ**server.ccl.net>
X-Original-From: Alexander Martins Silva <alex.msilva[-]uol.com.br>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Date: Tue, 25 Oct 2005 21:04:12 +0000
MIME-Version: 1.0
Sent to CCL by: Alexander Martins Silva [alex.msilva]=[uol.com.br]
Alexander Martins Silva wrote:
>
> Hi,
>
> I was trying to execute a parallel job with Gaussian. The job
> starts but not run at the nodes.
>
> After g03l inputefile:
>
> $ g03l 453.inp &
> setenv GAUSS_EXEDIR
> /usr/local/g03/linda-exe:/usr/local/g03/bsd:/usr/local/g03/private:/usr/local/g03
>
> g03 453.inp
> [2] 18731
> $ ntsnet: using global map file
> "/usr/local/g03/linda7.1/intel-linux2.4-rh8/../common/lib/tsnet.map"
> ntsnet: couldn't find user map file "/home/alex/.tsnet.map"
> ntsnet: checking for /usr/local/g03/linda-exe/l302.exel
> ntsnet: using executable file /usr/local/g03/linda-exe/l302.exel
> ntsnet: maxlicense is 1000000
> ********************************************
> appl config file: /usr/local/g03/linda-exe/tsnet.config-l302_exel
> user config file: /home/alex/.tsnet.config
> global config file:
> /usr/local/g03/linda7.1/intel-linux2.4-rh8/../common/lib/tsnet.config
> user map file: /home/alex/.tsnet.map
> global map file:
> /usr/local/g03/linda7.1/intel-linux2.4-rh8/../common/lib/tsnet.map
> local node: at1
> remote nodes: at1101 at1102 at1103 at1104
> nodelist: at1101 at1102 at1103 at1104
> nodefile: tsnet.nodes
> minwait: 600
> maxwait: 600
> workerwait: 900
> masterload: 1.000000
> workerload: 1.000000
> fallbackload: 0.990000
> maxprocspernode: 1
> delay: 0
> maxnodes: Number of available nodes
> minworkers: 3
> maxworkers: 3
> loadperiod: 5
> high: True
> suffix: True
> verbose: True
> veryverbose: True
> distribute: False
> cleanup: True
> getload: False
> translate: True
> useglobalconfig: True
> useglobalmap: True
> application name: l302.exel
> application lookup name: l302_exel
> local executable directory: /usr/local/g03/linda-exe
> distribution directory: /usr/local/g03/linda7.1/intel-linux2.4-rh8/
> maxlicense: 1000000
> ********************************************
> ntsnet: trying to schedule 3 workers
> ntsnet: scheduled a total of 3 workers
> ********************************************
> node name: xxx
> (...)
> node name....
> (...)
> ntsnet: starting master process on at1
> /usr/local/g03/linda7.1/intel-linux2.4-rh8/bin/linda_sh
> /usr/local/g03/linda-exe/l302.exel 65536000 /home/alex/Gau-18733.chk 0
> /home/alex/Gau-18733.int 0 /home/alex/Gau-18733.rwf 0
> /home/alex/Gau-18733.d2e 0 /home/alex/Gau-18733.scr 0
> /home/alex/Gau-18732.inp 0 junk.out 0 +LARGS 5 0 -kainterval 1 -master
> 30997 -tsnetport 42530 -maxworkers 3 -minworkers 3 -minwait 600
> -maxwait 600 -nodename at1 -kaon
> ntsnet: starting 1 worker on at1101.ime.eb.br
> /usr/local/g03/linda7.1/intel-linux2.4-rh8/bin/linda_rsh
> at1101.ime.eb.br -r ssh /usr/local/g03/linda-exe/l302.exel 65536000
> /home/alex/Gau-18733.chk 0 /home/alex/Gau-18733.int 0
> /home/alex/Gau-18733.rwf 0 /home/alex/Gau-18733.d2e 0
> /home/alex/Gau-18733.scr 0 /home/alex/Gau-18732.inp 0 junk.out 0
> +LARGS 5 1 -maxworkers 3 -chdir /home/alex -worker at1:30997
> -workerwait 900 -tsnetref 1 -nodename at1101xxx
> ntsnet: starting 1 worker on at1102.ime.eb.br
> /usr/local/g03/linda7.1/intel-linux2.4-rh8/bin/linda_rsh
> at1102.ime.eb.br -r ssh /usr/local/g03/linda-exe/l302.exel 65536000
> /home/alex/Gau-18733.chk 0 /home/alex/Gau-18733.int 0
> /home/alex/Gau-18733.rwf 0 /home/alex/Gau-18733.d2e 0
> /home/alex/Gau-18733.scr 0 /home/alex/Gau-18732.inp 0 junk.out 0
> +LARGS 5 2 -maxworkers 3 -chdir /home/alex -worker at1:30997
> -workerwait 900 -tsnetref 2 -nodename at1102xxx
> ntsnet: starting 1 worker on at1103.ime.eb.br
> /usr/local/g03/linda7.1/intel-linux2.4-rh8/bin/linda_rsh
> at1103.ime.eb.br -r ssh /usr/local/g03/linda-exe/l302.exel 65536000
> /home/alex/Gau-18733.chk 0 /home/alex/Gau-18733.int 0
> /home/alex/Gau-18733.rwf 0 /home/alex/Gau-18733.d2e 0
> /home/alex/Gau-18733.scr 0 /home/alex/Gau-18732.inp 0 junk.out 0
> +LARGS 5 3 -maxworkers 3 -chdir /home/alex -worker at1:30997
> -workerwait 900 -tsnetref 3 -nodename at1103xxx
> ntsnet: exec'ing
> /usr/local/g03/linda7.1/intel-linux2.4-rh8/bin/LindaLauncher
> /home/alex/tmp/viaExecDataaCXV13
> $
>
> After that, the job apparentelly freeze at link l302.exel.
> Nothing is running at nodes.
> $ps
> PID TTY TIME CMD
> 15837 pts0 00:00:00 bash
> 18731 pts0 00:00:00 g03l
> 18732 pts0 00:00:00 g03
> 18733 pts0 00:00:03 l302.exe
> 18735 pts0 00:00:00 cLindaLauncher
> 18738 pts0 00:00:01 l302.exel
> 18739 pts0 00:00:00 ssh
> 18740 pts0 00:00:00 ssh
> 18741 pts0 00:00:00 ssh
> 19034 pts0 00:00:00 ps
>
> To test the execution I have past the folowing command
> line (from ntsnet: starting 1 worker on at1101xxx)
>
> $/usr/local/g03/linda7.1/intel-linux2.4-rh8/bin/linda_rsh
> at1101.ime.eb.br -r ssh /usr/local/g03/linda-exe/l302.exel 65536000
> /home/alex/Gau-18733.chk 0 /home/alex/Gau-18733.int 0
> /home/alex/Gau-18733.rwf 0 /home/alex/Gau-18733.d2e 0
> /home/alex/Gau-18733.scr 0 /home/alex/Gau-18732.inp 0 junk.out 0
> +LARGS 5 1 -maxworkers 3 -chdir /home/alex -worker at1:30997
> -workerwait 900 -tsnetref 1 -nodename at1101.ime.eb.br
> failed to look up given name -workerwait
>
> What can I change to work withou problems?
>
> Alexander.
>
From owner-chemistry@ccl.net Tue Oct 25 19:59:00 2005
From: "TJ O Donnell tjo#%#acm.org" <owner-chemistry%a%server.ccl.net>
To: CCL
Subject: CCL: help needed
Message-Id: <-29739-051025194813-17188-5EyOrwSab3Xz7iijV9k+mw%a%server.ccl.net>
X-Original-From: "TJ O'Donnell" <tjo,acm.org>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Date: Tue, 25 Oct 2005 16:47:33 -0700
MIME-Version: 1.0
Sent to CCL by: "TJ O'Donnell" [tjo-#-acm.org]
For serious use of linux under Windows, I would recommend VMWare.
http://www.vmware.com - look for the Workstation product.
It is an excellent product and very reasonably priced - about $185 if I
remember correctly. The speed is about 90% of a native
linux machine, except for high-speed interactive graphics.
As a programming, X-windows and development tool it is great.
It works better if you have additional memory so it can have a big
chunk all for itself. I use VMWare on my 1Gb laptop every day,
with Windows XP and the host, Fedora as the guest operating system.
TJ O'Donnell
http://www.gnova.com
Perry E. Metzger perry^piermont.com wrote:
> Sent to CCL by: "Perry E. Metzger" [perry-#-piermont.com]
>
> "Dr Seth Olsen s.olsen1(_)uq.edu.au" <owner-chemistry.:.ccl.net> writes:
>
>>You may find your options are greater if you use something like Cygwin
>>under windows. I couldn't get that technical about the description, but
>>Cygwin essentially gives you a linux-imitating environment that runs
>>under windows.
>
>
> It is a reasonable emulation environment, though you have to recompile
> anything you need to use, and the edges are a bit sharp. I use
> it when I'm forced to set stuff up in Windows environments -- it is a
> lot more productive than the usual Microsoft tools.
>
> Microsoft also has a similar offering that they give away called
> (inexplicably) "Services for Unix". I've heard they're getting rid of
> it though.
>
> Performance under Cygwin will not rival performance on a real Linux
> box -- Windows has architectural issues that make it perform pretty
> badly in general, and Cygwin is just a POSIX scrim over Windows.
>
>
>>Cygwin is free. I think the site is www.cygwin.org or somesuch.
>
>
> http://www.cygwin.com/ actually.
>
> Perry>
>
From owner-chemistry@ccl.net Tue Oct 25 21:41:00 2005
From: "Mark Thompson mark|a|arguslab.com" <owner-chemistry-#-server.ccl.net>
To: CCL
Subject: CCL: A new ArgusLab 4.01 bug fix release is now available
Message-Id: <-29740-051025205840-17911-oqEwjiEV4zsx98DZWEnpBg-#-server.ccl.net>
X-Original-From: Mark Thompson <mark[]arguslab.com>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Date: Tue, 25 Oct 2005 17:58:23 -0700
MIME-Version: 1.0
Sent to CCL by: Mark Thompson [mark*o*arguslab.com]
I've just uploaded a new ArgusLab 4.01 installer with a couple of bug fixes:
1. Some bugs in the RMSD code fixed.
2. The installer is fixed to allow the user to install for anyone using
the computer (if they have admin priviledges) or for the current user.
Apologies for the inconvenience for so many users having to wait for #2
Cheers,
Mark
--
----------------------------
Mark Thompson
mark/./arguslab.com
http://www.arguslab.com
----------------------------
From owner-chemistry@ccl.net Tue Oct 25 22:16:00 2005
From: "Yubo Fan yubofan~~mail.chem.tamu.edu" <owner-chemistry(0)server.ccl.net>
To: CCL
Subject: CCL: cubegen problem in G03
Message-Id: <-29741-051025184157-19540-WvFt4MyVgp7uy6bjdvGOOA(0)server.ccl.net>
X-Original-From: "Yubo Fan" <yubofan-.-mail.chem.tamu.edu>
Content-Type: multipart/alternative;
boundary="----=_NextPart_000_0043_01C5D982.F9C5E8C0"
Date: Tue, 25 Oct 2005 16:41:40 -0500
MIME-Version: 1.0
Sent to CCL by: "Yubo Fan" [yubofan{=}mail.chem.tamu.edu]
This is a multi-part message in MIME format.
------=_NextPart_000_0043_01C5D982.F9C5E8C0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi, everyone,
I've done some CIS calculations about some chromophores with G03. Then =
I've got some problems to generate density transition cube files with =
cubegen. I've got some references but the results were calculated with =
G94.
In G94
#p CIS(read,direct,Nstates=3D2)/3-21G* pop=3Dreg scf=3Ddirect =
guess=3D(read,only) density=3D(transition=3D1,check) geom=3Dcheck nosymm =
cube=3D(density,full,cards)
can generate the cube files based on the previous CIS calculations.
BUT, in G03, the keywork cube is not available, so I have to utilize =
cubegen program (I guess it's better). In the command line, I input =
"cubegen 0 Density=3DSCF test.fchk test.cube 0 h". Is the INPUT correct =
compared to G94? If not, what should I type in?
Thanks in advance,
Yubo
=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
Dr. Yubo Fan Email: yubofan(a)mail.chem.tamu.edu
Department of Chemistry Tel: 1-979-845-5237
Texas A&M University
College Station, TX 77843
=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
------=_NextPart_000_0043_01C5D982.F9C5E8C0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.2769" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Hi, everyone,</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2>I've done some CIS calculations about =
some=20
chromophores with G03. Then I've got some problems to generate density=20
transition cube files with cubegen. I've got some references but the =
results=20
were calculated with G94.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2>In G94</FONT></DIV>
<DIV><FONT face=3DArial size=3D2><FONT face=3DArial =
size=3D2></FONT></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2><FONT face=3DArial size=3D2>#p=20
CIS(read,direct,Nstates=3D2)/3-21G* pop=3Dreg scf=3Ddirect =
guess=3D(read,only)=20
density=3D(transition=3D1,check) geom=3Dcheck nosymm=20
cube=3D(density,full,cards)</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><FONT face=3DArial =
size=3D2></FONT></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2><FONT face=3DArial size=3D2>can =
generate the cube files=20
based on the previous CIS calculations.</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><FONT face=3DArial =
size=3D2></FONT></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2><FONT face=3DArial size=3D2>BUT, in =
G03, the keywork=20
cube is not available, so I have to utilize cubegen program (I guess =
it's=20
better). In the command line, I input "cubegen 0 Density=3DSCF test.fchk =
test.cube=20
0 h". Is the INPUT correct compared to G94? If not, what should I type=20
in?</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><FONT face=3DArial =
size=3D2></FONT></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2><FONT face=3DArial size=3D2>Thanks in=20
advance,</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><FONT face=3DArial =
size=3D2></FONT></FONT> </DIV>
<DIV><FONT face=3DArial size=3D2><FONT face=3DArial =
size=3D2>Yubo</FONT></FONT></DIV>
<DIV><FONT face=3DArial size=3D2><FONT face=3DArial=20
size=3D2>=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<BR>Dr. Yubo=20
Fan &nbs=
p; =20
Email: <A=20
href=3D"mailto:yubofan(a)mail.chem.tamu.edu">yubofan(a)mail.chem.tamu.edu</A>=
<BR>Department=20
of Chemistry Tel: 1-979-845-5237<BR>Texas =
A&M=20
University<BR>College Station, TX=20
77843<BR>=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</DIV></FONT></FONT></BODY></HTML>=
------=_NextPart_000_0043_01C5D982.F9C5E8C0--
From owner-chemistry@ccl.net Tue Oct 25 22:51:00 2005
From: "Cory Pye cpye]|[crux.smu.ca" <owner-chemistry-,-server.ccl.net>
To: CCL
Subject: CCL: rigid scan in Gaussian
Message-Id: <-29742-051025194809-17152-yZM5MJ7CuxR8oUvQcY8Kpw-,-server.ccl.net>
X-Original-From: Cory Pye <cpye%x%crux.smu.ca>
Content-Type: TEXT/PLAIN; charset=US-ASCII
Date: Tue, 25 Oct 2005 20:47:56 -0300 (ADT)
MIME-Version: 1.0
Sent to CCL by: Cory Pye [cpye|,|crux.smu.ca]
Hello,
this setup is inherently asymmetric along the reaction coordinate, although the
reactant and product could be the same. Any constraint here would be
artificial, but may be adequate for your needs. It reminds be of a water
mediated 1,3-hydrogen shift in triazene, see Can J Chem 80 p447 (2002) .
If the reactant and product are the same, then you could use a bit of vector
algebra and specify the coordinates of the atoms using Cartesians. I don't
think that there is an automatic way of doing this.
Suppose q1i are the 3D coordinates of the molecule before the shift, and
let q1f be the coordinates after. Then q1(l)=(1-l)*q1i + l*q1f are the
coordinates along the dimensionless reaction coordinate 0<l<1 . As long as the
atom ordering remains the same, everything should be OK.
On Tue, 25 Oct 2005, Mima Staikova mima.staikova _ gmail.com wrote:
> Hi everyone,
>
> I need to model a reaction path of hydrogen abstraction, mediated by water
> molecule.
> A hydrogen from the substrate is transferred to the water molecule and
> simultaneously one of the water hydrogens is transferred to the acceptor. I
> am interested in the energy along this pathway without reoptimizing the
> substrate and the acceptor at every step along the reaction coordinate. The
> reaction coordinate is a straight line.
>
> The Scan keyword (rigid scan) is going to give me a potential surface, not a
> potential curve, if I specify the coordinates of the two hydrogens, that are
> moving simultaneously, calculating at every coordinate change of the first H
> the all the changes for the second H.
>
> How can I do the calculation in such a way that I have for each change in
> the first H distance to the substrate just one change in the distance for
> the second hydrogen, without having to resort to multiple single point
> calculations?
>
> Thank you.
> Mima Staikova
>
************* ! Dr. Cory C. Pye
***************** ! Associate Professor
*** ** ** ** ! Theoretical and Computational Chemistry
** * **** ! Department of Chemistry, Saint Mary's University
** * * ! 923 Robie Street, Halifax, NS B3H 3C3
** * * ! cpye^crux.stmarys.ca http://apwww.stmarys.ca/~cpye
*** * * ** ! Ph: (902)-420-5654 FAX:(902)-496-8104
***************** !
************* ! Les Hartree-Focks (Apologies to Montreal Canadien Fans)