LWN.net Logo

Using open-source tools for documenting research

January 18, 2006

This article was contributed by Carl Bolduc

Introduction

Getting published is a major concern for students conducting graduate studies in science. I'm a PhD student in molecular biology and I started using Linux at the beginning of my graduate studies. Public science research mostly looks like open-source software development. You work hard and give your methods and results to everyone through publications in scientific journals. Ironically, the majority of people working in the field of science use only proprietary software. I myself work in a Microsoft Windows environment.

A typical scientific article will require the use of several tools to reach its final published state. First, most researchers use Microsoft Word and Excel for text writing and tables. They also use EndNote to manage and create the bibliographies you will find in every scientific article. Finally, scientists use a graphics suite, such as Adobe's Photoshop, for figures and PDF creation. This software listing scales up to more than one thousand dollars. It's practically impossible for the regular student to purchase such a platform. In some laboratories, when the head researcher is kind enough, you will find a computer where most of these tools are installed and shared by all members of the team. But what if you could create your own open-source research writing box for free? In fact, you can. You can accomplish the entire array of tasks associated with scientific writing with any good Linux distribution.

The easiest step

One of the most popular open source application that has boosted the Microsoft to Linux transition is certainly OpenOffice.org. For anybody working in science reporting, it is a first and easy step that enables you to step out of proprietary software and remain compatible with Microsoft Office formats. In addition, several journals will ask that the submissions should be in the .doc or PDF format. OpenOffice.org saves you a lot of trouble with its useful PDF export tool.

Although OpenOffice.org can complete a fair portion of the job, it doesn't contain a bibliographic manager tool such as EndNote yet. Such a facility is necessary for academic writer and OpenOffice.org is supposed to fill the blank with some bibliographic extensions in its next version. For now, there is a commercial web-based tool called WriteNote which offers a 30 day free trial and enables you to produce a bibliography with RTF files created by OpenOffice.org.

LaTeX

While OpenOffice.org may be a first step toward writing scientific articles under Linux, the true power resides inside LaTeX. As it is mentioned on the latex project website: "LaTeX is a high-quality typesetting system, with features designed for the production of technical and scientific documentation". "LaTeX is the de facto standard for the communication and publication of scientific documents."

Some of the LaTeX features include insertion of tables and figures as well as the capacity to create complex mathematical equations. Additionally, there are tremendous advantages in learning to write with LaTeX. In fact, BibTeX could get you out of proprietary software tomorrow. You can gather your bibliographic references in a simple text file with the BibTeX syntax and easily insert quotations inside your LaTeX documents, automatically generating a bibliography at the end of your articles.

While the LaTeX format requires a minimum of learning, you can rely on the useful TeXmed web-based tool to query NCBI PubMed and generate BibTeX entries for you. You must specify, in your LaTeX document, a bibliography style to format it according to the journal's recommendations. In fact, many journals now offer their bibliography style on their website. If you can't find the format that you need on the web, you can use custom-bib to create the style you need.

LyX and friends

LaTeX basics can be learned quite easily, but you might need to read a lot from the web or buy some books (like I did) to use its full potential. But do you really need to go through all this trouble? That's where LyX comes into play.

LyX is a GUI front end to LaTeX. Though it has its own file format, it can import and export to LaTeX. LyX looks like a word processor while taking care of all the formatting, just like LaTeX. LyX is fully featured and let you insert figures, tables, mathematical equations and more. Though managing a bibtex text file is very easy, you can rely on graphical tools here too. Software like gBib and JabRef will help you deal with your numerous references and even let you insert them in LyX, just like EndNote does with Word.

Gnuplot

Continuing in your path to build an open-source research writing box, you need a powerful tool to generate plots from your precious experiment results. That is where Gnuplot enters the scene, with its almost limitless possibilities. Gnuplot is a command-line plotting utility with easy to learn commands that enable you to create high quality 2D and 3D plots suitable for scientific publications. It can output LaTeX and EPS code which can be inserted in your LaTeX documents. You can check out this demo that shows the wide variety of Gnuplot's capabilities.

Inkscape

One thing that was really missing in Linux in the past was a good vector graphics editor. I had to install Adobe Illustrator under Wine to be able to draw high quality figures showing various metabolic pathways. Now, with Inkscape, I have everything that I need to create high quality vector graphics which can be exported to EPS and inserted in my LaTeX documents. Inkscape can draw shapes, paths, text and can also export to PNG.

The Gimp

To complete your open-source research writing box, you need a powerful image manipulation program to process your photos and to generate figures from them. That's where The Gimp comes into play. With The Gimp, you can process gel photos, crop the area that you like, obtain negatives of your originals and add labels where you want, all with a few mouse clicks.

Linux drawbacks

While this path can be rewarding, a significant effort will be required. The first thing you need to do is to install a Linux distribution. This might seem frightening to the newcomer, but there are powerful Linux distributions such as Mandriva, Fedora and Ubuntu which are very easy to install and have packaged most of the tools mentioned it this article.

You also need to learn how to use new software. A few of the applications mentioned above only have a command-line interface, but most operations can be performed using GUI-based tools. There is plenty of documentation online, and you can always join an IRC channel to get live help. In a short time, you will become very functional, and you will reach new levels of productivity.

The worst drawback of using Linux in a Microsoft-based environment may involve compatibility issues with your coworkers. Since my boss insists on working with .doc files, I have to convert my papers to RTF using latex2rtf before I send him anything, even if PDF is the most portable format out there. But this doesn't stop me from benefiting of the LaTeX functionality.

Finally, you must rely on the Internet for support. Most of the system administrators in the research field don't know much about Linux (at least not in Quebec, where I'm working) and won't be able to support you if you have problems.

Linux superiority

Beside the fact that Linux and all this software is free, there are many advantages in building an open-source research writing box. Linux provides a robust environment that is a virtually virus-free. Interoperability among applications is quite good, all of the applications mentioned in this article can share data through the LaTeX and EPS file formats.

With little experience, you will start working faster and more efficiently. Serious page formatting issues found in Windows-based WYSIWYG software will be gone. Finally, you will be able to easily share your work by creating high quality PDF files.

An example screenshot of my desktop publishing environment can be seen here.


(Log in to post comments)

You can reach me here...

Posted Jan 19, 2006 3:52 UTC (Thu) by carlabi (guest, #34248) [Link]

I'm the author of this article, you can reach me here:

DrCurl at gmail.com

;)

Nice article!

Posted Jan 26, 2006 6:45 UTC (Thu) by moxfyre (subscriber, #13847) [Link]

I'm also a grad student (physics) and, like you, I've figured out my own set of tools for productive science work under Linux:

LaTeX
BibTex
Octave (MATLAB clone, super awesome!!) for crunching data and graphing
Gnuplot
Xcircuit for drawing schematics (free as in speech)
Eagle for drawing schematics (free as in beer)

Another important task for me is automated data taking. I use Perl with GPIB modules to control programmable lab instruments. For me, this is vastly better than using LabView which is proprietary, and in my opinion damaging to the brain of any real programmer. :-)

LaTeX is utterly, completely, and totally superior to Word for technical writing. You can't make a Word document too long before you start getting all kinds of problems with graphs mysteriously jumping around, images not scaling properly, table of contents not building correctly, etc. With LaTeX you get exactly what you want and once you've had a bit of practice you can write very quickly.

I took a lab course last semester where the professor did all his handouts in Word. Most of the graphs were anti-aliased to the point that they were barely legible (bad rasterization?) and there were lots of weird formatting quirks. He had no way to generate an index for the 100-page lab manual. Me and other LaTeX-using students would hand in perfectly formatted 30 page reports with flawless graphs and schematics and he'd look on enviously...

xmgrace

Posted Jan 19, 2006 8:24 UTC (Thu) by vblum (guest, #1151) [Link]

Grace (xmgrace) deserves a mention as a powerful 2D plotting tool with all sorts of extra capabilities.

Using open-source tools for documenting research

Posted Jan 19, 2006 8:29 UTC (Thu) by jwb (guest, #15467) [Link]

Might as well mention the R Project, which aside from providing a wonderful statistics package to help you complete your actual research, also has facilities to generate attractive graphics of very high quality.

Using open-source tools for documenting research

Posted Jan 19, 2006 8:36 UTC (Thu) by jrbeire (guest, #21244) [Link]

"One thing that was really missing in Linux in the past was a good vector graphics editor."

Similarily to the article's author I too started using open-source tools when commencing my graduate studies (and converted completely to open-source by mid 2000). Even then, however, there was an extremely powerful vector graphics programme for linux known as xfig. IMHO xfig remains a superior tool to inkscape by a considerable margin at present (aside from .svg support).

Regards, John.

Using open-source tools for documenting research

Posted Jan 19, 2006 10:21 UTC (Thu) by dvrabel (guest, #9500) [Link]

Xfig may be powerful but it has a horrific user interface.

For example: I fired on xfig, drew a rectangle and now I can't find how to pick the rectangle so I can resize it...

Using open-source tools for documenting research

Posted Jan 19, 2006 10:26 UTC (Thu) by dvrabel (guest, #9500) [Link]

Whilst I felt like shooting xfig after spending more than 30 seconds hunting for the required button (resizing with the 'move points' button is a non-obvious step) I clearly meant "I fired up xfig...".

(LWN really needs an 'edit post' feature.)

Using open-source tools for documenting research

Posted Jan 20, 2006 9:12 UTC (Fri) by jrbeire (guest, #21244) [Link]

Yep - can't deny that xfig makes some getting used to. But when you do.......

Using open-source tools for documenting research

Posted Jan 26, 2006 13:40 UTC (Thu) by rjforster (guest, #3375) [Link]

For any readers not understanding what makes XFig so hard to get to grips with, it uses a 'verb then noun' structure rather than a 'noun then verb' structure like Corel Draw does.

This means you choose what you want to do before you choose what you want to do it to, ie click the 'resize' button before the rectangle you wish to change the size of.
This is instead of clicking on the object you wish to change then looking for the control to make the change.

Using open-source tools for documenting research

Posted Jan 19, 2006 10:15 UTC (Thu) by dambacher (subscriber, #1710) [Link]

This is what I experienced on doing my doctors thesis:

latex was superior to all other word processing tools in that it does not fiddle with your layout once you set it up. You just write. No headaches. And if it comes to doing a printed book, you can generate high quality pdf or ps files wich can be imported directly to the (commonly mac-based) printing systems, including propper page sorting!

Openoffice spreadshead graphics was worse for me, unusable. I have to check if versino 2 is better in this. I had to do them using wine/excel and I hated it, because you need special tools to extract the graphics from wmf to eps for high quality printing.

Using open-source tools for documenting research

Posted Jan 19, 2006 10:45 UTC (Thu) by nix (subscriber, #2304) [Link]

And documents pretty much never break. I just re-LaTeXed a document I wrote in 1994. It worked, and it looked the same as it did back then. (This will always be true for the underlying TeX layer, of course, because it's nearly maintenance-dead upstream with an author known as an utter paranoid for identical results no matter what. But even e-TeX produces nearly identical results for nearly all documents.)

Using open-source tools for documenting research

Posted Jan 24, 2006 5:16 UTC (Tue) by roelofs (guest, #2599) [Link]

And documents pretty much never break. I just re-LaTeXed a document I wrote in 1994. It worked, and it looked the same as it did back then.

That was not my experience. LaTeX changed rather massively between 2.09 and 2e; I found I had to make several changes to my dissertation (ca. 1995) to get it to "compile." And it sounds (or sounded, anyway) like 3.0 will/would be an even bigger break. Of course, these days my usage is limited to the occasional one-page letter, so I'm not terribly worried about future compatibility...

Greg

Using open-source tools for documenting research

Posted Jan 25, 2006 16:09 UTC (Wed) by nix (subscriber, #2304) [Link]

Yes; 2.09 -> 2e was a large jump, but *2.09 still works*, because it runs atop an unchanging core (TeX). So you can use both at once for different documents. :)

Using open-source tools for documenting research

Posted Jan 25, 2006 16:26 UTC (Wed) by roelofs (guest, #2599) [Link]

Ah, thanks for the clarification.

Of course, from a practical standpoint, that means one needs to compile and install 2.09 first--and web2c before that, most likely--since no distribution of which I'm aware ships both. If you have lots of older documents and need only to do a one-time conversion to PDF (for example), that may be the simplest approach. But in general I'd expect most people would find it easier to modify the document(s). ;-)

Greg

Using open-source tools for documenting research

Posted Jan 26, 2006 14:45 UTC (Thu) by fgrosshans (guest, #35486) [Link]

I think not: Latex2e has a compatibility mode, pretending to be latex 2.09 when it processes documents starting with \documentstyle instead of \documentclass .

Disclaimer: I never tested this functionnality.

Fred

Using open-source tools for documenting research

Posted Jan 26, 2006 19:49 UTC (Thu) by nix (subscriber, #2304) [Link]

The compatibility mode isn't terribly capable. It can handle simple documents.

Using open-source tools for documenting research

Posted Jan 19, 2006 10:16 UTC (Thu) by hingo (guest, #14792) [Link]

Not being familiar with EndNote I can only wonder why the OOo Tools > Bibliography Database is not good enough?

Mostly a matter of the network effect

Posted Jan 19, 2006 12:10 UTC (Thu) by fredrik (subscriber, #232) [Link]

From what I here from others, it's not a matter of which utility has the technical edge, but _mostly_ a matter of wanting to use "what all the others use". The network effect is important, people need to share bibliographic information, and people co-write papers. And since EndNote appears to have a firm grip of the unfortunate Windows and Word users out there, everything else that is not 1:1 compatible will have a hard time competing. Even if it could solve the task at hand better than EndNote.

Fredrik

network effect, BibTeX

Posted Jan 19, 2006 13:18 UTC (Thu) by pjm (subscriber, #2080) [Link]

http://bib2web.djvuzone.org/bibtex.html describes BibTeX as the “de facto standard for publications in several fields of the ‘hard sciences’ (physics, mathematics, computer science and engineering”. I can confirm this for Computer Science: the name EndNote is unfamiliar to me, whereas BibTeX is used by CiteSeer.

The BibTeX format is plain text and well-understood: see e.g. http://en.wikipedia.org/wiki/BibTeX. The aforementioned http://bib2web.djvuzone.org/bibtex.html points to converters between BibTeX and numerous other important bibliographic formats.

The EndNote web site refers to ftp://support.isiresearchsoft.com/pub/bibtex/ for (apparently Macintosh-only) conversion between BibTeX and EndNote; where one finds ftp://support.isiresearchsoft.com/pub/bibtex/bibtex_expor... which gives the impression that all of the limitations in converting between the two formats are due to limitations in the EndNote format rather than in the BibTeX format.

It may thus be more profitable to seek compatibility between OpenOffice and BibTeX than trying to track a proprietary product like EndNote.

See http://wiki.services.openoffice.org/wiki/Bibliographic_So... for OpenOffice.org's wiki page on bibliographic software.

RefDB

Posted Jan 21, 2006 21:36 UTC (Sat) by liamh (subscriber, #4872) [Link]

You might look into RefDB, a reference database and bibliography tool. It is accompanied by conversion tools. There are some subtle points about bibliographic references that makes it hard to define a standard and perform intelligent conversions, but as far as I've seen, the RefDB people have put the most effort into it.

OOo biblio database SUCKS!! (at least in 1.1.3)

Posted Jan 23, 2006 1:11 UTC (Mon) by hazelsct (guest, #3659) [Link]

I just got through with an attempt to write a proposal with citations using OO.o 1.1.3 and its bibliographic database. The goal was to easily produce .doc files which my collaborators could edit and return with revisions.

It was excruciatingly painful. Bad. Horrendus. So dreadful, that after the edits to the text, I copied and pasted the whole thing into emacs/LaTeX, and re-did the bibliography using BibTeX.

First, entries such as author, title, etc. were length-constrained to about THIRTY CHARACTERS! What the hell??? Almost all of the articles I cited had longer titles or lists of authors.

Second, the default formats absolutely SUCKED! And when a field was missing (e.g. booktitle, editor), it just put two commas with blank space between them: "Author, Title, , , 2004.!" At least BibTeX can sanely deal with missing fields.

Third, the bibliography did not auto-regenerate when new entries were added, and there was no way to re-generate the bibliography!

Fourth, I share config files among a bunch of machines using rsync, and some machines used the database in .openoffice/1.0.1/database/biblio, others in 1.1.1, others in 1.1.0. What the hell??? All of these machines are using identical versions of Debian sarge and identical .openoffice tree, why can some of them open the database and others not??

Version 2.0(.1) may be much better, but that they could let such a piece of crap get released has forever soured me in using OOo for any document involving a database, which for me is just about everything.

Sorry about the all-caps, this was such a nuisance to me for the past few days I'm a bit emotional about it just now. Meanwhile, I'm upgrading to Debian etch with OOo 2.0. In the meantime, BibTeX users considering a migration should "caveat scriptor".

most scientists?

Posted Jan 19, 2006 13:44 UTC (Thu) by jbw (guest, #5689) [Link]

In my experience, most computer scientists and mathematicians use LaTeX and BibTeX and do not use Word. I notice this in particular when I serve on conference program committees and review submitted research papers: it is visibly obvious (from the use of the Computer Modern font in many cases) that LaTeX was used to produce most of the submissions.

Joe Wells

most scientists?

Posted Jan 19, 2006 15:04 UTC (Thu) by dambacher (subscriber, #1710) [Link]

Yessss .-)
I never found a better tool to typeset a page of formula.
Even if you use other fonts. First formula and you know wich program was used for typesetting.

most scientists?

Posted Jan 19, 2006 16:23 UTC (Thu) by carlabi (guest, #34248) [Link]

In biology, genetic and medical sciences, LaTeX is the exception. In math and physic fields, LaTeX must be a lot more popular...

Carl

most scientists?

Posted Jan 19, 2006 20:56 UTC (Thu) by kamil (subscriber, #3802) [Link]

You might be right with conference papers, since many conferences will prefer LaTeX submissions, and besides, such papers are often prepared by students and young researchers, who tend to be more "geeky".

However, in my experience (I've worked at a CS department at a Dutch university for over five years until last year), a large majority of professors/faculty/permanent staff/you name it prepare their documents in Windows/Mac environment, and will only tolerate LaTeX and such if they themselves don't need to edit a single line of it.

A tendency I have observed is that the more senior they are, the more Windows-only they get. Kind of strange when you consider that many of the ones I know have physics background, and physicists tend to be down-to-earth people who favour simple tools they can understand.

A cynic inside me would say that the more managerial responsibilities they have, and the more contact they have with business people and with big research money, the more Windows-loving they become... Coincidence?

Also, even among the LaTeX users, PowerPoint is a de-facto standard for presentations. I could count on one hand the people I know who use something else.

most scientists?

Posted Jan 27, 2006 0:15 UTC (Fri) by jbrusey (guest, #35501) [Link]

This is a great pity (that there are few LaTeX users using something other than Powerpoint) as there are a number of good packages for creating great presentations with LaTeX and LyX. I have been using a package called "beamer" for some time now and can thoroughly recommend it.

http://latex-beamer.sourceforge.net/

They have plenty of great samples, but here's a set of slides that I did for a lecture on Petri net based control:

http://www2.eng.cam.ac.uk/~jpb54/met2pn-2005.pdf

Most of the diagrams have been done with MetaPost.

most scientists?

Posted Jan 26, 2006 20:04 UTC (Thu) by arcticwolf (guest, #8341) [Link]

Actually, among mathematicians at least, LaTeX is not only the most widely-used typesetting tool, it's pretty much the *only* typesetting tool used. Most mathematical journals (at least the ones I'm familiar with) will not accept submissions in formats other than LaTeX [1].

1. Or possible AMS-TeX / AMS-LaTeX or so, but I think that's sufficiently close to LaTeX to be included in there.

LaTeX/BibTeX bibliographic notes for arts/humanities and law scholars

Posted Jan 19, 2006 15:02 UTC (Thu) by dune73 (subscriber, #17225) [Link]

If you favour footnotes over endnotes and need more custom citation formats as used in arts and among law scholars, then you should look into the jurabib macro package.

I wrote my PhD in medieval history in LaTeX. This was made possible by Jurabib.

See http://www.jurabib.org

See a humanities screenshot here: http://www.jurabib.homelinux.org/jurabib/screenshots/scre...

Christian

Graphics and analysis

Posted Jan 19, 2006 16:20 UTC (Thu) by vondo (guest, #256) [Link]

ROOT from CERN and IDL are two very full featured data analysis and plotting packages. ROOT has its basis in high energy physics, but it would seem useful for other fields as well. One very interesting thing about it is that the scripts are written in C++ which is interpreted. Once you've debugged them, you can compile them and include them in your programs or just run them stand-alone for better performance.

Graphics and analysis

Posted Jan 19, 2006 20:01 UTC (Thu) by kmccarty (subscriber, #12085) [Link]

ROOT from CERN and IDL are two very full featured data analysis and plotting packages.

I was just about to mention ROOT! It's one of the main analysis tools I've been using for my thesis. (Experimental physicist here.) It has a large user community and great documentation available at the website. Even better, it's recently been put under the LGPL. There are unofficial Debian packages of it at mirror.phy.bnl.gov and it ought to be entering Debian proper within a few months. (I know this because I am the Debian developer who intends to sponsor the packages :-)

Don't forget TeXmacs

Posted Jan 19, 2006 22:58 UTC (Thu) by jeskritt (guest, #4092) [Link]

TeXmacs is a wysiwyg latex editor and is often considered to be better than lyx

http://www.texmacs.org/

Don't forget TeXmacs

Posted Jan 20, 2006 16:22 UTC (Fri) by tzafrir (subscriber, #11501) [Link]

LaTeX or LaTeX-like? AFAIR it does not use or reimplement TeX. Rather, it has its own rendering.

Using open-source tools for documenting research

Posted Jan 20, 2006 9:10 UTC (Fri) by MortenSickel (subscriber, #3238) [Link]

I haven't seen R mentioned anywhere, although it is a statistical packages, it is also great for doing plotting. Morten

Scientific Python and Matplotlib

Posted Jan 20, 2006 10:14 UTC (Fri) by tnoo (subscriber, #20427) [Link]

Very convenient and easy to use tools for data analysis, programming and plotting are Scientific Python (Scipy) and Matplotlib.

As for writing I concur that there is nothing more productive than LaTeX. Better spend some time learning the markup in the beginning of a thesis, than reformat everything by hand one hour before the deadline (which I did for a former girl friend who used Word). best, tnoo

docutils & reStructuredText

Posted Jan 22, 2006 13:16 UTC (Sun) by kirr (subscriber, #14329) [Link]

I'm surprised, no one had mentioned docutils/rst

See:
http://docutils.sf.net/
http://docutils.sf.net/rst.html

docutils & reStructuredText

Posted Mar 15, 2006 12:09 UTC (Wed) by peterb (guest, #36485) [Link]

I think the reason is that reST doesn't support bibtex. Or any alternative bibliographic handling.

I'd love to be proved wrong, because I have half my masters thesis in reST format, and am approaching the stage where I will need to convert it to LyX to continue... which I would prefer not to do, as I am really enjoying writing plain text!

Using open-source tools for documenting research

Posted Jan 24, 2006 23:35 UTC (Tue) by wolfgang.oertl (subscriber, #7418) [Link]

I also wrote my thesis (medical) using LaTeX. There always is statistics involved, so the results of statistical tests are mentioned in the text. When the thesis of a few dozen pages is almost finished, and then some data change -- how to update this quickly and reliably?

So I embedded comments (which do not show up in the end result) with statistical tests and their intended results, just like asserts in a program, just before or after each paragraph that contains statistical results. A Perl script that extracts and runs the tests in GNU R makes updating the text when the data change very easy (which actually happened)!

Using open-source tools for documenting research

Posted Mar 28, 2006 7:52 UTC (Tue) by yarikoptic (subscriber, #36795) [Link]

Nice startup on introducing people to the world of proper technical writing. I wanted to add about existance of pyblink which allows to "merry" pybliographic and OO.org, providing a very easy way to create references in OO while using bib files...

To keep ideas and information up-to-date, I've decided to start up a wiki page on scientific writing. Any recommendations are more than welcome...

Using open-source tools for documenting research

Posted Dec 27, 2006 16:58 UTC (Wed) by Lea (guest, #42427) [Link]

I wrote thesis with latex and when starting to write an article discovered
that the journals in my subject want only Word documents with embeded
figures... (I can't imagine that learning to use latex could be difficult
anybody who is able to deal with science.)

Now I'm looking for tips how to write an article so that I could submit it
as Word document. Is OpenOffice Writer the best choice? Do you have
experiences with exchanging doc-files (which include figures and tables)
with MS users? Is there any place where to learn tips about using
OpenOffice Writer for submitting scientific article?

Thank you,

Using open-source tools for documenting research

Posted Jan 15, 2007 16:07 UTC (Mon) by srbjunk (guest, #42772) [Link]

In reply to:

> Now I'm looking for tips how to write an article so that I could
> submit it as Word document.

Alternative A:
---------------

Write it in LaTeX (or lyx), convert it to rtf using latex2rtf

You have at least a couple of choices:

* contribute your work as rtf (readable by Word)
* open your rtf file in OpenOffice and save as .doc (readable by Word)

Alternative B:
--------------

I think this works as well

1. Write your document in LaTeX
2. Convert from latex to html using latex2html
3. Rename your document to .doc (ex: foo.html renamed to foo.doc)

This .doc-file which is actually a html-text, will trick Word to believe it is a Word-document. When Word opens it, Word discover it is actually a html-file, and will convert it.

Alternative C:
--------------

Write in LaTeX, publish as pdf

Copyright © 2006, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds