Should this be a substitute for publishing your paper in a journal
(electronic or paper)? No.
Journals (both electronic and paper) have three obvious advantages:
There are several methods, all with strengths and faults, discussed
below. As mentioned above, right now it appears that
pdf is the best method.
The process of getting your TeX file into a pdf file is now easy
with the latest TeXLive CD and Adobe Acrobat;
see Recent News.
The AMS now has PDF as one of the choices for retrieving its math papers.
The results are nice.
It is easy to print or save the pdf file from the Acrobat Reader program.
The references below give additional information and alternative methods
of make pdf files.
To use idvi, you run idvi on a dvi file. It produces an html file and
some other files which you put in a subdirectory of your ~/public_html
directory (details below). When this html file is viewed, say with netscape,
it is an applet which is in many ways similar to xdvi but over the net. The
user can jump around, make the scale larger or smaller, etc. It has
support for (some) encapsulated PostScript diagrams. If you use the
latex with the hyperref package, idvi automatically generates html links
in you paper. So if you save "By Lemma 4 ..." the user can click on the
4 and go directly to the page with that lemma.
You can see
one of my papers
on the net to see how it looks. Or you can go to
my papers page
to find other examples.
Plusses
Minuses
Speed. When you first access an idvi document the java code,
about 200K, is downloaded. Over my 14.4 modem this takes one
minute 15 seconds. Then it downloads the fonts in pk format. This means
the first pages is displayed slowly. But after the first page
most of the fonts will already be there so displaying subsequent pages
will be much faster. Also if you view an entirely different document the
java classes will not need to be reloaded.
If idvi becomes popular, many math departments will install it so
their members can display their papers. The system administrator can
copy the classes for the idvi viewer to the same place the browser
looks for its java classes. After this is done the java program will
be loaded locally, greatly improving the startup time.
When the <MATH> tags are available a program such as this may become
important but for now it is not suitable to the most mathematicians.
Ross Moore has some good documentation on LaTeX2HTML and also on Xy-pic:
This is a good way to allow others to retrieve your
your papers over the net, although pdf is better.
Of course not all copyright agreements may allow you to put your
paper on the web. If you are uncertain you might want to check with
the publisher. For future papers you might write in something like
the above sentence on the copyright agreement.
Author:
Ralph Freese
On this page we present various methods currently available
of putting mathematics on the web together with some instructions.
Our primary focus is on methods for researchers, who are not experts
on the technologies involved, to put their mathematical papers on
their web homepages. For the most part we are concerned with taking
a paper prepared in TeX and putting in on the web but we also discuss
future format of electronic papers.
Essentially I view this as a much more convenient way of distributing
your preprints and reprints. Other mathematicians can gain instance
access to them; they can read them on the web and decide if the paper is
what they are looking for and print out a copy if they like. This has
obvious benefits for mathematical research.
Journals are of course very valuable to the mathematical community.
Putting your papers on your web site should not replace publishing them.
Copyright issues are discussed below.
Some of the desirable properties of a program to put mathematical
papers on the Web include:
PDF stands for Portable Document Format. It is a special PostScript
file format. It can include hyperlinks both within the document and
to outside URL's.
Adobe makes a program, Acroreader, which can display
pdf files. This program is free and this format is fast becoming a
standard. It supports searching and high quality printing.
If you try to access a pdf file through Netscape on a PC,
Netscape will ask if you would like to add Acroreader as a plugin.
On Macs it will ask if you want to add it as a helper and on unix systems
a system administrator can easily add it. This means most people can
easily access a pdf file through the net.
One warning: If you have Acroread 2.1 or older,
you need to upgrade to version 3.0 or newer.
Idvi is essentially a dvi viewer for the internet. It was written
by Garth Dickie and is described in more detail on his
IDVI PAGE. You can
also go directly to his
USERGUIDE.
I have installed it on our system and grabbed the
simple basic instructions for
preparing a document with idvi.
There is also a
simple reader's guide
with instructions for navigating
and changing magnification within an idvi display.
This is a perl script that parses a latex file and makes an html file.
For all but the simplest math it generates a gif and places it in the
html file. It is very fragile and difficult to get to work correctly.
For a paper with much mathematices each html page must contain a multitude
of gif files. This makes the downloading slow and means the reader has
to size
his web browser's font to fit the characters in the gifs. It can only
deal with parts of latex. For example, it is likely to not understand
theorem and proof environments and put them into big gifs. You will
almost certainly have to maintain a separate version of your tex file
for the latex2html version.
This uses java to allow you to write an html page using the math extensions
that are in html3.0. Version 2.0 will be released soon; see
Recent News. Although pdf is (currently) the best
for getting mathematical papers written in TeX on the web,
for situations when you need to include mathematics directly in html
such a page for students which includes html forms or java or javascript,
WebEQ is very nice and gives a glimpse of the future.
For more information visit the
WebEQ page.
This is a netscape plug-in that is able to interpret a subset of
TeX. There has been a lot of work on this project since I last
updated this page. Version which will run on Macs and Unix machines
(although there is some delay with linux) will be available very soon.
It now handles a larger subset of TeX including some LaTeX.
Since it does not handle all of TeX you are likely to have to maintain
a separate TeX file for your web page and so it is not ideal for putting
a TeX paper on the web but it is very good including math on a html
web page, say with interactive forms for your students.
You can find out more about this program directly from
their URL:
Techexplorer
homepage.
You can put links to such files on your page. (I did this on
the page with some of
my papers.) If a user clicks on such a link
to one of your papers, what will happen will depend on how his
browser is setup. It may launch ghostview or xdvi or it may just
download the file to the user's directory. With netscape you can click
with the right bottom. This gives you a menu and you can choose
to save the link. This all works ok but may be a little confusing to
the user. Also ghostview has it own method to save the file but
xdvi doesn't.
When you write a paper you become the copyright owner (at least in most
countries including the US). When you publish it you sign a
copyright agreement with the publisher which usually is an assignment
of the copyright to them. Often the agreement gives the author
and the scientific community the right to make copies. (Essentially
this is licensing back some of the rights to the author.) For
example the AMS copyright agreement includes the sentence
"The Work may be reproduced by any means for educational and scientific
purposes by the Author(s) or by others without fee or permission with
the exception of reproduction by services that collect fees for delivery
of documents." The
AMS
Copyright Agreement is online.
Last modified: Aug 11, 1997.
Please send comments by email to
ralph@math.hawaii.edu