Electronic processing of IJM papers
This page describes the system of processing papers electronically
that is being implemented at the Illinois Journal of Mathematics.
The system is modelled after one described by Princeton Computer
Scientist Andrew Appel in
"How to edit a journal by e-mail", Journal of Scholarly Publishing 27(2),
January 1996, 82-99. (The software implementation, though, is
different, using Perl instead of awk as programming language, and pine
instead of mh for processing email.)
This page gives an overview of the system.
A technical description is
given on a separate page.
- Centralized processing. All routine correspondence
(such as acknowledgements, referee inquiries, or thank-you letters)
will be conducted from the IJM Office, rather than individually by
each editor. Most of this correspondence will be done with form
letters that are generated automatically from a pool of available
templates (though with additional manual editing, if necessary). On
request by an editor, the IJM Editorial Assistant can customize the
standard letters in various ways, for example, by paraphrasing a harsh
referee's report using more diplomatic language.
- Electronic communication.Whenver possible,
correspondence will be carried out
electronically, using postscript files to
transmit papers to referees and editors.
- Efficient and reliable record keeping. All correspondence
regarding a particular paper will be saved (usually automatically)
in an easy-to-retrieve form, in a directory under the ijm account at
UIUC Math Department's Unix computer network.
- Substantially reduced burden on editors. Editors do not
have to deal with the hassle of keeping track of the status of each
paper they are handling, sending inquiries, reminders, and thank-you
letters to referees, sending referee reports to authors,
or digging out email/mail addresses of referees.
The editor's duties will be limited to tasks that require their expertise,
such as selecting a referee for a paper, and making decisions on
acceptance based on a referee's report, but will not include the
overhead involved in carrying out routine correspondence.
(Of course, editors are free to carry out such correspondence
themselves if they prefer; this might be appropriate, for example, in
cases where the referee is a colleague or student of the editor to whom the
paper can simply be passed along.)
- Substantially reduced clerical work.
With the new system, nearly all communication is conducted via email.
This is much more efficient and less timeconsuming than preparing hard
copy letters, especially with tools for generating form letters such as
the "makeletter" program described in the
Technical Details page.
For example, sending a paper to a referee
by mail requires preparing a hard copy cover letter to the referee,
making a Xerox copy of the paper, and enclosing this copy, along with the
"Referee Form" in an envelope, and addressing that envelope.
By contrast, using the makeletter program to generate an electronic
referee request requires nothing more
than filling in the "To:" address in the email header and the
salutation in the body of the letter. The makeletter
program attaches both the paper
(as postscript file) and the referee form, and creates an appropriate
email message from a template, with the
author/title/editor information already substituted.
As a result of this automation, routine clerical work takes up less
time, allowing the IJM Editorial Assistant to concentrate on dealing
with situations where a customized reply is called for. Such tasks
might include editing a referee's report for language problems before
forwarding the report to the author or paraphrasing a harshly worded
referee's report using more diplomatic language, if an Editor so
- Reduced processing time. Sending papers to editors and
referees electronically as ps files eliminates the delay involved in
sending paper copies of papers. This is especially significant in case
the other party is overseas, or if several potential referees have to be
contacted before one is found who is willing to serve as referee.
- Better manuscript tracking. Since all editorial records are
kept in a central location under the ijm unix account, the entire editorial
history of a paper can be instantly retrieved, and an author inquiring
about the status of the status of a paper can be sent an appropriate
reply from the IJM office. In the past, such inquiries would often have
to be forwarded to the editor handling the paper.
- Easy and efficient report generation.
The centralized record keeping makes it easy to create tools to
generate reports of various kinds, such as lists of papers that have
been in the hands of the referee for more than a specified amount of
For the system to work efficiently, all parties involved in the
editorial processing must be able to easily handle electronically
transmitted files. For the most part, this
has not proven to be a problem, and the trend is clearly towards an all
electronic form of communication.
submission of papers. The vast majority of papers submitted
to the IJM come in electronically, via a dedicated email address,
email@example.com. Virtually all of the remaining papers
have been prepared in some form of TeX, and thus are, in principle,
available in electronic format. If a paper that has obviously been
prepared in TeX is received as hard copy,
the author will be sent an email requesting a copy of the tex file,
and, if possible, a postscript version.
Though sending an electronic copy is not an
absolute requirement for being considered for publication,
most authors are willing and able to comply, in the interest
of expediting the refereeing process.
- Electronic transmission of
papers to editors and referees.
My experience with sending postscript
versions of papers to editors, referees, and (in the form of Galley
proofs) authors, has been overwhelmingly positive. These days, nearly
everyone in the academic world,
it seems, can print out a postscript file sent as an
attachment. Even with people located in
countries that are far behind the west in their degree of
computerization, I have rarely encountered problems; in many of these cases,
sending papers by regular mail would be far less reliable (and, of
course, much slower) than doing so by email. When sending papers by email,
it is important to make things as easy as possible for the receiving
party. This means:
- Sending postscript files rather than tex source files.
tex files can be a pain, may require a considerable degree of TeX
knowledge, and it may not work at all if the local TeX installation
lacks some of the style files needed for processing the paper.
For this reason we ask authors to provide a postscript version of their
paper, in addition to the tex file.
Postscript is a standardized language, and a postscript
document should, in principle, be understood by, and
produce identical output on, any
postscript capable printer or viewer. (This is a bit of an
oversimplification, but in practice this has proven correct.)
Dvi files represent a compromise between ps and tex files. In most
cases, a dvi file can be converted to a ps file using dvips or similar
tools. One drawback of dvi files is that they may not display figures
embedded in a paper. Sending a dvi file
may be appropriate if,
for some reason (usually because the postscript file is too large),
sending a postscript file is not possible.
- Sending the files as attachments, not in the body of the
message. Most email readers understand attachments and
automatically split an email into attachments, making it easy to
print or view the attachment. If a ps/tex file is simply incorporated into the
body of the email, one has to first save the message, then
manually edit the message to extract the file.
- Providing web access to files as an alternative.
By making a file (usually ps or pdf) available through a web address
(URL) which is made known
to the other party (referee or editor), but is not linked or
otherwise publicized, one can make the file available to
this party while keeping it, in effect, non-public. By opening the URL
with a browser, the other party can dowload the file. The advantage of
this method is that there are no inherent size limitations.
- Electronic submission of referee reports.
Referees should be encouraged (but not required) to submit their
reports by email. If a report is available only as a hard copy, it
will have to be sent to the author in that form as well, thus slowing
down the process. In some cases - for example, when the referee has
marked up the paper - sending a hard copy is unavoidable, but in the
vast majority of cases referee reports have been prepared in tex and
thus can be sent electronically as tex files.
Reports that do arrive by email are
often sent as tex files incorporated into the body of the message.
In this case, the message has to be saved and edited to extract and
process the tex file. The problems mentioned above regarding sending
tex files are rarely an issue with referee reports since such reports
tend to be short, TeXnically straightforward documents, and thus
rarely present problems when compiling.
Last modified Mon 31 May 2004 05:42:28 PM CDT