...Economics

We would like to thank, without implicating, Mark Dickie, James T. Lindley, and Frank Mixon, and participants at the Second International Conference on "Computing in Economics and Finance" in Geneva, June, 1996 for helpful comments.

...written.

When we began this paper (February 1996), we had not thought about ``conference'' editing of a document. In the beginning of May 1996, we added this idea. That month, Microsoft announced Microsoft NetMeeting, and we used it to edit both this paper and PowerPoint presentations.

...computer.

Thus, the emphasis in a networked world is not so much on reading material on the computer screen, but rather easy access to material. Of course, you may prefer to print the material locally rather than read it on the screen. Note that since very few readers will print everything they access, an electronic medium should save paper.

...form,

Even today the amount of information available on-line is amazing. At live Internet demos, we search the upcoming on-line version of back issues of the AER, and the NBER Macro Historical Database. We have found 6 articles and many data series which mention the proverbial ``kitchen sink.''

...library.

Some might say that the ultimate version is one published copy of a journal, with copies faxed to all readers and the publisher compensated by copyright fees paid by each reader.

...databases,

The former include working papers on central servers like EconWPA , and the later include databases of working papers WoPEc and programs (CodEc) and (ELSA) .

...computers

Such as a 133 Mhz Pentium, which is now-late 1996-an ``entry-level'' machine, which rivals the power of mainframes of not long ago.

...knowledge.

When access to information is restricted at universities, culture clashes often develop. For instance, some universities have had conflicts with their microbiologists who do not publish their results in order to protect trade secrets.

...refereeing,

Some journals pay a token honorarium, but this is not compensation for the effort expended on refereeing a journal submission. Referees could earn more per hour raking lawns.

...reprints).

On the other hand, academic textbook writers are ``trade'' authors

...change.

Note that this does not necessarily mean the material is not copyrighted; rather, the restrictions on use would simply fall-the rights holder permits copying. This sort of copyright is common in some software, and the best example is the GNU (GNU) ``copyleft''-copying is permitted, and the program can even be resold, but resellers cannot restrict further copying. What is claimed to be the world's second most popular version of Unix, Linux, which is built by thousands of volunteers around the world, is ``copylefted.''

...index.

Ginsparg (1996) explains many of it features, philosophy, and outlook for the future.

...success.

Success can be measured in a number of ways-number of accesses to the archive, typically 70,000 accesses per day from over 35,000 users in 70 countries; number of submissions, typically 200 per month to the most successful area, hep-th; number of preprints stored, over 20,000. Most if not all high-energy physicists do not consult hard copy journals anymore-only the archive. Also some senior physicists sometimes don't bother with paper publications-presumably their sponsors have learned how to value their contributions to the archive with measures such as citations.

...e-mail;

The archive operated in its first year solely as an e-mail based server.

...profession;

They called their ``working'' papers preprints, but that was not an optimistic name. Most papers are published, and published within 6 months of submission with very few revisions.

...peer-review;

But he wholeheartedly proposes that all scientific writing should be made electronically available, before peer review and possibly to cause peer review.

...publication.

Science journals commonly have page charges which were originally implemented to cover typesetting and copy editing costs, but now those charges are a general levy to all articles and not just those that are poor copy or in the wrong style.

...style.

Some math journals reject any submission that is not in AMS-TeX and most psychology and education journals reject submissions that are not in APA format.

...back    http://www.maths.warwick.ac.uk/gt/.

The EJC requires submission in AMSTeX, and upon final editorial acceptance of the manuscript, it is electronically published automatically without human intervention.

...sponsors,

The surprisingly low average cost of a paper in the E-Print Archive at Los Alamos is detailed below.

...paper.

The grant was for software development with little for ongoing maintenance. Ongoing costs per article might be one order of magnitude less, if not two.

...article.

Even though you might have the journal, it is faster and easier to search for an article with an online journal than to look through the shelves and each year's index of the paper journal.

...weeks.

Even a relatively common journal like the Southern Economic Journal has less than 3,000 subscribers (almost evenly split between institutional and personal ones), requiring many to experience the IIL delay.

...customers.

Individually and together they are working on a number of projects that add distribution of their material by networks. The NSF, NASA and ARPA are funding the $24 million Digital Libraries program at Carnegie Mellon, the University of California (Berkeley), the University of Michigan, the University of Illinois, the University of California (Santa Barbara), and Stanford University (NSF); Johns Hopkins Press along with its collaborators has its Muse Project (MUSE); Elsevier and nine U.S. universities have Tulip (TULIP); Academic Press has IDEAL (Press); and Pica is a joint project with Kluwer and Dutch libraries (Pica). Perhaps the largest move is a number of British publishers will make hundreds of their journals available on-line through U.K. libraries this year ((Hitchcock, Carr, and Hall 1996)).

...version.

Project MUSE at Johns Hopkins Press charges additionally for the electronic version though the subscription is institutional, so that any faculty of a subscribing library can access the electronic journal.

...exists,

Web browsers such as Netscape either with their standard html, or with a freely available Adobe Acrobat reader, are sufficient to display almost anything one may write. The CD-ROM version of the JEL uses Adobe Acrobat, and the reader is distributed on each CD-ROM. And Adobe Acrobat technology and the web are being used in a move in the U.K. by a number of U.K. publishers to put hundreds of journals on-line in U.K. libraries ((Hitchcock, Carr, and Hall 1996)).

...publications?;

The UK has legislated that in review of grants, electronic publications must be weighed equally with hard copy publications.

...year

The 20,000 may be conservative, since with 270 journals indexed in EconLIT and say 40 to 50 articles per journal, there are possibly 10,000 articles published each year.

...articles.

Papers in Informational Economics, .

...developement.

See http://www.sims.berkeley.edu/resources/collab/.

...papers.

However, it is important to realize the current methods of selection are inherently messy: waiting for working papers to arrive, talking to local and remote colleagues, checking the JEL, waiting for the latest journals to arrive in the mail, and visiting a library for the latest journals. It may be unrealistic, and perhaps unfair to expect the on-line world to be a panacea.

...documentation.''

citeNAnderson, p. 83.

...data,

Proprietary or confidential data raises some interesting replication issues. Should articles be accepted for publication which are based on data which no other researcher can access-ever? There are arguments on both sides of the issue.

...credit.

One significant exception is the Penn World Tables (Summers and Heston (1991)).

...``GAMS''

Designed for numerical analysts, it lists roughly 10,000 useful programs. Besides being a nice database in and of itself, it is a very nice example of the utility of such databases.

...libraries,

As a counterexample, the Duke University Library used to have two entirely separate card catalogs that varied depending upon the date of publication of the work.

...backward.

For example, the paper hep-th/9304128, has thirty clickable citations to it. It takes only a moment to click forward and backward through the literature to discover a great deal of literature written about ``Black Holes'', the subject of this paper. The value of this paper is in part measured by its thirty citations.

...testing.

In fact, work has begun on replacing one of the key components of the Internet, the Internet Protocol (IP), which transfers packets from one host to another. Test code is already being run using the new IP version 6.

...code.

The Internet Society, Internet Standards .

...journal

html stands for ``hypertext markup language.'' Web pages are written in it, so it describes their elements: text, graphics, and links to other web pages and various media. Unfortunately, it is not the ideal method of displaying technical material since html does not natively support many mathematical symbols. PDF (portable document format), invented by Adobe Corporation as part of their ``Acrobat'' product line, is designed as ``digital paper.'' Thus, it can accurately replicate any sort of table or mathematical expression, and even supports color, movies, and sound; it is ideal for technical papers. Adobe gives away ``readers'' for PDF that work closely with web browsers (they generate revenue from Acrobat by selling products that create PDF files). The programs that generate PDF files are quite flexible and work with many different programs.

...economists.

And they could charge whatever the additional cost is for other word processing formats. Authors who insist on using their own should bear that cost.

... journal.

The Student Economic Review, requires its student authors to write and submit with Microsoft Word using its own style files. The on-line and print versions then require little copy-editing.

...journals.

After all, the AER states that its policy is to only publish papers with data that is ``clearly and precisely documented and readily available.'' Putting data on-line would simply implement this policy.

Accessed times.