From loeb@grace.laas.fr Tue Sep 11 10:13:58 1990 Received: from bikini.cis.ufl.edu by fish.cis.ufl.edu (5.61ufl/4.12) id AA01289; Tue, 11 Sep 90 10:13:55 -0400 Received: from inria.inria.fr by bikini.cis.ufl.edu (5.61ufl/2.34) id AA22071; Tue, 11 Sep 90 10:13:17 -0400 Received: by inria.inria.fr (5.64+/89.0.8) via Fnet-EUnet id AA21933; Tue, 11 Sep 90 15:51:01 +0200 (MET) From: loeb@grace.laas.fr Received: from grace.laas.fr by laas.laas.fr, Tue, 11 Sep 90 15:32:50 +0200 Received: by grace.laas.fr, Tue, 11 Sep 90 15:39:32 +0200 Date: Tue, 11 Sep 90 15:39:32 +0200 Message-Id: <9009111339.AA05289@grace.laas.fr> To: fishwick@bikini.cis.ufl.edu Subject: Report6 Status: R No problem. Here is another copy of my report. \documentstyle{article} \begin{document} \begin{titlepage} \title{An Electronic Journal of Mathematics: Feasibility Report 5} \author{Daniel Loeb\\ loeb@geocub.greco-prog.fr\\ D\'epartement des Math\'ematiques et de l'Informatique\\ Universit\'e Bordeaux I\\ 33000 Bordeaux France} \end{titlepage} \maketitle \section{Introduction} Compared to the months and years that traditional publishing takes, the electronic network has a tremendous advantage. Considering that, as we will show, the ``journal \dots in its printed bound form has an existance which is unjustified except as a consequence of habit.'' \cite{JS} However, for the most part no real research is going on over the network. \cite{CR,WH} The reason is that up to now there is little motivation. There are several distinct and independent steps that can aided by use of the electronic medium \begin{enumerate} \item Research \item Submission \item Refereeing \item Distribution of Preprints and Abstracts \item Publication and Archiving\label{pub} \end{enumerate} As we will see, each of these steps can be carried out independently of the others. Publication (step 5) is perhaps the most ambitious, and Research (step 1) is already carried out electronically to some extent, so these steps should be delayed. The most likely implementation, is to start with electronic distribution Preprint and Abstracts of papers submitted elsewhere, and then after a year start accepting electronic submission, and referee them electronically. Only then would a program of electronic publication be feasible. Steps to help electronic research should be differed to the even more distant future. The most serious problem in getting started is that of acquiring the necessary motivation.\cite{SHA,ML,FR} How shall we motivate mathematicians ``to take the risk of expending effort to write and review for an electronic journal which has no established prestige-granting rating in the scientific community?'' \cite{TH} They are judged on their number of publications in journals, and it might not be clear to them if some new journal (especially an innovative electronic one) would count.\footnote{Several previous attempts \cite{BS,TH} at electronic journals have failed exactly because of this lack of confidence on the part of the authors.} It is for this reasoning, we suggest delaying part \ref{pub} of the project until the necessary prestige was developed since that is the stage in which prestige is the most critical. Also, it would help if the journal was started under the editorship of one or more famous mathematicians interested in the electronic network, and other celebrities were among the early contributors to the system. \cite{TEA,TH} However, as an electronic journal encorporates more and more of the features listed above, the rewards sharply increase. Improvements in each area of academic publishing will facilitate and draw attention to the need for improvements in other areas. Another general problem is competition. In order to benefit from the non-linear aspects mentioned in the paragraph above of this new medium, it is important to consolidate efforts. Therefore, there should probably be at most one electronic journal in each specialty of mathematics. There is a mathematics journal (the {\em Ulam Quarterly}) \cite{AO,WH} which encourages its contributors to submit their articles in electronic form. This is a significant first steps towards a completely electronic journal. Moreover, the American Mathematical Society sponsored a program called {\em Mathinfo} in order \begin{quote} ``to aid mathematicians in their research and teaching activity by giving them easy electronic access to each other (through specialized electronic bulletin boards and news services), to a restricted set of databases (including tables of contents for many journals, combined membership lists, and archived articles provided by users), and to public domain software available for specific machines and specific applications.'' \cite{AMS} \end{quote} If successful, this new program will embody steps 1 and 4 of our program. It is not yet anticipated that Mathinfo will include any original articles,\footnote{Although they do write that their conferencing facility could ``serve as a prototype for a refereed electronic journal.'' \cite{AMS}} so it would not qualify as an electronic journal on in its own right. Nevertheless, it is an extremely important step in the right direction. \section{Research} Mathematicians are scattered all over the world and work under a wide variety of conditions rangings from ``one-person departments in a two-year college to a center with a full compliment of research mathematicians and other scientists working together in a setting totally committed to research.''\footnote{Annual AMS-MAA Survey of New Doctorates, Notices of the AMS, {\bf 36}, 1989, pp. 1155--1160.} \cite{AMS} In order to encourage an exchange of information between these groups, many tactics are used such as visiting professorships and symposiums. However, none of them can compare in terms of speed, flexibility, and cost with the electronic network. Already some research is done over the network. The Usenet group Sci.Math and its subgroups do occasionally address unsolved problems; however the signal to noise ratio is fairly high, and little work gets done in the end. Much better results should be expect from {\em Mathinfo} when it gets started. During the recent fireworks, over cold-fusion and warm-superconductors, research was done on a day to day basis. Classical methods of communication - especially journals - can not cope with this pace. \cite{CHE} In the meantime, work is done privately by individual mathematicians who send letters to their collaborators via the network. Often \LaTeX{} files are sent back and forth, and an article is co-authored in this manner. Even when the letter is not a complete \LaTeX{} or \TeX{} documents, the \TeX{} notation is often adopted as a convenient means of expression. Any electronic journal should include various ``levels'' of publication. \begin{itemize} \item The lowest level would have no editorial control whatsoever. It would probably be the equivalent of Usenet's Sci.Math newsgroup. In addition, it could serve as a place to ``publish'' articles that have been encoded in order to guarantee priority of discovery before being released to the general public. \item The next level would be controlled by an ``editor.'' However, his role would be merely to control the level of noise, and eliminate any non-mathematical discussion. Hopefully, this group would be a conducive environment for doing research.\footnote{Such a level has been dubbed by some ``Scholarly Skywriting'' \cite{SHA} or ``The Paper Fair'' \cite{TH}.} \item A slightly higher level would consist of all papers currently being considered for review. These papers will eventually move down if they are rejected, or up if accepted. Each paper here will be associated with another file into which readers can place comments. These comments can be used by the author to improve his paper, and by the referees to help determine the importance of the paper. \cite{CHE} \item The highest levels would be the electronic equivalent of ordinary mathematical journals. It would consist of several levels each more selective than the one below it. Even in these levels, comments will be kept recorded for each article, and in addition extra space can be inserted as need be for corrections by the author, and for citations by other authors. \cite{CHE} \end{itemize} \section{Submission} A large part of the delay in printing articles is do to the fact that the publisher must retypeset every article he receives -- even if the submitted article was extremely well done. The most obvious electronic submission is that of sending a file typed by the professor or his secretary via electronic mail. However, there are several other possibilities. The file could be sent via modem over the telephone, or on a tape or diskette by the mail. Alternatively, the paper could be sent to the editor by the mail in printed form, and then either read in by an optical character recognition reader, or typed by the editor's secretary. In all cases, ``the author will of course subsequently check and revise his text'' \cite{BS}. Thus, ``the only people we will have to provide with terminals and the associated equipment are those without them who are members of the Editorial Board or the Editorial Staff.'' \cite{JS} In any case, the legal paper work required (copyright transfers {\em et cetera}) should be sent on paper via the Postal Service. Electronic submission is clearly compatible not only with electronic journals, but also with the tradition print medium where time can be saved via the use of electronic peer review, and by omitting any unnecessary retypesetting. \section{Refereeing} Peer review can be done electronically whenever the submission was handled electronically. Each referee can receive the paper electronically, and submit his comments to a mail-daemon which will automatically forward them to the author anonymously. Having achieved an immediate gain in speed, one might chose to do the refereeing in series.\footnote{Traditionally reviewing is done is parallel after a quick reading by the editor. Peer review is series is slower, but results in higher quality papers being published with the use of fewer referees.} Each referee (usually three) sees the paper in turn. That way, if the paper is unworthy of further consideration, then a minimal number of referees have been disturbed. Furthermore, the comments of each referee can be used to choose the most appropriate next referee. Moreover, if an electronic fileserver is being used, then the paper can made generally accessible before and during the reviewing process. The comments the paper receives during this period should greatly ease the job of any referee. \cite{CHE} The list of available referees can be managed by computer and automatically updated to indicate the speed and performance of each referee. The referees would be selected among a list of volunteers and contributors, since each qualified author should be expected to act as referee for the publication of someone else. This would hopefully avoid the potential problem of a lack of referees; as this is a serious problem even for ``real'' math journals and one of the principal reasons for delays in publication. \cite{AM} If the research was done electronically, the thread in question would also be sent to the referee in order to verify that the appropriate people get credit. If an electronic archive \cite{TEA} is being used, the referees would also be responsible for ascertaining the relevancy of any referenced articles.\footnote{Nowadays, it is very rare for referees to check references, so we are now suffering from bibliographies inflated to the point of inutility. With an electronic archive as proposed in \cite{TEA} such reference could be verified instantly.} One problem is that many of the most qualified referees may not be able or willing to print-out files in particular formats such as \LaTeX{} which we will see may be necessary for a quality publication. \cite{AM} Another problem is that occasionally it is difficult to send electronic mail from one site to another. The solution to both these problems is to acquire a team of volunteers over the world who can print out and transmit articles when necessary for the referees. These volunteers need not be mathematicians---they merely need be interested and eager. The {\em Mathinfo} project will have an advisory committee of six people appointed by the council of the American Mathematical Society. This committee will select moderators\footnote{A post equivalent to associate editor.} for various subjects. These moderators will be expected to recognize certain postings as being of very high quality. This alternate form of peer review taken to its logical conclusion would result in hierarchical pyramid structure to our journal/archive. \cite{Meg,RCR} The lowest levels would be unmoderated and would contain the virtual free-for-all of inquiries and commentaries now seen on Usenet. Slightly higher up would be groups that would work on particular problems under the control of a moderator who would try to limit the noise, by referring the less relevant comments to the level below. Even higher up in the pyramid, would be papers accepted by a panel of referees to be of publishable quality. Readers might be interested in a wide range of the choice material on the highest levels of the pyramid, but only that material on the lowest levels directly impacting on their current work. Papers could be updated as new results become available, and moved up the ladder when as a result. Similarly, papers would be moved up when modified according to the comments of the referees, or it was refereed to in many important publications or achieved a wider than anticipated readership. \cite{Meg,RCR} One common criticism\footnote{A. P. Melton, personal comments, 1973, \cite{JS}.} of the electronic journal, is that without cost dictated page limitations the system becomes a means of ``transferring garbage from one desk through a sewer to another desk.'' However, with the system described above, the reader can easily avoid seeing articles of less than desired quality by restricting himself to the ``upper'' levels of the filesystem. It is important to keep a tight control on the quality of papers on these levels, since ``scientific journal is not merely a way of dissemenating information, but is also a central element in the social control system of science. The peer review process serves to filer and validate scientific findings.'' Moreover, ``it serves as one of the central mechanisms in the allocation of scientific prestige and of resources for scientific research. Having one's work published in a prestigious journal terns to increase the resources subsequently available to a scientist for conducting further research.'' \cite{TH} \section{Distribution of Preprints and Abstracts} Papers submitted to other journals should be automatically distributed electronically in form of either preprints \cite{FR} or abstracts. This has already been done for {\em Psycholoquy}\footnote{Distributed by the Bitnet Psychology Newsletter via Listserv (psych@vm.tcs.tulane.edu) and Usenet to academic psychologists on the network.} \cite{SF,SHA} published by Stevan Harnad, and sponsored by the American Psychological Association. Similarly, the AMS's {\em Mathinfo} project intends to distribute at least the table of contents of most mathematical journals. The letter of acceptance should be sent to the editor as confirmation. In addition, it might be necessary to talk to the journal before republishing their paper (or its abstract). Moreover, it might be possible to create an arrangement with a journal to arrange for automatic distribution of the titles (and abstracts) of their newly accepted papers. \cite{WH,CR} The distribution could take place via a list Sci.Math.Abstracts. Each abstract would then be followed by a code which could be sent to a mail-daemon (or anonymous FTP) in order to receive the complete text of the article. \cite{CR} Already the publication of abstracts and the distribution by the author of preprints is used by many as a means of short-circuiting the delays involved in publication. Thus, often by the time something appears in a journal (or even a conference proceedings) it is outdated. With high-speed electronic distribution Usenet and Bitnet is the ideal medium for distribution of preprints. The only problem is that certain authors might prefer sending preprints only to certain colleagues in order to avoid plagiarists and cranks. \cite{FR} One major problem is the integration of text with figures.\cite{PF,JG} We can offer a number of options such as the \LaTeX{} picture mode or a digitalized bitmap, but in many cases (in particular graph theory) the soft-copy will have to forgo some of its figures. Instead a brief description of the figure and instructions on how to acquire a copy of the figure for a nominal fee could be included. Another problem concerns the choice of format. Clearly, a plain-text format filled with expressions such as \begin{center} \begin{verbatim} n ------- 2 \ i + i X i = ------ / 2 ------- i=1 \end{verbatim} \end{center} is virtually impossible to read. Other choices such as distribution in the form of \LaTeX{} files might turn away readers unfamiliar with that standard or with computers in general. \cite{TEA} In addition, \LaTeX{} files are difficult to read without making hardcopy. \cite{WH} Once the technology is perfected, the best solution will be to accept all sorts of submissions and then electronically distribute the material as a digitally stored image. University libraries would then be equipped with special graphics packages and high quality color laser printers with which to print the final product. \cite{CHE} \section{Publication and Archiving} Although this is the most ambitious part of the project, it is not without precedent. \begin{itemize} \item Recently there was a proposal \cite{CHE} for a computer network publication scheme for all academic research, not just mathematics providing for refereeing of papers, classification of them as to what kind of advance they made, permanent archiving of those judged by a council of referees to be worth saving, etc. \item The Mathematical Reviews database (MathSci) is available through DIALOG and BRS. A CD ROM (Silverplatter) version is also available. \cite{AMS} \item Moreover, the ``Systems Eletter'' edited by Sontag and Dickinson is an example of an electronic journal. It includes announcements of meetings, theses and reports giving addresses for further info. Papers in \LaTeX{} format are e-mailed by request from Rutgers. \cite{DE} \item ``Many trade and financial journals (e.g. {\em Harvard Business Review}) are now available on-line through Knight-Ridder's {\em Dialog} or Dow-Jones. \item ``Case reports of judicial decisions are available on-line through the LEXIS service.'' \cite{TEA} \item Since 1978, 25 state legislatures have jointly sponsored a system ``EIES: Legitech'' for exchange scientific information. \cite{TH} \item ``Several publications of the Association for Computing Machinery\footnote{Fox, E. (1989), The Coming Revolution in Interactive Digital Video, {\em Communications of the ACM}, {\bf 31}, 872--881.} Elsevier Science Publishers, and the American Chemical Society are now available on-line.'' \cite{TEA} \item Experimental (and abandoned) journals in the field of ``Computer Human Factors'' by the British Library \cite{BS} and the NSF \cite{TH}. \item Future projects are in progress sponsored by the European Economic Community \cite{CEC}, the United States \cite{OTA}, and the American Mathematical Society \cite{AMS}. \item ``Search and retrieval systems are \dots funded by the National Science Foundation for a variety of academic and scientific disciplines---psychology, linguistics, biology, chemistry, and so on.'' \cite{JS} \end{itemize} However, this is not an either-or situation. Electronic publication could be in additional to traditional publication. If used, traditional publication would be the major source of revenue for the project. In addition, many people will feel more secure knowing that there will be a hard copy of the article. During the difficult start-up period, authors may be reluctant to commit their articles to an unknown electronic journal. Their worries may be put at ease by allowing them to simultaneously publish in a traditional journal. If this journal is not a ``sister journal'' to our electronic one, then certain rules should be followed to avoid abuse: \begin{itemize} \item Print publication should occur after electronic publication. \item The print journal should be informed of the parallel publication. \item The article in the print journal should include a footnote to the title such as this one on \cite{BS}: ``This article has been refereed, accepted, and archived (in electronic form only) in the British Library R{\&}DD experimental journal ``Computer Human Factors.'' Because of the interest widely expressed in this British Library-sponsored research, the editors of {\em The Computer Journal}, {\em Ergonomics}, and the {\em Journal of the American Society for Information Science} have kindly accepted this article for parallel publication.'' \item Should the electronic journal refuse to ``publish'' the article within a reasonable period of time, the author is free to publish elsewhere without any of the above limitations. \end{itemize} There is a clear economic justification for an electronic journal. Traditional journals recover start-up costs in five or six years\footnote{Woodworth, D. (1979), {\em Financing Serials from the Producer to User,} Oxford: Basil Blackwell's.}, but while combined electronic-print journal might conceivably take longer, one most note that non-human computing costs halve every few years, while printed journal costs increase faster than inflation. \cite{TEA,JS,OTA} Meanwhile, ``potentially valuable papers are not published or are severly delayed in publication by an arbitrary peer review process which must reject the majority of papers submitted to prestigious journals because they cannot afford to publish them. At the same time, readers pay for journals in which the majority of articles are not of interest to them.'' \cite{TH} With existing methods, the average university library spends over 60\% of its budget on academic journals leaving very little for books. \cite{SO} Moreover, this figure can be expected to rise, since subscription prices increase generally 10\% per year while the number of journals doubles every 30 years.\footnote{\cite{CHE,TH}, and Lambert, J. (1985), {\em Scientific and Technical Journals,} London: Clive Bingley.} Even the size of these journals has also gone up, so the actual shelf space used up by journals (at nearly \$100 per square foot) is rising at an ever increasing rate. In comparison, electronic media is fairly inexpensive, and can be updated, corrected, or replaced without discarding the actual media which itself maybe be located at a central site rather than at each university. \cite{SO} With time, the electronic journal's position improves economically with respect to that of the traditional bound journal. In addition, there is a sizable economy of scale. ``If we try to make electronic a small journal of about 900 pages per year with no more than 2,500 subscribers, both private and institutional, the economics of the situation indicate a crossover point in the middle or late 1990's. However, if we set up a system to handle the entire body of English language scientific publications at once, the crossover point has already been reached.'' \cite[written in 1976]{JS} In addition, one must bear in mind that the purely economic features do not reflect the advantages offered by electronic journal: saved time, and increased quality. When these factors are taken into account, the crossover point was around 1971. \cite{JS} Some conceivable objections to virtual journals, is that they would be tie the researcher to his computer, and that there are no ``margins'' in which to take notes. \cite{SHA,TEA,JS} However, all of these problems can be avoided by special software, or by printing out the article when necessary. ``Most of the expressed concern stems from familiarization and habituation. Perhaps users' acceptance will have to await the retirement of the older scientists.'' \cite{JS} If there is not a parallel print-journal, then one problem is leaving a permanent record, since journal are supposed to archive their field and it is for that reason that scholars have a duty to put any important work there. Most electronic newsletters are not well archived, however one solution is to have all (or the best) published articles archived and referred to in a standard way. \cite{FR} In fact, William Gardner \cite{TEA} has proposed that the principle advantage of electronic publishing is not the electronic journals themselves but the electronic archives they create. He has proposed an advanced hypertext system that would distribute articles to users of all levels of computer expertise. Such a system would truly be aware of the interconnections between various articles. \cite{DP} It would be more useful than a traditional journal since users would only pay for the articles which interested in them. Probably the journal would specialize in one branch of mathematics, for if not be far too many postings to read \cite{CR} of course the subscribers (on Usenet or Listserv or through a private daemon) would be able to subscribe to a particular topic or articles containing particular keywords and lower the load to a reasonable level. Under the discussion of research over the network, we proposed that any journal should be divided into various levels according the prestige awarded to articles placed in those areas. In addition, other areas include letters to the editors, puzzle columns, editorials, book reviews, advertisements, minutes of recent mathematical conference, notices of upcoming conferences, help wanted, announcements of new system features, etc. \cite{TH} Most probably the journal would not be published according to any fixed schedule unless it was felt that deadlines were helpful for psychological reasons. In addition, certain articles can be distributed along with code (written for example in Mathematica) that could recreate the figures used in the article. By changing the programs and rerunning these programs, reader with access to the necessary hardware and software could understand the material more easily. \cite{JoCl} This packaging technique can also be used to include audio annotation, animation, and live video with the article when desired. \cite{DP} \begin{thebibliography}{99} \begin{center} Articles. \end{center} \bibitem{AMS} American Mathematical Society, ``Mathinfo Project Summary,'' NSF Grant Proposal, Principal Staff: William B. Woolf wbw@math.ams.com, David L. Rodgers, Donovan H. Van Osdol. \bibitem{LeB} * - Leslie Burkholder, ``Ejournals,'' {\em Notices of the AMS}, 37, 5, May/June 1990, lb0q+@andrew.cmu.edu \bibitem{JoCl}Jon Claerbout, jon@hanauma.stanford.edu, personal communication. \bibitem{CEC} * - Commission of the European Communities, {\em Electronic Publishing: The New Way to Communicate,} London: Kogan Page, Ltd. 1987. \bibitem{DE}Dave Elliott, delliott@cec1.wustl.edu, Washington University, St. Louis, Personal Communication. \bibitem{PF}Paul Fishwick, fishwick@fish.cis.ufl.edu, UF CIS Department, Personal Communication. \bibitem{SF}Stephen Furner, smf@hfnet.british-telecom.co.uk, personal communication. \bibitem{TEA} William Gardner wpg@virginia.edu, ``The Electronic Archive: Scientific Publishing for the 90's,'' Electronic Publication: Psycholoquy, Sunday, 1 April 1990, 17:15:22 EDT, and Traditional Publication: Psychological Science, To appear. \bibitem{JG}John George, george@altair, personal communication. \bibitem{SHA} Steve Harnad harnad@princeton.edu, ``On the Scholarly and Educational Potential of Multiple EMail Networks,'' Commed, and personal communication. \bibitem{WH}William Hammond, wfh58@leah.Albany.Edu, personal communication. \bibitem{ML}Michael Larsen, larsen@ginger.Princeton.EDU, personal communication. \bibitem{Meg} Ganesh Mani, ``Electronic Archival of Scientific Journals: The Megajournal ganesh@cs.wisc.edu, Psycholoquy, Sunday, 22 April 1990, 19:56:03 EDT. \bibitem{AM}Andrew P. Mullhaupt, amull@Morgan.COM, Morgan Stanley and Co., personal communication. \bibitem{SO}Steven L. Obegard, n8243274@unicorn.WWU.EDU, Western Washington University, Personal Communication. \bibitem{AO}Andrew Odlyzko, amo@research.att.com, Personal Communication. \bibitem{OTA} * - Office of Technology Assessment, {\em Informing the Nation: Federal Information Dissemination in an Electronic Age}, United States Congress, 1988. \bibitem{DP} Dick Phillips rip@lanl.gov, ``Interactive SIGGRAPH Proceedings: A New Form of Publications,'' Computer Graphics, {\bf 24}, 1, January 1990, p. 59--61. \bibitem{CR}Carl Riehm, riehm@maccs.dcss.mcmaster.ca, Personal Communication. \bibitem{CHE} Sharon J. Rogers. and C. S. Hurt. ``How Scholarly Communication Should Work in the 21st Century,'' The Chronicle of Higher Education, October 18, 1989, p. A56. \bibitem{RCR} * - R. C. Roistacher, ``The Virtual Journal.'' {\em Computer Networks}. 2: 18--74; 1978. \bibitem{FR}Flavio Rose, flavio\%mitisft@sun.com, Personal Communication. \bibitem{BS} B. Schackel, The BLEND System: Programme for the Study of Some ``Electronic Journals,'' {\em Journal of the American Society for Information Science,}'' {\bf 34}, pp. 22--30, 1983. \bibitem{AC} * - L. Seiler, ``The Future of Scholarly Journals,'' Academic Computing, March 14, 1989. \bibitem{JS} J. Senders, An on-line Scientific Journal, {\em The Information Scientist}, {\bf 11}, 3--9, 1977. \bibitem{SSMNGM} * - R. Sheridan, J. Senders, N. Moray, J. Stoklosa, J. Guillaume, D. Makepeace, ``Experimentation with a Multidisciplinary Teleconference and Electronic Journal on Mental Workload,'' Unpublished report to National Science Foundation (Division of Science Information Access Improvement), 320. pp; available from Prof. T. B. Sheridan, Room 1-110, MIT, Cambridge, MA 02139), 1981. \bibitem{AS} * - A. Singleton, The Electronic Journal and Its Relatives, {\em Scholarly Publishing}, {\bf 12}, 3, pp. 3--18, 1981. \bibitem{OS} * - O. Standera, {\em The Electronic Era of Publishing: An Overview of Concepts, Technologies, and Methods}, New York: Elsevier, 1987. \bibitem{TH} M. Turoff and S. Hiltz, The Electronic Journal: A Progress Report, {\em Journal of the American Society for Informational Science}, {\bf 33}, pp. 195--202, 1982. \end{thebibliography} \end{document} From loeb@grace.laas.fr Tue Sep 11 10:34:16 1990 Received: from inria.inria.fr by fish.cis.ufl.edu (5.61ufl/4.12) id AA01334; Tue, 11 Sep 90 10:34:02 -0400 Received: by inria.inria.fr (5.64+/89.0.8) via Fnet-EUnet id AA23456; Tue, 11 Sep 90 16:34:45 +0200 (MET) From: loeb@grace.laas.fr Received: from grace.laas.fr by laas.laas.fr, Tue, 11 Sep 90 16:22:17 +0200 Received: by grace.laas.fr, Tue, 11 Sep 90 16:28:58 +0200 Date: Tue, 11 Sep 90 16:28:58 +0200 Message-Id: <9009111428.AA05364@grace.laas.fr> To: fishwick@fish.cis.ufl.edu In-Reply-To: "Paul Fishwick"'s message of Tue, 11 Sep 90 10:25:59 -0400 <9009111425.AA01326@fish.cis.ufl.edu> Subject: Report6 Status: R You are welcome. No problem. Yours, Daniel Loeb ------------------ TOULOUSE BORDEAUX EMAIL loeb@laas.laas.fr loeb@geocub.greco-prog.fr (or else try loeb@frbdx11.bitnet) MAIL (H)13, boulevard des Fontanelles 150, cours Victor-Hugo; Appt B21 31240 L'Union France 33000 Bordeaux France MAIL (W)Group O. L. C. Departement des Maths et d'Info L. A. A. S. Universite de Bordeaux I 7, avenue du Colonel Roche 351 cours de la Liberation 31077 Toulouse Cedex France 33405 Talence Cedex France PHONE(H)xnational+33 61 09 68 15 xnational+33 56 31 48 26 (W)xnational+33 61 33 63 17 xnational+33 56 84 60 88 FAX xnational+33 61 55 35 77 xnational+33 56 80 08 37