i Current Comments@ EUGENE GARFIELD INSTITUTE FOR SCIENTIFIC lNFORMATIOtW 3501 MARKETST PHIWDELPHIA PA 19104

Reaftlrming the Central Role of the Scientific Literature: Joshua Lederberg Discusses the Print and Electronic Media

Number 1 January 4, 1993

Joshua Lederberg, Rockefeller Univer- sity, New York, has been a frequent con- tributor to Current Contents w (CC@) on a wide variety of topics. We have reprinted his papers on scientific biography,l post- mature scientific discovery,z medical sci- ence and infectious disease, j scientific progress,i and other subjects. CC readers probably know him best as a Nobel laure- ate geneticist and past president of Rocke- feller. But as a friend of over 30 years, I also know Josh has a strong and long-stand- ing interest in the entire process of scien- tific discovery and creativity, and how the fruits of that process—the research pape-- make their way into print. He is also fasci- nated by technologies designed to manage and access the literature. Joshua .Qderberg In fact, it was with Josh’s early encour- . agement and support that the Science Cila- deputy editor, west, JAMA—Journul of/he fiorr Index o (SC1’) was able to get off the American Medical Association. In addition ground. He wrote to me in 1958 to ask to Dr. Lederberg, the participants included whatever became of the idea for a citation six other Nobel laureates: David Baltimore index for sciences His suggestion that I (Medicine, 1975); Sir John Kendrew apply for a National Institutes of Health (Chemistry, 1962); Leon Lederman (Phys- grant eventually led to the Genetics Cifa- ics, 1988); Bernard Lown (Peace, 1985); fion Index expenment.G Over the years, Josh Melvin Schwartz (Physics, 1988); and has also contributed to ISI@’s success as a James Watson (Medicine, 1962). member of its Board of Directors and the I chaired a symposium on technology advisor-y board of the SC1. that covered hardware, software, networks, In October 1991, I participated with Josh and satellites as they relate to global scien- in a conference at Woods Hole, Massachu- tific communication. Some of the discus- setts, on “Science Editing in the Age of sants included Roald Sagdeev, Soviet Space Global Communication.” It was sponsored Program, and Danny Hillis, Thinking Ma- by the International Federation of Science chines Inc., of parallel computer fame. My Editors, whose founder and president is introductory comments provided a personal Miriam Balaban, International Science Ser- historical view on the field of information vices, Rehovoth, Israel. The conference was retrieval and some of the technologies— organized by Kenneth Warren, Biofield both visionary and practical—that enable Inc., New York, and Drummond Rennie, us today to both recover and discover in-

206 formation with high recall, precision, and concern to me. In fact, NSFNet has recently relevance. launched an experiment to mount issues of Josh gave a keynote presentation entitled The ScienriSt @, the newspaper for the ‘ci- “Communication as the Root of Scientific ence professional founded seven years agog Progress.”T His insights into the problem Although graphics and photos are not avail- of scientific communication—from the able with current technology, at least the reader’s perspective-struck me as being scientific community will have access to very perceptive. At the time, he had re- the full text of each biweekly issue. Addi- cent] y retired as president of Rockefeller tional issues of The Scientist will be University and was making the transition mounted and available as published. You back to full-time lab research. After 12 can access the file by typing “ftp years as president, he was in a sense re- nnsc,nsf.net” and use the Iogin “anony- turning to the lab for the “first time” and mous” and then your usemame@bitnet ad- faced the same practical problem most dress as the password. Then type “cd the- bench scientists struggle with daily. That scientist” and “get the-scientist-92 1109” is, how to keep up with tbe literature or, or “get the-scientist-9211 23’’-the numbers more precisely, spend one’s time most ef- correspond to year, month, and date of pub- fectively by locating and retrieving the lication. By simply typing “get*” you will “must read” papers. see a listing of all NSFNet files in The While Josh holds the printed journal in Scientist directory. special reverence as a public record, At the moment, Balaban is preparing the arch~ve, and forum of scientific communi- proceedings of all the conference presenta- cation, he is equally intrigued by the new tions. She hopes to have it available by possibilities posed by electronic publish- July 1993. In the meantime, readers inter- ing.g Provided they are as rigorously peer- ested in obtaining copies of the proceed- reviewed as their print counterparts, he sees ings should contact: Gerda Helbig, Secre- the value of directly iinking an original tary/Treasurer, ISSZ Kiebitzrain 84, paper in an electronic journal with the many D-3000 Hannover51, Germany (Tel.: 49- commentaries that may be made on it. He 511 -604-5956; Fax: 49-5 11-604-4507) or feels this “electronic dialectic” is how the Susan Eastwood-Berry, Neurosurgery Edi- scientific process works at its best. After torial Office, University of California, reading the text of Dr. Lederberg’s talk, I 1360 Ninth Avenue, Suite 210, San Fran- felt that it deserved much wider dissemi- cisco, CA 94122 (Tel.: 415-476-3272; nation by being reprinted below. Fax: 415-476-970 1). In my own way, electronic networks and bulletin boards have also been of special @ 1s1 1993

REFE NCES 1. Garffeld E. Scientific. biography-contemporary reflections on The excitement mdfascinadon ofscierrce. Essays of an information scientist: journalology, Key Words Plus, and other essays. Philadelphia: 1S1Press, 1991. Vol. 13. p. 29-37. 2------Postmature scientific discovery and the sexual recombination of bacteria-the shared perspectives of a scientist and a sociologist. Ibid. VOL 12. p. 16-23. 3------AIDS and beyond: living with the inevitability of virulent diseaac. Ibid. p. 145-7. 4------Joshua Lederbarg on tbe namre of scientific prngress. [bid. p. 335-44. 5------Citation indexes for science. Science 122:108-11, 1955. (Reprinted in: Ibid., 1984. Vol. 6. p. 468-71.) 6. Sher I H & Garfteld E. llre Genetics Citation Index experiment. Proceedings of the A?rwrican Documentation Instiwe, 26th Aonual Meeting. Chicago, IL: American Documentation Institute, 1963. p. 63-4. 7. Lederberg J. Communication as the rwt of scientific pragress. presentation at the Sixth International Conference, hrtemationrd Federation of Science Hltors, Science editing in the age of global communication. (Bataban M & Warren K, eds.) Wcmds Hole, Maasachuaetta. October 13-17, 1991. 8------Digital communications and the cnnduct of science: tbe new literacy. Proc. IEEE 66:1314-9, 1978. 9. Garffeld E. Introducing The Scierrrisr atlast, a newspaper for the science professional. Op. cit., 1988. Vol. 9. P. 222-5.

207 Communication as the Root of Scientific Progress Joshua Lederberg Rockefeller University New York, NY 10021-6399

Lecture presented at the Sixth International Conference “Science Editing in the Age of Global Communication” International Federation of Science Editors Woods Hole, Massachusetts, October 16, 1991

Introduction for what is published under a given person’s I am very interested in scientific infor- name. Just look at the daily headlines. It is mation. I don’t do very much editorial work the essential ingredient to make scientific these days; I’m back working in the labo- work responsible in the sense that one can- ratory after a lapse of 12 years and that has not readily retreat from assertions that have kept me very busy trying to reacquaint my- been signed, delivered to the printer, and self with the literature of my own field. So made available to thousands. I will offer the perspective of a scientific These publicly asserted claims also play reader. Now some people tell me that’s a an extremely important role in the alloca- vanishing species ! For anyone to say that, tion of resources, the ability of different even with some sense of irony, is an atroc- scientists to survive in the competition with ity. other legitimate claims for expenditures, for One of my main functions with my own support of laboratories, for positions at the laboratory group is that I try to be its prin- institutions, for space in the journals, for cipal reader. If something goes on in the the attraction of students and collaborators. world outside and none of us has heard All these rest on those claims, the evidence about it for two or three weeks, I‘m the for which in the end is in the public record. one who feels responsible. I want to be Both author and audience benefit from the alert to events that might have a very im- successful assertion of those claims: espe- portant bearing on the way we think about cially credibility, that one doesn’t have to our own research, our planning, of the data spend an inordinate amount of time reex- coming in, of the sources of error. amining every detail of an individual’s out- put if that person has established credibil- ity through prior publication and exposure. The Literature as Public Archive and Publication also results in a repository, Open Forum constructing the tradition of science. Up to Let me begin with a few truisms, just to this point it can hardly be anonymous in be sure that we are operating on a common order to perform the functions that I have ground of reverence for the publication pro- just indicated. But as time goes by, we have cess. Publication is, to start with, just that! the reassimilation of the content of scien- Public-ation. It converts private to public tific work, and as it settles in and survives knowledge, in the service of registering a the criticism that it should have had at its private claim of original authorship-in sci- early stages of the process, it becomes the ence, of discovery. Above all, the act of common tradition, the unquestioned shared publication is an inscription under oath, a wisdom-often becoming anonymous by testimony. It is accepted as valid until firm obliteration. evidence to the contrary; and there is an The literature is also a forum. It’s a gladi- extremely high standard of accountability atorial arena for competing claims, resolv-

208 ing discrepancies in data or interpretation. leisure today to read a book from cover to There used to be oral duels, and we revel cover. A few biographies command atten- in stories like Pasteur’s confrontation with tion. I just finished a proof copy of Carl Pouchet that finally put spontaneous gen- Djerassi’s life story: The Pill, Pygmy eration to rest in 1864. Today, our battles Chimps, and Degas’ Horse; another of that are more often fought out in print, which is genre was Fran~ois Jacob’s revelation of indeed appropriate because the testimony the development of his scientific work: The then becomes available to the universe, not Statue W’irhin.These are obviously not very simply to the immediate onlookers. contributory to the details on how to do my Despite the opportunity for very broad next experiment, but they tell me a lot about dissemination, there is the paradox, never- the scientific personality, providing object theless, that broadcast restricts individuals’ lessons and models for emulation. access to feedback. The publication sys- Rarely, I do see a work that compels tem, at least in principle, should allow a total ingestion—for example, Physiology of dialectic to appear in more symmetrical the Bacterial Cell by Neidhardt, Ingraham, terms where anyone with something pur- and Schaechter. This is such a magnificent poseful to say has a way to get into the synthesis at a fairly elementary level of ex- system. position that I really marveled at the delib- If the literature is a forum, it is also a eration and distillation that went into the rumen, a place for the digestion and as- telling. Wonderful books like that are rare. similation of the variety of inputs where In printed form they surely will be the sur- scientific claims go through a period of sea- vivors of any electronic revolution. soning, modification, modulation. Even the At an intermediary level of indispens- truths look different 5 or 10 years later re- ability as books in print format are the An- gardless of explicit criticisms. We can ex- nual Reviews. They are reference works pect a process of reinterpretation, a post- for whatever you have to look up; but they historical reexamination of the meaning of also give a chance to browse through an their terms. enormous literature with some coherence. And now I only need to remind you of Compare an Annuul Reviews of Genetics the term “imprimatur” (a wonderful meta- with current issues of the journal Genetics. phor): the imprinted witness that, an article Even if I had the time to read every articIe, having appeared in a refereed journal, it I wouldn’t have the background to be able had survived a critical process, a conspiracy to place each one of them in the appropri- if you like, of the editors and the publish- ate context of what comes through. And I ers and the referees—that something has regard this as my home discipline! People appeared which is worthy of the shared in- will spend varying amounts of their time terest and precious attention of the com- and energy as well in trying to understand munit y. what is going on in science beyond the window of their own specific work in their research and teaching. Keeping Up with the Literature There are about a dozen journals that I May I tell you what I do as a reading subscribe to and maybe seven or eight of scientist today? Reading the scientific lit- them that I do scan from cover to cover erature has been my primary vocation for Nature, Science, Proceedings of the National 50 years. Academy of Sciences of the United States of Books play a diminishing role. Today America—those are the very general ones. they are mostly for targeted reference. In The Journal of Bacteriology, Microbiologi- the scientific domain, we rarely have the cal Reviews, Genetics, Biochemistry. I pick

209 up a hot paper now and then from The out reasonably well, but have to be embel- Scientist’, and look at The Sciences, New lished from time to time. You discover new Scientist, American Scientist, and Scientific keys, other notations that authors insist on American for general scientific culture. in changing fads and idiosyncrasies of lan- That is a textual sampling, not immer- guage. I can warrant that my profiles re- sion. You couldn’t read every article in a cover on a current basis about 90~o of what critical and detailed fashion in just the jour- I have read or would want to read. God nals I have listed within the number of hours help me if I lose my notes on the rest! that there are in the week. What you can Then, how to keep up with what’s clos- do within a couple of hours a day is to scan est to my immediate specialty? Acquiring that range of material and try to pick out a couple or three papers a day is not hard. those things that might be of interest. To And even with a fairly detailed critical ex- follow the structure of argument just in amination down to checking the points on one’s own specialty, you must go to the the graphs and so on, reading them as they detail of trying to check the numbers on come in is entirely doable. My problem is the graphs and see if they match the au- the arithmetic of accumulation. After a de- thors’ assertions-an arduous task. cade, I’ve got about 10,000 papers that I We are well served by those kinds of have got to keep track of the texts and my journals in terms of maintaining a general marginal notes and so on. currency about what is going on in the field. And here my system is absolutely bro- And they match very well the energy and ken down ! A technological fix is on the interest and intellectual acuity that our sci- way: document scanners that can store page entific readers are able to put into the pro- images and digitize scripts on searchable cess. I see no occasion for those to be al- media. One or a few CD-ROMs will take tered. Most scientists are very grateful for care of the storage. But what a lot of bother them: what thousands of scientists will for information, yes full text, that I should share as common currency, to carry in their be able to acquire electronically in the first briefcase and read on the airplanes and the place. The more so for specialty journals commuter rides, with all the convenience and references to be searched on demand. of the present print format. Within a given specialty there are usu- ally one or two journals that specialists must Selective Retrieval and Managing a see. There may be only a couple of hun- Personal Library dred people who have such a level of inter- est that they will look at every article. There My main problem is: How do you reac- are the journals of broad appeal, and then a quire, retrace that intellectual traffic? What very flat distribution of the other sources. do you do with all of your marginal notes For my part, an additional 30 articles a and how do you synthesize a coherent sys- month—perhaps half of them come from tem of what you’ ve read? Well to try to about 15 journals; you can probably ex- deal with this on a current basis, I have trapolate with Bradford’s law to the rest. Gene Gartleld’s wonderful products. I get Ninety percent will come from about 35, the weekly Current Contents on Diskettem and then there is a gradual asymptote out with all of its embellishments. to the vanishing returns from the total cov- I eagerly await the five or six diskettes erage that the system is going to offer. that have to be loaded, every week, and Every now and then an article does pop sometimes am impatient about how long it up from an obscure place whence you had takes to load them and get going with that no systematic way to recover it; but in ret- week’s literature. My stored profiles work rospect it was really quite important.

210 So each of us faces the task of selective being drowned, I have a great deficit of retrieval from a cosmic domain of stuff that specific and detailed knowledge of exactly every other eager beaver in the world has what happens in such and such a system been busily putting into the repository. Our with such and such reagents, and so forth. present technology enables an approxima- Our systems for acquisition of that kind tion with reasonable confidence. Keeping of material are not perfect. But they are track of what you have accumulated on getting a lot better. And with devices like pieces of paper is the frustration. That’s keyword searching, articles related by bib- not your bedside reading, well served by liographic coupling, full abstract searching, the print on paper version. The next step, which is just about what the technology to integrate that into your own private li- does offer today, 1can feel reasonably con- brary of useful knowledge, is simply not fident that explicit matters of factual de- achievable with last year’s technology. tail—--whether somebody has done that par- ticular experiment-can be retrieved, but General Information Flood vs. Specific often only with a lot of effort. Information Drought Much more difficult, has anybody else had a good idea that would be pertinent to The fact is that scientific literature in- my search? Those keys are so much more herently has grown beyond the scope of difficult to catalogue. Often it takes a great any hundred people to have understood it creative act to recognize that a concept de- and gone into it in some depth. It is built in veloped in one context really is pertinent to the growth of knowledge that past im- to another. So there will never be a guaran- provements in communication and storage tee that those can all be acquired. But there aren’t going to alter. is at least the hope of finding it in that What are the consequences? For one literature, and it is a very important hope thing, the problematic of assessing the lit- to try to preserve. erature reinforces all the other drives to specialization. The ambitions of scientists Taking the Literature Seriously have changed, to focus on ever narrower targets. It’s just too much hard work to There are different adaptations to that master an interdisciplinary area on top of flood, and more and more we see what I all the other institutional obstacles. Never can only describe as a scandal-that the mind the intellectual conceptual problems. scientific literature is not always taken se- Never mind the problem of getting funding riously any more. In polls of scientists, or moral and fiscal support, just to get hold many will say that the primary source of of the necessary expertise and information! their information about scientific work in But that impediment is in principle reme- their field is not the published literature. diable. It’s by word of mouth, it’s by telephone At the same time, are we drowning in networks, by attendance at meetings, and information, inundated by the numbers of so on. People have got to do what they’ve journals? You know, when you come to got to do. any specific issue, when there is some im- But I find those kinds of sources so un- portant special fact that you would like to reliable ! I feel very uncomfortable when know all about, the shoe is very often on the only place that I have heard something the other foot. My usual experience in ask- is by word of mouth. If I can’t pin it down, ing a new question: the odds are that the if I can’t hold its source accountable by exquisite detail needed to take the next step saying it was in a published item, I can’t has just never beerr done. So here, far from look at it in detail, ruminate about it, think

211 through what second order reactions I drop a journal, you would think that every would have. I don ‘t know whether my col- professor was reading every issue of every leagues share that. They may feel that they journal in the library! For the operational don’t have any alternative except to pick procedures by which libraries cart make sen- up what’s in the rumor mill. But I think sible decisions about acquisition priorities, great mistakes can be promulgated in that they could get any number of technical aids fashion. on that point. * But it does put them in a The telephone is a wonderful instrument, very tough spot. Besides the budgetary But when I try to use it to get information, crunch, the libraries are also running out of people who have what I am looking for are space. The older stuff is deteriorating any- all pretty busy. I hate to impose on their how! Maybe ink on paper was not a totally time, and if I do, there is usually a round of bad idea, for that reason alone, provided telephone tag of three or four attempts to one clean copy remains avai Iable. Unfortu- catch somebody before you actually do get nately, things don’t always work out that hold of them for the information. If it’s a way. reference, I am delighted. If it’s an attribu- tion, it cannot be pinned down more defi- Editorial Review: The Essential Value- nitely than, “You know, this is what I Added of Print and Electronic Journals think.” I don’t feel like I have made a great advance over what I have had before. One direction things could take if we Not taking literature seriously reinforces don’t reform the system is that invisible the trend that libraries in desperation are co[leges will take over as the principal but canceling subscriptions to journals that they unreliable routes of communication, Archi- don’t see being very much read locally. val copies of material will eventually be And it doesn’t make any sense to have a sent in to the repository. But there will be a local copy of a serial where perhaps one in limbo of material that doesn’t know if it is a hundred titles will ever be examined by going to go to hell or heaven for four or anybody in that institution. Some of these five years, while it is still cooking and un- journals, de facto, are approaching the point accountably available, on a basis far from where they might as well only print one equitable. So, in due course there has to be copy, send it to the National Library of a wholehearted exploitation of the new tech- Medicine or some other repository, and let nologies and I don’t have to plead for it. it redistribute reprints by interlibrary loan. It’s happening because electronic networks The economics obviously are insupportable. are becoming more and more available, de The fundamental problem is trying to foist facto, to people working in a variety of an inappropriate number of vehicles on an fields. A couple of dozen of them now op- outmoded mechanism for the purpose of erate with a routine exchange of preprints. dissemination. So that would fall of its own The central problem facing the journal has weight. You can see what I’m leading to: been a radical change in the economics and go from 1,000 to 1 to Oprint copies. technology of printing, without an adequate Meantime, the libraries are in a great di- recognition of the essential value-added in lemma trying to figure out exactly what to the journal process. do. They get a fight from the faculty— From Gutenberg’s time until mechanized what a librarian hears when they want to and computerized composition, that was

* For The Rockefeller University library, I used the Science Cifaion /mfe& to preparean index to the frequency with which different journals were cited in their published papers. That would be a bad algo- rithm for acquisition of books and review journals.

212 providing the capitrd and the entrepreneur- worth capturing the priority of my atten- ship and the organization to facilitate a pro- tion. cess whereby an expensive and precious printed article was the product. It was char- A For-Profit and Nonprofit Alliance for acterized by rather high capital investment Electronic Publishing in the initial composition of any material. Once it was composed, there was a rather I think it will be the societies that pro- low variable cost for further dissemination. vide the most likely framework for the or- We had a market system for determining ganization of those functions, It won’t make what was worthy of that degree of capital any money to start with. But the econom- investment, Well today the capital invest- ics and the technology will converge with ment on the printing technology is almost the social necessities for this kind of im- zero. provement. Technically, we don’t need The important value-added is the edito- much more than what we now have. There rial process including issues of selection, are a few problems about transmitting then of editorial work and improvement. graphics and formatting manuscripts. Some And that very precious imprimatur. When standards have to be established and some something comes out in a journal of high minor fixes, especially on the graphics. But repute (to make a circular argument), that’s we are basically right there. a journal worth my time and worth my at- Machines with gigabyte storage and ever tention. If it is just thrown up in the air ;maller 25-meg processors are very routine without having undergone that kind of edi- oday. You will find them by the hundreds torial review, it will not have been refined n the laboratories and the libraries and so in terms of both the presentation, and per- m, with a doubling of capabilities per unit haps even substance of the argument. And :osts every couple of years. So in 10 years, it won’t have the imprimatur of other oday’s “super-computer” will be available people, whose judgment I trust, that it’s :ertainly in every institution, and to a large worth reading and can be relied upon for Iegree in every laboratory. Communica- accountability. tion links won’ t grow quite as fast as that, Whether the article then gets into p-in? Jut if you consider the bandwidth of a pack- is almost an irrelevancy at this point. Any tge of CD ROMs, you have a variety of of a variety of media of communication echnologies for all the communication we could follow on that editorial process. What ~eed. So those are not limiting factors ei- we need to see more than has happened so her. lltey are not very expensive. far is the marriage-of that editorial role, The machinery, the social framework the on the one hand, with a production role Iecisions involved, the wetware, the distri- that uses the electronic technologies rather mtion channels, the marketing, and so on, than the print, on the other. eally are all that stand in the way. There And that’s where the spontaneous bulle- ue not the same kinds of profit incentives tin boards don’t quite make it. They quickly hat drive paper publishing; so I think the get tilled up with obscenities, literal and lot-for-profit institutions will start taking otherwise, for lack of that sort of control. I wer. Perhaps the for-profit publishing don’t mind the obscenities as long as I don’t louses wilI provide the essential technical have to plow through them. But I’d like a ervices because they can have the economy truth-in-advertising framework that tells me, If scale, the organization, the hardware, as I say, what’s worth reading. I’d like to nd so on, and then contract that out to the know that X, Y, or Z editorial committee ocieties for providing the other elements had been established as a guide for what is ,f the equation.

213 This partnership could be a very produc- Global Access to the Literature tive one for the entire scientific commu- Let me make one further comment about nity. One feature of this kind of a system global access: something very dear to my to which we have only a crude approxima- heart. There was a remark in my letter of tion today, is feedback, dialectic. It invitation: “You may feel like you are in a shouldn’t take a federal case for reactions flood, but people in the Third World are in to a paper to be elicited from the scientific a real drought. They never get the journals commuttit y—not just on the rumor network, that you complain of getting too many of.” but some place where everybody else can And so forth. see it. This is the bulletin board system of The economics of sharing will shift dra- commentmy and would complement what matically with these new media. For trivial the fixed board of editors would have to marginal costs you can provide 100 CDs a say. year which would far exceed the total vol- If there is a good dialectical system and ume of publication that they could ever the critical community has an opportu- hope to get in any other way. There is no nity to express its views, even ex post other way in the world that we can redupli- facto, that’s how the scientific process cate all the paper libraries that we now have works at its best. Here the economics and as a privileged treasure. the technology for dialectic give a great Another feature about globality that elec- edge on the electronic systems over the tronic systems will offer is built-in transla- printed ones, if for no other reason than tion aids. I am not talking about the nir- how to get propinquity. I mean, if an ar- vana of automated, perfect, smooth trans- ticle has been printed and then a little lation. Most of us here have a smattering later on I write a critical reaction to it of two or three foreign languages; a few of (even in the rare case that the journal ac- you are great linguists. But when I am read- cepts that sort of commentary and further ing an article in German, which I am fairly dialogue), they do not adjoin one another fluent in, wouldn’t I love to have a built-in on the shelves. It’s a nuisance trying to dictionary to help out when I run into a find them. phrase that I didn’t understand? Let’s say I write something six months I’11take the risks of that crude transla- ago; Gene Garfield wrote a blistering cri- tion. It may come out with some of the tique sometime after that. How are the ridiculous puns that you all know about. two of them going to be brought together? Again, this becomes trivially easy in terms That kind of reshuffling of the units is of its marginal cost. And it will greatly very hard with printed paper. It’s trivial, extend the global accessibility of literature of course, to do it with electronic media to a wide variety of people whose com- via the networks of linkage of material mand of the current international standard and commentary. This potential for reag- English may not be perfect. So these are gregation stands just after mechanized some of the arguments for the reforms that search and tempo of availability as the I hope you share with me and I would like greatest advantage that these new kinds to see brought about. of media can offer.

214