CODES USED IN THIS DOCUMENT: A{-} = LONG A {I} = START ITALICS I{-} = " I {i} = END ITALICS N{-} = " N {B} = START BOLD O{-} = " O {b} = END BOLD U{-} = " U {SC} = START SMALL CAPS M{.} = M W/DOT UNDER {sc} = END SMALL CAPS N{.} = N " {BI} = START BOLD ITALICS T{.} = T " {bi} = END BOLD ITALICS S{.} = S " S{'} = S W/SLASH OVER N{~} = N W/TILDE OVER {THE ASIAN CLASSICS INPUT PROJECT (LOGO)} {B}DEMONSTRATION DISKS{b} {B}Ten Major Texts from the Kangyur and Tengyur Collections of Classical Sanskrit Literature in Tibetan Translation{b} including Native Catalogs to the Kangyur and Tengyur with a special supplement: the Complete Cataloged Listings of Tibetan Works of the United States Library of Congress produced and distributed without charge through support from {I}The David and Lucile Packard Foundation The Packard Humanities Institute Sera Mey Tibetan Monastic University The Andin International Diamond Corporation The Mahayana Sutra and Tantra Center Rashi Gempil Ling Kalmyk Mongolian-American Society{i} {Following down through "Limits of Liability" should be in a reduced point size on back of title page so to fit attractively on one page} Digital form of "Ten Major Texts from the Kangyur and Tengyur Collections of Classical Sanskrit Literature in Tibetan Translation, including Native Catalogs to the Kangyur and Tengyur," and "Selected Errata to the Delhi Edition of the Derge Kangyur and Tengyur" copyright {copyright symbol} The Asian Classics Input Project, 1990. {I}With the written permission of the Asian Classics Input Project, this data may be freely copied, used, and distributed for non-profit private and institutional research or education. It may not be copied and re-sold for any purpose. Any modification of the data prior to its re-distribution requires written permission of the Project and a prominent notice of such modification.{i} Digital form of the "Complete Cataloged Listings of Tibetan Works of the United States Library of Congress" copyright {copyright symbol} The Library of Congress, 1989. {I}This data has been licensed from, and a fee paid to, the United States Library of Congress. Requests for permission to copy or distribute the records in any form should be referred to the Cataloging Distribution Service of the Library of Congress, Washington DC, USA 20541.{i} "LIST" file viewing and browsing utility copyright {copyright symbol} Vernon D. Buerg, 1983-89. All rights reserved. {I}See copyright/license/warranty section on the accompanying documentation file.{i} "SeekEasy" search program copyright {copyright symbol} Correlation Systems, 1984-87. All rights reserved. {I}See registration information on the accompanying documentation file.{i} {I}Registered trademarks:{i} "Magellan" is a trademark of the Lotus Development Corporation. "ZYindex" is a trademark of the ZYLAB Corporation. "Norton Utilities" is a trademark of Peter Norton. "WordPerfect" is a trademark of the WordPerfect Corporation. "Microsoft Word" is a trademark of the Microsoft Corporation. "Professional Write" is a trademark of the Software Publishing Corporation. "Go{-}fer" is a trademark of Microlytics, Inc. {I}Limits of liability and disclaimer of warranty:{i} The staff of the Asian Classics Input Project have to the best of their ability assured that the data, printed information, and programs contained or reviewed in this complimentary package are accurate and effective. ACIP though makes no warranty of any kind, expressed or implied, with regard to these materials, and shall not be liable in any event for incidental or consequential damage in connection with, or arising out of, the furnishing, performance, or use of the package. {The Asian Classics Input Project (logo)} {leave four-line space here for pasting in a short line of Tibetan later} "Torch of knowledge Lamp for the world." --the Man{~}jus{'}ri{-} Na{-}ma Sam{.}gi{-}ti, c. 500 BC {B}I. The Project Goals{b} The Asian Classics Input Project has been organized for the purpose of preserving and furthering the study of important examples of Asian literature, through the creation and distribution of inexpensive computer disks containing these works in a simple and accurate digital form. The initial goal of the project is to input the Kangyur and Tengyur collections of classical Sanskrit literature in Tibetan translation. The 4,500 works of these collections represent the cream of Asian philosophical thought from the period of 500 BC to 900 AD. With few exceptions the Sanskrit originals have been lost, and survive only in faithful Tibetan renderings protected over centuries by the natural barrier of the Himalaya mountains. The project aims to make these texts, along with research tools such as dictionaries and bibliographies, easily accessible on the personal computers of researchers around the world. This will stimulate the translation of the collections, and gradually open this treasure of knowledge to the general public. This influx of the great ideas of the other half of mankind will deeply enrich our Western culture, and inevitably lead to greater understanding between the peoples of the world. The choice of Tibetan-language texts is also dictated by a concern that recent political events may quickly erase the Tibetan people themselves and the rich store of Asian classics they have carefully collected and preserved over a complete millenium. The entire input work has therefore been accomplished at a traditional monastic university by native Tibetan refugee scholars trained, equipped, and salaried by the project. {B}II. The Project History{b} The project's roots lie in a 1977 effort by its organizers at the Pyne Hall quarters of Princeton University's Department of Classics to define rules of Tibetan verb transformation with computer-aided search and Tibetan alphabetization programs. This building was also a focal point for users of the Thesaurus Linguae Graecae (TLG) data, a massive collection of important Greek literature made available on a CD-ROM laser disk with support from the David and Lucile Packard Foundation and the Packard Humanities Institute. Exposure to this database and the Ibycus software tools for its manipulation naturally led to a proposal to create an equivalent for the Eastern classics. During the next eight years individual texts were input, transcription systems refined, and the world's first Tibetan- language word processor and printer developed. Important contributions were made by native expert Geshe Lobsang Tharchin, software designer Stephen Bruzgulis, and the chairman of Princeton's Department of Classics, Prof. Samuel D. Atkins. In 1986 a meeting was held at Princeton with Dr. David W. Packard, a Greek scholar and developer of much of the TLG work, to help chart the course of an Asian equivalent. On Dr. Packard's advice, a survey of nearly 100 scholars of Asian literature throughout the world was undertaken to determine what texts should be input first, in what format, to accomplish which immediate and long-term goals. Project organizers spent months reviewing and evaluating similar efforts by other groups: Dr. Theodore Brunner of the University of California at Irvine freely shared his experience and contacts as the director and driving force behind the TLG project; Mr. John Bradley of the Computer Input Corporation of Redondo Beach, California, and a major TLG contractor, supplied much information on data standards and conventions; a visit was made to KICO, the Korean Information Company, in Seoul to study their operator training and methodology for early TLG work; and the impressive database, search, and indexing tools developed for the classics of Pali in Thai translation by Dr. Supachai Tangwongsan and other researchers were examined first-hand at their computer center in Mahidol University, Bangkok. On the basis of the international survey, it was decided as a first step to target 25 texts from the Derge edition of the Kangyur and Tengyur, which was printed from woodblocks carved in eastern Tibet circa 1730. Scanning of the texts by optical character recognition (OCR) methods was researched, but rejected since the irregular outlines, stray ink marks, and see-through of carved xylograph letters printed on rice paper would with present technology lead to a high number of errors, and necessitate costly and time-consuming editing. It was decided to key in the texts; this would be prohibitively expensive in the United States or Europe, so a number of Asian data entry firms were researched. Estimates were obtained in Korea, the Philippines, Hong Kong, and Thailand, but it quickly became clear that the difficulty of mastering and interpreting the often ambiguous Tibetan characters would best be left to native Tibetan scholars. The abbot and board of directors of Sera Mey Tibetan Monastic University, a traditional academy of 500 monk- scholars founded by refugee professors who had escaped the 1959 bombardment of the original Sera Mey University in Tibet, were invited to meetings in Bangalore City, south India, to discuss a possible entry project. They immediately pledged their support, and quickly made arrangements for the necessary building facilities. Funding for the project was finalized in 1988 through a cooperative effort of a number of institutions. The David and Lucile Packard Foundation, the Packard Humanities Institute, the Andin International Diamond Corporation of New York, Sera Mey Tibetan University, and the Mahayana Sutra and Tantra Center at Rashi Gempil Ling, a Kalmyk Mongolian- American Society in New Jersey, made generous financial and material grants. Manuscripts of the targeted texts were obtained and a group of 30 Tibetan students selected for training at Sera Mey. Ven. Thupten Pelgye, a member of the board of trustees of the Sera Mey Scholarship Fund, assumed responsibility for the management of the computer center. An early decision was made to utilize locally-produced hardware and entry instructors from nearby cities, since this would facilitate the repair of equipment and solution of operational problems in the primitive conditions at the refugee university. A diesel generating station was built adjacent to the computer facility, to supply backup electrical power during the frequent local blackouts. ACIP staff from the United States rotated over to the center in the first year according to specialty required. Dr. Robert Taylor, assistant project director and head of the Washington area office, directed the set-up and training phases. Michael Roach, project director, followed with transcription instruction and standards implementation. John Malpas, chief project programmer from ACIP New York, organized the actual entry and editor education stages. Dieter Gewissler, quality assurance manager, came next with revised systems and training procedures, as well as oversight of the reference materials input phase. These ACIP staff members all donated their time; the Tibetan entry operators and editors are paid a modest wage. An extra four dollars for every dollar paid to the young men are set aside in a scholarship fund for them and their classmates at the refugee university. The two shifts are limited to four hours each per day, so that computer and language skills can be sharpened, but without detriment to the young people's important traditional study of the Asian classics themselves. At the very outset of the undertaking, it was agreed by all participants and sponsors that--so long as funding allows--the data created should be offered to the international community without charge, for the betterment of mankind. {B}III. Contents of the Demonstration Disks{b} {BI}A. The Ten Texts{bi} The following represent the ten most often requested titles from the Kangyur and Tengyur collections of Sanskrit classics in Tibetan translation, according to the ACIP international survey conducted in the fall of 1987. A summary of texts and filenames for easy reference is located below in the section on file nomenclature conventions. {B}1. Abhisamaya{-}lam{.}ka{-}ra, the "Jewel of Realization"{b} {I}author: Maitreya as related to Master Asan{-}ga title in Tibetan:{i} mNgon rtogs rgyan {I}date of composition: c. 350 AD location: Derge Tengyur, section on the Perfection of Wisdom, volume 1 (ka), ff 1B-13A file name: TD3786F.ACI size: 41,000 bytes synopsis:{i} Is there anything truly good about our life, from birth to death? Can knowledge change the content of this life? Are there limits to human knowledge? Must a concern for humanity accompany knowledge, for its fullest development? The text answers these questions through a presentation of three types of knowledge, four stages of related understanding, and a discussion of the final goals of human thought. {B}2. Madhyamaka{-}vata{-}ra, or "Entering the Middle Way"{b} {I}author: Master Candraki{-}rti title in Tibetan:{i} dBu-ma la 'jug-pa {I}date of composition: c. 650 AD location: Derge Tengyur, section on the Middle Way, volume 23 ('a), ff. 201B-209A file name: TD3861F.ACI size: 65,000 bytes synopsis:{i} Do things exist the way we think they do? And if they didn't, could they not exist at all? If we realized that we had never had an accurate perception in our entire lives, would life be meaningless? Would good and evil still matter? The text answers with a resounding "Yes!", leading us through the six perfect virtues: charity, morality, patience, effort, concentration, and especially wisdom. {B}3. Abhidharmakos{.}a, the "Treasure of Knowledge"{b} {I}author: Master Vasubandhu title in Tibetan:{i} Chos mngon-pa mdzod {I}date of composition: c. 350 AD location: Derge Tengyur, section on Knowledge, volume 61 (ku), ff 1B-25A file name: TD4089F.ACI size: 86,000 bytes synopsis:{i} What are we made of? What atoms and elements build our physical bodies? How do our senses work, and why do we all have the same basic mental functions: from simple awareness, to attention, on down to particular emotions like jealousy or consideration for others? How did the world begin, and what made it start? What brought each of us his own life? Is it a good life? Can we change it? This enduring masterpiece is tightly packed with the precious answers, the entire classical knowledge of Asia in eight concise chapters. {B}4. Prama{-}n{.}ava{-}rttika, the "Commentary on Valid Perception"{b} {I}author: Master Dharmaki{-}rti title in Tibetan:{i} Tsad-ma rnam-'grel {I}date of composition: c. 650 AD location: Derge Tengyur, section on Valid Perception, volume 95 (ce), ff 94B-151A file name: TD4210F.ACI size: 204,000 bytes synopsis:{i} How do we know--anything? Just how much of what we think we know is only assumption, just what everyone has always said? Beyond the basic objects of our physical senses, how are we to perceive the ideas and principles which have the power to ruin or exalt us? What rules shall we apply in our thinking; how do we know a conclusion is true? The four chapters of this text provide answers so powerful that, like all the classics from our list, it is still studied and debated on a daily basis in the traditional Tibetan monastic academies. {B}5. Vinayasu{-}tra, the "Sutra on Vowed Morality"{b} {I}author: Master Gun{.}aprabha{-} title in Tibetan:{i} 'Dul-ba'i mdo {I}date of composition: c. 500 AD? location: Derge Tengyur, section on Vowed Morality, volume 80 (wu), ff 1B-100A file name: TD4117F.ACI size: 375,000 bytes synopsis:{i} What is right? What is wrong? Why should we care, if we can get away with it? Where did our moral instinct come from? Are there some moral rules our culture has completely missed? Who decides the rules, and with what authority? Are there exceptions, and who decides them? What would it feel like to live an honestly examined moral life, for the whole rest of one's life? The text replies with the considerable detail that our human frailty demands. {B}6. Mu{-}laprajn{~}a{-}, the "Root Text on Wisdom"{b} {I}author: Master Na{-}ga{-}rjuna title in Tibetan:{i} rTza-ba shes-rab {I}date of composition: c. 200 AD location: Derge Tengyur, section on the Middle Way, volume 17 (tza), ff 1A-19A file name: TD3824F.ACI size: 64,000 bytes synopsis:{i} How could anything be what we think? When you look at a pot on the stove, do you see a pot or a collection of handle, bottom, part of the round edge, some metallic gleam? If you are seeing the handle, how can you be seeing the pot? But how can you see a pot if you don't see the handle? With the earth-shaking simplicity of a child's questions, these few pages overturned the entire history of Asian philosophy, and may someday our own. {B}7. Uttaratantra, the "Higher Line"{b} {I}author: Maitreya as related to Master Asan{-}ga title in Tibetan:{i} rGyud bla-ma date of composition: c. 350 AD location: Derge Tengyur, section on Mind Only, volume 14 (phi), ff 54B-73A file name: TD4024F.ACI size: 69,000 bytes synopsis:{i} Where will knowledge take us? Does the wisdom of the ages really make any man a sliver less liable to the inescapable descent into old age, senility, and death? Is there something that, if we knew it, could protect us? Is the perfection of human understanding possible? What would it be like? The five chapters of the text give brief and eloquent answers that have survived the test of centuries. {B}8. Catalog to the Kangyur, Derge Edition {I}author: Si-tu Pan{.}-chen Chos kyi 'byung gnas title in Tibetan:{i} bKa'-'gyur dkar-chag, sDe-dge par ma {I}date of composition: c. 1730 AD location: Derge Kangyur, section on the Catalogs, volume "Laks{.}mi{-}," ff 113A-157B filename: KD4568FP.ACI size: 162,000 bytes synopsis:{i} An invaluable index to the 1,100 works of the Kangyur section to the traditional collection of Sanskrit classics in Tibetan translation, finally codified from nearly two millennia of accumulated masterpieces by the Tibetan savant Buton Rinchen Drup in the 14th Century. The native catalog comes in two versions: expanded and concise. The concise version is presented here and lists title, section, volume, author, Tibetan translator and Indian collaborator, editors for important later revisions, and brief notes about editions in various languages, cases of disputed authorship, treatment in other collections, and the like. As any student of classical Asian literature early learns, the names of texts and their authors may appear in three or four very different forms. This fact has made hard-copy indices to the catalogs oftentimes unworkable, since a user must anticipate which name might have been used for the alphabetical listing. The tremendous advantage of having this and the other indices on disk is that the names may instantly be located by a single key word, regardless of any alphabetical or section order. A disadvantage is that the native indices were composed in running Tibetan prose; the format provides a trove of information for the more advanced researcher, but may prove a stumbling block for newer students whose Tibetan is not yet fluent. Moreover, titles in the native Derge catalogs are referenced only to a particular section and volume of the collections; no page number is given, although the order of the text relative to other texts in the same volume is listed, and a second pagination included for each work within its larger volume. One reason for including the index to the Lhasa edition of the Kangyur on these distribution disks is that it also includes starting folio and line locations for the same texts. {B}9. Catalog to the Tengyur, Derge Edition{b} {I}author: Zhu-chen Tsul-khrims rin-chen title in Tibetan:{i} bsTan-'gyur dkar chag, sDe-dge par-ma {I}date of composition: c. 1730 location: Derge Tengyur, section on the Catalogs, volume "S{'}ri{-}," ff 337A-468B file name: TD4569FP.ACI size: 559,000 bytes synopsis:{i} Again the concise catalog, but here to a massive collection of 3,400 works written largely in commentary to the Kangyur itself. The capacity to research this material with a personal computer will be appreciated by any researcher who has spent hours flipping through the folios seeking a title; but again the user will need considerable fluency in the language to make full use of the file. Only section, volume, and relative position within the volume are given in the native catalog, and the author in its final folios names a handful of volumes for which he has not listed very brief liturgical works by their separate titles. {B}10. Catalog to the Kangyur, Lhasa edition{b} {I}author: "composed at the behest of the 13th Dalai Lama" title in Tibetan:{i} bKa'-'gyur dkar-chag, Zhol par-ma date of composition: 1934 location: Lhasa Kangyur, section on the Catalogs, index volume (ka-tho), ff 1A-49A file name: KL4568FP.ACI size: 139,000 bytes synopsis: This is a native catalog to the most recent carving of the Kangyur, executed at the Shol printing works near the base of the great Potala palace, home of the Dalai Lamas of Tibet. The edition is similar to the Derge but carved in larger and even more clear characters than the masterful Derge. Precise data for the starting page and line are given for each work, and volume numbers (actually Tibetan letters) are more clearly indicated with a special carved ornament, which has been included in the file and will greatly facilitate machine searches. The preparation of the corresponding Tengyur section of this edition was put to an abrupt end by the reckless savagery of the Chinese armies who in 1959 invaded and still occupy Tibet, endangering the very existence of a great portion of classical Asian literature. {BI}B. The United States Library of Congress Tibetan-Language Listings{bi} {I}File name: LCLIST series combined size: 1,700,000 bytes {i} During the period from 1965 to 1984, the United States Library of Congress conducted an unprecedented program of acquiring copies of important Asian classics as they flowed down into India with Tibetan refugees escaping the destruction of their country. This effort was made possible with excess foreign currency funds provided by the United States Congress, under Public Law 480 (PL 480) and later the Special Foreign Currency Program (SFCP). The Library's representative in New Delhi, Mr. E. Gene Smith, worked tirelessly over this period to organize the purchase and complimentary distribution of some 8,000 titles (not counting the Kangyur and Tengyur collections) in over 2,000 volumes to each of 18 participating American libraries. It can be claimed without exaggeration that the Library's foresight and Mr. Smith's personal efforts thereby stimulated the largest and most important publishing of major classical literature ever undertaken in the history of Asia, and perhaps the world. Participant libraries welcomed the massive influx of materials but were ill-equipped to process and catalog it; because of the previous inaccessibility of Tibet, there were few experts in Western countries and nearly no catalogs of the treasures the Tibetan had collected over the centuries. In fact the long distinctive cases for rice-paper blockprints of the PL480/SFCP Tibetan materials can still be seen stacked in arrearage areas of many libraries, usually with a dedicated senior cataloger struggling to complete the backlog, assisted by one or two of the many upcoming younger scholars whom the materials themselves have helped to spawn. In order to share the efforts made by their own catalogers, the Library of Congress through its Cataloging Distribution Service has licensed the Asian Classics Input Project to distribute with this complimentary demonstration package their completed listings of Tibetan-language materials. Included too are many works outside of the PL480/SFCP acquisitions program, although it is very important for users to note that a large block of texts has yet to be added to the database by the Library, so that a given work may well have been acquired but not yet be listed. Researchers should therefore continue to utilize their traditional resources for locating particular texts from the program, including the larger commercial bibliographical databases available through the terminals of most major libraries, as well as the card and on-line catalogs of materials already cataloged by the individual institutions. ACIP has arranged to receive regular updates of newly cataloged works from the Library of Congress and will continue to make them available without charge. The listings are an invaluable resource for the proper names of authors and titles, their dates, and notes on the contents and sources--for these we must again thank Mr. Smith, who pioneered much of this trail into uncharted territory. The data on LC catalog numbers and dimensions of the volumes themselves will greatly assist users to find the books in participant libraries. Institutions who hold the PL480/SFCP collections, by the way, are the following ("C" indicates comprehensive, and "B" basic, coverage): The Library of Congress (C); University of California, Berkeley (B); University of California, Los Angeles (B); Center for Research Libraries, Chicago (C); University of Chicago (C); Columbia University (C); Cornell University (B); University of Hawaii (C); University of Illinois (B); Indiana University (B); Institute for the Advanced Study of World Religions, New York State (C); Naropa Institute, Colorado (C); New York Public Library (C); University of Pennsylvania (C); University of Virginia (C); University of Washington (C); University of Wisconsin (C); and Yale University (C). The data also contain serious drawbacks which should be frankly noted. The most serious is the inconsistency, inaccuracy, or imperfect conversion of the transcription method used by the Library. This has resulted from the lack of generally accepted transcription standards among the scholarly community; from simple errors by entry operators confronted with long strings of the very difficult Tibetan transcription; and from the fact that diacritics in general do not behave well between systems; users of commercial library databases are well aware that the diacritics sometimes fail to convert at all, so that (in ACIP transcription) the different Tibetan characters {SC}NA, NGA, NYA,{sc} and n{SC}A{sc} may all appear as simply {SC}NA{sc}; the operator has had to remember this in calling up data, or he misses the entry. As a general rule then users should study the quirks of this database and any others they intend to use, and be aware that they are more often confirming the presence of a work than establishing its absence: do not assume a work has not yet been acquired because you cannot find it on the first few tries. (Some useful notes on successful computer searches of Tibetan data are included below.) Another weakness of the listings, well known to regular users of the material, is that individual titles within the {I}sungbum{i} or collected works of the major Tibetan writers have not been listed separately. Tibetan philosophers have traditionally undergone decades of training in a rigorous and widely diverse curriculum, and have commonly written commentaries on a broad selection of the original Asian classics. The standard entry in the database however will only be a single one for the general collection of these works, and in fact the 4,500 separate titles of the Kangyur and Tengyur collections appear as but two listings. It would have been humanly impossible for the Library's limited staff of experts to complete all these titles, and in fact a small army of researchers is at work throughout the world to do so. Fruitful efforts have been initiated by individual Tibetan institutions in India, native Tibetan scholars in China and Tibet, and a number of libraries and projects in the United States and other countries. We can thus look forward to more refined and complete databases in the coming years, yet take advantage in the meantime of the first raw data for personal computers here presented. We are pleased to note that the Library of Congress has recently announced its intention of resuming these very valuable efforts in the acquisition of the important Asian classics in the Tibetan language. Suggestions for texts to be included into the program, requests to receive copies of the acquisitions themselves, or other inquiries may be forwarded to Mr. E. Christian Filstrup, Chief, Overseas Operations Division, The Library of Congress, Washington DC, USA, 20540. {BI}C. Selected Errata to the Delhi Edition of the Derge Kangyur and Tengyur{bi} {I}filename: DKTERR.ACI size: 6,000 bytes{i} In the early stages of the ACIP, scholars throughout the world were polled to determine their choice of editions for the Kangyur and Tengyur to be input. The nearly unanimous selection was the Derge, since it has a reputation for clarity and accuracy, had been the subject of a number of cataloging attempts, and had recently been reprinted in whole or part by offset photography, in several countries. Entry work in the computer center at Sera Mey Tibetan Monastic University began from the Delhi reproduction of the Derge, which is the most prevalent printing and is also included in the PL480/SFCP collections. Upon proofreading it was found that a significant number of errors were reaching the final edit stage, which seemed impossible since each text had been typed in twice and then compared by machine--and no two operators would have made the same unusual errors. A review of the relevant sections of the Delhi edition revealed that the errors had been introduced by native scribes retained to touch up light spots in the original blockprint prior to offset publishing. The errors in some texts were so serious and prevalent that they preclude the use of particular titles for anything but general reading. Samples of errors, and the corrected readings, are included in this file to demonstrate the problem and also assist researchers in their correction. A comprehensive listing of errors was begun but abandoned due to the scope of the problem. As experienced students know, it is nearly the nature of reproducing Tibetan blockprint manuscripts that problems will be encountered. If the camera setting is too dark, material from the obverse side of the rice paper will bleed through and obscure the reading--as occurred with the microfiche edition of the Chone edition. If the setting is too light, entire lines may disappear--as was the case with the recent California reprint of the Derge. If the setting is intermediate and a scribe hired to fill in the incomplete characters, he introduces new errors. The errors of the Delhi edition were painstakingly corrected by the ACIP domestic editors, and as a precaution new entry is being made from a direct blockprint of the Lhasa edition of the Kangyur. Negotiations for the purchase of an original Derge edition of both collections are currently under way. It is extremely important to note that attempts to reproduce the Kangyur and Tengyur to date, though imperfect, have been an invaluable contribution to world scholarship. Without these materials, younger scholars would not have been trained, and without new scholars, no new material can be properly edited. A critical point in fact already exists where the qualified older survivors of the Tibetan holocaust are too few to properly proofread any large number of new editions of the Asian classics. {B}IV. Data Transcription Method{b} Disagreements over how to transcribe Tibetan, once a fruitful source of journal material for us students of Asian literature, are thankfully obviated by computers since a worker can now search a database for any offending transcription of a particular character and replace it with that of his own choice, automatically, in a number of seconds. The transcription method chosen here corresponds with a few exceptions to that developed by Prof. Ngawangthondup Narkyid and other researchers at the Library of Tibetan Works and Archives of the Tibetan Government. The method is unambiguous; requires no diacritics, overstrikes, or "dead" keys; is input on a standard keyboard with (so far) no control or alternate keys; observes absolute economy in number of keystrokes; and conforms as nearly as the English language allows to the actual sounds represented. The goal is that an entry operator types exactly what he sees in the Roman letters he would naturally expect, without memorizing a long list of key combinations. These points are critical in a large data entry project such as the ACIP. The Tibetan operators have been trained from zero (most had never seen a computer before), on locally available and serviceable hardware, with the lowest possible capital investment. A major aim of the project is to provide these refugee scholars with a marketable skill, data entry, which will contribute to their continued survival--and the continued survival of the Asian classics is tied to theirs. Original entry and the automated comparison/edit steps are performed in Roman characters on the normal QWERTY keyboard, with transcription rather than native script on the screen as well, so that the refugees are incidentally prepared for later contract work in Western languages. It has further been our practical experience over the years that almost no one who uses a computer regularly and who can be expected to conduct research with the ACIP database will not be familiar with Roman letters and transcription. A well developed toggle switch between Roman and native script, such as that already available with Thai word processors, will of course be the ideal. Economy and ease of keystrokes will allow more data to be entered within the project's budget limits: elimination of a single unnecessary keystroke in transcription will over decades represent a substantial savings in labor. Tibetan syllable punctuation dots, for example, have been replaced with a space, and a separate program written for the reconstitution of these marks, according to the rules of Tibetan orthography, in the print-out of native script. This adjustment and ACIP's unique on-screen format of maintaining the original page and line layouts facilitate reading the transcription from a computer monitor. The transcription method here also conforms closely to that required for print-out and on-screen representation of native script utilizing the Tibetan Text Processing System (TTPS), the world's first integrated software for processing Tibetan, developed for IBM-compatibles by a number of ACIP's original staff. In recent years a number of additional programs have appeared, each with its own particular advantages and disadvantages, so that a researcher may want to experiment and find the system best fitted to his needs. For Apple users there is the widely used "Tibetan on the Macintosh" program from Mr. Pierre Robillard of Toronto, Canada, available at present through Snow Lion Publications of Ithaca, New York. For IBM-compatibles there are, besides TTPS, the Atisha program developed on Chiwriter software by Dr. Peter Ebbatson of Oxford, UK; as well as a newly announced system (also available through Snow Lion) from the Tibetan Computer Company of Colorado, USA. A number of other interesting programs are currently under different stages of development in various countries, and we can look forward to a day when all the amenities of modern text application software are available for Tibetan, as they are already for Japanese, Thai, and other non-Roman scripts. ACIP data has purposely been supplied in "raw," unformatted ASCII characters, which means that it can easily be converted by the user to the transcription form required by any given Tibetan word processing program. A small program for creating a TTPS-readable file is available without charge on request from ACIP, and we would be pleased to make conversion programs for other Tibetan processors available to our subscribers as they are developed and submitted by users of the data. A few notes of explanation to the transcription chart attached {transcription chart should be on page facing the following paragraphs}: first, all letters are input as capitals except for the retroflex series, indicated in Tibetan with a reversed character and in ACIP data with the lower case. A lower case "m" represents the Sanskrit {I}anusva{-}ra{i}. The {I}'a-chung{i} subscript used in Tibetan for a long Sanskrit vowel is simply typed, with the normal apostrophe, like any other subscript: just after the consonant and before the affected vowel, since in this position it cannot be mistaken for the larger suffix version. The classic ambiguity between a subscript {SC}YA{sc} below {SC}GA{sc}, and a main-letter {SC}YA{sc} following a prefix {SC}GA{sc}, is resolved by tying {SC}GYA{sc} for the former and {SC}GA-YA{sc} for the latter, since the TTPS program will automatically parse any combination joined by a hyphen as side-by-side consonants. A similar principle applies to the very rare cases where a Sanskrit stack creates an ambiguity; a plus sign is used to indicate the stack, so that for example with the only two cases in the database so far where a palatal nasal is a Sanskrit stack of {SC}NA{sc} above {SC}YA{sc} rather than simple {SC}NYA{sc}, one should search for the combination {SC}NA+YA{sc}. {B}V. File Nomenclature Conventions{b} Following is a summary of how ACIP filenames are presently applied to the different texts. We can expect modifications of these conventions as the database grows, particularly with the anticipated standardization of categories for native Tibetan commentarial literature, which we refer to with the native appellation Sungbum ("collected works" of major authors). A. Filename Space 1: Collection code--(K) Kangyur, (T) Tengyur, (S) Sungbum, or (R) Reference Space 2: Edition code--at present only (D) Derge or (L) Lhasa Spaces 3-6: Tohoku or Sera Mey catalog number Space 7: Status letter as follows-- A = first input version B = second input version C = first edited version D = second edited version E = third edited version (final overseas edition) F = domestic proofed version for distribution (in ASCII) G-J = distribution copy modified for printing in Tibetan characters using current version (G) or future updates (H-J) of TTPS (Tibetan Text Processing System) K-N = distribution version in customized format or converted for particular commercial word processors at user's request O = (not used because of similarity to zero) P = (as space 8) only part of the text cor- responding to the Tohoku or Sera Mey catalog number Q-U = later annotated or otherwise marked versions of distributed copy V-Z = reserved for unforeseen needs Spaces 7-8: Operator or editor identification numbers (with pre-distribution copies only, for assessment of error penalties; collection/edition codes truncated) B. Extension Spaces 1-3: Batch number (normally of five pages each, with pre-distribution copies), or "ACI" (with distribution copies, to identify authorized ACIP data for public use) C. Sample filename: TD3786F.ACI "T" indicates that the text comes from the Tengyur collection of Asian classical literature "D" indicates that the Derge edition of the Tengyur collection has been used for input "3786" indicates a Tohoku catalog text number, which happens to be the first text that was targeted for entry: the {I}Abhisamaya{-}lam{.}- ka{-}ra{i} of Maitreya and Asan{-}ga "F" indicates that this is a final edited version, derived as shown in the uppermost box of the accompanying data verification procedures flow-chart "ACI" indicates that the file has been released for public use; it should not be modified by users so long as this extension is retained D. Notes The ACIP file convention format allows a DOS wildcard search for a text using its catalog number, as well as a quick wildcard search for copies of a particular text at defined levels of editing. The use of letters prevents ambiguity in number searches, avoids long numeric strings, and allows more parameters per single space. The presence of an initial collection code in the filename allows limited searches for texts or phrases later by collection, as well as updates of texts completed by collection. Users of the demonstration disks will want to keep a copy of the chart below near their computer, for quick reference. The same chart is available on demonstration disk #1 under the file name "TITLES.ACI". E. Summary of file names, ACIP demonstration disks {B}File name Contents title TD3786F.ACI Abhisamaya{-}lam{.}ka{-}ra TD3824F.ACI Mu{-}laprajn{~}a{-} TD3861F.ACI Madhyamaka{-}vata{-}ra TD4024F.ACI Uttaratantra TD4089F.ACI Abhidharmakos{.}a TD4117F.ACI Vinayasu{-}tra TD4210F.ACI Prama{-}n{.}ava{-}rttika KD4568FP.ACI Derge Kangyur Catalog KL4568FP.ACI Lhasa Kangyur Catalog TD4569FP.ACI Derge Tengyur Catalog{b} {B}VI. ACIP Editorial Policy{b} The organizers of the successful Thesaurus Linguae Graecae project for the input of the classics of Greek literature onto a master CD-ROM developed sensible editorial policies that have been emulated by ACIP. Every work is input exactly as it appears in the original woodblock edition, without any attempts to modify or correct the text. Obvious errors introduced by scribes touching up the plates for later photographic editions have though been corrected insofar as a careful visual inspection of a suspicious reading clearly indicated a mistake written over the original reading. Despite the above "no editing" policy, when a reading in the original blockprint appeared certainly in error, proofreaders in selected cases have suggested an alternate reading in the format [*alternate reading] following the word in question. When the alternate reading could be in doubt, a question mark is added in the form [*alternate reading?]. A doubtful or particularly unusual reading is indicated with a simple question mark in the form [?]. These suggested corrections are only meant as an aid in conducting successful searches of the data; it is beyond the current scope and resources of the project to provide annotated editions. All other editorial comments are included in brackets as well. They are summarized on the attached transcription chart and most often indicate additional material on the xylograph prints (such as ornamental Sanskrit titles or line drawings of early authors, with captions) that would not be considered a part of the text as it was originally created. This is to allow accurate statistical analyses of the material by means of removing bracketed material from a second copy of the file prior to running the check. The physical format of each page is also strictly maintained. Folio or page numbers are indicated by an "each" sign "@" followed by the number, and an "A" or "B" to indicate obverse or reverse face in the traditional unbound stack of rice paper folios. The original lines are maintained and separated by a double return, which greatly facilitates checking any questionable reading against the original blockprint. A sample of original transcription and print-out in native text using the TTPS system is attached. {Sample page of letters and Tibetan should face this area.} {B}VII. How to Use the Disks{b} The text data supplied here on the ACIP demonstration disks was input in standard ASCII format with a simple DOS editor. This means that the files contain no specialized formatting codes and you can use virtually any editor or word-processing program to read them on your personal computer. If your word processing program requires it, follow the instructions from its manual for importing an ASCII file; otherwise simply use it to call up the desired file straight from the disk. If you don't have a word-processing program, put disk #1 in your computer and type {SC}LIST{sc} followed by the "enter" key. Follow the instructions below for the {SC}LIST{sc} program, to view and search through text for particular strings of words. This program will stay in the memory of your computer until you turn it off, and you can insert the other disks to look at and search any text you want. The disks will work on any standard IBM or compatible machine, with or without a hard disk. Users with hard disks can create directories tailored to their needs and load all the files in. Macintosh users will have to make the standard conversion to bring up the disk on their current operating system; even without conversion, the disks can be copied to a 3-1/2" floppy and used directly in the newer generation of Apples. As with any data, we recommend that you immediately create back-up copies of the disks in case the originals are damaged, or accidentally changed or erased. Follow the directions in your DOS manual to do so. ACIP disks are supplied with a write-protect tab to prevent any modification of the data on the originals distributed by the project. Please do not modify these originals. If you choose to copy files to another location for manipulation of the data, we request that you maintain the original filename (letters preceding the dot) and change the extension (letters after the dot) from "ACI" to an extension of your choice. The single most common use of this data will be to search a file or files for a desired string of text--whether a quotation, name of a particular philosophical concept, person, place, or so on. Scholars of Asian literature in the past have spend days and even months sifting through rice- paper folios for an important citation, often mentioned in another text without proper attribution, or with no attribution at all. Entering the texts on disk eliminates this tremendous expenditure of time, and makes searches more thorough--if they are executed properly. We will make some suggestions about undertaking successful searches and then comment on a number of search programs we have reviewed for use with ACIP data. Most commercial word processors on the market at present include well-developed search programs. The user typically calls up the file he wishes to search, hits the necessary control or function key, types in the string he is looking for, and hits a carriage return or similar key to begin the search. After the first find, striking another key continues the search on to the next match. The key to a successful search is to ask for the right thing in the first place. Study the ACIP transcription chart to be certain you are typing in the correct Roman letter equivalents. Bring up a file and launch a search for something you can see in the first few lines as a test. Remember that most ACIP transcriptions are in upper case; even better, set your search program (if possible) to be case insensitive so you don't have to remember the capitals. (Remember too that the file of Library of Congress listings of Tibetan literature contains varying transcription methods according to date and place of creation, and searching this data will take some creativity and patience.) The single most important rule in searching is to limit your search request to the absolute minimum needed to sufficiently narrow the results. Many search programs cannot locate a string of characters that carry on to the following line, which means that you may miss even a short two-syllable match if the second syllable continues on the next line (this is because the search program reads the carriage return as an additional and non-matching character). The ideal then is to obtain a program that will locate across lines (several are noted below), or else search by a single Tibetan syllable. Which syllable? Choose the most uncommon syllable in the string, or you may have to search all day. If you must, input enough additional syllables to narrow down the possibilities. Remember that Tibetan orthography allows for variant spellings much more frequently than English: you may not find a string containing the extremely common {SC}LA SOGS PA{sc} ("etc") in an older text if you don't remember to check as well for {SC}LA STZOGS PA{sc}, an antiquated variant. Sanskrit words in Tibetan transliteration are particularly tricky: the word {I}ki{-}rti{l} ("fame") will most often be found as {SC}K'IRTI{sc}, but can also appear in two syllables as {SC}K'I RTI{sc}, and in its Tibetanized version might be reduced to {SC}KIRTI{sc}. Avoid using spaces around single-syllable searches since you may miss one with a case ending such as the genitive {SC}'I{sc}. Remember too that endings such as the instrumental {SC}SA{sc} may replace part of the original syllable, precluding an exact match. Sometimes it's a good idea simply to search for the basic root of a word, {SC}TANG{sc} for example, in order to locate a string including {SC}BTANG, BSTANG, STANGS,{sc} or {SC}GTANG{sc}-- since the distinctions between spellings for different verb tenses and related nouns are blurred in progressively more recent citations of an older text. Many search programs allow "wildcard" replacements for a single letter, so that a search for {SC}MKHA?{sc} turns up both {SC}MKHA'{sc} and {SC}MKHAS{sc}, preventing a missed match due to inflection. Other programs go one better, allowing a wildcard such as * to stand for a multiple string of letters within a specified length of text. The most sophisticated programs allow a so-called "fuzzy" or approximate search, and we strongly recommend their use. As any experienced seeker of citations in classical Asian literature can testify, the original version of a quotation handed down from teacher to student over centuries can differ substantially from the current form. The problem of finding a variant of any citation is largely overcome by the fuzzy search capacity, which locates anything even similar to the desired string, and also very importantly can find string despite its being broken by a carriage return. Many search programs further permit a user to define a logical relationship between two key parts of a string with operators such as "and," "or," "not," or "nearby"--with the ability to define how close the two parts should be for a find. Another factor you will want to consider in choosing your search program is whether or not it provides you with the full context of the text around the string you've found. Some basic programs, and even the DOS system that runs the computer at all, have the capacity to locate a string but may tell you only the name of the file where it resides, and also perhaps the line number, with no further context. Other early search programs were restricted to about 20 characters of context, or else a few lines. Ideally you will want to use a program that loads the whole text, takes you to the desired string and highlights it, then allows you to scroll up and down to read the entire context. Remember that most of these programs begin the search from the present position of the cursor: if you're already half way through the text, the first half will not be searched. Most programs allow a search backwards or forwards in the text; we suggest that users get into the habit of returning to the top of the file and doing forward searches only for most purposes. Deciding which file to search for the desired citation or term is another important decision. Many search programs require first that the user specify a single file for searching, or at least a specific subdirectory or group of files. Other programs allow wildcards in the names of the files to be searched; ACIP filenames, as noted above, have been created to facilitate these wildcard searches so that one can direct the program to the Kangyur only, a specific text, defined section of catalog numbers, etc. The more powerful programs currently appearing on the market enable one to conduct a fairly quick search through all the files on a disk, or even on all drives in the machine: that is, they will seek the phrase throughout every file available on the computer, even if there are thousands. This is ideal for a database like ACIP's, although the usefulness and speed of such a search will depend on the computer's storage capacity (some kind of hard disk is minimum), and processing power (a 286 or higher is preferred, although even less will work). Like the TLG project, ACIP aims at producing a single CD-ROM containing the entire database of Asian classics. At present, these laser disks are usable only on a special reader which is plugged into the back of the computer. Their capacity is so large that the entire Greek classics occupy no more than half a disk. Encyclopedias and other large bodies of data are being released on CD-ROM in ever-growing numbers, and it appears that personal computers with built-in CD drives will soon be a market norm. For the time being, ACIP data is being supplied on double-density (360 KB) 5-1/4" disks, since these can be used in the vast majority of personal computers. Future updates prior to the initial CD-ROM releases will also be offered on 5-1/4" high density (1.2 MB), and 3-1/2" double (720 KB) or high-density (1.4 MB) disks upon request. All of these options promise much more powerful search and other text manipulation tools for the near future. {B}VIII. Search Tools Supplied with ACIP Demonstration Disks, and Other Available Options A. Programs Supplied with the Disks The demonstration disks in this package include two very useful search programs provided, like the disks themselves, free of charge. They are easy to start and use. If you are not familiar with the basic operation of a personal computer, we recommend that you have a more experienced friend help you for the first few minutes to get the programs up and running. You'll find that you are then able to explore these exciting tools by yourself, without any further training. 1. The LIST program written by Mr. Vernon D. Buerg of California, USA, is a simple, well-designed, and powerful utility for bringing text up on the screen, browsing, and searching it. Initiate the program from disk #1 by simply typing {SC}LIST,{sc} then selecting the appropriate menu item. Help is available on screen by hitting the {SC}F1{sc} function key. Be forewarned that the program will not find a string if it is broken up by a carriage return. This program is an example of "shareware," which means that the author has granted permission for free copying and distribution, so that users may try out the program before deciding to purchase it. If you find the program useful, we encourage you to submit the requested donation of $20 to the author at 139 White Oak Circle, Petaluma, California, USA 94952. This registration process for most shareware programs also entitles one to regular updates and often more detailed instruction manuals. Please note that this and the following program have been selected for inclusion solely on merit, and their authors are not associated in any way with the ACIP or its staff. 2. The SEEKEASY program for text searches is from Correlation Systems of California, USA, and is included for several important reasons. Although it displays only two lines of the context surrounding a search string, the program is capable of executing a "fuzzy" search: it will locate what you request even if you have spelled it wrong or typed it incompletely. Very importantly, it can therefore locate a string broken by a carriage return. Finally, the program is unusually fast and simple: just insert disk #1 in your computer and type {SC}SEEKEASY{sc} with a return, then follow the easy instructions for finding your string. Help is available in a short screen that appears when you type {SC}SEEKEASY ?{sc} at the DOS prompt. Again this is a shareware program, and if you find it useful we encourage you to submit the registration fee of $30 to Correlation Systems, 81 Rockinghorse Road, Rancho Palos Verdes, California, USA 90274. B. Other Available Programs Please note that the following is only a partial list of many fine programs available. ACIP would be pleased to review and include any other useful text tools in future updates; information can be sent to one of the two American offices listed below. Inclusion of programs is by merit only and does not constitute a commercial endorsement of any product by ACIP. 1. MAGELLAN, developed by the makers of the popular spreadsheet program Lotus 1-2-3, is a commercially available search program that we would advise any serious researcher to purchase. The program searches the entire hard disk (and any other disks) of your computer for the desired string, so you don't need even to guess the text where your citation might be found. The search can be made for an exact or only approximate ("fuzzy") match. Files in which the desired string appears are listed in a window, by rank of match probability: the first is most likely to contain the particular string you need, and the last has something similar but not very close. As you scroll down the list of these files in the window, their contents pop up on screen, in correct format regardless of which of the more popular software programs you have used to create them. Matches can be gathered with their context to another file, or printed out so you have a brief summary of every relevant instance of your string. Magellan does all this work with amazing speed due to a special indexing feature. The routine is also taught to skip some English "fill" words such as "a" or "that" in order to increase the search speed. This can be a little annoying when the English skip word also happens to be an important Tibetan word, but the file called "MAGELLAN.SKP", which defines these words, can be modified (we emptied ours altogether) without much loss of speed. If you become a serious user, you may want to have a computer-literate friend assist you in this change. Actual instances of this problem are though extremely rare and the program as is represents an important tool for scholars. Magellan has only come out this year, and like any program in its first version is a little sharp around the edges. You will need to invest some learning time to take full advantage of its powers. The program is available through any computer dealer or software store for about $100. 2. ZYINDEX is a product similar to Magellan that may in many respects be preferred by users. It works on the same principle of indexing the text, skipping filler words (this file can again be modified) for more speed. The program comes in three versions depending on the quantity of files to be searched; the basic "personal" version allows a search across up to 500 files, while the "professional" version can cover up to 5,000 files. The "plus" package supports searches of up to 15,000 files. One drawback is that these files are limited to about 600K each (larger files must be split up), but then only a few ACIP files will ever exceed this size. A great advantage is that searches can be tailored by logical operators, so that any combination of terms can be sought within a user- specified distance of one another using the logical operators "and," "or," "not," or "within". (Magellan by contrast operates on a simple "or" and therefore at times finds too many matches.) This capacity is especially useful for searching Asian texts with many variant spellings. The program will locate across a line. ZYindex is a product of the ZYLAB Corporation and the three versions sell for about $100, $300, and $700 respectively (the first version will serve most users' purposes at present). 3. THE NORTON UTILITIES is one of the most useful programs a personal computer user can ever buy. Aside from its excellent routines for keeping your hard disk healthy and recovering lost or accidentally deleted files, the package contains a small jewel called the NORTON EDITOR. This is a very basic but intelligently designed word processor that takes up minimal disk space and can be thrown in on any file of ACIP data that you choose to carry around to work on. The editor's search capacities are simple, quick, and thorough; the entire context is given, but strings are not found across lines. The entire package is widely available for about $75. 4. WORDPERFECT is the largest selling of any word processor and can be used for search and other operations on ACIP data. Input the text as an ASCII file from the appropriate file index and use the various {SC}F2{sc} function key search routines. WordPerfect and other more advanced word processors have the power of searching and replacing commands as well as text, which can be very useful as you learn more about manipulating ACIP data for your special needs. Remember though that any specialized commands in a particular word-processor's format may cause trouble with the various Tibetan printing programs. As for searching across lines, the search wildcard key (control V-X) does not seem to allow it, so we suggest you try to restrict searches to key syllables. WordPerfect is universally available for about $220. 5. MICROSOFT WORD is deservedly the leading competition to WordPerfect and contains a search command that, as with many word processors of its kind, allows forward and reverse searches, case sensitivity adjustment, partial word or character strings, and repeat of the most recent search. Use of the caret control-key symbol ^ (not the control key itself) in the desired string allows searches for commands and other special characters; a caret with "w" typed between syllables outsmarts the line-break by searching for any kind of "white" space in these spots--whether it be spaces, line or paragraph markers, or so on. Microsoft Word is also available almost anywhere, for about the same price as WordPerfect. 6. PROFESSIONAL WRITE is chosen here to represent a large number of "executive" word processors--less powerful but cheaper and easier to learn than the two just listed, so probably more practical for many users. Its search feature is quite "intuitive"--meaning you execute the keystrokes you would naturally guess you should--and wildcards are available in the form of simple dots. The program has been intelligently written to automatically search for strings across lines. Get it through any major software store or computer magazine advertising section, for about $125. 7. GO{-}FER is an excellent memory-resident search program that is always available in the background regardless of what program you are using to work with the ACIP data. It is a pleasure to use and also fast (though not as powerful as Magellan), and again allows the user to define logical relationships between parts of the string: we suggest for example that you choose two or more key distinctive words, set the "exactness" for even approximate matches, and use the "or" or "nearby" logical operator to find all instances of these key words within a few lines of each other. Context is given in full, and finds can easily be gathered to another file or printed out directly. Go{-}fer is available for both IBM-compatibles and Mac machines, and at $40 has to be the best value of all the commercial programs listed here. As of this writing a new version has been announced which is claimed to be twice as fast as the already speedy original. We would suggest that any user of ACIP data obtain this program. It is marketed by the Microlytics Corporation and can be ordered through any computer or software dealer. {B}IX. ACIP Data Verification Procedures {Flowchart of verification procedures should face this section} The attached flowchart summarizes the process by which ACIP attempts to assure high-quality data. This is a priority since input errors will prevent search matches, and important citations will then never be located. Texts on disk, that cost only cents apiece, will naturally gain widespread prevalence, as will any corruptions they contain. To help prevent errors, each text is entered in its entirety twice, by two different operators. The two versions thus produced are compared to each other independently by two different editors, using a compare/edit program named "Active Comparator" and developed especially for the project by ACIP chief programmer John Malpas. Any letter that fails to match exactly in the two files is checked against the original and corrected on an edited version. The two edited versions that result are then themselves compared character by character, by yet another editor. Entry operators for particular texts are identified by a special code in the filename; they are charged for their own errors and receive a bonus for documenting any errors in the original blockprint. Shipments of data received at the Washington area office of ACIP are first screened through an error-detection utility written into the TTPS Tibetan processing system. Most unallowable combinations of Tibetan letters are isolated at this point. Statistical counts are then run on the data using SYLLIST and the other ACIP UTILITIES written by assistant project director Dr. Robert Taylor. These alert domestic editors to apparent discrepancies in the number of lines per page or vowel and consonant counts per line, improper page or line divisions, and obvious vocabulary problems. The text is then printed out in the native script on TTPS, and examined in full by a primary and then secondary domestic proofreader. Possible errors are marked and checked against the original input edition and, if necessary, corroborated against an alternate edition. The master files are then corrected, briefly reviewed for obvious errors, and re-named for distribution. Despite the above procedures, the finished files will contain some errors. It is possible (though unlikely) that two operators, or editors, will make the same error at the early computerized steps, and for proofreaders to overlook it. In many cases, as noted above, an error has been introduced by scribes into a reprint of the original woodblock. We are happy to note that errors in these woodblocks themselves were extremely rare, and discrepancies in the Tibetan transcription of Sanskrit words (especially in the Kangyur and Tengyur catalogs) much fewer than researchers have previously estimated. The carving work itself was clearly executed, except for the traditional difficulty of distinguishing between {SC}DHA{sc} and {SC}NGHA{sc}, and we urge users to keep this in mind while conducting automated searches. The difference between a printed reproduction and a digital version of this material is that the latter cannot be ambiguous; the operator is obliged to make a decision one way or another about which key to strike for every character, however unclear. And this decision can sometimes be mistaken. William Dwight Whitney in the 1879 preface to his timeless Sanskrit grammar observes that "In every such case, it is necessary to make a compromise between measurably satisfying a present pressing need, and doing the subject fuller justice at the cost of more time." ACIP has had to balance the need to complete a reasonable amount of data with a desire for absolute accuracy, all within a modest budget and even more modest staff, almost totally volunteer. ACIP would like to call for the participation of any researcher who can make a serious volunteer commitment of time and effort to the project. The material is invaluable and the rewards self-evident; if you would like to sign on, please contact either the New York or Washington area office listed below. All regular users of ACIP data are encouraged to enroll in the Participating Correction Plan. As noted in the following section on the future of the project, a significant amount of additional data has already been input. The data will be released only once or twice a year to most users. You can receive individual disks immediately as they come out if you enroll in the Participating Correction Plan. To do so, keep a running record of any errors you notice as you use the ACIP data. When you have located any 10 errors, submit them to ACIP on the enclosed form. For every 5 errors that have not already been located by another user, you will be sent one free advance disk containing a recently completed text, before it is released to the general public. You may then submit corrections to this disk and again receive any newly created data in advance of public release. Users who do not submit lists of errors will not be eligible for the advance releases. Like the TLG project, ACIP will re-issue the past texts from time to time in versions corrected with the help of errors submitted by users. {B}X. Creative Applications for ACIP Data{b} Searching for citations and the locations of texts in the native collections are basic uses for the ACIP data discussed above. Here is a short list of ideas for other applications. Print out a list of syllables as the basis for a comprehensive dictionary. Create a concordance of definitive occurrences of specialized terms, their literal explanations, or members of the traditional Asian lists (the "five this," the "four that"). Input a Tibetan-English or Sanskrit- English dictionary that resides below a text so definitions of unknown words can be called up in a window. Compare the native indices for the Kangyur and Tengyur to each other and isolate differences. Search them for the word {SC}THE TSOM{sc} ("doubt") or {SC}ZER BA{sc} ("claim") to find occurrences of disputed authorship. Find alternate names for Asian authors under {SC}MING GZHAN{sc} ("a.k.a."). Create a genealogical tree of the early scholars with proximity searches of names. Isolate examples of the oldest language by searching for {SC}RNYING{sc} ("old") near any variant of the root {SC}GYUR{sc} ("translate"). Find out which texts were translated from the Chinese ({SC}RGYA DPE{sc}) or other non-Indic languages, and compile a list of those that are described as possibly being of native Tibetan authorship ({SC}BOD MA{sc}). Construct a geographical index of older place names with {SC}YUL{sc} ("state"), {SC}PHYOGS{sc} ("area"), {SC}GRONG KHYER{sc} ("city"), and {SC}GTZUG LAG KHANG{sc} ("monastic center"). Do statistical analyses of the ratios between common particles, to help resolve questions of authorship. Immediately delineate how different philosophical schools treat a common concept such as causation ({SC}RTEN 'BREL{sc}). Help date authors and texts by confirming inclusion of excerpts from one work in another. Do metrical analysis with a study of spaces and punctuation spaces. Create a generative grammar for Tibetan by analyzing verb forms and derived nouns in older literature, then searching the entire database for anticipated additional forms. Find the odd {L}hapax legomenon{l} (supposed single occurrence of an unknown word) again, in a context which defines it. Compare different editions of the same classic on screen, for insight into the ancient translators' methods, and thereby refine our own. Use the transcription as a basis for a new annotated edition. Print out inexpensive copies of portions of text for use by a small group or class. Call up a colleague and send an interesting section over the phone by modem. Copy the disks and keep a collection in your department or library for free public use. {B}XI. Future of the Project{b} The immediate future goals of ACIP are well under way as you read this brochure. The massive {I}8,000 Verses on the Perfection of Wisdom (A{-}rya{-}s{.}t{.}asa{- }hasrikaprajn{~}a{-}pa{-}ramita{-}),{i} {this material in parentheses should all be on one line if possible--if not possible, I will split for you} perhaps the most important classic of Asia, has been entered in full and received for domestic editing. (Again, ACIP users may receive an advance copy of the data if they fill out and return the enclosed Participating Correction Plan form.) A large commentary to Master Haribhadra's explanation of the {I}Jewel of Realization (Abhisamaya{-}lam{.}ka{-}ra),{i} listed above, has also been completed and received. ACIP quality assurance manager Dietrich Gewissler is currently at the overseas center in Sera Mey Tibetan Monastic University, working with center manager Ven. Thupten Pelgye to organize the input of related reference tools, beginning with an extensive Tibetan dictionary. Other on-line tools await development; a status bar for cursor location within the native catalogs would be wonderful, as would the addition of Sanskrit names and (where necessary) page number and other background data. We also aim at continued correction of the first form of the data as more original prints become available. Long-range goals of the project, should it continue, are the ongoing entry of the Kangyur and Tengyur collections of Asian classics in Tibetan translation, followed by a body of major native Tibetan commentarial literature to be defined in a third collection called the Sungbum. Concurrent with these important works of philosophy, linguistics, the fine arts, and science, will be entered further supporting reference works such as Tibetan-Sanskrit lexicons, histories, geographical and bibliographical resources. These latter stages assume sufficient demand for and support of this type of database; not necessarily financially, since the project has sufficient funding for its current activities, but philosophically. Undertakings such as ACIP can be perpetuated only if they perform a useful function for a sufficiently broad group of people. The staff of ACIP would like to receive your comments and suggestions on the improvement of the data, and especially concerning which texts you personally would find useful to have on disk in the coming years. For this purpose, a reply form is included here for you to complete and return. Submitting this form will also assure that you receive new data as it comes out, in the format you have requested. Again we welcome any offers of participation in the project. There are no honorary posts and to be accepted you will have to make a substantial commitment of time and effort, without pay, but very possibly with financial support for any expenses you incur in taking part. ACIP is also willing to entertain requests for expenses support from any researcher who proposes a creative contribution to the data and tools for using it. Contact the appropriate address below for any needed information, and please be patient in waiting for your reply. We have only a small staff with little extra time. To order disks, brochures, ACIP programs, print- outs in native script, or for general project updates, contact: The Asian Classics Input Project Washington Area Office 11911 Marmary Road Gaithersburg, Maryland USA 20878-1839 telephone: (301) 948-5569 For questions on editorial content, text input schedules, project participation, submission of errors found, or support of proposals for related efforts, contact: The Asian Classics Input Project New York Area Office c/o The Princeton Club of New York Box 57 15 West 43rd Street New York, New York USA 10036 telephone: (201) 364-1824 {B}XII. Credits{b} {This section should be put on a single last page by itself. Smaller point size OK. If it is practical, put it in two parallel columns which I'll call A and B here. I suggest all italics} We would like to recognize the contributions of the following individuals and institutions towards the success of the Asian Classics Input Project to date. We apologize in advance for any omissions. {In column A, on the left side of the page, list:} Ven. Geshe Lobsang Tharchin, editor-in-chief The David and Lucile Packard Foundation Dr. Colburn Wilbur, director The Packard Humanities Institute Dr. David W. Packard, chairman of the board and president Dr. John Gleason, (former) associate director Dr. Stephen V.F. Waite, associate director Dr. Brigitte Comparini, administrator The Andin International Diamond Corporation Mr. Ofer Azrielant, president Mrs. Aya Azrielant, executive vice president The Mahayana Sutra and Tantra Center Rashi Gempil Ling Kalmyk Mongolian-American Society The United States Library of Congress Asian Division Mr. Louis Jacob, head, southern Asian section Ms. Susan Meinheit, Tibetan specialist, southern Asian section Overseas Operations Division Mr. E. Christian Filstrup, chief Mr. Frederic Protopappas, cataloging coordinator Loan Division Mr. Christopher Wright, acting chief Cataloging Distribution Service Ms. Susan Tarr, chief Ms. Deborah Ramsey, supervisor, marketing unit Mr. Lloyd Lewis, library customer services analyst The New York Public Library Oriental Division Dr. John Lundquist, chief Mr. Francis Paar, Tibetan specialist *ACIP would like to express particular thanks to the staff of NYPL for continually facilitating access to necessary texts throughout the project to date The Monmouth County, New Jersey, Public Library Mrs. Tracey Lyon Curtis, request librarian {in column B, on the right side of the page, list:} The Board of Directors of Sera Mey Tibetan Monastic University Ven. Geshe Lobsang Ngudup, abbot Ven. Geshe Lobsang Pende, member Ven. Geshe Jampel Lodro, member Ven. Geshe Gendun Gyaltsen, member Ven. Geshe Lobsang Gyaltsen, member Ven. Thupten Paljor, secretary Ven. Tenzin Drakpa, member Ven. Chophel Yonten, member Ven. Lobsang Chophel, member The Trustees of the Sera Mey Scholarship Fund Ven. Jampa Rabgye, director Ven. Geshe Konchok Palden Ven. Ngawang Kelsang Ven. Thupten Pelgye The Staff of the Sera Mey Computer Center Ven. Thupten Pelgye, manager Ven. Ngawang Rigdrol, leader, team A Ven. Jampa Namdrol, leader, team B Ven. Ngawang Jungne, assistant leader, team A Ven. Jampa Tsering, assistant leader, team B Entry team A: Ven. Tenzin Nyima, Ven. Ngawang Gyatso, Ven. Ngawang Tharlam, Ven. Jampa Kelsang, Ven. Thupten Choden, Ven. Jampa Yonten, Ven. Jampa Dondrup Entry team B: Ven. Lobsang Nyingje, Ven. Thupten Tsondru, Ven. Ngawang Thupten, Ven. Jampa Monlam, Ven. Ngawang Konchok, Ven. Tenzin Donyo, Ven. Thupten Nyima, Ven. Jampa Lekchok, Ven. Lhundrup Dargye The Apollo Computer Company, Ltd, of Bangalore Mr. Ajay Patel, president Mr. Y. Subramanyam, entry instructor Mr. S. Sundeep, entry instructor Mr. M.D. Vasudevan, entry instructor Project Advisory Staff Prof. Samuel D. Atkins, chairman (retired), Department of Classics, Princeton University Prof. William R. LaFleur, chairman, Department of East Asian Languages and Cultures, University of California at Los Angeles The Asian Classics Input Project, Washington DC Area Office Dr. Robert Taylor, assistant project director Mrs. Barbara Taylor, Asian fonts design Mrs. Margaret Brenner, printing assistant The Asian Classics Input Project, New York Area Office Mr. Michael Roach, project director Mr. John Malpas, chief programmer Mr. Dieter Gewissler, quality assurance manager Mr. Dejung Gewissler, shipping manager {the following will be on one side of a loose single sheet of ACIP stationery included in the package): ASIAN CLASSICS INPUT PROJECT Date ______________ PARTICIPATING CORRECTION PLAN Please list below a minimum of any ten errors you have found in the ACIP data received to date. Whenever you have submitted any five errors not already located by another researcher, you will receive from ACIP a complimentary advance disk of the latest data, prior to its release to the general public. Users who do not submit this form will still receive the regular issues as they come out, but will not be eligible for advance copies. file name text name folio line error correction ========= ========= ===== ==== =========== =========== 1___________________________________________________________ 2___________________________________________________________ 3___________________________________________________________ 4___________________________________________________________ 5___________________________________________________________ 6___________________________________________________________ 7___________________________________________________________ 8___________________________________________________________ 9___________________________________________________________ 10___________________________________________________________ 11___________________________________________________________ 12___________________________________________________________ 13___________________________________________________________ 14___________________________________________________________ 15___________________________________________________________ 16___________________________________________________________ 17___________________________________________________________ 18___________________________________________________________ 19___________________________________________________________ 20___________________________________________________________ Please provide the following information: Name_________________________________________________________ Mailing address______________________________________________ Phone________________________ Fax __________________________ Type of computer_____________________________________________ Preferred disk: ___5-1/4" DD (360K) ___3-1/2" DD (720K) ___5-1/4" HD (1.2M) ___3-1/2" HD (1.4K) Areas of interest (we will try to provide something relevant to your specialty)_________________________________________ {the following will also be on an enclosed loose sheet, printed back to back} ASIAN CLASSICS INPUT PROJECT RESPONSE FORM DEMONSTRATION DISKS, 1990 Dear Asian specialist: We have invested considerable time and funds to produce and distribute the enclosed complimentary demonstration disk and brochure for the computer version of the Asian classics in Tibetan translation. We would very much appreciate your taking a few minutes to share your suggestions by completing and returning this form. By doing so you will automatically be registered for mailings of new data in the coming months. These will include the famed {I}8,000 Verses on the Perfection of Wisdom{i} (c. 500 BC) and commentaries to the {I}Jewel of Realization{i} by Maitreya (c. 350 AD). Your response will also help us determine whether we should continue with this type of project over the long term. {I}Address information{i} Name________________________________________________________ Institution______________________________Title______________ Office address______________________________________________ Office telephone and fax____________________________________ Home address________________________________________________ Home telephone and fax______________________________________ {I}Computer information{i} What kind of computer do you use?___________________________ What kinds of software do you regularly work with?__________ ____________________________________________________________ If the project continues, would you like to receive new data as it comes out?__________________________________________ What kind of disk should we send you? ___5-1/4" DD (360K) ___3-1/2" DD (720K) ___5-1/4" HD (1.2M) ___3-1/2" HD (1.4M) {I}Comments and Suggestions{i} Did you find the data on the demonstration disks accurate? (Please also refer to the enclosed Participating Correction Plan form.)___________________________________ ____________________________________________________________ Were you able to make the disks work properly without much trouble?_________________________________________________ ____________________________________________________________ (please see back) Did you find the format clear and easy to use?______________ ____________________________________________________________ What classical texts are you interested in seeing made available on future ACIP disks?__________________________ ____________________________________________________________ What reference materials would you like to see made available?________________________________________________ _____________________________________________________________ Can you name three specific uses you intend to make of these demonstration disks?______________________________________ _____________________________________________________________ Do you think that the effort to produce these disks will have any important value? And how so? ________________________ _____________________________________________________________ _____________________________________________________________ Would you like us to provide a similar package to any colleague? Please give name and correct address. ________ _____________________________________________________________ _____________________________________________________________ Do you have any other questions or comments about the project? _________________________________________________ _____________________________________________________________ _____________________________________________________________ Please submit to: Asian Classics Input Project Washington Area Office 11911 Marmary Road Gaithersburg, Marlyand USA 20878-1839 telephone: (301) 948-5569