CANTERBURY TALES — Volkswagen Pointer Variant II

14 Мар 2015 | Author: | Комментарии к записи CANTERBURY TALES — Volkswagen Pointer Variant II отключены
Volkswagen Pointer Variant II

An Electronic Reading-Text of Chaucer’s Tales

By Paul G. Remley

1993 by the author.

T he capability of to mark the beginning of the third by inaugurating a new phase of humanistic was recognized long ago. addressing the subject of computing in the have appeared frequently the course of the past twenty-five 1 and advanced research on the use of computers in scholarship is in its third decade.2 The Computers and the Humanities was established in and numerous serial publications since followed its lead.3 In the of computer-assisted pedagogy, there has a proliferation of computational innovations the curriculum.4 Nevertheless, after than a quarter-century of intensive much of the initial promise of use in higher education remains For a number of reasons, notably the discrepancy between the pace of in computer technology and increases in resources for academic hardware the classrooms of the future at most are still in early phases of Progress in another practical that of computer-assisted textual has also been slow. intensive work in the field, few procedures for the treatment of texts the aid of computers have achieved any of wide acceptance.5 Plans to computers in the production of critical of texts, catalogues of manuscripts, literary databases, stemmatic of textual variants, digitized of manuscript leaves, and so on, have all announced, but specific applications remain accessible only by groups of specialists or have to emerge altogether.6 With the exception of the field of computer-assisted which has already succeeded in generating a wide range of and producing a substantial amount of results,7 it seems reasonable to that humanistic computing is at an early stage.

Many of the that might be adduced to for these circumstances are wholly the public at large has had very access to powerful, multifunctional and the cost of disk storage dear, especially in comparison to of traditional media. To date, simply has been no pervasive of traditional printed texts by encoded text-files for purposes of the and consultation of texts. It is still not clear that the impact of the reading of electronically transmitted will ever be as great as occasioned by the transition from, oral to literary transmission of or the supersession of the scriptorium by the printing-plant.8

are a few signs, however, that changes may be imminent. As the footnotes to article attest, recent have seen a huge in the number of publications on computer-assisted textual research, and related A significant amount of recent on computer-assisted analysis of texts has undertaken in service of research on literature. 9 The publication in computer-readable of such standard (and, sometimes unwieldy) reference as the second edition of the Oxford Dictionary. the Modern Language International Bibliography. the Corpus (CETEDOC), the Riverside Chaucer. and conspicuous diskette releases major university presses. A few and almost indispensable textual have only appeared in form (e.g. the IBYCUS and Thesaurus Linguae Graecae materials assembled for the Dictionary of Old etc.). The past few years also seen distribution of scholarly journals such as the Mawr Classical Review international electronic-messaging networks and a and diverse group of computer in all parts of the world have in become acclimated to reading on computer in the course of their exchanges of electronic mail. All of developments have profound that may well become widely recognized as the years Since 1990, moreover, has been an enormous increase in the of generally available microcomputing as well as a concurrent drop in It appears that the point may be reached at which computer-literate and researchers will be able to powerful applications suited to own work on an ad hoc basis. The general of flexible authoring systems has access to a wide variety of computational routines in the hands of an number of users.10

The main of the present discussion are pedagogical. The goal of the project outlined is the production of an electronically-encoded reading-text of Canterbury Tales. specifically to assist undergraduate students in this medieval work in its (Middle English) language. The text differs fundamentally say, the diskette release of the Chaucer in that it lays no to the status of a critical edition and is to serve mainly as a tool in the The following discussion, however, also address a number of that have arisen in the of the development of the text that more directly on areas of in textual studies (such as homography, and full-text searching by as well as several kindred disciplines (e.g. lexicography and editing). Above all, I tried to incorporate into the essay a bibliographical summary of in these areas that appeared over the course of decades. Indeed, this may be one of the opportunities to undertake such a in the space of a single article.

The of Chaucer’s Canterbury Tales

The set of that distinguishes the classroom of the Canterbury Tales described from other editions is circumscribed. The appearance of the user described in more detail in most essentials resembles of a conventional printed book of The Middle English text of the normally provides the only visible to the user on the display of the Students may obtain a Modern explanation of any word in the Chaucerian at any time by selecting it on-screen by of a trackball, mouse, or similar device. This explanation or regularly contains a generous and selection of possible synonyms for the term and a brief linguistic (e.g. noun or comp. Students may, in addition, through passages containing all of a word or phrase of their by searching for a specified string of Passages containing groups or of several different terms may be tracked down easily in the text. It is only necessary to multiple terms when the search. A search for multiple will only succeed the terms are found within a user-specified number of lines. So apart from the circumstance it can only be consulted on a computer, the described here would to offer few capabilities beyond already available in any standard edition of Chaucer containing a set of marginal glosses and a decent The reading-text includes two specific however, which set it apart the standard classroom editions. although all searching results in the of passages of Middle English, the specified for a search may be drawn the vocabulary of either Middle or Modern English. This capability is made possible by the that all searching should place simultaneously in (1) the Middle text and (2) an off-screen buffer, below the shadow-text, which the Modern English treatments of terms that appear in the glosses described above. The process may be termed proximity-searching by 11 One beneficial result of this is that the specification of a particular of Modern English terms in almost every instance a number of different passages individual consideration. Passages are on the basis of their inclusion of corresponding to all of the terms specified for the and their proximity within a number of lines. As the merits of must be evaluated independently by the the whole system has a tendency to students to explore the text of the at will and to evaluate discrete critically with little or no on the part of the instructor.

In the course of development, each of the two features outlined here—e-screen of vocabulary and proximity-searching by collocation—has previously unsuspected benefits. The correspondence resulting from the of supplying each word of the Tales with its own gloss in greater detail below) has in defined a generalized textual (or document architecture ) capable of a scholarly apparatus of virtually any of complexity, precision, or erudition. to the glossing apparatus, however, in every case be introduced the labors of an expert, so this flexibility should for the moment be a long-term benefit of the system. The to carry out multiterm searches of a English text by specifying terms in Modern English a benefit that is more available. The passages set out for the user’s in many cases embody amount to conceptual equivalents of the semantic range of the terms for the search. The Middle English is revealed in a way hitherto unattainable the use of conventional glossaries and lexicons. It an easy matter for the student (or to search the literary text for an infinite variety of themes, topoi, examples of medieval descriptio, and the like, without the of undertaking an arduous program of

Beyond the fairly restricted set of uniquely available within the environment, the model used for the reading-text is conventional and largely in with the history of texts back to the codex and scroll. The computational metaphor of the project is of the book. After launching or the reading environment containing the readers are first confronted a decorated cover illustration. may then turn to a Table of containing coded pointers to the of individual texts, allowing to move to the beginning of a particular end-link, or Tale by selecting its with the pointing device. the unit of the book, the fundamental of the electronically encoded text are the and the line. 12 The sense in which terms are used here little from that in references to traditional parchment or documents. There are twenty-five to a page; page-breaks are mainly page-breaks occur only at the of discrete head-links, end-links, and to the dimensions of the page and the custom that appears on the computer A variable-width, Times Roman-based is aligned with the left (or both margins in the case of prose); pages are also with graphical approximations of running heads and page One non-traditional feature that make special use of the computer is the provision of limited scrolling on individual pages. Although the turning to a particular page for the time, is presented with twenty-five lines of text, up to additional lines immediately and following the visible excerpt are off-screen in buffers or scrolling This scheme offers two benefits. First, the reader, on the bottom of a page, may choose to one or several additional lines jumping to another part of the of the Canterbury Tales or concluding a session altogether. (If, the reader decides to go to the next after all, the words at the top of that page will be immediately following the last text on the preceding page.) the inclusion of hidden regions scrolled text on each a brute-force solution to the problem of for collocations across page-breaks.

the project summarized here is essentially as a fait accompli, the phases involved several starts and a large amount of and error. The specific procedure during the developmental process may, in practice, never and some technical specifications for the itself may now be set out concisely. Although it be stressed at the outset that is proposed here is essentially a system, the developmental text for the project was based on that in two nineteenth-century printed editions, those of W. W. Skeat and Thomas The developmental text was encoded by my collaborator Eric Juvet and to constitute what amounts to a prepared private edition.13 We to release a publicly-distributed form of the as soon as possible, perhaps by the end of the This will embody a new edition of the Middle English of a critical standard, which to the current design specification comprise a semi-diplomatic edition of the based on the readings of a single

The ASCII-encoded text file by Juvet occupied 968,252 (inclusive of punctuation), thus almost a full megabyte of The file contained a total of words. The first step in the phase of the effort involved the of this full text of the Tales to produce a lexicon of strings through a simple of pattern-matching. (If the program has seen the before, it was discarded; otherwise it was to the lexicon of unique strings.) this sorting process had completed, the one-megabyte text nearly two hundred thousand had yielded only 12,181 strings. These strings then sorted into a database (described in greater below), i.e. an arrangement of containing, among other a single headword (or lemma ) and a accounting of all inflected and variant of the word encountered in Middle texts to date.14 (The of building and maintaining the lemmatizing is in fact a protracted process; the will surely continue to capable of receiving further almost indefinitely.) When all of the strings had been integrated the lemmatizing database, the records a vocabulary of around six thousand for Chaucer. The figure may seem low in to a vocabulary of about thirty words for the typical Ph.D. but it may in fact seem rather when it is noted that the is distributed throughout a single work.

It is perhaps too early to say where the reading-text of the Canterbury described here stands in to the main progression of computational of medieval texts.15 As far as I can ascertain, few applications of computer technology to the canon have been to date, and in most of these the English text supplies the example for a fairly restricted of a particular methodological approach.16

and Homography

The first step in the mechanism allowing users to glosses to individual words of the Tales involves a process of sorting generally known as Although researchers who have the issue at all invariably conclude the term describes the nature of the to which it refers imprecisely at lemmatization is regularly taken to the association of all forms of a given that occur in a continuous whether inflected or not, a normalized citation form or which is analogous to the headword or element in a dictionary entry.17 will note that the use of the lemma in this sense comparison with the nuance of the Latin term interpretamentum well as certain aspects of the glossa ). Systems of lemmatization attracted a substantial amount of in recent years, mainly in with analyses of texts in modern languages (notably Dutch, and French)18 and attempts to systems capable of performing translation by machine.19 The lemmatization of texts, however, has been less often, presumably of its frequent status as a target in machine-translation experiments.20 In recent research, the process of lemmatization has crucial in attempts to achieve interpretation of continuous prose in advanced applications, speech) by most of which fall the rubric of natural-language processing.21 scholars and lexicographers have had occasions to address the issue for reasons. An exhaustive lemmatization of a constitutes a preliminary stage in the of a lexicon or, say, a glossary for a edition. It is in this connection most of the work to date on the of medieval vernacular texts has studies exist for Medieval Old French, Old and Middle High several dialects of medieval and Old Icelandic (but not, to my Middle English).22 In the course of the of the reading-text of the Canterbury Tales. provided a preliminary means of the precise definition of specific of the Middle English text, facilitating the intelligent interpretation of passages.23 Indeed, lemmatization may be as the main component of the whole since it is only through the of a given string with a that the on-screen text can be with its proper gloss. the link has been established, the of the electronically encoded text is in capable of accommodating an expanded capable of achieving virtually any of scholarly precision.

As all words in the text of the Canterbury are treated initially by the lemmatizing as discrete strings of characters, arise whenever two distinct share a single spelling. All in the field of lemmatization have that the phenomenon of homography the concomitant ambiguity of meaning it engenders) constitutes a major to performing an efficient and accurate of any continuous text in a single The problem of homography as a whole may be into many subcategories. are special problems, for example, in the of terms from certain categories (e.g. participles, and multiple terms written in English as a single discrete Cases of homographic ambiguity traditionally been resolved in one of two The most direct method, the intervention of an expert reviser, is the most time-consuming. Beyond complex algorithms have developed that achieve a degree of precision in the resolution of according to context in the treatment of languages. No system of lemmatization, to my has yet laid claim to a level of accuracy.26 In the treatment of medieval moreover, it is by no means clear a system of fully automatic is a practical or even desirable The corpus of texts preserved in a medieval vernacular may in most be viewed as finite and fairly Given the variation in orthography, and punctuation observed in most vernaculars (and many texts as well), the development of an system of algorithmic contextualization require scarcely less than a thorough appraisal by an reviser. The whole question of the of orthographic ambiguity in the computer-assisted of medieval texts deserves attention that can be offered

No ideal method for dealing homographic ambiguity has come to in the course of the present project and is likely to emerge at any point in the future.27 In accordance with the of other literary scholars who used computers in the preparation of however, a system of semiautomatic has been adopted as a compromise The first stage in this which may be termed prelemmatization, the assignment of groups of generalized, definitions to homographic strings. this is an unending process; in the case of medieval literature, homographs continue to come to throughout the course of the preparation of the database.) For example, the Middle homograph glede, which represent a form of an adjective cheerful, radiant), a verb (to glad), or any one of several nouns burning coal, or kite). The glede would thus be automatically at the prelemmatization stage a group of possible single-word glad cheer happiness bird. The substitution of the general fire for burning coal and for the more specific kite a significant feature of the design of the resource. By assigning a group of glosses to a specific homograph displays a broad semantic it is possible to begin the preliminary of proximity-searching by collocation (employing English search terms) at the prelemmatization stage.

Although it not remove the necessity of the eventual of an expert reviser altogether, the strength of the system of prelemmatization out here is that it remains a automatic process right up to point. It allows the textual within hours of receiving a new of a computer-readable text, to associate word of that text its appropriate lemma and gloss. The drawback to the scheme is that it results initially in the association of character strings with (and thus definitions) are incorrect and must eventually be by an expert reviser. In practice, the of precision incurred by the association of definitions with individual is not great. The process in fact the expert reviser at the outset a predefined set of choices to assist in the precision of the electronically encoded By supplying homographic strings single-word definitions whose compass as broad as possible, it has possible in practice to carry out proximity-searching once the initial of prelemmatization has been completed. The has convenient access to a range of for a given Middle English and the literary scholar is immediately in a good position to search for concepts, or topoi.

The lemma, in the described above, serves a role in the production of the Chaucerian It provides a handle allowing information—most commonly definitions, and the like—to be linked to particular of Middle English. The information in viewed collectively, has been to here with some vagueness as the gloss. It is worth that the gloss in the electronically text of the Canterbury Tales for that matter, any other prepared in a similar manner) in theory contain any type of Though in the present instance the of the gloss have been to the part of speech and a range of synonyms, expanded versions of the might accommodate a single translation, a detailed etymology, a set of variants, a pedagogical commentary, a personal notes, or any kind of illustration, or animation.

Shadow-Text and

One crucial aspect of the reading-text has yet to be the procedure allowing individual to be associated with specific in the Canterbury Tales. It was noted that the main part of the encoded text may be described in of two fundamental structural units, the page and the line. This architecture in fact carries to the glossing apparatus. The glosses in every instance on the same as the continuous Middle English to which they refer, but are hidden from the reader in an field or buffer that is envisioned as a grid or matrix of that stand in the background of the text. The collected glosses for a page of the Canterbury Tales are to here as the shadow-text. The process of the constituent elements of the normally shadow-text to the individual words in the consulted by the reader requires the of a third basic structural in the reading-text: the word. Each of the shadow-text includes precisely the number of lines and words as the page of the reading-text. Words in the text of the Tales are defined as of characters bounded by blank (regularly assigned an ASCII of 32) or, in cases of line-breaks, hard (ASCII 13). Words in the however, are defined as strings of of any length and containing any number of (in the traditional sense), spaces, and the like, separated by one of the less used seven-bit computational (in the developmental version, ASCII These are accessed in response to generated by the user. For example, if a selects word 4 of line 8 of 3 in the on-screen reading-text, the glossing immediately fetches the corresponding in the shadow-text and renders it visible on the display.

One immediate objection to the scheme set out would be that it greatly the size of the text file to store the text of the Canterbury Every word in the main has its own gloss and many of these are identical and ostensibly redundant, as they do in response to multiple of a given string or lemma. The ASCII file used to the developmental version of the reading-text here expanded to more eight megabytes when with the full apparatus of the An arguably more efficient would involve a greatly shadow-text that only pointers to individual records in a which would presumably be in the form of a flat-file database. objection has been set aside for two reasons. First, the proposed architecture is intended to be capable of an apparatus able to achieve a degree of scholarly precision, in the treatment of unusual senses of and idiomatic phrases, which in practice comprise any number of in various lines of the reading-text. The of glosses in a database would not preclude the achievement of this of precision, but it would greatly both the initial developmental and any future revision of the text. to the plan set out above, however, any or phrase in Chaucer’s text may receive any sort of treatment an expert reviser deems Second, the redundant and seemingly repetition of similar glosses at points in the shadow-text is in fact the feature of the system that searching for themes, concepts, and so on, to be by the reader on an ad hoc basis. The key to the system of it will be recalled, is the availability of of terms. One of the most striking unexpected) conclusions to emerge the present course of research is the that the arrangement of the fairly lemmatized synonym sets constitute the glosses described in the form of a shadow-text immediately an extremely powerful searching that provides useful even before the intervention of an reviser. This conclusion may ramifications for future work on medieval texts.

The Relationship of the to Hypertext

The association of the so-called to the text of the Canterbury Tales comprising fairly concise of synonyms and near-synonyms) with terms in a shadow-text effectively in a greatly expanded form, a new approximation of the Chaucerian lexicon. the expansive qualities of this after some consideration the hypertext has been excluded the present discussion in favor of the contrived, if slightly awkward, electronic [or, better, encoded’] reading-text. The longest-standing of hypertext, primarily associated the visionary theorization of Theodor Nelson, consists in two main (a) an addressing scheme for the whole of literature and every other of text, intended to facilitate the and retrieval of digitally encoded of every kind; and (b) a system for text in compound documents.29 The formulation of hypertext amounts to may be termed a full-text model in no instance of text is actually to or reproduced in a linked document. might appear at first to be a quotation of, say, five of the Canterbury Tales is in fact a of the original edition of the text of lines, stored in a critical, encoded master text of the canon; it is not a copy or representation of lines. In another formulation, the simply points to two addresses in the text, representing the beginning and end of the excerpt and providing a window, as it to those five lines. who wish to investigate, say, a interpretation of the lines in question may simply go into the citation, is, to move from the citation to the text of the Canterbury Tales which it is drawn, where may read the five cited in their full original Similarly, references to the work of scholars in the footnotes of our hypothetical would supply pointers to the texts of published versions of the research. Clearly this of hypertext is far removed from the reading-text described here, in terms of specific design and anticipated imminence ofavailability. text might be modified to one node in such a system, but it be held to constitute an example of in its own right.

In recent literature, the of the problematic term hypertext, an infelicitously contrived Greco-Latin that continues to dissonate in the of many scholars, has become blurred.30 The introduction of methods of sound and graphic art into presentations has exacerbated the situation, rise to an even more neologism, i.e. hypermedia. 31 The for distancing the Chaucerian text discussion from these however, rests on criteria firmly grounded than aesthetic objections. At least the appearance of Cortazar’s multistranded Hopscotch. experiments in hypertext involved the introduction of specific of passages by an author. Such may be said to rely on a system of links. 32 The shadow-text has been so as to facilitate the association of words, and concepts through the use of implicit than explicit links. Table of Contents mentioned contains the only instance of the use of links in the system.) This substitution of implicit links for links—obtains in the cases of both the of the gloss to the reading-text and of entire to groups of search terms.

though it cannot be taken in as an example of hypertext, the approach out here may serve to counter of the objections to the viability of hypertext have been raised in years. There are, all, many purely difficulties involved in the establishment of an system of explicit links in a system. A single page of a text or academic publication in an ideal system, require if not thousands of links to be drawn up. As far as has yet been able to ascertain, links would have to be one at a time either by the author of the document or an expert reviser. raises certain questions of since the reader’s subjectivity is in placed in the hands of unseen whose levels of expertise in any area that may be of interest to the are unknown. Even if a comprehensive of explicit links could be successfully in a hypertext system, is a second, even more affective side effect from the use of such links. I characterize this as a buried or secret door syndrome. systems have the potential to a sense that there is just out of reach, some or interpretation that lies the passage at hand but whose nature is hard to fathom. In an indexed system, the sheer of choices might well unsettling. It would be hard for the to know which way to turn.

Hypertext, while abandoning the model of text promoted by the scriptorium and printing press, may produce on occasion the unfortunate of promoting compartmentalization and producing a sense of isolation in the user, conveying a sense of being by an unseen, external agent. The prospect of achieving something omniscience—or, at any rate, synchronous to a diverse range of texts—by texts on computers renders the of existing hardware and software all the frustrating. Most users are forced to consult single of textual data sequentially.

The used here for the production of the of the Canterbury Tales addresses of the concerns raised above. its employment of a process of prelemmatization, it a virtually infinite number of links without the intervention of an agent. Readers are free to these at will by formulating in any manner that they The fairly simple system of glossing introduced in the present produces a result that is, on the one absolutely predictable—the reader who a word will invariably a group of terms in a gloss exemplifies a consistent style—without the variety of choices available to the and, arguably, greatly their number. Most I think would agree the ability to flip through the of a book at random, to scan the surrounding the one you are seeking in the shelves of a stack, to notice the title of an quite by chance on the cover of a is every bit as important as the systematic of sequences of continuous prose. It be noted, however, that books and libraries reached present state over of trial, error, and refinement. of textual computing will address their present and continue to evolve. The project here is intended as a small and contribution to the task at hand: the use of computers more precisely in with the way students and scholars in their daily routines.33

University of Washington


1 include Computers in Humanistic ed. Edmund A. Bowles (Englewood Prentice-Hall, 1967); Computing in the ed. Serge Lusignan and John S. (Waterloo, Ontario: University of Press, 1977); Computing in the ed. Richard W. Bailey (Amsterdam: 1982); George M. Kren and Christakes, Scholars and Personal Microcomputing in the Humanities and Social (New York: Human Press, 1988); Rudy Steve Smithson, and Diane Microcomputers and the Humanities: Survey and (New York: Ellis 1990); Humanities and the Computer: New ed. David S. Miall (Oxford: Press, 1990); Scholarship and in the Humanities. ed. Mary Katzen, Library Research (London: 1991); and Humanities Research Computers. ed. Christopher Turk Chapman, 1991).

2 On the use of computers in and textual studies, see Robert L. Computer Methods for Literary (Columbia: University of South Press, 1980; 2nd ed. Athens: of Georgia Press, 1984); in Literary and Linguistic Research. ed. L. and C. Peters, supplement to Linguistica 3 (1983) (Pisa: Giardini, L’ordinateur et les recherches littéraires et ed. Jacqueline Hamesse and Antonio (Paris: Champion: 1985); B. H. and T. Corns, Computers and Literature: A Guide (Tunbridge Wells: Press, 1987); Nancy The Relevance of Computational Linguistics to Studies, Computers and Texts. 1 Computers and Written Texts. ed. S. Butler, Applied Language (Oxford: Blackwell, 1992), John F. Burrows, Computers and the of Literature, at pp. 167-204; and M. Deegan, S. and C. Mullings, Computing in Textual Computers and Education. 19 (1992):183-91; Computing and Literary Criticism: and Practical Essays on Theme and ed. Rosanne G. Potter (Philadelphia: of Pennsylvania Press, 1989).

3 that may be consulted profitably for germane to the present discussion the ALLC [Association for Literary and Computing] Bulletin (Swansea: College et al. 1973-85) and ALLC (Cambridge: Cambridge University et al. 1980-85), succeeded by Literary and Computing (Oxford: Association for and Linguistic Computing, 1986-); Humanities Computing (Toronto: for Computing in the Humanities, 1987-); Technology. incorporating Language succeeded by Electric Word Language Technology BV et al. 1987-90); Computing (McKinney, Texas: Computing Publications, 1987-90); in Literature and Computers in Literature now Computers and Texts (Oxford: CTI in Teaching Initiative] Centre for Studies et al. 1990-); and Writing on the (Davis: University of California at Campus Writing Center,

4 For details of specific applications, see Hockey, A Guide to Computer in the Humanities (Baltimore: The Johns Press, 1980); Humanities Yearbook. ed. Ian Lancashire et al. (New Oxford University Press, and CTI Centre for Textual Studies Guide, March 1992. ed. Davis, Marilyn Deegan, and Lee (Oxford: CTI Centre for Textual 1992). For general discussion of involved in text-processing, see Peter Text Specific Workstations: A Problem, Academic Computing. 4.1 and 70-72; and Ronald F. E. Weissman, In of the Scholar’s Workstation: Recent and Software Challenges, Academic 4.1 (1988-89):28-30 and 59-64. On specifically topics, see John M. Slatin, and the Teaching of Writing, in Text, and Hypertext: Writing with and for the ed. Edward Barrett (Cambridge, MIT Press, 1988), pp. 111-29; J. Hockey, Jo Freedman, and J. Cooper, in the Study of Set Texts, in Humanities and the ed. Miall, pp. 113-22; and resources and reviewed in Bits and Bytes Reviews and News of Products and for Academic Computing (Whitefish, Bits and Bytes Computer 1986-).

5 See especially Peter Smith, An Introduction to Text (Cambridge, Massachusetts: MIT Press, G. Salton, Automatic Information and Retrieval (New York: Hill, 1968); Donald E. Searching and Sorting. vol. 3 of The Art of Programming (Reading, Massachusetts: 1973). See also W. Martin, B. Al, and P. van Text-Processing and Lexicographical Information—A of the Art, ALLC Journal. 2 For specific applications, see J. McNaught, Lexicography in the Context of a British Data Bank, in Lexicography in the Age. ed. Goetschalckx and Rolling, pp. Tove Fjeldvig and Anne Experiments with Language-Based in Information Retrieval Systems, Journal of Linguistics. 11 (1988), J. K. Proud, The Oxford Text British Library Research and Report, 5985 (London: Library, 1989); G. Chartron, Management Tools for Large Databases: The Leximet System, of Information Science. 15 (1989):339-44; J. and C. Grover, The Derivation of a Large Lexicon for English from in Computational Lexicography for Natural Processing. ed. Boguraev and Briscoe,

6 Jacques Froger, La critique des et son automatisation. Initiation aux nouveautés de la 7 (Paris: Dunod, 1968); J. Computertechnik im Dienst der Edition Texte, in Probleme der Edition und neulateinischer Texte. ed. Ludwig and Dieter Wuttke (Boppard: 1978), pp. 143-49; La pratique des dans la critique des textes. ed. Irigoin and Gian Piero Colloques internationaus du Centre de la Scientifique, 579 (Paris: CNRS, Peter L. Shillingsburg, Scholarly in the Computer Age: Theory and (Duntroon: University of New South 1984); Ulrich Müller, Computer, wissenschaftliche Manuskripte und Editio. 2 (1988):48-72; Gian Zarri, Some Experiments on Textual Criticism, in Miscellanea di in onore di Aurelio Roncaglia. ed. Antonelli, et al. 1 vol. in 4 (Modena: Editore, 1989), 1439-64; Bräuer, Historische Edition und Internationale Tagung von 26. bis 30. Oktober in Graz, Zeitschrift für Germanistik. 10 and Wilhelm Ott, Computers and Editing, in Computers and Written ed. Butler, pp. 205-26.

7 See essays in Lexicography in the Electronic Age. ed. J. and L. Rolling (Amsterdam: North-Holland, and Theorie und Praxis des lexikographischen bei historischen Wörterbüchern. ed. Herbert Wiegand, Lexicographica, series 23 (Tübingen: Niemeyer, 1987); see Willem Meijs, Computers and in Computers and Written Texts. ed. p. 141-65. Many groundbreaking in computational lexicography were in the course of the treatment of medieval and modern lexicons. See Computers and Old Concordances. ed. Angus Cameron, Frank, and John Leyerle, and A for the Dictionary of Old English. ed. Roberta and Angus Cameron, Toronto Old Series 1 and 2 (Toronto: University of Press, 1970 and 1973); F. Huntsman, Computers and Medieval Lexicography, Computers and the Humanities. 12 Jürgen Schäfer, Elizabethan A Computer-Assisted Study of the Beginnings of Lexicography, I, ALLC Bulletin. 8 and II, in Computers in Literary and Linguistic ed. Cignoni and Peters, pp. 235-42; La du latin médiéval et ses rapports les recherches actuelles sur la civilisation de Colloques internationaux du Centre de la Recherche Scientifique, 589 (Paris: 1981).

8 For some speculation in this see John Slatin, Reading Order and Coherence in a New Medium, in and Literary Studies. ed. Paul and George P. Landow (Cambridge, MIT Press, 1991), pp. 153-69.

9 The example of a medieval study I found, on a Medieval Latin is that of Anezka Vidmanová, textová kritika a pocítací Listy Filologické. 92 (1969):28-52; see Paul Tombeur, Le traitement des documents et l’étude de textes (Beckmann et al. 1981), pp. 329-39; Applications to Medieval Studies. ed. Mary-Jo Arn, The Systematic of Early Manuscripts in Computer A Proposal, in Historical and Editorial in Medieval and Early Modern for Johan Gerritsen. ed. Mary-Jo Hanneke Wirtjes, and Hans (Groningen: Wolters-Noordhoff, 1985), pp. Paul Tombeur, Informatique de textes médiévaux, L’homme et univers au Moyen Âge. ed. Wenin, Philosophes Médiévaux, 26; 2 (Louvain-la-Neuve: Éditions de l’Institut de Philosophie, 1986), 1, pp. 174-86; and C. M. Text in the Electronic Age: Study and Text Encoding Examples from Medieval Literary and Linguistic Computing. 6.1 Cf. also Helmut Droop, Lenders, and Michael Zeller, zur grammatischen Klassifizierung und maschinellen spätmittelhochdeutschen Texte. Forschungsberichte des für Kommunikations-forschung und Phonetik der Universität III: Linguistische Datenverarbeitung, 55 Buske, 1976); Maschinelle altdeutscher Texte. ed. Sappler and

10 See, e.g. Patrick W. The Beowulf Workstation (Morgantown: Virginia University, Department of [1991]) and materials produced by Allen Frantzen, Clare John Ruffing, and others in with their Seafarer (Chicago, New York, Ithaca, and 1990-). Although the document proposed here is intended to be all prototyping to date has been with HyperCard software, developed by Bill Atkinson for Computer, and its native programming developed by Dan Winkler. Workstation purchased in part with an from the University of Washington Workstation Initiative and optimized for included a Macintosh IIci, cache card and eight of RAM, operating under 6.0.8 in one-bit (black-and-white) mode. HyperCard, long as slow and cumbersome, suddenly sense in such an environment and as a versatile tool capable of unbounded manipulation of large of multiple-typeface, multiple-font text. the configuration described here now a mid-range system.

11 For an elegant of related issues, see J. M. Sinclair, Concordance, Collocation (Oxford: University Press, 1991); cf. Gerald Purnelle, Recherche de groupes verbaux recurrents et de dans les fichiers latins Revue: informatique. 25 (1989):157-92.

Volkswagen Pointer Variant II

12 For an defense of this approach, see Ott, Pages and Lines: on Some Fundamental Requirements of Processing Software, in Computers in and Linguistic Research. ed. Cignoni and pp. 227-33.

13 In recent months, the has benefited from substantial by Professor Joseph Monda of University.

14 For accounts of other in this area, see N. Calzolari, Definitions in a Computerized Dictionary, and Artificial Intelligence. 2 (1983):225-34; Seebold, Die Lemma-Auswahl bei einem Wörterbuch, in Theorie und Praxis des Prozesses. ed. Wiegand, pp. 157-71; and B. Extracting Lexical Knowledge Dictionary Text, Knowledge 1 (1989):113-37.

15 As noted above, in this area is at an early The only volume-length collections I noted to date are Computer to Medieval Studies. ed. Anne Studies in Medieval Culture, 17 Michigan: Medieval Institute 1984), and Le médiéviste et l’ordinateur. de la Table ronde (Paris, 17 november 1989). L. Fossier, (Paris: CNRS, 1990).

16 include Charles Moorman, Housman’s Fleas: A Statistical of Manly’s Landmark Manuscripts in the Prologue to the Canterbury Tales , for Literary and Linguistic Computing 3 (1982):15-35, also printed in International Conference on Computers and the ed. Sarah K. Burton and Douglas D. (Rockville, Maryland: Computer Press, 1983), at pp. 431-46; M. Logan and Barry W. Miller, A for The Book of the Duchess. A Semantic of Sentence Structure, in Sixth Conference on Computers and the Humanities. ed. and Short, pp. 384-90; and Kari Rand Schmidt, Type/Token for Consecutive Units of Text as a in Authorship Studies: An Assessment Special Reference to the Attribution of The of the Planetis , in L’ordinateur et les recherches et linguistiques, ed. Hamesse and Zampolli, pp. More wide-ranging studies Walter S. Phelan, The Study of Vocabulary, Computers and the Humanities. 12 ibid. . From Morpheme to in Chaucer’s Canterbury Tales , in of the International Conference on Literary and Computing. ed. Zvi Malachi (Tel Katz Research Institute, 291-316; Eugene Green, Acts and the Art of the Exemplum in the Poetry of and Gower, in Literary Computing and Criticism. ed. Potter, pp. 167-87; M. Logan and Grace B. Logan, The of the Canterbury Pilgrims: Sentence and World View in Frag. I of The Tales , Literary and Linguistic 5 (1990):242-47; Charles Barber and Barber, The Versification of The Canterbury A Computer-Based Statistical Study, Studies in English. new series, 21 and 22 (1991):57-83.

17 For an early introductory of the subject that remains today, see M. L. Hann, Principles of Lemmatization, ITL: Review of Linguistics. 49 (1973):3-22.

18 On applications in the of texts in modern languages, see Dietrich, Automatische Textwörterterbücher. zur maschinellen Lemmatisierung verbaler des Deutsche (Tübingen: Niemeyer, Tove Fjeldvig and Anne Automatisk rotlemmatisering—et lingvistisk for tekstøking (Oslo: Universitetsforlaget, Annette Ostling Andersson, automatique des lexèmes du français Acta Universitatis Upsaliensis, Romanica, 39 (Uppsala: Uppsala Press, 1987). See also S. Spraycar, Automatic Lemmatization in ALLC Journal. 1 (1980):55-59; de Kock, De la lematización, Lingúistica actual. 9 (1987):255-56. For details of projects, see Hans Eggers, et al. Ein Verfahren zur automatischen Lemmatisierung Texte (Tübingen: Niemeyer, Christine Schneider, Lemmatisierung im JUDO, ALLC Bulletin. 8 Wolfgang Krause and Gerd Lemmatizing German Newspaper with the Aid of an Algorithm, Computers and the 15 (1981):101-13; Gerd Willée, des Algorithmus LEMMA2 zur Lemma-tisierung Wortformer, in Computers in Literary and Research. ed. Cignoni and Peters, pp. Normand Beauchemin and Michel MICRO-SOLIVO. Un Lemma-tiseur semi-automatique le québécoise parle, Revue de linguistique. 3 (1984):19-38; Nicoletta Maria Luigia Ceccotti, and Roventini, A Lexical Data for Interactive Lemmatization, in L’ordinateur et les littéraires et linguistiques. ed. Hamesse and pp. 107-14; Étienne Evrard, Le Revue: informatique. 25 (1989):206-7; Masereeuw, Les travaux de l’Universite Revue: informatique. 25 (1989):207-11.

19 For of recent advances in machine see William John Hutchins, Translation: Past, Present, (Chichester: Ellis Horwood, Derek Lewis, Computers and in Computers and Written Texts. ed. pp. 75-113; and Muriel Vasconcellos, et al. of the Art: Machine Translation, 18.1 (1993):152-86.

20 For a general see Edward F. Kelly and P. J. Stone, Recognition of English Word (Amsterdam: North-Holland, 1975).

21 Lexicography for Natural Language ed. Bran Boguraev and Ted Briscoe Longman, 1989); Natural Processing. ed. M. Eilgueiras, L. Damas, N. and A. P. Tomás (New York: 1991); Geoffrey Sampson, Language Processing, in Humanities Using Computers. ed. Turk, pp. Terry Patten, Computers and Language Parsing, in Computers and Texts. ed. Butler, p. 29-52. On applications, see Boris Katz, Processing with the START Language System, in Text, and Hypertext. ed. Barrett, pp. 55-76; F. M. Russo, M. T. Pazienza, and P. Velardi, A for Text Analysis and Lexical Acquisition, Data and Knowledge 4 (1989):1-20.

22 On lemmatization in the treatment of texts, see Hans Fix, Normalisierung: Vorarbeit zur Lemmatisierung diplomatischen altislandischen Textes, in Verarbeitung altdeutscher Texte. ed. Sappler and Erich Strassner Niemeyer, 1980), pp. 92-100; H. Die automatische Lemmatisierung frühmittelalterlicher DAI. 42C.4 (1981):697 4543C); René Pellen, Construire un dictionnaire lemmatisé l’informatique. Texte d’experience: Los milagros de Nuestra Senora , La 7 (1983):197-231; Bernard Derval, A System of Text Lemmatization to the Romances of Chrétien de Troyes, ed. Doutrelepont, in Computer Applications to Studies. ed. Gilmour-Bryson, pp. 31-44; and Najock, Lemmatization of Latin and Concordances and Word-Indexes: Problems and in L’ordinateur et les recherches littéraires et ed. Hamesse and Zampolli, vol. 2, pp. For an early modern study, see P. S. di Homographs and Lemmata in Thresor de la francoyse by Jean Nicot: A Perspective?, Quaderni di semantica. 8

23 For comparable approaches, see K. Devine and F. J. Direct File Organization for Text Retrieval, Information Research Development Applications. 3 G. David Huffman, Dennis A. and Royal G. Bivins, Generating with Lexical Association Term Uniqueness, Information and Management. 26 (1990), pp. 549-58; P. S. G. R. Krupka, and L. F. Rau, Lexico-semantic Matching as a Companion to Parsing in Understanding, in Speech and Natural Proceedings of a Workshop. ed. P. Price Alto: Morgan-Kaufman, 1991), pp.

24 See the monographs of Hans Dieter Homographie und maschinelle Sprachübersetzung. Arbeiten, 8 (Saarbrücken: Germanistisches 1969) and M. Boot, Homographie. Ein zur automatischen Wortklassenzuweisung in der Computerlinguistik Rijksuniversiteit, 1979); see also M. Homography and Lemmatization in Dutch ALLC Bulletin. 8 (1980):175-89; Beauchemin, Homographie et solutions en lemmatisation minimale, in Méthodes et informatiques dans l’étude des En hommage a Charles Muller. Etienne Brunet, 1 vol. in 2, de linguistique quantitative. 35 (Geneva: 1986), pp. 25-36.

25 A. Duro, Un problème de lemmatisation: le traitement du in Proceedings of the Second International Table Conference on Historical ed. W. J. J. Pijnenburg and F. de Tollenare (Dordrecht: 1980), pp. 117-42, and Peter J. Computer Assistance in the Editorial of Contractions in Middle English ALLC Bulletin. 9.3 (1981):9-10.

26 attempts include Philip J. Some Association-Based Techniques for Disambiguation by Machine. Computer Department Technical Report (Rochester, New York: University of 1977); Yaacov Choueka and Lusignan, Disambiguation by Short Computers and the Humanities. 19 (1985):147-57; and Taylor, Wordz that Match, Computer Language. 3 1986):47-59. On the larger linguistic see Lexical Representation and Process. ed. Marslen-Wilson (Cambridge, Massachusetts: MIT 1989); N. Calzolari and A. Zampolli, and Tools for Lexical Acquisition, in Language Processing. ed. Eilgueiras et al. at pp. For details of specific applications, see J. A. and J. L. Mitchell, SPAN: A Lexicostatistical and Some Applications, in Computing in the ed. Lusignan and North, pp. 59-71; J. On The Complexity of Lexicographic Sorting and Aplikace Matematiky. 26 (1981):432-36; L. A. Brandenburger, and E. Dekel, An Overview of Choice under Uncertainty, of Operations Research. 19 (1989):247-72.

27 One approach to the problem would existing algorithms designed for analysis; see, e.g. V. W. Zue and D. P. Computer Recognition of Isolated from Large Vocabularies: Access Using Partial Information, in Proceedings of the International on Advanced Automation, 1983. T. Tou, chair (Taipei: of Information Science, 1984), pp. Jim Howell, An Alternative to Soundex, Dr. Journal. November 1987, M. D. Riley and A. Ljolje, Lexical with a Statistically-Derived Phonetic in Speech and Natural Language. ed. pp. 289-92.

28 Maria Assumpta i Alavedra, Lematització semiauto-matitzada i gràfica de Tirant lo Blanch , Universidad de Barcelona, 1990; see DAI (1990):338-C (no. 1414).

29 Holm Nelson, Literary 90.1. rev. ed. (Sausalito: Press, 1990). Nelson’s scheme in some respects the Dewey decimal system as any address (e.g. 768.1004.3.345620987) can the basis of another by means of the of one of its numerical subsections or the interpolation of decimal point. Nelson the origin of his scheme back to an by Vannevar Bush, As We May Think, Monthly. July 1945:101-8. He cites Douglas Engelbart as a influence in the de-velopment of the concept of but Engelbart’s papers do not seem to been issued in any readily form.

30 On current notions of see Text, Context, and Hypertext. ed. George P. Landow, Hypertext in Education, Criticism, and Scholarship, and the Humanities. 23 (1989):173-98, reissued as Texts, Changing Readers: in Literary Education, Criticism, and in Reorientations: Critical Theories and ed. Bruce Henricksen and Thäis E. (Urbana: University of Illinois 1990), pp. 133-61; Hypertext: Systems, and Applications. ed. N. Streitz, A. and J. André, Cambridge Series on Publishing (Cambridge: Cambridge Press, 1990); Jay David Writing Space: The Computer, and the History of Writing (Hillsdale, New Erlbaum, 1991) and ibid. . Writing: Hypertext and the Electronic Space, in Hypermedia and Literary ed. Delany and Landow, pp. 105-32; Berk and Joseph Devlin, ed. The Handbook 1991 (New McGraw Hill, 1991). See [Bibliography:] Hypertext and Hypermedia, in CTI for Textual Studies Resources March 1992. ed. Davis, and Lee, p. 68, and Adam Hodgkin, of Hypertekst, tr. Harald Engelstad, 45.2 (1991), 62-64.

31 On see Hypermedia and Literary Studies. ed. and Landow; J. Bradley, Research in Information Technology: Hypermedia, Humanities Computing. 3.1 (1989):4-8; and Younggren, Using an Object-Oriented Language to Create Audience-Driven Events, in Text, Context, and ed. Barrett, pp. 77-92.

32 On the problems of see Terence Harpold, The Contingencies of the Link, Writing on the Edge 2.2 1991):126-38.

33 References to current on computing, a discipline that is on systematization, are only accessible a surprisingly diverse group of The longest-running annual bibliography is Literature, and the Computer, Annual of English Language and Literature. 46- Researchers, particularly those on medieval topics, will need to supplement this by reference to specialized bibliographies, Wilhelm Ott, Bibliographie. im Editionswesen, in Probleme der Edition und neulateinischer Texte. ed. Hödl and pp. 175-85, reissued in the Italian as Bibliografia. Uso del computer nella editoriale, in La critica dei testi medievali e umanistici. ed. H. Furhmann and A. (Rome: Jouvence, 1984), pp. Joseph Rudman, Selected for Computer Courses in the Humanities, and the Humanities. 21 (1987):245-54; Pauline Literature and Computers: A Short 1980-87, College Literature. 15 S. N. Matsuba, Computer Applications in the A Reading List, Canadian Computing. 4.1 (1990):1-8; Heyward An Interdisciplinary Bibliography for Computers and the Computers and the Humanities. 25 (1991):315-26; and in CTI Centre for Textual Studies Guide, March 1992. ed. Deegan, and Lee, 64-76. See references indexed under and other terms in the MLA International and International Medieval Bibliography Maney, 1977-). There is an important section treating elettronica dei dati in the annual Medioevo Latino (Spoleto: italiano di studi sull’alto 1980-). I would like to Marilyn Deegan for the opportunity to this paper at the 1992 Congress on Medieval Studies at Michigan University, and the staffs of the of Washington Engineering Library and the Library at Redmond, Washington, for assistance in tracking down hard-to-find items.

Volkswagen Pointer Variant II


Other articles of the category "Variant":

Our partners
Follow us
Contact us
Our contacts

Born in the USSR

About this site

For all questions about advertising, please contact listed on the site.

Volkswagen all cars catalog with specifications, pictures, ratings, reviews and discusssions about cars Volkswagen.