Early Modern Literary Studies
Iter: Where Does the Path Lead?
William R. Bowen
University of Toronto
bowen@chass.utoronto.ca

Bowen, William R.  "Iter: Where Does the Path Lead?" Early Modern Literary Studies 5.3 / Special Issue 4 (January, 2000): 2.1-26 <URL: http://purl.oclc.org/emls/05-3/bowiter.html>.

 

  1. Iter was conceived in 1995 in response to the demand by scholars who were beginning to appreciate the potential of electronic publication for more timely and effective access to Renaissance bibliography for articles. At that time, the printed research tools had common faults which Iter was designed to avoid. For example, most bibliographies were produced as annual volumes which were usually several years behind the publication date of the articles which they reported. Also, they had limited searching capability, largely because of rudimentary subject indexing and the practical restrictions common to printed bibliographies. Moreover, by designing the bibliographies as annual rather than as cumulative reports, scholars were precluded from being able to do comprehensive searches without having to consult each individual volume. This format had a further consequence in so far as omissions from one year were not remedied in later volumes; the inconsistent coverage of the literature was most apparent in bibliographies which relied on fragile, voluntary networks of scholars and research institutions around the world. And finally, given the uneven quality of the products, the available bibliographies seemed expensive.[1]

  2. In the process of defining Iter's mandate, we quickly moved beyond the initial concern for developing better bibliographical access to articles on the Renaissance to including other forms of scholarly communication in both print and electronic media, and to building digital collections, all within a much larger historical period. Thus Iter is now broadly defined as a not-for-profit partnership created for the advancement of learning in the study and teaching of the Middle Ages and Renaissance (400 to 1700 AD), primarily through the development and support of electronic resources. The name of the corporation suggests a way or a journey, appropriately characterizing the goal of the partnership as a gateway to Medieval and Renaissance studies, and serves as a reminder of Iter Italicum, Paul Oskar Kristeller's unique and highly prized finding list of uncatalogued or incompletely catalogued humanistic manuscripts.

  3. To begin, Iter concentrated on establishing an online bibliography for all scholarly materials. This ambitious initiative was founded on and continues to be guided by two observations. We saw a need for a high-quality research tool which is truly comprehensive, timely in indexing recent publications, and easily accessible to the scholarly community. (Accessibility is measured in relation to the use of technology and in terms of the price of subscriptions.) And we saw an opportunity to contribute directly to the education of the next generations of scholars by hiring forty to fifty research assistants each year to index and analyze the materials under the supervision of faculty and staff. Where this vision has taken Iter and, more specifically, the online bibliography, will be explained in the following three sections. I will first give a history and overview of Iter, then I will look at the current status of the bibliography and, finally, I will outline our plans for the future.

    A Brief History and Overview of Iter

  4. After several months of discussion, Iter was conceived in September 1995 as a partnership between the Renaissance Society of America (RSA), one of the largest academic societies dedicated to the study of the Renaissance, and the Centre for Reformation and Renaissance Studies (CRRS), a research centre in the University of Toronto.[2]. For the first eight months, we set ourselves fairly basic tasks. Our primary concern was to investigate the practical challenges of creating an online database. This database was to be limited to standard bibliographical information for articles printed in twenty-five to thirty journals going back to 1945 or back to their inception if they began after 1945. We were also to investigate procedures for downloading public access bibliographical databases and to work on subject classifications.

  5. Two developments during this start-up period dramatically altered Iter. In February 1996, the Arizona Center for Medieval and Renaissance Studies (ACMRS), a research centre in the Arizona State University, joined the project as an equal partner. In making this commitment, the Arizona Center agreed to match the CRRS contribution which, at that time, was sixteen hundred hours per year of student time, and to cover their own indirect expenses. Also, both research centres assumed responsibility for providing equipment and space for the project, a commitment which was met in large part by the generous support of the libraries of both the Arizona State University and the University of Toronto. The second development was also of great importance for, in March 1996, the Andrew W. Mellon Foundation (New York) gave Iter a tremendous boost by awarding it a grant of $125,000 US to be used for the bibliography.

  6. The rapid expansion of Iter continued during the remainder of 1996. By this point it was clearly evident that we needed two things: additional professional expertise in order to design a robust system for subject access to the bibliographical records; and much more sophisticated computer equipment. Hence Iter invited and was soon joined by two new partners. The Faculty of Information Studies in the University of Toronto accepted in September 1996, giving Iter the benefit of its resources for building and maintaining the bibliography according to international information standards. And the John P. Robarts Library of the University of Toronto joined in October -- shortly before our test site was made available to the public -- providing Iter with access to its central computer system.

  7. The impact of these developments is crudely evident in the accelerated growth of the articles database. Beginning with approximately ten thousand records from the startup-period (1995-96), within twelve months the database grew to include more than sixty thousand articles from about two hundred journals. What this comparison does not show, however, is the substantial amount of time required to develop the procedures and quality control for the database of articles; to bring the records created during the start-up period to the desired standard; and to begin collecting records for notes, book reviews, and review articles, so that the bibliography would gradually provide full coverage (as appropriate to its claim of comprehensiveness) of the entire runs of journals. Also, the additional resources allowed us to make substantial progress in developing procedures for downloading Machine Readable Code (MARC) format records for books; in addressing issues of pricing, marketing, and our legal structure; and in pursuing opportunities for cooperation with internationally recognized societies and institutions.

  8. In 1996-97, we were also able to make an intensive study of subject classifications, a matter of critical importance to the future of Iter's bibliography. A pilot project was started in June 1996 involving faculty, staff, and students of the CRRS and FIS. The final report of the pilot project was completed and then implemented in the summer of 1997 with the support of a grant of $15,000 US from the Gladys Krieble Delmas Foundation (New York). In brief, subject analysis for Iter is based on internationally accepted information standards, including both Library of Congress Subject Headings (LCSH) and Dewey Decimal Classification (DDC). Using these two systems creates different access points for finding information. Further, by supplementing LCSH with specific dates for the time period and "uncontrolled" keywords which capture the latest research vocabulary, and by providing multiple DDC notations for each item rather than a single notation which locates a physical artifact (i.e. a book or journal) in a library, the user is given even more powerful tools for isolating relevant materials.[3]

  9. In 1997-98, Iter entered a new phase in its life in so far as its legal status was defined through its incorporation in New York State. But, more particularly, the bibliography moved out of its infancy on 22 January 1998 when we separated the main database for journals from a much smaller sample which we call the guest database. This division marked the start of our campaign to offer institutional (as well as individual) subscriptions at modest rates. Finally, our application to the Mellon Foundation for a second grant was successful and in March 1998 we received $420,000 US to continue our work on the bibliography. This grant could well see us to the point where the bibliography is self-sufficient.

  10. The period since the summer of 1998 has been used primarily for consolidation of the bibliography which is gradually gaining recognition. Indeed, after an expected slow start, universities and research institutes in North America and abroad are now subscribing at the rate of two new subscribers per week so that in mid-November 1999 we reached 115 institutions.  Yet many initiatives were started during this period which will have a profound effect not only on the bibliography, but also on Iter's other projects. Some of these matters are outlined in the ensuing sections of the article which report on the current status of the bibliography and on Iter's future.

    Current Status of the Bibliography

  11. Sketching a simplified version of the work flow for the articles database will serve to illustrate how the various components of the Iter partnership currently fit together, at least with regard to the bibliography.[4] What this sketch does not articulate are the quality control checks which are in place or the procedures for revision and correction, all of which are carefully tracked for each record. Nor does it indicate how the partners co-ordinate their contributions to other projects in order to take advantage of their individual strengths and resources.

  12. Policy with respect to standards, journal selection, and available resources is set by the partners and the Director, with day to day management left in the hands of the full-time project manager and her assistants who run a central office located at FIS: see the organizational chart. Instruction in Iter's methods and the assignment of specific journals is coordinated by the central office in order to match the work required to the expertise of the research assistants in the various fields of Medieval and Renaissance studies. Collection of data begins in the libraries affiliated with the ACMRS and CRRS where research assistants input basic bibliographical information using templates designed in EndNote, a personal bibliography software program. For recently published articles, research assistants not only input the citation, they also read the articles and provide subject information. Once this is done, the electronic files are transferred to FIS where they are first converted into MARC format (using Data Magician) and mounted on the system maintained by the Library of the University of Toronto. FIS research assistants then edit the records in accord with international information standards. This includes using the Anglo-American Cataloguing Rules for descriptive cataloguing and for establishing subject name authorities.

  13. In sum, with this process we have created an effective and efficient marriage of the technical, library (or information), and subject expertise which is provided by the partners in the project. And we have encouraged the training of students under the supervision of faculty and staff: at this time, there are about forty-five students working on the bibliography. Through this marriage and the diligence of our research assistants, the bibliography has grown to contain 293,859 records for articles, books, and reviews: the articles and reviews are taken from 379 journals. Since the system for subject analysis was introduced fairly recently, that is, late in the summer of 1997, the number of article records with full subject access using controlled vocabularies is considerably smaller. Nevertheless, we are rapidly reaching a point where we will have full subject access to all articles dating from the present back to 1990 in the full database, having completed all of the journals in the guest database.

  14. What the user can see and do with Iter's bibliography is best shown in a brief tour of the new search interface.[5] A customized version of Data Research Associates' (DRA) Web2, this new interface will be made available to subscribers early in 2000 and will replace the current interface based on WebZ, a product of the Online Computer Library Center (OCLC). The move to DRA Web2 brings immediate gains in terms of sophistication and power for searching the bibliography. The future benefits will be noted after the tour which again, for the sake of simplicity, will focus on the articles database.

  15. It seems reasonable to begin by looking at what information about an article is available to the public: tracking and certain other information is hidden. The sample full record describes an article published in Studies in English Literature 1500-1900. Note that, in addition to the basic information for the author, title, and journal, the record includes a field for the language of the article and the time period covered by the article. Moreover, subject access to the article is provided through six Library of Congress Subject Headings, two uncontrolled keywords (that is, supplementary words not accommodated by LCSH and used by specialists in the field), and three Dewey Decimal Classification notations. The number of subject access points is about average for Iter's records. For comparison, you may want to look at full records for articles in The Sixteenth Century Journal and PMLA. In addition to the fields found in these three sample records, some records also have notes. These usually draw attention to the inclusion of material in addition to the basic text (e.g. maps) or indicate some other important feature (e.g. that the article is published in parts and where the parts are to be found).

  16. The header for the new interface offers three different buttons to start a search for information contained in the bibliographical records. The basic search screen (an alternate view of this screen is available) includes two text boxes for words to be searched. For each box, the user can choose either to do a keyword search of all fields in the records, or to limit the search to one of six listed fields; the truncation sign, #, is also available. The two boxes are linked by a boolean operator (i.e. and, or, not) menu, so that the two search strings can be connected in different ways. Finally, the search is limited to one of three databases, that is, articles, books, and reviews.[6]

  17. The advanced search screen (an alternate view of this screen is available) enables more complex searches by doubling the number of text boxes, adding three more record fields, and allowing limits by language and publication year or range of years. For those users who are comfortable with typing in their own commands, including field designators and boolean operators, there is a button at the bottom of the page which takes you to a version of the advanced search screen with an open box for composing your own command line (an alternate view of this screen is available).

  18. The third way to start a search is to press the browser button found in the header. Although the browser pages have not been fully designed yet, it is possible to describe the functions which we propose to place there. Two choices will be offered initially. Users will be able to browse through our authority files for names or through the hierarchy of subject fields (i.e. words or phrases) defined by the DDC notations (i.e. numbers) in the records themselves. Clicking on the names or subject fields will initiate searches.

  19. Once a search has been started, the interface facilitates subsequent searches in three ways. First, the search results screens include buttons at the top and bottom of the page to revise the current search or to start over. Second, clicking on hyperlinked names or subjects in the full records starts a search for the same information in the same field in other records. In the case of the LCSH, the user can choose how many levels in a heading are to be searched. For example, if a heading is "Presbyterianism-Controversial literature-History and criticism," the user can choose to search "Presbyterianism," "Presbyterianism-Controversial literature," or "Presbyterianism-Controversial literature-History and criticism." And, third, the program remembers the searches which have been tried. By pressing the search history button in the header, the user is taken to a list of the previous searches which can then be deleted or combined via boolean operator menus in new searches.

  20. The user is given a great deal of control over how search results are displayed, and the opportunity to toggle from one format to another. For example, the advanced search page (an alternate view of this screen is available) allows choices as to the number of hits displayed at one time and whether the default record display is in brief, full, or customized format. The last is an unusual feature in so far as the program allows the user to design their own display format, including the number and order of twelve fields found in the record. The capability of personalizing the interface is further enhanced by the preferences page (an alternate view of this screen is available) which allows subscribers with individual passwords to set their preferences not just for the current session, but for future sessions as well. The preferences page gives options for the choice of startup screen, how the search results are displayed (i.e. brief, full, or customized), and so on.

  21. The user is also given some ability to manipulate the search results. For example, in all search results pages which list more than one hit, the user is able to sort the records by title, author, or publication date (ascending or descending).[7] Records can be selected by marking them, that is, by clicking on the little box beside the record. Then the selected records can be isolated from the full list by clicking on the view marked button. This system allows the user to refine search results while reading through the records and, once the records are marked, to mail or download them using the buttons on the marked list page.[8]

  22. This brief tour has pointed to many essential features of the new interface which are described in greater detail in the help pages and search tips. In addition, it should be mentioned that the interface includes a notice board for the announcement of further developments. Finally, search results pages will have an automated system which will allow subscribers to contact the central office about any problems which may arise.

    Plans for the Future

  23. There is much that we hope to accomplish in the immediate future. With respect to the bibliography, we expect with our current resources to add 100,000 new records per year and to place increasingly greater emphasis on subject access: it is our goal to have 500,000 records by the end of 2001. Further, the new interface will allow us to create a multi-lingual environment. This will be accomplished at three levels: the bibliography will be able to display latin as well as non-latin based character sets; users will be able to choose from half a dozen different languages for our various web pages; and, perhaps more importantly, they will be able to browse through the DDC subjects in the language of their choice. At the same time, we are planning to expand the scope of the current bibliography by mounting several additional databases for collections of essays (with a detailed account of each essay), online resources, research societies and institutions (modeled on the International Directory of Renaissance and Reformation Associations and Institutes. Toronto, CRRS), and scholars and their research projects. As far as possible, the supplementary databases will be linked to the main bibliography so that the user will move seamlessly from one kind of information to another. For example, it is our plan to enable the user to search an author's publications in the bibliography and at the same time to consult the author's profile in the directory of scholars and research projects.

  24. We are also taking preliminary steps toward the creation of a digital library. For example, we have been working for some time with E.J. Brill to offer an up-to-date, online version of Paul Oskar Kristeller's Iter Italicum which will supersede their CD-ROM product.[9] And we have more recently started a pilot project to co-publish a substantial bibliography with ACMRS in their Medieval and Renaissance Texts and Studies (MRTS) series. However, beyond the online publication of databases, the real challenge for Iter lies in publication projects which take advantage of the Web's ability to link pages or documents. A few modest steps have been taken in this direction with pilot projects. And we are examining how we might link the records of the main bibliography and related databases to online documents within our domain or outside our system, although the latter possibility is a matter which will require considerable study and negotiation.

  25. Speculating on the distant future reveals a variety of opportunities, especially given our long-term interest in enhancing study of all facets of Medieval and Renaissance culture by creating a multimedia digital library. To this end, we are proposing to design a sophisticated interface, to develop digital collections comprised of electronic texts and the gamut of audio and visual materials (e.g. sound recordings, videos), to assist in the development of tools for the analysis of multimedia materials, and to link as smoothly as possible to other digital collections. The library will be constructed with the consent, and indeed the active participation, of copyright holders.

  26. One of the initial uses of such an interface and collections would be for instruction. For example, the architecture of the digital library might allow teachers and students to incorporate their notes and interactive discussions with assigned materials. In effect, individuals could wrap their commentary around digital files provided by Iter. Eventually, scholars may find this a suitable way to do their research. But it is also true that, for both teaching and research, having access to digital files would allow the combination and analysis of materials in new and exciting ways which are not always possible or practical in a print environment. Thus, given the scope of the enterprise, the project would inevitably enhance teaching and stimulate research. Furthermore, it would encourage the application of new technologies to multimedia studies and would be a test case for numerous issues currently debated in the field of information studies. In this way, Iter would effectively find its place as a world-class leader in Medieval and Renaissance studies in the evolving digital era. And, at the same time, Iter would begin truly to realize the meaning of its name as a journey, a road, or a path.

Works Cited

Notes

1.  For example, see one of the few printed bibliographies dedicated to the period as a whole: Bibliographie Internationale de l'Humanisme et de la Renaissance (BIHR)The situation has not improved significantly in so far as the most recent volume of BIHR, volume 29, lists 1994 publications and was published in 1998. However, in 1999 Droz announced forthcoming CD-ROM and online versions of BIHR which will address some of the problems associated with printed bibliographies.

2.  Technical support for the pilot project was given by William Barek, Director of the Centre for Instructional Technology Development, University of Toronto at Scarborough. Funding was provided by the two partners and by Heather Munroe-Blum, Vice-President of Research and International Relations, University of Toronto.

3.  An early account of the cataloguing standards is found in "Maintaining Web-based Bibliographies: A Case Study of Iter, the Bibliography of Renaissance Europe," a contributed paper by Tracy Castell, the first project manager for Iter.

4.  The first management system devised for Iter's bibliography is explained in Castell's "Maintaining Web-based Bibliographies." Although the general principles outlined in this article are still upheld, many details have changed over the past few years.

5.  For subscribers who do not have access to a web browser, Iter is developing a simple Lynx interface so that they can still use the online bibliography.

6.  The next version of this interface will likely allow the user to search more than one database at a time.

7.  We have set a limit on the number of records which can be sorted. At present, the limit is 2000 records.

8.  The page for a single full record also gives options for mailing or downloading the record.

9.  In November 1999, the Gladys Krieble Delmas Foundation granted Iter $21,000 US which should be sufficient to complete this project.


Responses to this piece intended for the Readers' Forum may be sent to the Editor at L.M.Hopkins@shu.ac.uk.


1999-, R.G. Siemens and Lisa Hopkins (Editor, EMLS).

(WB, DR, RGS, 19 December 1999)