International Association of Libraries and Museums of the Performing Arts

Société Internationale des Bibliothèques et des Musées des Arts du Spectacle

The SIBMAS International Directory of Performing Arts Collections and Institutions Online

The Transition from Printed Volume to Online Service
How it was done, why was it done, how is it being received?

Paul S. Ulrich, Berlin

Performing Arts Collections and Their Treatment
Le collections des arts du spectacle et leur traitement

International Association of Libraries and Museums of the Performing Arts
24th International Congress, Rome, 3-7 September 2002

Société Internationale des Bibliothèques et des Musées des arts du spectacle
24ème Congrès International, Rome, 3-7 septembre 2002

At the beginning of the 20th century theatre was recognized as a field of study in it’s own and not merely a branch of other fields, most particularly literature. The problem of locating the primary sources soon became painfully clear: what had been preserved was scattered in numerous museums, archives and libraries throughout the world. Just as no one knew how many theatres and performances there were throughout the world, likewise no one knew where the multitude of records and traces of the theatre could be found. The multi-faceted institution theatre, encompassing theatre, music, art, architecture, technology, public administration, to mention only a few of the aspects, meant that the necessary sources were scattered, often hidden within collections which would initially seem to have nothing to do with the theatre itself. The question for the theatre researcher was thus: where are the materials for research? The extensiveness of this problem was illucidated in 1960 by Julien Cain, administrateur général de la Bibliothèque Nationale: “The staggering amount of material in the field of the performing arts which has accumulated throughout time, whether it be handwritten, printed, designed, engraved, photographed, or microfilmed, such a variety of documentation, where even the most divers objects, set designs, and costumes themselves, have their part, all this material deserved to be more widely known so that more use could be made of it.”¹

History of the Directory

The late André Veinstein (1916-2001), founder and former President of SIBMAS, George Freedley and Rosamond Gilder convinced the members of the International Section for Performing Arts Libraries and Museums of the International Federation of Library Associations² that producing an international directory with information about these institutions should be a continuing project for the association. The first edition finally appeared after seven years of preparation: André Veinstein, Marie Françoise Christout and Denis Bablet (eds.): Bibliothèques et musées des arts du spectacle dans le monde = Performing Arts Collections, An International Handbook . Paris: Centre national de la recherche scientifique 1960.

In her introduction to this first directory, Rosamond Gilder described the work which others had previously done to make information about performing arts collections available:

“[… ] One of the fields that seemed to her³ of primary importance was the theatre collection- As early as 1933 Theatre Arts ran a series of articles on the great theatrical collections of Europe and America. In 1936 it published the first handbook on the subject: Theatre Collections in Libraries and Museums, by Rosamond Gilder and George Freedley. The studies made to secure the material for this modest volume both by correspondence and by the actual visits of George Freedley, Curator of the Theatre Collection of the New York Public Library, to the major collections of the continent and England revealed an extraordinary wealth of material that few had realized was so extensive or so important both in quality and quantity. So keen was the interest in the subject that the handbook published in 1936 was soon out of print.
A new edition, revised and brought up to date has long been needed and work on the American section was undertaken in 1952 by Paul Myers as Chairman of the Library Project of the American Educational Theatre Association. Shortly there after, Dr. André Veinstein of the Bibliothèque de l’Arsenal, Paris, became keenly interested in the whole subject of bibliography and collections in the field of the performing arts and as President of the Committee of the International Federation of Library Associations dedicated to this subject, launched early in 1955 an exhaustive, world-wide inquiry on theatre libraries and theatre collections. His questionnaire, cast in a somewhat different mould from that of the Gilder-Freedley handbook, which, with the approval and cooperation of its authors he used as a springboard, was addressed not only to the collections listed in that volume but to other libraries and museums both public and private as well.”⁴

From the very beginning Mr. Veinstein was very clear as to how the directory was to be used:

“The use of the present work will permit librarians to offer their readers an enlarged scope of information, to increase their network of foreign correspondents, and to know better the activities of certain of their colleagues on the national level in order to avoid duplication of effort. Moreover, the diversity and the particular nature of the documents brought together in the different fields of the performing arts and the use made of the various institutions by certain categories of specialists will facilitate the spreading of solutions which have been made with respect to classification, cataloging, and ordering of places; will allow the organization of meetings designed to try to resolve problems of a professional order; and, finally, will go forward with collective projects; e.g., the editing of catalogs, bibliographies, etc. Still, using the present book as a point of departure, useful observations will be drawn from the orientation taken by certain collections (the rejection of the principle which accords precedence to books over documents, of the radical separation of museums and libraries, and of the documentary value of such recent supports as films and magnetic tapes); from the increase in holdings and reference systems in response to the expansion of sciences, besides history, which are interested in the performing arts; e.g. aesthetics, psychology, sociology, pedagogy, phonetics, and electro-acoustics; and from the growing needs of artists and technicians with respect to authentic documents, that is to say, those which are not, interpreted and are not foreign to the production of their art.
Scholars and students will be able to find in this book subjects bearing, for example, on the geographic distribution of documentary resources, the conditions under which they were set up, the conditions and general make-up of holdings, in different countries, devoted to certain subjects or personalities, and the particular nature of certain institutions; e.g., those set up by the theatres themselves; those devoted to the documenting of professionals (artists, technicians, or administrators); or those recently created by radio and television stations.”⁵

From the very beginning, Mr. Veinstein and those assisting him strove not only to provide the basic information about the respective institutions (name, address, telephone numbers, opening hours), they also strove to provide a glimpse into the actual holdings of the institutions. In order to gather all this information, extensive - 7 page - questionnaires (both in French and in English) were mailed out to all those institutions which were known to have holdings relating to the performing arts. The institutions were encouraged to write descriptions about the history, holdings and other interesting facts about their institution.⁶ As is always the case with questionnaires, the recipients did not always return them to the editors, hence an extensive apparatus was needed to check up and remind those who had not returned the questionnaire, that information from them was missing.⁷

In keeping with the bilingual (French/English) policies of SIBMAS, all of the replies not only needed to be edited to correspond to editorial guidelines, they also had to be translated into either English or French. This procedure was maintained for the first 4 editions,⁸ the only major changes being in modifications in the wording of the questions asked.

The Directory, which was published with a blue cover, was affectionately known within SIBMAS as the "Blue Book".⁹ At the biannual SIBMAS congresses the "Blue Book" Commission would discuss the problems of the preceding volumes and make decisions on the next edition.

At the 1988 SIBMAS Congress in Mannheim there was a general consensus within the "Blue Book Commission", that it was becoming increasingly difficult to continue the work in the present form. A major concern was the extensive time required to make the translations for a bilingual publication; this time factor decreased the currentness of the content. Furthermore, the translators - even in the first edition - noted the problems in making adequate translations.¹⁰ It was suggested that for many sections the use of symbols would streamline the presentation and also make repetition in French and German superfluous.¹¹

For the 3^rdand 4^th editions, André Veinstein passed on much of the editorial work to Alfred S. Golding, who retained the general content and structure of the publication. After the appearance of the 4^th edition, Alfred Golding turned over the editorship to a new editorial team,¹² which began implementing the suggestions which had been discussed at Blue Book Commission meetings since 1988 in Mannheim.

Already at the 1992 Blue Book Commission meeting in Stockholm, the structure of a very simplified, one-page questionnaire¹³ was accepted with minor changes by the members and the necessary preparation was done to transfer the information from the 4^th edition to this new form. The instrument for gathering information was no longer called a questionnaire, but rather a proof sheet. Institutions would be asked to review the information already known about their institution and make the necessary corrections. The information on the proof sheets was what would be printed if the institution did not make corrections.¹⁴

There were numerous reasons for streamlining the information-gathering instrument:

Erroneous information would be more likely to be corrected, if the institution was aware that the proof sheet was the basis for a publication.¹⁵
It made it as easy as possible for the institutions to provide the information: If all the head of the institution had to do was make corrections of what was wrong with the information available to the editorial board, the chances were very great that he would be able to do this very quickly. Furthermore he would not have to spend time formulating texts.
It made it as easy as possible to make additions and corrections to the database.¹⁶ This reduced the editorial preparation of the 5^th edition to less than 4 months (compared to the several years needed for the 4^th edition)!

Richard M. Buck succeeded in finding a new publisher, who would be willing to finance the directory, and in 1994 at the SIBMAS Conference in Antwerp, the decision was made to sign a contract with Jim Emmett of Emmett Publishing in England. Emmett Publishing agreed to finance the mailings to the institutions, cover the costs of publishing and distribution.¹⁷

For a directory of German-language institutions,¹⁸ a different approach to gathering information had been used, which seemed transferable to the collecting and verifying the information about the institutions which would be included in the Directory. Instead of sending out a questionnaire, the institutions were sent an extremely simplified one-page proof sheet with information about the respective institution. The institution only had to make the necessary corrections and return it to the editors. The institutions were informed that the information the way it was on the proof sheet would be printed "as is", unless the institution made corrections. Failure to return a proof sheet would be interpreted by the editors to be approval of the content. Instead of the usual 30% return rate for most questionnaires, the German directory achieved over 60% returns.¹⁹

Prior to sending out the proof sheets, the information about the institutions in the 4^thedition was standardized and put into a structured ASCII flat file²⁰ for later import into a database. Other regional performing arts directories which had been prepared by national/regional members of SIBMAS were used to verify and augment this information.²¹ Furthermore, numerous international directories²² were consulted and relevant material was added to the information already present. In contrast to previous editions, which primarily focused on the major international institutions with material on the performing arts, the editors of the 5^th edition²³ tried to identify less-known institutions with smaller collections,²⁴ since it was particularly these lesser-known collections which were of particular interest to scholars. The result was a database of over 5000 institutions.

The database was sent to Emmett Publishers, where the preparation, printing and mailing of the proof sheets was managed. The institutions were asked to return the proof sheets to Paul S. Ulrich in Berlin, who made the necessary changes to the information in the database, and after all the proof sheets had been received and the corrections made in the entries (some of the members of the Blue Book Commission put pressure on institutions which hadn't sent in their proof sheets, and this made many of the entries better than they would have been), the final database was sent to Emmett Publishers, which then printed the directory, which had a new name (SIBMAS International Directory of Performing Arts Collections) and was printed with a gray cover.

Institutions which were members of SIBMAS were indicated as such in the directory.²⁵ Cross-linking²⁶ was already present in a passive form in the 5^th edition with inclusion of their URLs and email addresses. Since then many institutions now have web sites which have had to be included. Even more important, additional databases with information about collections have become available on the web.²⁷

The Migration to a Web-Based Directory

Indications of potential long-range problems became increasingly apparent during the preparation of the 5^th edition. The number of institutions which needed to be sent proof sheets was so large (over 5000), that postage costs were high. Even though the layout and content had been streamlined, the space needed for the increased number of institutions and the indexes was such that the price for the printed edition could not be kept under USD 100. The price of the printed version (USD 150) was too high for many institutions to afford. This resulted in a low number of purchases. Parallel to this, the international publishing scene was experiencing major changes by increasingly using electronic methods of distribution (CD-ROM and internet) particularly for reference works.

Already in 1996 there were numerous requests for an electronic version of the Directory, primarily from academic institutions in the USA and the United Kingdom. Within SIBMAS there was still a feeling that the printed version should be maintained for institutions which were not capable of accessing electronic versions. Supporting this argumentation became increasingly difficult when one examined the sales of the printed version: the institutions which did not have the resources for accessing an electronic version of the directory were not purchasing the printed version, and those institutions which were purchasing the printed edition were also those institutions desiring an electronic version.

Following the publication of the 5^th edition in 1996, there were numerous discussions about the future of the directory. Transferring the directory to CD-ROM was rejected, since many of the problems of the printed directory would merely be transferred to the new medium. In addition new problems would need to be addressed, among them creating at least two versions (Macintosh and Windows) and developing an appropriate user interface which could be used internationally.

The increasing acceptance and accessibility of the internet seemed to offer an alternative which might solve many of the problems: no printing costs, ease in making corrections and keeping the information current. Emmett Publishing already had made several of its databases accessible by subscription on the internet, so such a move would be possible. They already had a structure which could be transferred to an online version of the directory. The question still remained, whether or not institutions would be willing to pay for online access to the directory, or whether the non-use would continue because of an unwillingness to pay for online databases. An early decision was made that should the directory be offered as a subscription database, a demo version would be made available free-of-charge for potential users to evaluate the usefulness of the offering. In order to increase the attractiveness of being a SIBMAS member, it was decided that only institutions which were members of SIBMAS would be accessible in this demo database.²⁸

With the cessation of Emmett Publishing in the summer of 2000, SIBMAS was forced to reevaluate the continuation of the directory. At the SIBMAS Conference in Paris in September 2000, the decision was made to migrate the directory to the web and to make it available free-of-charge on the SIBMAS web site.²⁹ The administrative costs of maintaining a fee-based directory were such that it was unrealistic to expect that it would be able to generate income for SIBMAS.³⁰ Furthermore usage would probably not be very high, since most institutions felt that what was on the internet should be free-of-charge anyway.

Maria Teresa Iovinelli, the webmaster of the SIBMAS web site, indicated that there was sufficient space available on the server to house the web-based directory.³¹ Paul S. Ulrich was given the task of migrating the information in the database (which had in the meantime been increased to include almost 7000 institutions) to the web.

At the Executive Committee meeting in October 2001, the web directory was presented to the committee members for approval and it was proposed that it be mounted on the web site prior to the end of the year. The members of the committee had numerous suggestions for improving both the presentation and the content. It was then decided to postpone the start of the web-based directory until March 2002.³² After the official start of the directory on the web, the URL was sent to numerous mailing lists and online news services, so that it would be known. Since it’s appearance on the web, there has been a noticeable increase of the usage of the SIBMAS web offerings.

The Structure of the Directory

Users of the directory have different needs at different times. The printed directory already provided the basis for the types of access needed for the web-based directory: geographic (what institutions are in a city?), name of institution (where is information about a specific institution?) and names of special collections (which institutions have material on a special topic?). Transferring these points of access to the web meant that the information in each web page be broken down in the smallest possible units:³³ geographic by city, institution and collection names alphabetical in many short lists rather than one long list.

The web version of the Directory has over 3000 web pages. With this many pages (=files), the organization on the server plays an important role, both in generating the files from the database as well as for the navigation and later use.

The main information about an institution is contained in the geographic (city) files. These are grouped by geographic region,³⁴ each region having it's own directory on the server. Each regional directory has an index file (“index.html”³⁵ ) for all the countries in the region and each country has it's own index file for the cities. All file names have no more than 8 characters plus the ".html" extension. The file names for the cities all have the same structure: ISO country code, first letter of city name, 3 digit number padded with "0" to indicate how many other cities from the country have the same beginning letter. These file name conventions make it very easy to isolate and copy files with wild cards for the administrators of the system.³⁶ It also makes it very easy to create file names by the program which exports the information from the database to html-files. This convention also guarantees that a name will not be given twice to a file when, for example the name of a city is the same in two countries (for example, San Francisco in California and in Spain).

There are two additional indexes on the server which make it possible for a user to go directly to the main entry: Institutions and Collections. In addition to the “index.html” file, which lists all the files in the respective directory, the names of the individual files have the following structure: the first two characters of the name of the institution or the collection. These files are also easily generated by the program.

Findability and Usability

The first important decision was that the Directory would have to attract as many possible visitors as possible and offer as many points of access as possible. This meant that the content could not be presented as a database³⁷ on the server, which would only offer one point of entry, but rather all the information would be on individual pages, i.e. the presentation on the web would be with static rather than dynamic pages. Furthermore, frames would not be used, since the content of frames provide indexing problems for search engines, and it is very difficult to link to a specific page within a frames environment.

Much of what is necessary to make it easy find a page is not really new or special, rather it is a matter of logic. In order to create a usable web site it is necessary to understand how search engines gather their information about a web site and how they process it for presentation in their lists of “hits”. Likewise it is important to be aware of how people using the internet react to the presentation of web offerings. This knowledge must then be transferred to the actual presentation so that it supports rather than discourages use.

There are generally two ways that people find a web site:

They find a reference to it an a thematic directory (such as Yahoo, the Open Directory), in a clearinghouse or in a link collection. In such instances they are directed to the home page of the offering. These link collections are manually maintained and links are added when those managing the directories are made aware of the presence of a new site. As a rule a link is only made to one location - the home page -, and the web offering may or may not reflect where the creator of the offering feels it should be placed.³⁸
They use a search engine (such as Google, AltaVista, AllTheWeb, etc.) to find it; more specifically they find a page with words corresponding to terms they are looking forward to, whether or not the terms are on the home page or somewhere within the offering.

These search engines are fed with information picked up by robots. Each individual page is analyzed and the words on the page are entered into the search engine database, i.e. each word on the page (or combinations of the words) serve as points of entry for someone looking for information. Complicated mathematical algorithms ultimately determine how easily or poorly one's page is rated, and this determines whether or not one lands among the first 10 to 20 offerings.

After a page has been found, it must also be used. The easier a page can be used, the greater the chances are that it will be used again in the future. A page which is difficult to use discourages users from returning to the site in the future, since negative experiences are counterproductive.

Within the past few years the concept of usability is of central concern to designers of high-usage web sites. They realize that the greater emphasis they place on making a web site usable, the greater the usage will be. The recommendations of such usability studies have been incorporated in the SIBMAS directory.

There are many factors which must be taken into consideration when planning the presentation. Too frequently designers are so interested in making a web presentation look nice that they forget the behavior of themselves and others when actually using the web. Before a web site is used, it must first be found. It is not enough to inform search engines that a site is present, it must also land in the top 20 results of a search engine for it to be found. This means that choice of page title, choice of wording, coding the text all play a role in the findability.

Minimum hierarchical depth. From the home page to a page with the sought information, a user should ideally only have to click a maximum of three times.³⁹
Navigational elements at all points. Every page should provide information to the surrounding environment and have links to the various points of entry.⁴⁰
Each page has it's own distinctive title, with the page specific information at the beginning of the title.
The title plays a role not only in being found, but also when navigating within a web site. The most import differentiating part of the title of the page should be placed first, not at the end of the title. The title has several functions:
- It is always indexed by all search engines.
- The words in the title have added weight in mathematical algorithms for search engine placement.
- The title is what is automatically given for the entry in the favorite (Internet Explorer) or bookmark (Netscape Navigator) listings
- The title is what appears in browser “history” listings
- The title is printed on all print-outs.⁴¹ For this reason, the title of each main entry page begins with the name of the city, followed by the country and ending with “SIBMAS International Directory of Performing Arts Collections and Institutions".
Page renewal coding to guarantee that the current version of a page is viewed by the user. Most designers forget that if a user has already viewed a page recently, his browser will load the version from the cache and not the version on the respective server. By putting a command in the html text,⁴² it is possible to override the use of the local cache and be more certain that the viewed information reflects the current status of a page.
Index at top and bottom of each page . Not only does this facilitate navigation on the page, it also has a few other important advantages: The user sees at the beginning of the page, what is on it and can quickly jump to the entry which he is interested in. Equally important, it means that for all the institutions on the page the name appears three times; this in turn means that there is automatically a greater weight of the page for search engines.
Names of institutions coded as heading. Since search engines give greater weight to terms listed on a page as “headings”,⁴³ the name of the institution in the main entry is coded as a heading and is not merely in bold text.⁴⁴
URL on every page. On every page of the directory the URL is included on the page. This is equivalent to professional journals including the name and issue of the journal on every page and guarantees that on every printout or download of the page, that it's source will be retained. Furthermore, having the URL on each page is good public relations and it guarantees that the source is always provided with the page content - particularly when it is downloaded.⁴⁵
Content on the left side of the page, navigation elements on the right. Just as most people read from left to right, so do the content evaluation programs of search engines. Text on the left-hand side of a page has greater weight than text on the right-hand side. The navigation elements are important, but should not be given additional weight for search engines.
Putting the main text on the left side of each page also makes the page easier to use. This is particularly true for handicapped viewers who use auxiliary programs to magnify the text: it is easier to jump to the left or right side of a page than to locate text which is in the center of the pager.⁴⁶ Equally important is what information about the page appears on the upper left portion of the screen, since this is the first information about the page and very often determines whether the whole page is viewed or not.
Quick loading. The file must load quickly, otherwise the viewer will click to another page⁴⁷ - no one has time to wait for a page to be loaded. As a general rule the designer of a page must do everything in his power to reduce the loading time of a file to a minimum. This means that the file should be less than 50 KB in size and have as few graphics as is necessary.⁴⁸

Use of the Directory

After spending so much time designing the web site so that it would be visible on the web and also usable, the big questions remain: is it being found, is it being used?

An examination of the log files⁴⁹ of the SIBMAS server shows very clearly that since the presence of the directory the usage of the SIBMAS web site has increased considerably. Not only have the number of visits to the server by search engine robots increased considerably,⁵⁰ the number of visits to the directory by users has increased dramatically. Log files also give a good indication of how the Directory is being accessed. Although many visits are over the regional pages, the largest number of visits and usage is over the indexes of the special collections.⁵¹

Likewise, regular checks for web sites linking to the Directory has pleasantly revealed an increasing awareness of webmasters who are including links to the Directory. This is doubly important:

1. The more pages on the internet linking to the offerings, the greater the chances of somebody finding it.
2. Particularly in the last year, and in particular popularized by Google, the placement of a web page in a list of hits by a search engine is to a large extent governed by how many pages link to it: the more links, the closer to being number one on a search engine hit list.

A further indication of the use of the web-based Directory is evident in the number of emails which the editor receives from visitors to the web site. Almost daily emails requesting information about the holdings of various institutions are received.

Even more interesting are the emails which report corrections for the Directory. Whereas in the past all modifications were either identified by the editor or members of SIBMAS, now potentially all the users of the Directory are potential sources for corrections or additions to the information. Many institutions have noticed that their listing is not correct and have directly contacted the editor and requested that corresponding changes be made to their entry.

Future Plans for the Directory

It would be wrong to assume that the Directory should remain in it’s present form. By reviewing usage of the presentation, changes are continually being made to the layout: some of the changes are major and easily recognizable, others are minor and not noticeable. Changes in the web presence of information from others means that this needs to be regularly reviewed and either incorporated in the Directory or made available via SIBMAS in other forms.

New browsers and new expectations of users mean that on a regular basis the layout and page coding must be carefully examined to see that it conforms to the current standards and still not only be viewable by the most popular browsers.

Currently SIBMAS is not capable of financing extensive mailings to update the information on the various institutions. Since the number of institutions with email is rapidly increasing, it would seem logical that at regular intervals these institutions be sent emails in which they could be requested to update the information about themselves.

Currently the editor makes the changes to the database on his computer and then generates the pages for the web, which then must be manually updated to the server. In the near future it will be necessary to move this database to the server directly and install programs which will automatically update the pages on the server. Such a move would also make it easier to email requests for corrections to the institutions in the directory. Likewise it would make it possible for numerous persons to do the actual editing of the database.

Conclusion

From it’s inception as a book in 1960 to the present electronic web version, the SIBMAS Directory has been an important instrument for locating which institutions had material on the performing arts and where specific material could be found. The many changes which have been made reflect not only the rise in the number of institutions collecting this material, it also reflects the changes in how information is made available in an increasingly changing world. The use of the SIBMAS International Directory of Performing Arts Collections and Institutions both in the past and at the present is an indication of the need of such resources. In the future the need for such a tool will not decrease, but will be even more necessary. The content and the presentation will not necessarily remain the same. It will, however, reflect the needs of both the institutions for making their holdings known and for those looking for these holdings.

Footnotes

¹Julien Cain: “Preface”, in: André Veinstein, Marie Françoise Christout and Denis Bablet (eds.): Bibliothèques et musées des arts du spectacle dans le monde = Performing Arts Collections, An International Handbook . Paris: Centre national de la recherche scientifique 1960, p.9. [return]
²This section within the International Federation of Library Associations was the beginning of SIBMAS. [return]
³Edith J.R. Isaacs, editor of Theatre Arts Monthly. [return]
⁴ Rosamond Gilder: “Theatre Collections: Present and Future”, in: André Veinstein, Marie Françoise Christout and Denis Bablet (eds.): Bibliothèques et musées des arts du spectacle dans le monde = Performing Arts Collections, An International Handbook. Paris: Centre national de la recherche scientifique 1960, pp. 15-16. [return]
⁵ André Veinstein: “The Book and Its Use“, in : André Veinstein, Marie Françoise Christout and Denis Bablet (eds.): Bibliothèques et musées des arts du spectacle dans le monde = Performing Arts Collections, An International Handbook . Paris: Centre national de la recherche scientifique 1960, p.24. [return]
⁶ A major problem here is that institutions do not like to respond to such questionnaires, since it takes a long time to formulate answers. For institutions where English or French is not a native language, there is either a hesitancy to avoid answering such forms, put it off indefinitely, or to have someone translate a text into either French or English - the quality of the translations often being very poor. [return]
⁷ Experience with other mailings of questionnaires had shown, this procedure requires extensive follow-ups to get the questionnaires returned. [return]
⁸ André Veinstein, Marie Françoise Christout and Denis Bablet (eds.): Bibliothèques et musées des arts du spectacle dans le monde = Performing Arts Collections, An International Handbook . Paris: Centre national de la recherche scientifique 1960, 761 pp.
André Veinstein and Cécile Giteau (eds.): Bibliothèques et musées des arts du spectacle dans le monde = Performing Arts Libraries and Museums of the World. 2.ed. Paris: Centre national de la recherche scientifique 1967, 803 pp.
André Veinstein and Alfred S. Golding (eds.): Bibliothèques et musées des arts du spectacle dans le monde = Performing Arts Libraries and Museums of the World. 3. ed. Paris: Centre national de la recherche scientifique 1984, 1181 pp.
André Veinstein and Alfred S. Golding (eds.): Bibliothèques et musées des arts du spectacle dans le monde = Performing Arts Libraries and Museums of the World. 4. ed. Paris: Centre national de la recherche scientifique 1992, 740 pp. [return]
⁹ Not to be confused with the FIRT/IFTR-SIBMAS Bulletin, which also had a blue cover. The Bulletin was published several times a year and contained current information about conferences, publications, exhibitions and other matters pertaining to the performing arts. It ceased publication in paper form in 2001. [return]
¹⁰ This problem was prominently mentioned in George Miller’s “Translator’s Note” in the first edition: “The particular difficulty for the translator in the present book has been that it is a work which describes specific things, unknown, for the most part, to the translator. For this reason, where alternative translations have presented themselves, particularly in the many lists and catalogs of holdings, educated guesses have been more the rule than the exception. Inevitably, some of these have been unfortunately incorrect. The translator can only hope that no one will have journeyed half way around the world because of one of these presumed mistranslations.” [return]
¹¹This suggestion was partially realized in the 4^th edition with the use of abbreviations. [return]
¹²This editorial team was composed of Richard M. Buck (New York Public Library of the Performing Arts), who as general editor was also given the task of finding a publisher who would help finance the project, Paul S. Ulrich (Amerika-Gedenkbibliothek, Berlin) and Ruth Freydank (Märkisches Museum, Berlin) and Roger Rennenberg (Stadsbibliotek Antwerp), who were to do the actual preparation of the next edition. [return]
¹³ Prepared by Ruth Freydank and Paul S. Ulrich. [return]
¹⁴ In a cover letter the institutions were informed that not returning the proof sheets would be interpreted as approval of the information. In the printed directory those institutions which had not returned the proof sheets were thusly indicated. [return]
¹⁵ No one likes to have erroneous information printed about themselves, and they will do what they can to correct errors. This increases the possibilities of returns being made. [return]
¹⁶ This was possible because only the fields containing positive information needed to be entered in the database. Furthermore extensive simplification in the number of characters entered in each of the fields reduced both data entry time and typographical errors. [return]
¹⁷ Because of the high production costs - primarily arising from mailing 5000 proof sheets -, the original intention to keep the price of the directory under $50 was not possible. [return]
¹⁸ Peter Schmitt and Paul S. Ulrich (ed.): Archive, Bibliotheken, Museen, Sammlungen und Gedenkstätten mit Beständen zum Bereich „Darstellende Künste“ in der Bundesrepublik Deutschland =Archives, Libraries, Museums, Collections and Memorials with „Performing Arts“ Holdings in the Federal Republic of Germany (SIBMAS-Brief. Sonderheft 1.) Erlangen: Bundesverband der Bibliotheken und Museen für Darstellende Künste 1992. [return]
¹⁹ Of the 350 institutions sent the questionnaire, 310 responded. [return]
²⁰ A flat file is a text file, with the information so structured that it can be automatically transferred to the respective categories in a database. The advantage of working with information in this form is, that it is independent of the database used, it can be processed in any word processor, it does not need a data-entry interface, it is easily transported and can be imported into almost any database. [return]
²¹Joke Elbers (ed.): Theater collecties in Nederland en Vlaanderen. Amsterdam: Nederlands Theater Instituut 1991.
Lennart Forslund (ed.):Teatersamlingr i Norden. Katalog over teatermuseer, -bibliotek og -arkiv i Danmark, Finland, Island, Norge og Sverige. 3., reviderte opplag. Umea: Nordisk Center for Teaterdokumentation 1991
Heather McCallum and Ruth Pincoe (ed.): Directory of Canadian Theatre Archives. London: Vine Press 1992. (Dalhousie University, School of Library and Information Studies. Occasional Paper Series.)
Roger Rennenberg (ed.): Bibliotheken, Archieven, Musea en Documentatiecentra over de Podiumkunsten in Vlaanderen en Brussel = Libraries, Archives, Museums and Documentation Centres of the Performing Arts in Flanders and Brussels. Antwerp: National SIBMAS Centrum Vlaanderen 1992.
Theatre Communications Group (ed.): Theatre Directory. Vol. 20. New York: Theatre Communications Group 1992-93.
Nicole Leclercq (ed.): Le Bibliothèques et musées des arts du spectacle en communauté française de Belgique = Libraries and Museums of the Performing Arts in Belgium’s French-speaking Community. 4. ed. Bruxelles: Centre SIBMAS Communauté française-Belgique 1994. [return]
²² In particular the World Guide to Libraries, Museums of the World and the World Guide to Special Libraries. [return]
²³ SIBMAS International Directory of Performing Arts Collections. General Editor: Richard Buck, Editors: Paul S Ulrich, Ruth Freydank and Roger Rennenberg. Haslemere: Emmett Publishers 1996, 651 pp. [return]
²⁴ This change was also reflected in the title which replaced the words “Libraries” and “Museums” with “Collections”. [return]
²⁵ This has also been retained in the web-based directory, where all SIBMAS institutions are indicated by a blue “S”. [return]
²⁶ Active links from the Directory to web sites of the institutions and email addresses. [return]
²⁷ At the SIBMAS Congress in Rome, it was decided that links to these databases should be included in the entries. The addition of this material has begun to be implemented, beginning with those databases presented at the SIBMAS Congress in 2002 in Rome. [return]
²⁸Since it could be assumed that institutions with large holdings were SIBMAS members, this would give the demo database a high feeling of quality and usefulness and thereby be a good way of promoting the value of the complete database. [return]
²⁹ http://www.theatrelibrary.org/sibmas/sibmas.html [return]
³⁰ Among these problems are managing payments (collecting the money), controlling access to the directory (entering and maintaining passwords) and maintaining statistics for the users. [return]
³¹ It was estimated that the directory would require between 30 and 50 megabytes of disk space. [return]
³² The web-based directory was actually available on the web in December 2001. Initially only members of the Executive Committee had the URL. This permitted them to check the content and send corrections to Paul S. Ulrich, who then made the respective changes. When several institutions linked to this URL, it was picked up by the robots of several search engines, which meant that prior to the “official” presence of the directory on the web, it was already being used by people who had found it via search engines. This put pressure on SIBMAS to officially make the presence known. [return]
³³There are numerous reasons for having so many files: it corresponds to the needs of the users: if he is interested in the institutions in a town, providing him with additional information about other towns distracts him from what he is looking for. The more towns in a file, the larger it will be; this means that the user will be forced to find what he is really interested in on the page. A large file also takes longer to download. Furthermore, by not having long lists on one page it makes it more difficult for institutions looking for lists of addresses for commercial mailings to simply download the addresses and make them commercially available. [return]
³⁴ Africa, Asia, Australasia, Caribbean, Central America, Europe, Middle East, North America and South America. [return]
³⁵ This corresponds to the default file name of the server. Using this name means that it is not necessary to provide the file name in the URL, since it will automatically be used by the server. For example, the URL of the home page of the directory is http://www.theatrelibrary.org/sibmas/idpac/index.html. The URL http://www.theatrelibrary.org/sibmas/idpac/ automatically loads the actual file and is shorter to use. [return]
³⁶ A negative aspect of this convention is that the name of the file does not reflect the content. If the name of the file contained the name of the city, it would increase the ratings on search machines. Creating an algorithm to make use of this negative aspect would not be easy. This disadvantage is not so serious, however, that the administrative advantages of the file name conventions are overridden. [return]
³⁷ Databases belong to the “invisible web” and their content is not indexed by search engines. Hence someone looking for information on a search engine is not directed to the individual entries. This is comparable to library OPACs on the internet, their content can not be found with a search engine. [return]
³⁸ This is comparable to the traditional way of finding information in a library via the catalogs: the primary source of information is what is found on the title page of the book. One must “guess” what information is hidden within the book, since it is not indexed in the catalogs. [return]
³⁹ For practical reasons this principle was rejected for the geographic access, which requires four clicks to access the information: home page, geographic region, country, city. [return]
⁴⁰ Since each page of the directory is an point of entry when it is found by a search engine, it is important that navigation elements for the entire directory be present on each page. [return]
⁴¹ Assuming that this is not intentionally suppressed by the user. [return]
⁴² “<meta http-equiv="expires" content="43200">”, whereby the value “43200” corresponds to the number of seconds in a day, hence the cached copy of a file on the user’s PC should automatically be replaced with the current file from the SIBMAS server if it is older than one day. [return]
⁴³ The html coding is “<h1>…. </h1>” to “<h6>…</h6>” as opposed to” <b>…</b>” for bold face. [return]
⁴⁴ To the naked eye there is no difference between the appearance of text marked as a heading or in bold face, the difference is the weight given to them by search engines. [return]
⁴⁵ When a file is downloaded to a disk, the URL is lost if it is not included in the file itself. [return]
⁴⁶ Since increasingly countries are passing legislation which require that everything should be done to make information equally available to handicapped persons, it is only logical that web designers should begin making usability for handicapped persons a prime requirement in their design. [return]
⁴⁷ A rule of thumb is that one has 20 seconds after a page has begun to be loaded to catch the attention of the viewer. If within these 20 seconds his interest is not caught, he will return to the page he came from and the page will not be viewed. [return]
⁴⁸ It makes good sense to keep the loading time short, since time is money: viewers expend a lot of valuable time waiting for information to become visible on a browser. Furthermore, in many countries the user‘s telephone fares are calculated by the length of time they have telephone contact to their internet provider. Long file loading times mean higher costs for the viewer. [return]
⁴⁹ Statistical information collected by the server. These records include information about where the visitor to the page came from, how long the page was viewed, where the user went from the page, errors encountered using the page etc. By regularly evaluating this information, corrections can be made to the web site in order to optimize usage. [return]
⁵⁰ The increased visits by search engines means that the information from the directory can potentially be found among the results of queries on the search engines. Likewise it means that they are regularly checking to update their information about the site. [return]
⁵¹ This has lead to an in-depth evaluation and standardization of the entries in the index. In particular the names of persons with special collections are being standardized and expanded to include full names (with variants) dates of birth and death, and professions. [return]