Clearinghouse of Image Databases
The Clearinghouse of Image Databases is an online directory of image databases housed on a gopher server at the University of Arizona Library. It was created in July 1994 by Stuart Glogofi, Assistant Dean for Library Information Systems at the University of Arizona, who continues to maintain it.
The Clearinghouse’s intended audience was librarians and archivists who were either planning or developing databases containing digitized images from their collections. Since its inception, however, the Clearinghouse’s scope and purpose has been expanded to gather technical and descriptive information about imaging projects as information resources, regardless of where these projects are being developed. Within this broad purpose is the expectation that, if this information is presented in a relatively non-technical fashion, it will he highly useful to people in planning and developing imaging projects. It will help avoid duplication and lessen the learning curve by identifying others who have worked with a particular storage media or original material. Because image databases are a relatively new endeavor, there are many information professional, scientists, and teaching faculty are new to this technology. The audience for Clearinghouse information is perceived to be vast and international in scope
The University of Arizona Library, like other large university libraries, has a mission to promote the dissemination of information. The Clearinghouse is one effort to address this need. In the past two years, the hardware, software, and telecommunications needed to create and distribute an image database has advanced to the degree that interest in image databases is becoming widespread. Technological developments have also raised new interest to create image databases for the preservation of rare and special collections.
At the time the Clearinghouse was created, it was hoped that it would prove to he a useful contribution. Although others have helped fine tune the data model and construct the WAIS index, the Clearinghouse is the result of the efforts of one individual. Crucial activities such as entering data, creating links, editorial functions, soliciting participation, and implementing new features continue to reside with this individual. This raises a continuing support issue: there is the risk that if this person leaves the University of Arizona Library, the Clearinghouse will not continue to be supported.
Content and Scope
The Clearinghouse is multidisciplinary and international in scope, although all entries to date are in the English language. While its primary orientation is to list projects, these listings are supplemented with information on related publications, keys contacts, the project’s purpose, and comments from the person submitting the information.
Projects are not limited to a particular media. For example, if a project is submitted to the Clearinghouse and it does not fall within an exiting media directory, a new directory is created. (See Appendix A for a list of the media currently included.)
Editorial Policies and Process
The Clearinghouse operates with an open invitation for submissions. Projects are submitted via electronic mail, preferably according to the format described in the Clearinghouse’s “How to Submit Projects.” Little editing is made to submissions. For the most part, the submission is fp’ed to a directory on the Library’s gopher server and cap files are created for the institutional name and project name listings. Submissions that neglect to enter complete information are still added to the Clearinghouse in the hope that even an abbreviated submission will be useful to someone. Like Blanche in A Streetcar Named Desire, the Clearinghouse depends upons the kindness of strangers.
The format for submissions is specified in “How to Submit Projects” (See Appendix B). In addition, World Wide Web users will find a form on the library’s web server for submitting entries to the Clearinghouse.
To maximize the voluntary participation of databases developers in listing their projects, the data model use5l in the Clearinghouse focuses on collecting only the most critical, basic information. Data fileds may contain highly teclmical information, but field headers use terminology that will clearly he understood by developers with a variety of backgrounds and levels of expertise. Clearly, asking people to complete the data element form may deter some from submitting their projects.
The Clearinghouse is available to anyone who has access to the Internet, regardless of the sophistication of their hardware and software. Users can access the Clearinghouse if they can execute a telnet or gopher command, or via a World Wide Web client.
At the time the Clearinghouse was conceived and implemented, gopher and telnet access was the best option. Since then, World Wide Web has exploded and access via lynx (for VT1OOftextonly access) and graphical Web browsers, such as Netscape, were added in the spring of 1995. (See Appendix C for access information.)
At the time this essay was drafted, the University’s Library Information Systems Team (LIST) was enhancing project displays via Web browsers. A PERL script was written that takes the ascii files retrieved from the gopher server and converts them to an html format when displayed in a Web browser. In addition, LIST is exploring a way to embed a hyperlink within project descriptions having a URL so users accessing the Clearinghouse via World Wide Web might connect to the actual image database listed in the Clearinghouse.
Clearly, a resource that is available in an electronic form must keep pace with technological change. Having the time and resources to continually upgrade a project like the Clearinghouse is a continuing challenge.
The Clearinghouse was constructed with the intention of presenting its information in an open, easily identifiable format. Each entry is listed alphabetically by institution name and again by institution name but this time sorted in a media–type directory.
In the spring of 1995, access to the Clearinghouse was added via the University of Arizona’s World Wide Web server. In addition, WMS keyword searching was added. This has enhanced finding information contained in the Clearinghouse’s project descriptions. With access available via World Wide Web, high-level access tools such as Netscape and Mosaic can be used. Persons with these clients can set bookmarks to facilitate navigating directly to the Clearinghouse directories.
The Clearinghouse does not use descriptors or a controlled vocabulary. No special training or academic education is necessary to makes submissions or to work proficiently with the database.
The data element resource identifiers offer the potential of building a separate database that would be easily searchable by standard database management software, such as MS Access, Paradox, or Quatro Pro. Such a database will only be useful if project descriptions provide nearly complete data element information. A fiiture issue regarding data elements pertains to the need to augment current fields to reflect changes new technologies may introduce. For instance, if watermarking becomes extremely important, it will be important to add a data element that collects this information.
Another issue related to the data elements concerns completeness. As mentioned earlier, a balance was struck between collecting enough information on a project without going overboard. The Clearinghouse focuses on showing cornmona1ities so users can easily identify the information they are seeking.
The Clearinghouse largely relies on posting messages to selected listservs and newgroups inviting subscribers to submit information on their projects to the Clearinghouse. On occasion, an image database is found that looks appropriate for listing in the Clearinghouse and its developer is contacted.
The Clearinghouse is one of many directories with multiple files on the Library’s Gopher and Web servers. The information is backed up regularly to tape according to standard procedures.
: Different Media Currently Supported By the Clearinghouse
:How To Submit Projects the Clearninghouse
Appendix C: Accessing the Clearinghouse of Image Databases
v2.0 sjg 8/13/95