[Archive copy mirrored from: http://www.cimi.org/about/history.html, August 1997]
|
|
| Since its founding in 1990, CIMI has been dedicated to encouraging the use of standards--by finding the standard, creating consensus around it, testing it, and disseminating it to the museum community. Our history, then, is mainly a history of projects and their results. |
Initial Agreement on Standards: SGML for structuring information and Z39.50 for information interchange
CHIO Structure: A Museum Demonstration of SGML
CHIO Access: A Museum Demonstration of Z39.50
Testing Z39.50 Across the World: The CIMI Interoperability Testbed
CIMI Case Study of Integrated Information Management
Initial Agreement on Standards
During the 1980s, information professionals in the museum community recognized that the key to long-term availability and wide electronic access to information lay in the use of standards. So, CIMI was born in 1990 as a research project to develop community standards to support the preservation of museum information in digital form and enhance the potential for scholarly information exchange. CIMI's first aim was to provide guidelines to museums on the types of information interchanges museums might engage in and how to be positioned to do those, taking advantage of information industry-wide developments in the sharing of information across systems.
For two years, representatives from all the major North American museum associations and network service providers attended meetings that contributed to the creation of a standards framework for electronic exchange of museum information, published in 1993. The CIMI Standards Framework, <http://www.cni.org/pub/CIMI/www/framework.html>. became a blueprint for how the museum community could not only adopt standards but test them through practical demonstration projects using real museum content.
The two key standards that CIMI endorsed in the 1993 document were Standard Generalised Markup Language (SGML) for structuring information and ANSI Z39.50 for search and retrieval. At the same time, through research and a series of consensus-building meetings, CIMI developed the CHITG Scope of Collections finding aid for museum content, which is equivalent to a collection-level record for archival content.
The natural next step was to develop a demonstration project to show how the standards could be implemented. In 1994, CIMI designed Project CHIO, Cultural Heritage Information Online, a two-part demonstration project to test these standards. The first part, CHIO Structure, funded by the Telecommunications and Information Infrastructure Assistance Program (TIIAP), set out to test the SGML standard. The second part, CHIO Access, funded by the National Endowment for the Humanities, set out to test the Z39.50 standard.
CHIO Structure: A Museum Demonstration of SGML
In CHIO Structure, different databases containing full text, object records, and images on CIMI members' computers were "marked up" using SGML. SGML makes it possible to add markup (extra, non-intrusive information) to existing documentation, database records, full-texts, catalogs, and images, making them available directly for electronic access. The Text Encoding Initiative (TEI), a NEH-funded project, endorsed and implemented SGML as a standard for the encoding of electronic texts for use by scholars. During CHIO Structure, CIMI built upon the TEI work by creating an application of SGML specific to museum content.
Adapting SGML to a given genre requires developing an agreed-upon set of categories and guidelines that will be used when marking up documents. In SGML parlance, this tool for enabling effective mark-up is known as a DTD (Document-Type Definition). The creation of the CIMI-DTD <www.cimi.org/products/cimi_products.html/cimi4.html> and accompanying Tagging Guide <www.cimi.org/products/cimi_products.html/tagguide.html> is one of the most significant contributions of Project CHIO to the museum community. Its design reflects a community-endorsed system for encoding museum text using SGML and provides a framework for any future work that CIMI or any cultural heritage organization might do in applying SGML to museum information.
To develop the CIMI-DTD (and to make effective searching possible as noted below), the CHIO Structure team needed to develop a set of access points. These access points are common themes in museum content (such as technique, material, nationality, etc.). Both the CIMI-DTD and the Access Points were very significant steps in the direction of establishing helpful, practical standards for museum information. For example, a major software developer, SoftQuad, is issuing a new humanities version of its SGML tool, Author Editor, which incorporates the CIMI-DTD.
Above all, we learned from Project CHIO that there are other factors beside the technical appropriateness of a standard that influence whether it is workable in real life. As we noted in the conclusions in the project's final report:
"SGML projects are not just about technology. Other factors come into play that have an impact on the implementation of SGML, including: organizational behaviour; factors of economics; institutional policies about information management, access, and distribution. Future work needs to include strategies for building a clear, simple economic model to point museums to the viability of using SGML." [The Consortium for the Computer Interchange of Museum Information, Project CHIO: Cultural History Online; Final Report for US Department of Commerce, Telecommunications & Information Infrastructure Assistance Program, (Washington, DC: US Department of Commerce, 1996).]
CHIO Access: A Museum Demonstration of Z39.50
Following on the work accomplished in CHIO Structure, team members working on CHIO Access, supported by an NEH Preservation and Access grant, used the CIMI Access Points to create an application of the Z39.50 standard for search and retrieval specific to museum content. To implement Z39.50 for a particular body of information, the information must be in a format that can be searched and retrieved, such as documents that have been marked up using SGML. Adapting Z39.50 to a particular content area involves the development of an "Application Profile." This profile specifies the kinds of queries that will be used and how they will be structured. During CHIO Access, team members developed the CIMI Profile <http://lcweb.loc.gov/z3950/agency/profiles/cimi2.html> for use with museum information.
Marking a positive new direction for Z39.50 application profiles, the CIMI Profile was developed as a companion profile to the Library of Congress's Z39.50 Profile for Access to Digital Collections. The Digital Collections Profile, as it is called, has been under development since August 1995 by a working group established and coordinated by the Library of Congress with active participation from CIMI. It specifies the use of Z39.50 for access to digital collections organized through structured descriptors defined in the profile. It provides the semantics needed to navigate digital collections and to locate and retrieve objects of interest.
The Collections Profile serves as a "base" profile, a foundation upon which other "companion" profiles can be built for search and retrieval in digital collections. As one such companion profile, the CIMI Profile is an independently developed profile that lays out compatible extensions to the Digital Collections Profile for search and retrieval of digital museum information. The Digital Collections Profile focuses on the access and navigation of digital collections, without regard to the content or domain of those collections. The CIMI Profile focuses on the search and retrieval of specific objects contained in the digital collections of museum information.
Testing Z39.50 Across the World: The CIMI Interoperability Testbed
As the final phase of CHIO Access, CIMI is testing the workability and effectiveness of the CIMI Profile through a Z39.50 testbed project. In March 1997, we sent out a call for participation in this testbed project. An overwhelming number of international organizations working with cultural heritage information responded. The strong response from organizations working with museum content confirmed the pressing need for work on interoperability in this field. CIMI's test will be carried out with participation from the largest museum collections management vendor in North America, from several groups involved in enormous cultural databanks throughout Europe, and from a vitally important national museum project in Taiwan.
CIMI was able to choose 5 participants and 2 alternates to take part in the CIMI Z39.50 Interoperability Testbed Project after painstaking review of applications from 42 highly qualified organizations. The chosen groups are working together with CIMI to further develop the CIMI Profile as an application of this international standard for search and retrieval appropriate to cultural heritage information. In order to accomplish this, they are receiving training in the use of the standard and CIMI's particular application. They are also receiving tools and assistance in implementing the standard on information from their own local systems.
CIMI Case Study of Integrated Information Management
Project CHIO made it very clear that different museums have different missions, resources, and approaches to their information. These differences need not be treated as confounding obstacles. Rather, they can be regarded as a natural (and healthy) state of affairs that any program of standards must adapt itself to. CIMI sees itself as providing standards, not strait jackets. Therefore, following directly on CHIO we decided to further the CIMI Guidelines for Practical Information Access through a case study project with eight museums. The goal of the case study is to verify that the standards-based information strategies described in the guidelines have validity in a "real life" institution.
