Working Group 2: TEI Text Encoding in Libraries Discussion Summary and Recommendations

Most of the members of Working Group 2 were from libraries and involved in electronic text projects, so there was a large base of shared experience. The librarians had many goals for their projects, but most centered on collection development and public service. Most attendees follow the TEI Guidelines in their projects, and most reported using the TEILite DTD. In general we agreed:

  1. Librarians value interoperability very highly, and chose TEILite in part for its promise of future interoperability among projects. Cross-collection searching, for instance, is seen as highly desirable but no one is sure how difficult that will be, given differences among individual projects' encoding practices. We felt that while the TEI provides a common set of rules for text encoding, further agreement will be necessary to allow for easy interoperability.
  2. Within the group, there is a range of encoding practice, particularly regarding depth of encoding. We felt that whatever guidelines are drafted should reflect this range of practice, and should allow for future enhancement of lightly-encoded texts.
  3. An umbrella organization must be found to allow for continued discussions among librarians who use the TEI for electronic text projects. Past experience with MARC development demonstrates that efforts such as this one should be international in scope in order to prevent divergent practices from developing.

Recommendations for immediate action (with deadlines and assignees noted):

  1. Establish a listserv for Working Group 2 members; invite members of other Working Groups to join. For the moment, restrict membership to conference attendees, and consider opening up the listserv membership at a later date. (July 15: Perry Willett, Indiana)
  2. Establish a Task Force of Group 2 members to draft guidelines for "best encoding practices" with library focus. Task Force members should already have drafted such documents for their local projects. The Task Force will consider the encoding guidelines from other projects (see #3 below), and draft a "Manual of Accepted Practice" that indicates areas of common agreement and recommends best practice. The Task Force will meet before mid-November 1998, perhaps at the Library of Congress, with travel and lodging funded by the Digital Library Federation. The Task Force would have a draft distributed to the entire group before ALA Midwinter. (Task Force named by August 15: Perry Willett, Indiana; Task Force draft recommendations by December 31.)
  3. Share any documents drafted for TEI projects that detail encoding, editorial and work practices. Share representative examples of full-text. (For those with documents already available: July 15; any project that wishes their documents to be considered by the Task Force must submit them by September 15: Michael Sperberg-McQueen, Univ of Illinois, Chicago)
  4. Identify the requirements of organizations which might serve as an umbrella. Several organizations were mentioned, including ALA (and the many sub-organizations within ALA), Canadian Library Assn., the DLF, and IFLA. (Chuck Spornick, Emory, and Doug Hodges, National Library of Canada)

Recommendations for future actions:

  1. Develop pedagogical tools for training in the use of the TEI. It was noted that the EAD provided a model for how to organize to provide such training.
  2. Use sample full-text from several projects to develop a testbed of documents for testing the viability of cross-collection searching.
  3. Identify forums for programs/meetings/articles to discuss the issues raised by and results achieved by this Working Group.

Perry Willett (Indiana University)


Page maintained by Kat Hagedorn
Last modified: 03/05/2013