General Information

1. Introduction
2. Project History
3. Participating Institutions
4. Technical Information

Introduction

The Irish Literary Collections Portal provides access to the finding aids of over 100 of North America's Irish literary manuscript collections. The collections represented in the portal range in date from the Irish Literary Renaissance to the present and contain a wide variety of materials including correspondence, manuscripts and typescripts of literary works, photographs, and printed material. Researchers should note that this site contains descriptions of collections rather than electronic versions of the individual items within these collections.

Finding aids are tools created by archivists to present detailed information about the content and organization of a collection. This information enables researchers to browse a collection and to locate specific materials of interest. Though finding aids vary in length and level of detail, they generally consist of the following: a biographical or historical sketch, a scope and content note, an outline of the collection's organization or series arrangement, relevant administrative information about copyright and restrictions, and a detailed container list of numbered boxes and folders along with a description of their contents.

This portal brings together finding aids from a number of institutions into a single searchable database. The "Search" function allows researchers to conduct keyword searches within the finding aids and to filter those searches by creator and/or repository. By using the "Browse" function, researchers may quickly view an alphabetical listing of all the collections available through the portal. Entire finding aids may be downloaded and printed in PDF-format.

Project History

In 2001, a two-year grant from the The Gladys Krieble Delmas Foundation helped Emory University and Boston College to digitize finding aids and develop a search interface that allows scholars both on- and off-site to explore the collections and quickly find relevant materials. In 2006, a second grant from the Delmas Foundation allowed the project to incorporate the finding aids of Irish literary manuscript collections at several additional repositories.

Participating Institutions

General information on the hours, location, and access to the collections contained in the Irish Literary Collections Portal may be found at the following web sites:

1. Boston College, John J. Burns Library, Archives and Manuscripts
2. Emory University, Manuscript, Archives, and Rare Book Library
3. Pennsylvania State University, Special Collections Library, Rare Books and Manuscripts
4. Southern Illinois University Carbondale, Morris Library, Special Collections Research Center
5. University of Delaware Library, Special Collections
6. University of Texas at Austin, Harry Ransom Humanities Research Center
7. Wake Forest University, Z. Smith Reynolds Library
8. Washington University at St. Louis, Department of Special Collections

Technical Information

The findings aids for this project have been encoded using the Encoded Archival Description document type definition (EAD.DTD), Version 2002, an XML-compliant data structure developed and maintained jointly by the Society of American Archivists and the Library of Congress.

Three EAD templates based on the complexity of the finding aid were created in X-Metal: a non-series template, a series template, and a subseries template. Using the RLG Best Practice Guidelines for Encoded Archival Description (RLG EAD Advisory Group, August 2002), we incorporated recommended ID attributes, encoding analogs and the normalization of dates in each EAD document.

The finding aids were created in Microsoft Word and the high-level elements were cut and pasted into the appropriate X-Metal template. Each component-level container list was converted to a text file and then run through a perl script which applied the appropriate tagging. The container lists were then cut and pasted back into the template in X-Metal.

Further transformations of the X-Metal data were made through a process written in Apache Ant: replacing bad data (invalid utf-8 characters introduced by Microsoft Word), changing entities to unicode, adding a unique id for most elements, and creating eXist database load files.

The data was loaded into an eXist database and then retrieved with X-Query calls. X-Query implements a subset of the XPath specification (a W3C recommendation). Web pages were dynamically generated with XSL Transformations (a W3C recommendation) and PHP.