As I wrap up cataloging the last few maps and polishing the last blog post for this phase of Council on Library and Information Resources (CLIR)-funded map cataloging for BHS, the time has come to let everyone know what we have accomplished in the last 17 months.
The purpose of a CLIR Hidden Collections grant is to ‘uncover’ ‘hidden’ collections, by making previously uncataloged collections available for discovery on the Web. For libraries, this goal is achieved by the creation of MARC (machine-readable catalog) records for each item in the collection for inclusion in local and international online library holdings catalogs. As libraries face an increasingly digital future, it has become highly desirable to expand these records to include more detailed machine-readable data to be ready for use by future online systems, some barely imagined yet. For map cataloging, this means creating and including formatted GIS information.
The ‘hidden collection’ for this phase of the grant project was the 20th century map collection. Since August 2013, records for more than 470 maps and 35 atlases owned by BHS have been added to Bobcat, the online catalog hosted by New York University. Our holdings are also included in the OCLC catalog, and are accessible through Worldcat. Not only have we made these holdings known, but we have enhanced the catalog records wherever possible by including detailed information usually not recorded in a catalog record to increase the likelihood these maps be discovered by those searching for such information. The rich detail of these enhanced records also makes it possible for a researcher to more reasonably assess whether a particular map will meet their needs.
To see a typical record enhancement, I’ll use the “Nester’s Brooklyn Maps” record in Bobcat as an example. The view shown below—which can be found by clicking ‘more bibliographic information’ in the Bobcat record display—reveals the expanded and detailed content note for the map, along with a note describing content found on the verso. These notes almost always include the phrase ‘covers’ to give a clear geographic description, and ‘shows’ to indicate what kind of information might be found on the map. These note fields are searchable in the catalog by keyword search, so anyone wondering about Brooklyn automobile routes in the 1970’s will find this map as long as they use the keyword search box–usually the default search in Bobcat. In addition, these details, if considered important, are also reflected in the subject headings assigned to the map and so become links in the record display. This map will collate with other maps with the subject heading ‘Streets — New York (State) — New York — Maps’ or ‘Downtown Brooklyn – Maps.’ In this way, this descriptive catalog record makes discoverable more features about the maps, while providing additional information with which to evaluate its content.
While borough-wide and city-wide maps are enhanced with attention to important details on the maps, maps covering smaller portions of Brooklyn have been analyzed by neighborhood. Below is a view of the “Borough of Brooklyn 52nd Assembly District, 1971” map record. In this instance, it is evident the cataloger took pains to identify all seven neighborhoods through which this meandering strip of a district made its way from the Sunset Park to the Fort Greene neighborhoods of Brooklyn.
These neighborhoods are documented in notes, in subject headings, and in machine-readable code found in the 052 field (see the ‘MARC tags view’ below). This identification of neighborhoods, painstakingly undertaken with Kenneth Jackson’s Neighborhoods of Brooklyn (1998) in hand, has proven to be a valuable enhancement for researchers here at BHS, who often seek information for specific neighborhoods.
Finally, as part of the CLIR project, we have enhanced catalog records by including geographic coordinates indicating the coverage of each map whenever possible. To return to “Nester’s Map of Brooklyn,” this time on the MARC tags view shown below, scale and coordinate statements are recorded in machine readable and normal formats in the 034 and 255 fields, respectively. Many maps do not include geographic coordinate information, and it was necessary to determine the boundaries for each map on a case by case basis and put the information into the catalog manually. For this work, the online Bounding Box tool was invaluable, for it made the process as simple as drawing the outline of the boundaries on the Bounding Box tool, and then copying and pasting the coordinates (already properly formatted) right into the catalog record. We were able to include this information as we created new records for maps which had never been cataloged before, and we have started to add them to maps which had already been cataloged without this information. This retrospective addition of coordinates to old catalog records will be an ongoing process for the cataloging community as we prepare to move to an increasingly digital (i.e. machine-readable data) environment. Here at BHS, we have begun making a contribution to this work both in our own catalog and in Worldcat.
To share this work, we are publishing a preliminary spreadsheet compiled from the records which have been completed on our catablog, Emma. The spreadsheet gives basic information about the maps which are helpful for identification: title, author (when available), date, and geographic subject headings. In addition, we have included the OCLC record number, which may prove useful as a unique record identifier as the library catalogs and other indexes move to an increasingly open linked data environment. Also included is the geographic information in the machine readable and standard formats mentioned above. The information on this spreadsheet can be easily edited and formatted into a CSV or KML file and uploaded as a Google Fusion Table where it can be shared, developed and used collaboratively, or imported into GIS applications such as ARCGIS or Google Earth, tools which allow users to display the geographic coverage of our maps, even for specific points in time. These are powerful visualization tools which will give users access to our maps in a way the traditional library catalog cannot.
The creation of the data in the spreadsheet we are posting to Emma is the first step in the process that will bring about such interactive GIS displays. We will continue to enhance our catalog records and expand this spreadsheet, in the hope that we will be ready with the data for future digital projects. It is our hope this preparation will make us attractive to collaborative partners and grant administrators alike. Watch the BHS blog for future posts on our work in this area.
In the meantime, be sure to see the interactive GIS-based map created to illustrate individual cases found in the Brooklyn, N.Y., Department of Law, Corporation Counsel records, 1843-1920 processed by my colleague, John Zarrillo. Although John utilized pinpoint GIS coordinates instead of bounding coordinates, as was appropriate for his collection, the map illustrates very well how information such as titles, dates, subjects, or item location information can be easily displayed with the wave of a mouse.
The “City, Borough, Neighborhood, Home: Mapping Brooklyn’s Twentieth-Century Urban Identity” project was spearheaded by Julie May (Head of Collection Management), Elizabeth Call (former Head of Reference and User Services), and Jacob Nadal (former Director of the Library and Archives), all of whom contributed to the success of the project. The grant also funded the processing of our Brooklyn, N.Y., Department of Law, Corporation Counsel records, 1843-1920. Processing archivist, John Zarrillo, provided a tremendous amount of assistance in assessing GIS applications for the collections processed and cataloged with this CLIR grant. Thanks are also due to Matt Knutzen, Geospatial Librarian at NYPL’s Map Division, who gave us an orientation on potential applications for GIS information in library records. Finally, we would like to thank the Council on Library and Information Resources (CLIR), whose generous funding made this project possible.
 For an succinct explanation of how such data can be converted into a GIS-based interactive index, see the PDF of the ALA presentation “Map Indexes: a Practical Application of GIS for a Map Collection” by Christopher J.J. Thiry of the Colorado School of Mines.