Various federal and states government agencies provide geographically referenced digital data to the federal depository libraries around the country with or without some viewing software packages. The libraries receiving these data faced new challenges of storing, maintaining, and providing access to these data. In response to these new challenges, research libraries explored various strategies to make the geographically referenced digital data accessible to library users. A few libraries took the initiative and introduced GIS service in their libraries; other libraries were uncomfortable introducing this new service in the library environment.
The real initiative started with the development of the ARL/GIS Literacy Project in June 1992. This project evolved when the Association of Research Libraries (ARL), in partnership with the Environmental Research Science Institute (ERSI), decided to develop GIS programs in various libraries. The project was very successful and now most of the major academic libraries in the United States and Canada are providing GIS services in their libraries.
The introduction of GIS service in the libraries has created a new profession called the GIS librarian. The main role of this profession is to collect, store, manage and provide access to geospatial data. There are different ways of providing access to a collection. In this presentation, I will discuss and share my experience of using different methods of making geospatial data accessible in the Princeton University Library.
Princeton University Library’s GIS service was started in the Fall of 1998. Before the formal start of a GIS service in the library, the library bought many commonly used geospatial data including the library-wide license of the Wessex Streets database. This database comes with software that will allow a person to extract the data. However, to provide library users access to the database, we need to either load all the database and software in every computer, or to load software in every computer and the database in a server. In addition, people need some level of training to extract the database from the Wessex Streets Installer software.
In order to solve the problem of giving access to Wessex Streets data online, we have used the PERL script to make the database accessible via a web page.
The Wessex Streets 5.0 database is organized into nine CD-ROMs, including a CD-ROM for setting up Installer software to extract ArcView format files from the CDs. In each CD-ROM, there are four to thirteen states, depending on the size of the state. All of the files are compressed and are organized in a folder. For example, Streets/State/County.
In order to make all the state and county theme layers accessible online via a web page, we copied all the state and county files on our server on the folder called Streets. Then we made a text file for each state that contains information such as state FIP code, state name, county FIP code and county names (34:New Jersey:34021:Mercer County) and saved them on the same folder where the Wessex Streets database was located. After that we wrote four PERL script routines and stored them on the CGI folder, that is, C:\InetPub\WWWRoot\cgi-bin. We also made two html documents and saved them on C:\InetPub\WWWRoot\. The first html document will allow a user to either download the national level compressed data files directly from the server through hyperlinks, or to select the state and county level data. Once the person clicks on the state and county level data, it will open a second html document, and this page will allow the patron to select one state. Once the state is selected, the user is given an option of selecting state or county level data layers.
This is how the four PERL scripts work: The first PERL script will allow a user to select a state and particular county in that state; the second script will allow a person to select either a state level or a county level data layer; the third script will allow a user to select state or county individual data layer, and it will also create a temp directory on the server to store the selected layers; the fourth script will uncompress the state and county files and then select the individual files that were requested by a patron. It will then compress those files as a regular zip file for all platforms or a self-extracting zip file for Windows95/NT and make them available for downloading. This PERL script provides a very powerful and easy way to download any Streets file provided by the Wessex Streets database. This has allowed our library to make the database accessible without installing the Wessex Street 5.0 installer software and database in any individual computer.
Providing the Wessex Streets database online via a web page helps us to make this database accessible to all university library users without installing any particular software; however, it does not provide information about the data. One of the main functions of a GIS Librarian is not only to give access to data, but also to catalog information so that people can find the data, and also find information about the data. We recognize this, and therefore, we have created metadata for the Wessex Streets database that are compliant with the FGDC standard and made it searchable through the FGDC Clearinghouse. Before creating metadata, we looked for words that are crucial for making the data accessible. We organized the metadata in such a way that the user can search each and every state, their county name, and their individual data layer. For example, each state has one metadata record but on a “Theme Keyword” we included all the themes that are in the individual state and their counties, and on a “Place Keyword” the states and their individual county names are listed. This will allow a person to search the data layer of a particular state and their county. We have also made sure, while creating the metadata, that Online Linkage on the metadata is linked to our Wessex’s Streets web page so that a patron can get access to the data even without having a good knowledge of our web page resources.