Terence R. Smith
Professor of Computer Science
Professor of Geography
University of California, Santa Barbara
Director, Alexandria Digital Library Project
D-Lib Magazine, March 1996
The goal of the Alexandria Project Digital Library (ADL) is to build a distributed digital library (DL) for geographically-referenced materials. A central function of ADL is to provide users with access to a large range of digital materials, ranging from maps and images to text to multimedia, in terms of geographical reference. An important type of query is ``What information is there in the library about some phenomenon at a particular set of places?''.
From the Internet, both users and librarians can access various components of ADL, such as its catalog and collections, through powerful, graphical interfaces without having to know where these different components are located on the Internet.
The main aspect of ADL's strategic approach involves:
1. A focus upon access to the many classes of collection items, including non-traditional items, by geographical reference;
2. The development of the user interface (UI) and catalog components of the DL architecture;
3. Accessibility to the ADL catalog and collections via the Internet for a wide variety of users;
4. Close interaction and interoperability with other DL activities by way of Internet-related technologies;
5. A process of incremental, evolutionary design and implementation of ADL that takes advantage of critical technological developments, and especially Internet-related technologies; and
6. Digitally-supportable extensions to traditional library functionality.
In particular, the transitions from our initial stand-alone rapid prototype (RP) in early 1995, to our first World Wide Web prototype (WP) in late 1995, to our publicly-available testbed in mid-1996, to our million-item operational library in mid-1997, are being made in incremental steps that build upon each other and upon a basic four-component architecture. This architecture involves: (1) user interface components that support graphic and text-based access to the other ADL components and services; (2) a distributed catalog component that metadata and search engines permitting users to identify holdings of interest; (3) a distributed storage component containing the digital holdings; and (4) an ingest component allowing librarians to store new holdings, extract metadata from the holdings, and add metadata to the catalog.
A variety of technologies being applied and developed in each of the four components. The graphical/geographical interface is supported by a variety of Internet-related technologies, such as browsers and programmable browsers. Access to the holdings is by way of a catalog component that supports spatially-based metadata models and content-based search techniques. Currently, such techniques employ gazetteers for map documents and texture features for image documents. Browsing and the delivery of large items is supported by progressive delivery techniques based on wavelet technology. The high-performance servers that support the library operation are being based in part on approaches that involve parallel computing.
The research and development in our first cycle activity, lasting six months, yielded a stand-alone rapid prototype (RP) testbed library, which was based on commercial database management systems and geographic information system technology. The RP was distributed for evaluation by major users as a CD-ROM.
The research and development in our second cycle of activity, lasting six months, led to a ``Web prototype'' (WP) testbed library accessible from the World Wide Web (WWW), but with current accessibility limited to a small number of major users and project partners.
This system involved the development of a complex WWW interface, a catalog for metadata, and the preliminary applications of research results relating to image processing and parallel computing technologies.
Conceptually, the WP UI is a collection of HTML "pages" implementing three major search capabilities: map browsing; gazetteer queries; general catalog queries.
As well as control/configuration and help/glossary links, the user interface (UI) is designed around a state transition model with each state representing a WWW form or page, some of which include partial or complete query results. The HTML code for the WP UI is generated dynamically by approximately 15K lines of Tcl code running in a NaviServer HTTP server.
The primary function of the both the map browser and the gazetteer pages is to allow the user to define spatial extents or regions for catalog searches. The map browser allows these search regions to be defined explicitly (by zooming and panning a base map), while the gazetteer defines them implicitly (as the footprints corresponding to place names and feature types.) Figure 1 shows a screen dump of the map browser.
The visible portion of the map browser's base map (the display window) is the default search footprint (the query window), but this relationship can be modified (e.g., the user may specify a subset of the display window, or may direct that the display window be completely ignored.) The base map is also the background on which the gazetteer and catalog query result footprints are drawn. The base map images are dynamically generated by a Common Gateway Interface (CGI) application based on the Xerox PARC Map Viewer [http://www.parc.xerox.com/map/], which we have modified to support generic labeling, fast panning, and graphic overlays.
Figure 1: The map browser component of the interface.
Figure 2: The browse graphic display returned from a query.
Gazetteer queries may interact with the map browser. For example, if the current map browser query window contains the USA but not Europe, then a gazetteer query with the place name set to "Paris" (and the query window enabled) will return Paris, Texas but not Paris, France. The map browser, in turn, may be directed to reset the query window to the minimum bounding geographic rectangle for the gazetteer query results.
Query windows resulting from gazetteer-map browser interactions are ultimately passed to the catalog page for incorporation into catalog queries. In addition to geographic footprints, the catalog page allows the user to search against any of the metadata fields (such as theme, time, or author) in the ADL catalog, expressed as textual or numeric values.
Catalog queries are assembled from user input into a generic conjunctive normal form (CNF) representation, and then translated to the specific query language (currently SQL) of the catalog DBMS . Query results are converted to HTML tables, with hyperlinks to browse images and on-line holdings. Query results are presented incrementally, with a subset of the metadata fields displayed initially and complete fields subsequently displayed for user-selected holdings. The format and fields used in the query results are completely user-configurable.
Queries may also return the footprints of ADL holdings, which may be displayed on the map browser base map. Unfortunately, it is common for many more footprints to be returned from a catalog query than can be shown intelligibly on the map browser's relatively small display. When footprints of multiple data holdings are displayed on the same map, it is difficult to distinguish which footprint is associated with which item. We continue to experiment with heuristics and visual aids (such as clustering and labeling) for disambiguating "crowded" footprints. In Figure we show examples of the browse graphics that may be returned as the partial results of a query.
The WP UI stores all user configuration parameters, query statements, and current query result sets in a separate (from the catalog) database maintained by the NaviServer HTTP server. This state information may also be stored on request on the client side in "hidden" HTML form variables. This allows a user to save an ADL "session" by using the browser's save-page feature. The session may be restored by reloading the saved page. Otherwise, state maintenance is handled entirely by the server, with only a minimal opaque handle used on the client side to identify the current session.
The WP is proving to be an excellent model for the testbed libraries that we are currently developing, and for an operational library that we are now designing. The main goal of the current cycle is to make the testbed accessible to anyone with a WWW connection by mid-1996. This involves developing major collections of DL documents and servers with appropriate power. A goal of the next cycle will be to construct an operational library of over a million items by mid-1997.
Recent accomplishments include the following:
1. Distribution of CD-ROM of rapid prototype of ADL, which is being tested and evaluated by many data users/producers.
2. Design and implementation of a web prototype of ADL, which is a system accessible from WWW and is being tested by variety of agencies.
3. Design and construction of metadata schema for ADL, which is based on FGDC/USMARC and is being used by several organizations.
4. Implementation of content-based retrieval in ADL, involving access to digitized maps by named earth surface feature and access to digitized images by named texture feature.
5. Implementation of progressive browsing and delivery of images.
6. Development of parallel servers based on multicomputers.
Our plans include the following:
1. Creation of a testbed accessible to anyone from the WWW, available by mid-1996 which provides access to significant collections.
2. The design and construction of a new component of the ADL interface that is based on a multi-level ``Alexandria Atlas'' and that supports graphical/geographical access to many classes of documents by geographic reference.
3. The design and implementation of advanced models of geographical access using complex and fuzzy geographic footprints.
4. The development of a general model of metadata and a catalog based on this model that integrates many forms of metadata.
5. The development of significant collections of maps, images, and text with the aim of having a million-item, operational library by mid-1997.
6. Interoperability with other DLs (University of California, Berkeley; Stanford University; University of Illinois, Urbana-Champaign).
7. Significant extensions of content-based search functionality based on gazetteers, and image texture, color, and multispectral images.
8. Support for a high-performance library based parallel computing technology.
We are developing our catalog and metadata component in close cooperation with a variety of partners, including CIESIN, Central Imagery Office, ERDAS, ESRI, Hughes, Library of Congress, NASA, Oracle, USGS, and US Navy. We are sharing the metadata schema that we developed with various groups (e.g. CIESIN). We are making our testbed facilities available to various organizations facing a need to store their collections in publicly-accessible form (e.g. Sierra Nevada Ecology Project, Mojave Desert Ecosystem Project). We are providing versions of our system for testing purposes in the applications of cooperating partners (e.g. USGS).
We are interacting closely with three of the DLI projects in developing interoperable libraries, and sharing our expertise in spatially-indexed information.
To date, the most significant event in the development of the Alexandria Digital Library has been the construction of a WWW-accessible prototype. This prototype integrated acceptable versions of all four basic components (interface, catalog, storage, and ingest) and employed significant image-processing and parallel processing technology. The system is currently being tested by various partners and is the basis for the publicly-accessible testbed system that is currently under development and that will be made available in mid-1996.
For further information, see http://alexandria.sdc.ucsb.edu
The Alexandria Digital Library Project is a consortium of universities, public institutions, and private corporations headed by the University of California at Santa Barbara, and supported by NSF, ARPA, and NASA under cooperative agreement NSF IRI94-11330.