Advances in Discovery
First Monday

Advances in Discovery: The Electronic Cultural Atlas Initiative Experience by Michael K. Buckland and Lewis R. Lancaster



Abstract
The Electronic Cultural Atlas Initiative was founded in 1997 by a group of scholars with a mission to advance education and research in the humanities and social sciences through increased attention to time and place. Knowing about context forms the basis for discovery and understanding. Our recent emphasis has been on developing a metadata infrastructure for the four facets What, Where, When and Who, each of which has special characteristics and display requirements, to advance discovery. WHAT requires thesauri of topics and tools to explore cross references within and between thesauri. WHERE needs place name gazetteers and map displays. Similarly, for WHEN we developed a directory that connects named time periods with calendar dates and a timeline or chronology. WHO requires best practices and standards for encoding the events in people’s lives, for contextualizing those events, and for displaying interpersonal relationships.

Contents

The Electronic Cultural Atlas Initiative (ECAI)
Time and place
ECAI’s mission
The ECAI metadata infrastructure
WHAT: Thesauri and cross references
WHERE: Gazetteer and map
WHEN: Time period directory and timeline
WHO: Biographical dictionary and personal relationships
Connecting the facets: What, When, Where, and Who
Lessons learned

 


 

The Electronic Cultural Atlas Initiative (ECAI)

In 1997 an informal meeting of scholars, mainly in the humanities, discussed the difficulties of analyzing and representing cultural changes across time and place. As one example, it would be expensive and unsatisfactory to use static, printed maps to show the ways in which Buddhism changed as it moved north out of India into the Himalayas then east into China, Japan, and Korea. To seek better methods they established the Electronic Cultural Atlas Initiative (ECAI) with a mission to advance education and research in the humanities and social sciences through increased attention to time and place. A special focus is the use of digital methods to create cultural maps showing historic movements and developments in a dynamic and interactive way. Scholars, librarians, educational information technology professionals, and others joined in. An administrative base was formed at the University of California, Berkeley, when ECAI was established as an academic unit reporting to the Dean of International and Area Studies. A major start–up grant from the Lilly Foundation supported the initial years.

Recent ECAI projects include a series of interactive maps concerning the Silk Road (http://ecai.org/silkroad), a catalogue of Sasanian seals (http://escholarship.cdlib.org/ecai/sasanian/sasanian_intro.html), and a temporal–spatial portal to resources on the history, cultural sites, and archaeological excavations of Iraq, ECAI Iraq (http://ecai.org/iraq/).

 

++++++++++

Time and place

Understanding requires some knowledge of context. Culture evolves in communities and communities develop in time and in places, and time and place are universal attributes for sharing cultural knowledge across different languages, disciplines, and points of view. For these reasons, time and place can be used to integrate data from museums, libraries, heritage sites, and research projects around the world.

 

++++++++++

ECAI’s mission

ECAI is not a project aimed at the creation of a singular, centralized digital cultural atlas, but an initiative intended to improve education and research. The mission of ECAI is to advance scholarship through increased attention to place and time. The more awareness of the benefits of paying serious attention to place and time, the more resources will be come available for sharing. The more that standards and best practices can be developed and adopted, the more easily different resources can be combined as needed. The better the tools made available, the better the uses that can be made of these data.

The mission of ECAI is to advance scholarship through increased attention to place and time.

ECA operates informally and at three levels:

  1. Advocacy for the use of place and time in research and education. Place and time provide a unifying framework across all disciplines and provides an organizing principle for bringing together scholarly resources of many different kinds;

  2. Encouraging the development of infrastructure through best practices, collaboration, standards development, and technology; and,

  3. Gaining practical experience and providing proof–of–concept.

To the extent that these goals are achieved and resources become network–accessible and interoperable, scholars will become able to compose temporally–dynamic maps for themselves drawing on each others’ resources.

ECAI exists informally as a worldwide collaboration of several hundred affiliated academics, information technology specialists, librarians, and others, and, formally, as a center within International and Area Studies at the University of California, Berkeley. ECAI maintains a clearinghouse of Internet–accessible geo–temporal resources, organizes two international conferences a year, promotes the development of best practices, standards, and software where needed, encourages online publications that incorporate dynamic maps, provides training workshops for digital cultural atlases, and undertakes enabling research and development (Buckland and Lancaster, 2004; Zerneke, et al., 2006; for further details see http://ecai.org).

 

++++++++++

The ECAI metadata infrastructure

To be discoverable, resources need to be collected and cataloged. But the existence of a large number of cataloged collections is not sufficient because the descriptions of resources vary greatly in completeness and form. Interoperability is needed not only between the very varied forms of topic representation but also between the descriptions of the locations, times, and people in each resource. A team of metadata and retrieval specialists in the School of Information, University of California, Berkeley, led by Fredric C. Gey, Ray R. Larson, and Michael K. Buckland, has collaborated with ECAI in addressing these issues.

To be discoverable, resources need to be collected and cataloged. But the existence of a large number of cataloged collections is not sufficient because the descriptions of resources vary greatly in completeness and form.

Prior work on the problems of searching across and between different genres, specifically between text files and socio–economic numeric data series, had revealed the difficulties of mapping between different topical vocabularies and the inadequacy of using place names to denote geographical areas. Our response was to develop a four–facet structure for topic, place, time, and people — What, Where, When, and Who — using tools and displays appropriate to each. We will discuss each in turn.

 

++++++++++

WHAT: Thesauri and cross references

Most well–edited resources contain careful topical indexing. (Here we use “thesauri” in a broad sense to include all kinds of systems for representing topics: Categories, classifications, indexes, ontologies, subject headings, and so on.) One problem is that there are many different systems, so most remain unfamiliar to the casual user. Even when verbal headings are used, one might not guess what they are. In the Library of Congress Subject Headings, for example, for martial arts films, you should look under “Hand–to–hand fighting, oriental, in motion pictures.” Such a heading is easier to recognize than to imagine.

Topical vocabularies tend to be stylized and a little antiquated. “HS 847120 Digital auto data proc mach conting in the same housing a CPU and input & output device” is the International Harmonized Commodity Classification heading for “Computer.” To make matters worse, not only does language change over time, vocabularies evolve within specialized communities, so no single vocabulary can suit everyone. In principle, every resource should have multiple indexes, one for each community of users (Buckland, et al., 2001; Petras, 2006).

Indexers rightly worry about establishing relationships between different terms within a thesaurus (“vocabulary control”), and the creation and display of cross references between related terms within a thesaurus receives a lot of attention. The much bigger problem of mapping between terms in different thesauri, an essential need in a networked environment, has received far less attention. However, the whole logic of a network environment is to make many different resources available, and different resources means, in practice, using many different vocabularies. To take a simple example, “Automobiles” are 3711 in the Standard Industrial Classification; 180/190 in the U.S. Patent Classification; TL 205 in the Library of Congress Classification; and, in the U.S. federal import and export statistics, “PASS MOT VEH, SPARK IGN ENG.” Fortunately, adherence to standards and automatic indexing can mitigate these difficulties (Buckland, 1999; Buckland, et al., 2002).

 

++++++++++

WHERE: Gazetteer and map

For many queries, especially in the humanities and social sciences, a place needs to be specified. Place names, however, are unreliable. There can be different forms (e.g. St. Petersburg, _____ _________, Saint–Pétersbourg), multiple names (e.g. Cluj, in Romania/Roumania/Rumania, is also called Klausenburg and Kolozsvar), ambiguous (several places are called Lafayette), and vague (e.g. the Midwest). Also, both names and boundaries change.

Spaces, however, have coordinates — longitude and latitude — and gazetteers are a well–established tool for linking place (a cultural concept) with space (a physical concept). A gazetteer is a list of place names, which notes what kind of a place it is (“feature type”), the spatial coordinates (latitude and longitude), other names for the same place, and, usually neglected, when that name was is use. The spatial coordinates allow the place to be displayed on a map.

Library catalogs support geographical search weakly because place names are used and political jurisdictions favored — both unstable. Connecting the library’s list of place names with a gazetteer allows better disambiguation, supports a variety of spatial relationships (e.g. near, between, within 50 kilometers), and, best of all, enables the catalog to have a map interface (Buckland, et al., 2004).

 

Connecting the library catalog with a map interface. A search for 'folklore' in the library catalog lists books about 'folklore' on the left and shows the locations the books about folklore refer to in the map on the right.
Figure 1: Connecting the library catalog with a map interface. A search for “folklore” in the library catalog lists books about “folklore” on the left and shows the locations the books about folklore refer to in the map on the right.

 

 

++++++++++

WHEN: Time period directory and timeline

Calendars provide a measure for time, but, in both speech and writing, people often mention events to denote time rather than calendar dates (e.g. “During the war ...,” “After graduating ...,” “Under Clinton ...”). This situation resembles the dual naming system of place and space, so we have developed a similar solution: a time period directory analogous to a place name gazetteer, thus:

Gazetteer
Place name — Type — Spatial markers (Lat. & Long.) — When
Period directory
Period name — Type — Time markers (Calendar) — Where

Note the symmetry and also that both use time and place.

A time period directory was constructed with 2,000 entries derived from the chronological subdivisions use in the Library of Congress Subject Headings associating event or period names with calendar dates and their geographic location (Petras, et al., in press; also Support for the Learner, 2006). A useful feature of this directory is that since the entries are derived from the Library of Congress catalog, they can be used directly as queries in catalog searches to find books relating to that period. The catalog records retrieved reveal what topics and persons where important enough to be the subject of published books.

Just as the spatial coordinates in the gazetteer allow places to be displayed on a map, a time period directory can display named events on a timeline or chronology.

 

++++++++++

WHO: Biographical dictionary and personal relationships

Biography is a genre of historical writing, and summary biographical descriptions are long–established in the form of biographical dictionaries and in the shorter “Who’s who” style. But, curiously, there appears to be little by way of best practices, let alone standards, for biographical texts that allow for interoperability and cross references across several resources. Librarians and archivists have standards for personal names, but not yet for representing the events in lives.

Just as WHAT, WHERE, and WHEN have distinctive, specialized display requirements for cross references, maps, and timelines, methods of displaying genealogies and other interpersonal relationships are needed for WHO.

To move beyond names to the individual’s lives, we are examining the formulation of personal life activities as events, and the feasibility of using the four facets of what [activity], where, when, and who [else] as an encoding structure for events. An attraction of this approach is that it could link back to the kinds of resources described above (see Figure 2).

 

Biographical description from Wikipedia and four examples of event facets.
Figure 2: Biographical description from Wikipedia and four examples of event facets.

 

 

++++++++++

Connecting the facets: What, When, Where, and Who

The logic of our technical approach is to construct useful resources building up from these four facets. The logic of the Internet is to not limit oneself to local resources but to select and use the best elements from anywhere — to be “webwise”! Drawing on diverse distributed resources will require a significant increase in interoperability.

The logic of the Internet is to not limit oneself to local resources but to select and use the best elements from anywhere — to be “webwise”!

As an illustration of what could be done, we can extend the example in Figure 3 in which we drilled down from a timeline in our time period directory and found War of Independence, 1285–1371, Scotland. Forwarding this entry as a query to the Library of Congress Catalog retrieves more than 60 records. The subject headings of each retrieved record can then be examined and one finds, for example, that several contain the personal subject heading Robert I, King of Scots, 1274–1329. Who was he? — and what did he do? Forwarding that biographical subject heading to the Wikipedia, another networked resource, retrieves a biographical article which identifies him as being Robert the Bruce, describes his life, and provides numerous links to additional contextual information.

 

A search for the event War of Independence, 1285-1371.
Figure 3: A search for the event War of Independence, 1285–1371, Scotland in the Library of Congress catalog retrieves 65 records with various biographical subject headings (1). Biographical subject headings can be looked up in Wikipedia (2). The Wikipedia entry (3) details the life of King Robert I and provides background information on the War of Independence (here called the Wars of Scottish Independence).

 

Figure 4 is a diagram showing some of the structural relationships among the resources, metadata vocabularies, and display options that we have discussed. The metadata vocabularies in capital letters constitute part of the scholarly infrastructure just as much as hardware, software and physical plant (Summit on Digital Tools for the Humanities, 2006).

 

Metadata vocabularies, resources and display options for the What, When,Where, and Who Metadata infrastructure.
Figure 4: Metadata vocabularies, resources and display options for the What, When,Where, and Who Metadata infrastructure.

 

 

++++++++++

Lessons learned

The work presented above draws on several years of exploration and experimentation based on the principle that understanding, unlike mere memorizing, means knowing about context.

Cross references between thesauri are needed as well as within each thesaurus.

Above all, only attention to standards and interoperability will allow the benefits of the Internet and attention to time and place can provide a unifying framework across all disciplines.

There are numerous standards for topical description (WHAT) but not for gazetteers, time–period directories, or biographical text.

The dual–naming approach of place and space using gazetteers and map displays can transform geographical searching in bibliographies and catalogs.

A similar approach for time — events and calendars — appears feasible and a prototype time period directory is now available.

We think that this four–facet approach will allow a more effective treatment of events and biographical texts.

The four facets — What, Where, When, and Who — do not exhaust the list. Why and How await attention. Nevertheless, we believe that a great deal of progress can be made by recognizing the significant differences between these four facets and the need to address each in its own right. Above all, only attention to standards and interoperability will allow the benefits of the Internet and attention to time and place can provide a unifying framework across all disciplines. End of article

 

About the authors

Michael K. Buckland is Emeritus Professor, School of Information, and Co–Director, of the Electronic Cultural Atlas Initiative, at the University of California, Berkeley. He has worked as a librarian, professor, and academic administrator. He has specialized in the design of library services, the organization of information, and the history and theory of documentation.

Lewis R. Lancaster is Co–Director, Electronic Cultural Atlas Initiative, Emeritus Professor of East Asian Languages and Literatures, University of California, Berkeley and President, University of the West. Prof. Lancaster is a specialist in the canons of Buddhist texts and in the digitization of ancient Asian religious texts.

 

Acknowledgements

The work presented was partially supported by two National Leadership Grants from the Institute of Museum and Library Services: “Going Places in the Catalog,” 2002–2004, and “Support for the Learner: What, Where, When and Who,” 2004–2006, and by the Henry Luce Foundation grant for “A Religious Atlas of China and the Himalayas.”

The projects and studies mentioned can be found at or through the ECAI Web site at http://ecai.org.

 

References

Michael K. Buckland, 1999. “Vocabulary as a Central Concept in Library and Information Science,” In: by Tatjana Aparac, Tefko Saracevic, and Peter Ingwersen (editors). Digital Libraries: Interdisciplinary Concepts, Challenges, and Opportunities. Proceedings of the Third International Conference on Conceptions of Library and Information Science (CoLIS3, Dubrovnik, Croatia, 23–26 May 1999). Zagreb: Lokve, pp. 3–12; also at http://www.sims.berkeley.edu/~buckland/colisvoc.htm, accessed 5 June 2006.

Michael K. Buckland and Lewis Lancaster, 2004. “Combining Time, Place, and Topic: The Electronic Cultural Atlas Initiative,” D–Lib Magazine, volume 10, number 5 (May), at http://www.dlib.org/dlib/may04/buckland/05buckland.html, accessed 2 June 2006.

Michael K. Buckland, Fredric C, Gey, and Ray R. Larson, 2004. Going Places in the Catalog: Improved Geographic Access: Final Report, at http://ecai.org/imls2002/imls2002-final_report.pdf, accessed 2 June 2006.

Michael K. Buckland, Fredric C. Gey, and Ray R. Larson, 2002. Seamless Searching of Numeric and Textual Resources: Final Report on Institute of Museum and Library Services National Leadership Grant, number 178, at http://metadata.sims.berkeley.edu/papers/SeamlessSearchFinalReport.pdf, accessed 5 June 2006.

Michael K. Buckland, Hailing Jiang, Youngin Kim, and Vivien Petras, 2001. “Domain–Based Indexes: Indexing for Communities of Users,” In: 3e Congrès du Chapitre français de l’ISKO, 5–6 juillet 2001. Filtrage et résumé informatique de l’information sur les réseaux. Paris: Université Nanterre Paris X, pp. 181–185; also at http://metadata.sims.berkeley.edu/papers/ISKObuck.pdf, accessed 4 June 2006.

Vivien Petras, 2006. “Translating Dialects in search: Mapping between Specialized Languages of Discourse and Documentary Languages,” Ph.D. dissertation, School of Information, University of California, Berkeley.

Vivien Petras, Ray Larson, and Michael Buckland, in press. “Time Period Directories: A Metadata Infrastructure for Placing Events in Temporal and Geographic Context,” Proceedings of Opening Information Horizons: Joint Conference on Digital Libraries (JCDL), Chapel Hill, N.C., 11–15 June 2006.

Summit on Digital Tools for the Humanities: Report on Summit Accomplishments. University of Virginia Institute for Advanced Technology in the Humanities: Charlottesville 2006; at http://www.iath.virginia.edu/dtsummit/SummitText.pdf, accessed 5 June 2006.

Support for the Learner: Time Periods, at http://ecai.org/imls2004/timeperiods.html, accessed 2 June 2006.

Jeanette L. Zerneke, Michael K. Buckland, and Kim Carl, 2006. “Temporally Dynamic Maps: The Electronic Cultural Atlas Initiative Experience,” Human IT, volume 8, number 3, pp. 83–94, at http://www.hb.se/bhs/ith/3-8/jzmbkc.pdf, accessed 2 June 2006.


Editorial history

Paper received 5 June 2006; accepted 21 July 2006.


Contents Index

Copyright ©2006, First Monday.

Copyright ©2006, Michael K. Buckland and Lewis R. Lancaster.

Advances in Discovery: The Electronic Cultural Atlas Initiative Experience by Michael K. Buckland and Lewis R. Lancaster
First Monday, volume 11, number 8 (August 2006),
URL: http://firstmonday.org/issues/issue11_8/buckland/index.html





A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2017. ISSN 1396-0466.