4W Vocabulary Mapping Across Diverse Reference Genres
Available from goo.gl
Page 1
4W Vocabulary Mapping Across Diverse Reference Genres
Michael Buckland
Ryan Shaw
4W Vocabulary Mapping Across Diverse Reference
Genres
Abstract
This paper examines three themes in the design of search support services: linking different genres of
reference resources (e.g. bibliographies, biographical dictionaries, catalogs, encyclopedias, place name
gazetteers); the division of vocabularies by facet (e.g. What, Where, When, and Who); and mapping between
both similar and dissimilar vocabularies. Different vocabularies within a facet can be used in conjunction, e.g.
a place name combined with spatial coordinates for Where. In practice, vocabularies of different facets are
used in combination in the representation or description of complex topics. Rich opportunities arise from
mapping across vocabularies of dissimilar reference genres to recreate the amenities of a reference library. In
a network environment, in which vocabulary control cannot be imposed, semantic correspondence across
diverse vocabularies is a challenge and an opportunity.
Introduction
The move away from cards and print to an online graphical user interface has
important consequences for search support. This paper draws on studies at Berkeley to
examine three different aspects of search support: overcoming the traditional separation
between different genres of reference resource; the dividing of vocabularies by facet
(e.g. What, Where, When, and Who); and mapping between different and dissimilar
vocabularies. Important relationships between these three themes are noted.
Our starting point is the principle that understanding requires a knowledge of
context and our approach to providing learners with contextual knowledge is to try to
recreate the functionality of a traditional reference library in a digital environment.
Different Genres of Reference Resource
Several different forms of reference work have evolved for specialized purposes:
Bibliographies, biographical dictionaries, chronologies, encyclopedias, library catalogs,
place name gazetteers, and others. In a print environment any entry in any of these
resources might cite any other, but, except for the library catalog, rarely tells one where
a copy can be found. With the technology of card and paper these resources are not
physically connected.
Sixty years ago, Ray Swank (1951) tried to combine the subject indexing of
bibliographies with the locations listed in library catalogs, but this was not practical
until the 1980s when digital technology became sufficiently available and affordable
for records in published bibliographies to be linked to call numbers in local library
records (Buckland 1988).
A digital, networked environment invites connectivity. Authority records on cards
mention authoritative sources; authority records online can connect in real time with
authoritative, explanatory resources, e.g. a place name authority record can link to a
place name gazetteer. A link to a specialized reference work is likely to provide more
and more up-to-date information than would be provided in an authority record. A
well-formed place name gazetteer will not only couple each place name with spatial
coordinates (latitude and longitude) for that place, but will also include a geographic
Ryan Shaw
4W Vocabulary Mapping Across Diverse Reference
Genres
Abstract
This paper examines three themes in the design of search support services: linking different genres of
reference resources (e.g. bibliographies, biographical dictionaries, catalogs, encyclopedias, place name
gazetteers); the division of vocabularies by facet (e.g. What, Where, When, and Who); and mapping between
both similar and dissimilar vocabularies. Different vocabularies within a facet can be used in conjunction, e.g.
a place name combined with spatial coordinates for Where. In practice, vocabularies of different facets are
used in combination in the representation or description of complex topics. Rich opportunities arise from
mapping across vocabularies of dissimilar reference genres to recreate the amenities of a reference library. In
a network environment, in which vocabulary control cannot be imposed, semantic correspondence across
diverse vocabularies is a challenge and an opportunity.
Introduction
The move away from cards and print to an online graphical user interface has
important consequences for search support. This paper draws on studies at Berkeley to
examine three different aspects of search support: overcoming the traditional separation
between different genres of reference resource; the dividing of vocabularies by facet
(e.g. What, Where, When, and Who); and mapping between different and dissimilar
vocabularies. Important relationships between these three themes are noted.
Our starting point is the principle that understanding requires a knowledge of
context and our approach to providing learners with contextual knowledge is to try to
recreate the functionality of a traditional reference library in a digital environment.
Different Genres of Reference Resource
Several different forms of reference work have evolved for specialized purposes:
Bibliographies, biographical dictionaries, chronologies, encyclopedias, library catalogs,
place name gazetteers, and others. In a print environment any entry in any of these
resources might cite any other, but, except for the library catalog, rarely tells one where
a copy can be found. With the technology of card and paper these resources are not
physically connected.
Sixty years ago, Ray Swank (1951) tried to combine the subject indexing of
bibliographies with the locations listed in library catalogs, but this was not practical
until the 1980s when digital technology became sufficiently available and affordable
for records in published bibliographies to be linked to call numbers in local library
records (Buckland 1988).
A digital, networked environment invites connectivity. Authority records on cards
mention authoritative sources; authority records online can connect in real time with
authoritative, explanatory resources, e.g. a place name authority record can link to a
place name gazetteer. A link to a specialized reference work is likely to provide more
and more up-to-date information than would be provided in an authority record. A
well-formed place name gazetteer will not only couple each place name with spatial
coordinates (latitude and longitude) for that place, but will also include a geographic
Page 2
2
description code (“feature type”) indicating what category of place it is (lake, city,
mountain, etc.). In turn, the gazetteer could contain links to more extensive descriptions
of the place in encyclopedias, pictures of it, map displays, socio-economic data series,
and mentions in literary works.
We may think of representation in a knowledge organization system as different
from description (as in an encyclopedia), but that distinction is not helpful here. The
important point is that the historic, physical separation of the catalog from reference
works need not and should not continue.
Divided Vocabularies
A general purpose knowledge organization system, such as a list of subject headings
or a library classification, will include all kinds of topics: activities, animals, events,
institutions, persons, places, and so on. However, it can be more effective and probably
more efficient to separate out categories of headings for separate treatment. In divided
library card catalogs, personal names for the subjects of biographies are ordinarily
separated from other subject headings and interfiled with the records for authors to
constitute a name catalog for persons, both as authors and as subjects.
A graphical user interface allows the display of different kinds of relationships. For
example, who, now, would want an online catalog without a map display showing
where each place is and its geographical relationship to other places? This is feasible if
the name of the place can be combined with its spatial coordinates of latitude and
longitude. (Buckland, Chen, Gey, Larson, Mostern and Petras 2007).
We have been working with the four facets of WHAT (topic), WHERE (place),
WHEN (time), and WHO (person) because these facets are different in kind and each
has distinctive characteristics and display requirements: Semantic syndetic structure for
topics; map displays and complex spatial relationships for places (e.g. Larson &
Frontiera 2004); and time-lines and chronologies for time periods. Displays of family
trees and other interpersonal relationships are desirable in biographies. Geotemporal
arrangement of biographies into “life paths” or collective analysis as prosopography
may be useful. There are other possibilities: We are currently considering the
separation of named events from other topics (Lancaster, Buckland and Shaw 2007).
Divided Vocabularies Combined in Representations
Facets are, in principle, different in their nature, but topics often cannot be expressed
in terms of a single facet. For example, 005.912=112.2(075)(410) representing
“German language primers on office management in the U.K.” in the Universal
Decimal Classification, is composed of a term from each of four facet vocabularies.
Library of Congress Subject Headings (LCSH) routinely occur as compound
representations in which the main heading is qualified by chronological and/or
geographical subdivisions (e.g. Architecture – Japan – Meiji period, 1868-1912). (For
discussion of LCSH and faceted interfaces see McGrath (2007)). Of course, textual
descriptions in encyclopedia articles or elsewhere will use all kinds of vocabulary.
As already noted a well-formed place name gazetteer will not only couple each place
name with the latitude and longitude for that place, but will also included a geographic
description code (commonly called “feature type”) indicating what category of place it
description code (“feature type”) indicating what category of place it is (lake, city,
mountain, etc.). In turn, the gazetteer could contain links to more extensive descriptions
of the place in encyclopedias, pictures of it, map displays, socio-economic data series,
and mentions in literary works.
We may think of representation in a knowledge organization system as different
from description (as in an encyclopedia), but that distinction is not helpful here. The
important point is that the historic, physical separation of the catalog from reference
works need not and should not continue.
Divided Vocabularies
A general purpose knowledge organization system, such as a list of subject headings
or a library classification, will include all kinds of topics: activities, animals, events,
institutions, persons, places, and so on. However, it can be more effective and probably
more efficient to separate out categories of headings for separate treatment. In divided
library card catalogs, personal names for the subjects of biographies are ordinarily
separated from other subject headings and interfiled with the records for authors to
constitute a name catalog for persons, both as authors and as subjects.
A graphical user interface allows the display of different kinds of relationships. For
example, who, now, would want an online catalog without a map display showing
where each place is and its geographical relationship to other places? This is feasible if
the name of the place can be combined with its spatial coordinates of latitude and
longitude. (Buckland, Chen, Gey, Larson, Mostern and Petras 2007).
We have been working with the four facets of WHAT (topic), WHERE (place),
WHEN (time), and WHO (person) because these facets are different in kind and each
has distinctive characteristics and display requirements: Semantic syndetic structure for
topics; map displays and complex spatial relationships for places (e.g. Larson &
Frontiera 2004); and time-lines and chronologies for time periods. Displays of family
trees and other interpersonal relationships are desirable in biographies. Geotemporal
arrangement of biographies into “life paths” or collective analysis as prosopography
may be useful. There are other possibilities: We are currently considering the
separation of named events from other topics (Lancaster, Buckland and Shaw 2007).
Divided Vocabularies Combined in Representations
Facets are, in principle, different in their nature, but topics often cannot be expressed
in terms of a single facet. For example, 005.912=112.2(075)(410) representing
“German language primers on office management in the U.K.” in the Universal
Decimal Classification, is composed of a term from each of four facet vocabularies.
Library of Congress Subject Headings (LCSH) routinely occur as compound
representations in which the main heading is qualified by chronological and/or
geographical subdivisions (e.g. Architecture – Japan – Meiji period, 1868-1912). (For
discussion of LCSH and faceted interfaces see McGrath (2007)). Of course, textual
descriptions in encyclopedia articles or elsewhere will use all kinds of vocabulary.
As already noted a well-formed place name gazetteer will not only couple each place
name with the latitude and longitude for that place, but will also included a geographic
description code (commonly called “feature type”) indicating what category of place it
Page 3
3
is (lake, city, mountain, etc.). Thus each WHERE is also categorized as a WHAT and,
since names and boundaries are often unstable, should also be encoded for WHEN.
The treatment of time is less well-developed than place. In ordinary discourse and in
metadata it is common to express time in geographical and cultural terms by using
named events adjectivally to denote period, e.g. Victorian literature, a Civil War
weapon, Louis Quinze furniture, and so on. LCSH chronological subdivisions
ordinarily define time using periods named in terms of the political history of the topic
(e.g. « Meiji period ») with calendar dates added for clarification. The use of named
events to denote spans of time has more evocative connotations than mere calendar
time because a cultural context is identified that might otherwise not have been
recognized when relying on dates alone. Calendar time is needed, however, to map
temporally between the periods of different cultures. (Petras, Larson and Buckland
2006).
Biographical texts are rich in actions, locations, dates, and other persons, but mark
up standards and best practices are still generally unsatisfactory (Text Encoding
Initiative 2006). We are examining the feasibility of decomposing lives into a series of
events and using 4-tuple of What, Where, When, and Who to characterize each life
event: an activity or event (birth, education, employment, etc.) in a place at some time,
sometimes with other people (Electronic Cultural Atlas Initiative 2006).
The use of terms from different facets in compound representations and textual
descriptions provides an opportunity to link any identifiable What, Where, When, or
Who to an explanatory description in some other reference work. The examples given
above can be briefly summarized:
Table 1. Components of examples different reference genres
WHAT (LCSH) : Topic – Geographical subdivision – Chronological subdivision
WHERE (Gazetteer) : Place name – Feature type – Latitude and longitude – When
WHEN (Chronology) : Period name – Period type – Dates – Where
WHO (Biogr. Dict.) : Personal name – Actions – Places – Dates – Other persons.
These examples provide the basis for a two-dimensional 4W array: The WHAT
records have WHERE and WHEN subdivisons; the WHERE records (place name
gazetteer entries) should have WHAT (feature type) and WHEN elements; WHEN
records in our design for a time period directory include elements for WHAT (period
type) and WHERE; and a WHO record in a biographical directory will contain multiple
WHAT (action, status), WHERE, WHEN, and WHO elements. We can note a WHAT
(Topic, Type, Type, Action) element in each row and also WHERE and WHEN in
each. Rearranging the components in each row yields a new table in which the facets
are aligned vertically.
Table 2. Two dimensional array of components of reference genres
WHAT (LCSH): What Where When
WHERE (Gazetteer): What Where When
WHEN (Chronology): What Where When
WHO (Biogr. Dict.): What Where When Who
is (lake, city, mountain, etc.). Thus each WHERE is also categorized as a WHAT and,
since names and boundaries are often unstable, should also be encoded for WHEN.
The treatment of time is less well-developed than place. In ordinary discourse and in
metadata it is common to express time in geographical and cultural terms by using
named events adjectivally to denote period, e.g. Victorian literature, a Civil War
weapon, Louis Quinze furniture, and so on. LCSH chronological subdivisions
ordinarily define time using periods named in terms of the political history of the topic
(e.g. « Meiji period ») with calendar dates added for clarification. The use of named
events to denote spans of time has more evocative connotations than mere calendar
time because a cultural context is identified that might otherwise not have been
recognized when relying on dates alone. Calendar time is needed, however, to map
temporally between the periods of different cultures. (Petras, Larson and Buckland
2006).
Biographical texts are rich in actions, locations, dates, and other persons, but mark
up standards and best practices are still generally unsatisfactory (Text Encoding
Initiative 2006). We are examining the feasibility of decomposing lives into a series of
events and using 4-tuple of What, Where, When, and Who to characterize each life
event: an activity or event (birth, education, employment, etc.) in a place at some time,
sometimes with other people (Electronic Cultural Atlas Initiative 2006).
The use of terms from different facets in compound representations and textual
descriptions provides an opportunity to link any identifiable What, Where, When, or
Who to an explanatory description in some other reference work. The examples given
above can be briefly summarized:
Table 1. Components of examples different reference genres
WHAT (LCSH) : Topic – Geographical subdivision – Chronological subdivision
WHERE (Gazetteer) : Place name – Feature type – Latitude and longitude – When
WHEN (Chronology) : Period name – Period type – Dates – Where
WHO (Biogr. Dict.) : Personal name – Actions – Places – Dates – Other persons.
These examples provide the basis for a two-dimensional 4W array: The WHAT
records have WHERE and WHEN subdivisons; the WHERE records (place name
gazetteer entries) should have WHAT (feature type) and WHEN elements; WHEN
records in our design for a time period directory include elements for WHAT (period
type) and WHERE; and a WHO record in a biographical directory will contain multiple
WHAT (action, status), WHERE, WHEN, and WHO elements. We can note a WHAT
(Topic, Type, Type, Action) element in each row and also WHERE and WHEN in
each. Rearranging the components in each row yields a new table in which the facets
are aligned vertically.
Table 2. Two dimensional array of components of reference genres
WHAT (LCSH): What Where When
WHERE (Gazetteer): What Where When
WHEN (Chronology): What Where When
WHO (Biogr. Dict.): What Where When Who
Page 4
4
Displayed this way, the opportunities for mapping vertically between vocabularies in
reference genres become apparent. Semantic associations can be established by vertical
vocabulary mapping and contextual associations can be established through horizontal
associations.
Mapping Between Similar Vocabularies
Ordinarily vocabulary control is within a single vocabulary and interoperability
through vocabulary mapping is between resources of a similar type, meaning at the
same horizontal level in Table 1. Similar resources commonly use different knowledge
organization vocabularies that are the same (or overlap) in scope but differ in form. The
Library of Congress Subject Headings (LCSH), the Dewey Decimal Classification and
the Library of Congress Classification are different and look different (e.g. for
Economics: “Economics,” “330,” and “HB1”), yet they are similar in that they are
general purpose systems for providing topical access. Vocabulary interoperability is
ordinarily based on cross-walks defining comparable fields in two or more similar
resources and mapping between terms in those fields.
Mapping Between Dissimilar Vocabularies
The attraction of linking between different genres of reference resource is the wider
range of descriptions and so the richer context. The vertical columns Table 2 reveal a
greatly expanded scope for mapping between dissimilar vocabularies. For example, the
feature type codes of a place name gazetteer can be mapped to the topics in a subject
heading list. We examined the 600+ Geographic Description Codes (GDC) of the US
federal National Geo-Intelligence Agency in relation to the far larger Library of
Congress Subject Headings and found that, despite stylistic differences, there was, in
most cases, a identifiable equivalent. This mapping allows a connection between a
category of geographical feature (e.g. a lighthouse) and literature about lighthouses. In
the other direction one could go from literature to physical examples, their locations,
and a map display showing where they are located. One could move, for example, from
the Geographic Description Code “school” to the corresponding library catalog subject
heading and on to related literature. Since a gazetteer is concerned with physical,
geographical features the semantic equivalent of the geographical feature « School » in
LCSH is “School buildings,” but, situationally, “Schools” (denoting schools as
institutions) may be preferred. In both GDC and LCSH the option of displaying related
terms would be highly desirable (Buckland, Chen, Gey, Larson, Mostern and Petras
2007). For examples of literature being related to maps see Moretti (1988).
Such mapping may seem inappropriate because subject headings are concerned with
topics and feature types with physical objects and these are different in kind and,
therefore, not comparable. This criticism is correct in principle, but misguided. It is
precisely this difference that enables one to move from literature to physical objects
and vice versa, a genuinely multimedia approach.
Why Vocabulary Mapping Is Hard
There is a widely-held assumption that search support in the Web should be
provided by ontologies and that these ontologies can be made interoperable. (DeRidder
2007 provides an overview). This seems reasonable until one tries to do it and/or one
Displayed this way, the opportunities for mapping vertically between vocabularies in
reference genres become apparent. Semantic associations can be established by vertical
vocabulary mapping and contextual associations can be established through horizontal
associations.
Mapping Between Similar Vocabularies
Ordinarily vocabulary control is within a single vocabulary and interoperability
through vocabulary mapping is between resources of a similar type, meaning at the
same horizontal level in Table 1. Similar resources commonly use different knowledge
organization vocabularies that are the same (or overlap) in scope but differ in form. The
Library of Congress Subject Headings (LCSH), the Dewey Decimal Classification and
the Library of Congress Classification are different and look different (e.g. for
Economics: “Economics,” “330,” and “HB1”), yet they are similar in that they are
general purpose systems for providing topical access. Vocabulary interoperability is
ordinarily based on cross-walks defining comparable fields in two or more similar
resources and mapping between terms in those fields.
Mapping Between Dissimilar Vocabularies
The attraction of linking between different genres of reference resource is the wider
range of descriptions and so the richer context. The vertical columns Table 2 reveal a
greatly expanded scope for mapping between dissimilar vocabularies. For example, the
feature type codes of a place name gazetteer can be mapped to the topics in a subject
heading list. We examined the 600+ Geographic Description Codes (GDC) of the US
federal National Geo-Intelligence Agency in relation to the far larger Library of
Congress Subject Headings and found that, despite stylistic differences, there was, in
most cases, a identifiable equivalent. This mapping allows a connection between a
category of geographical feature (e.g. a lighthouse) and literature about lighthouses. In
the other direction one could go from literature to physical examples, their locations,
and a map display showing where they are located. One could move, for example, from
the Geographic Description Code “school” to the corresponding library catalog subject
heading and on to related literature. Since a gazetteer is concerned with physical,
geographical features the semantic equivalent of the geographical feature « School » in
LCSH is “School buildings,” but, situationally, “Schools” (denoting schools as
institutions) may be preferred. In both GDC and LCSH the option of displaying related
terms would be highly desirable (Buckland, Chen, Gey, Larson, Mostern and Petras
2007). For examples of literature being related to maps see Moretti (1988).
Such mapping may seem inappropriate because subject headings are concerned with
topics and feature types with physical objects and these are different in kind and,
therefore, not comparable. This criticism is correct in principle, but misguided. It is
precisely this difference that enables one to move from literature to physical objects
and vice versa, a genuinely multimedia approach.
Why Vocabulary Mapping Is Hard
There is a widely-held assumption that search support in the Web should be
provided by ontologies and that these ontologies can be made interoperable. (DeRidder
2007 provides an overview). This seems reasonable until one tries to do it and/or one
Page 5
5
reflects on the nature of language and/or is concerned with representations of
knowledge rather than of widgets and/or contemplates the size of the task. Mapping
between knowledge organization vocabularies is hard for several reasons:
1. Vocabularies are descriptive. Assigning descriptive metadata (descriptors, index
terms, classification codes, category codes) is a language activity even through
documentary languages are more or less artificial. They are incorrigibly context-
specific and inherently obsolescent (Buckland 2007) .
2. Vocabularies are cultural. The classic definition by E. B. Tylor (1871, v.1, p.1)
states “Culture or Civilization, taken in its wide ethnographic sense, is that complex
whole which includes knowledge, belief, art, morals, law, custom, and any other
capabilities and habits acquired by man as a member of society.” More recent
definitions refer to integrated patterns of human knowledge, belief, and behavior, agree
that culture is not genetically transmitted but learned, and position knowledge as a part
of culture. It follows that knowledge organization is concerned with learned, cultural
entities and it is a characteristic of what is cultural (and of what is learned) that it is
subjective, more or less imprecise, and unstable. Discourse about culture tends to use
archetypes for lack of clear definitions and instances often form a smooth continuum
from one type to another. The implications for knowledge organization are significant.
In general terms, it must mean that precise, logical representations are unavoidably
Procrustean distortions. Vocabularies such as thesauri and classification schemes are
valuable and necessary. Nevertheless vocabularies become increasingly problematic as
their use is extended beyond a single, local application and/or are continued over time.
3. Language, being cultural, evolves within domains of discourse and it is not to be
expected that any given vocabulary will correspond precisely to one homogenous
community. Typically a knowledge organization vocabulary is a linguistic compromise.
Ideally, there would multiple indexes to any given resource, one for each community of
users, a challenge addressed by Petras (2006).
4. There is a lot mapping to be done! The whole logic of an internet environment is to
greatly increase access to resources with more-or-less unfamiliar descriptive metadata,
greatly increasing the need and scope for mapping between vocabularies. Mapping is
best done by well-qualified experts, but it is very labor-intensive. Fortunately,
statistical association techniques combined with natural language processing can
provide rapid, inexpensive search term recommender services where a training set can
be found (Buckland and others 1999; Gey, M. Buckland, A. Chen and R. Larson.
2001).
Conclusions
Understanding requires a knowledge of context so it is important to enable learners
to use diverse reference genres to find contextual information. In a traditional, paper-
based reference library the various resources are not physically linked but cite each
other and are physically collocated so that it is humanly possible to move from an entry
in one to an entry in another. That functionality has yet to be reconstructed in a digital
reflects on the nature of language and/or is concerned with representations of
knowledge rather than of widgets and/or contemplates the size of the task. Mapping
between knowledge organization vocabularies is hard for several reasons:
1. Vocabularies are descriptive. Assigning descriptive metadata (descriptors, index
terms, classification codes, category codes) is a language activity even through
documentary languages are more or less artificial. They are incorrigibly context-
specific and inherently obsolescent (Buckland 2007) .
2. Vocabularies are cultural. The classic definition by E. B. Tylor (1871, v.1, p.1)
states “Culture or Civilization, taken in its wide ethnographic sense, is that complex
whole which includes knowledge, belief, art, morals, law, custom, and any other
capabilities and habits acquired by man as a member of society.” More recent
definitions refer to integrated patterns of human knowledge, belief, and behavior, agree
that culture is not genetically transmitted but learned, and position knowledge as a part
of culture. It follows that knowledge organization is concerned with learned, cultural
entities and it is a characteristic of what is cultural (and of what is learned) that it is
subjective, more or less imprecise, and unstable. Discourse about culture tends to use
archetypes for lack of clear definitions and instances often form a smooth continuum
from one type to another. The implications for knowledge organization are significant.
In general terms, it must mean that precise, logical representations are unavoidably
Procrustean distortions. Vocabularies such as thesauri and classification schemes are
valuable and necessary. Nevertheless vocabularies become increasingly problematic as
their use is extended beyond a single, local application and/or are continued over time.
3. Language, being cultural, evolves within domains of discourse and it is not to be
expected that any given vocabulary will correspond precisely to one homogenous
community. Typically a knowledge organization vocabulary is a linguistic compromise.
Ideally, there would multiple indexes to any given resource, one for each community of
users, a challenge addressed by Petras (2006).
4. There is a lot mapping to be done! The whole logic of an internet environment is to
greatly increase access to resources with more-or-less unfamiliar descriptive metadata,
greatly increasing the need and scope for mapping between vocabularies. Mapping is
best done by well-qualified experts, but it is very labor-intensive. Fortunately,
statistical association techniques combined with natural language processing can
provide rapid, inexpensive search term recommender services where a training set can
be found (Buckland and others 1999; Gey, M. Buckland, A. Chen and R. Larson.
2001).
Conclusions
Understanding requires a knowledge of context so it is important to enable learners
to use diverse reference genres to find contextual information. In a traditional, paper-
based reference library the various resources are not physically linked but cite each
other and are physically collocated so that it is humanly possible to move from an entry
in one to an entry in another. That functionality has yet to be reconstructed in a digital
Page 6
6
library environment (Buckland 2008), but the adoption of online graphical user
interfaces allows many important developments. Dividing vocabularies by facet
simplifies the task of assigning descriptive metadata, facilitates specialized displays,
and constitutes a form of infrastructure (Buckland 2006). Combining divided
vocabularies is necessary for representation and descriptions. Mapping across both
similar and dissimilar vocabularies is a difficult but necessary part of a broadly based
learning environment.
Acknowledgments. This paper draws on the work of several other people, especially
Fredric Gey, Ray R. Larson and Vivien Petras. We are grateful for the support of the
Institute of Museum and Library Services National Leadership Grant for Libraries,
LG-02-02-0035-02; LG-02-04-0041-04; and LG-06-06-0037-06 and to both the
National Endowment for the Humanities and the Institute of Museum and Library
Services for an Advancing Knowledge grant PK-50027-07 (Electronic Cultural Atlas
Initiative. 2002, 2004, 2006 and 2007).
References
Buckland, M. 1988. Bibliography, library records, and the redefinition of the library catalog.
Library Resources & Technical Services 32: 299-311.
Buckland, M. 2006. Description and search: Metadata as infrastructure. Brazilian Journal of
Information Science vol 0 (2006). http://www.portalppgci.marilia.unesp.br/bjis/
Buckland, M. 2007. Naming in the library: Marks, meaning and machines. In: Nominalization,
nomination and naming in texts. C. Todenhagen & W. Thiele, eds. Tübingen, Germany:
Stauffenburg Verlag, pp 249-260.
http://people.ischool.berkeley.edu/~buckland/naminglib.pdf
Buckland, M. 2008. Reference service in the digital environment. Library & Information Science
Research Forthcoming.
Buckland, M., A. Chen, H-M. Chen, Y. Kim, B. Lam, R. Larson, B. Norgard, and J. Purat. 1999.
Mapping entry vocabulary to unfamiliar metadata vocabularies. D-Lib Magazine 5
http://www.dlib.org/dlib/january99/buckland/01buckland.html
Buckland, M., A. Chen, F. C. Gey, and R. R. Larson. 2006. Search across different media:
Numeric data sets and text files. Information Technology and Libraries 25: 181-189.
http://www.lita.org/ala/lita/litapublications/ital/252006/number4decembera/buckland.pdf
Buckland, M. A. Chen, F. C. Gey, R. R. Larson, R. Mostern & V. Petras. 2007. Geographic
search: Catalogs, gazetteers, and maps. College & Research Libraries 68: 376-387.
http://www.ala.org/ala/acrl/acrlpubs/crljournal/collegeresearch.cfm
Buckland, M., and L. Lancaster. 2004. Combining time, place, and topic: The Electronic Cultural
Atlas Initiative. D-Lib Magazine 10
http://www.dlib.org/dlib/may04/buckland/05buckland.html
DeRidder, J. L. 2007. The immediate prospects for the application of ontologies in digital
libraries. Knowledge Organization 34: 227-246.
Electronic Cultural Atlas Initiative. 2002. Going places in the catalog: Improved geographic
access. http://ecai.org/imls2002
Electronic Cultural Atlas Initiative. 2004. Support for the learner: What, where, when, and who.
http://ecai.org/imls2004
Electronic Cultural Atlas Initiative. 2006. Bringing lives to light: Biography in context.
http://ecai.org/imls2006
Electronic Cultural Atlas Initiative. 2007. Context and relationships: Ireland and Irish studies.
http://ecai.org/neh2007
library environment (Buckland 2008), but the adoption of online graphical user
interfaces allows many important developments. Dividing vocabularies by facet
simplifies the task of assigning descriptive metadata, facilitates specialized displays,
and constitutes a form of infrastructure (Buckland 2006). Combining divided
vocabularies is necessary for representation and descriptions. Mapping across both
similar and dissimilar vocabularies is a difficult but necessary part of a broadly based
learning environment.
Acknowledgments. This paper draws on the work of several other people, especially
Fredric Gey, Ray R. Larson and Vivien Petras. We are grateful for the support of the
Institute of Museum and Library Services National Leadership Grant for Libraries,
LG-02-02-0035-02; LG-02-04-0041-04; and LG-06-06-0037-06 and to both the
National Endowment for the Humanities and the Institute of Museum and Library
Services for an Advancing Knowledge grant PK-50027-07 (Electronic Cultural Atlas
Initiative. 2002, 2004, 2006 and 2007).
References
Buckland, M. 1988. Bibliography, library records, and the redefinition of the library catalog.
Library Resources & Technical Services 32: 299-311.
Buckland, M. 2006. Description and search: Metadata as infrastructure. Brazilian Journal of
Information Science vol 0 (2006). http://www.portalppgci.marilia.unesp.br/bjis/
Buckland, M. 2007. Naming in the library: Marks, meaning and machines. In: Nominalization,
nomination and naming in texts. C. Todenhagen & W. Thiele, eds. Tübingen, Germany:
Stauffenburg Verlag, pp 249-260.
http://people.ischool.berkeley.edu/~buckland/naminglib.pdf
Buckland, M. 2008. Reference service in the digital environment. Library & Information Science
Research Forthcoming.
Buckland, M., A. Chen, H-M. Chen, Y. Kim, B. Lam, R. Larson, B. Norgard, and J. Purat. 1999.
Mapping entry vocabulary to unfamiliar metadata vocabularies. D-Lib Magazine 5
http://www.dlib.org/dlib/january99/buckland/01buckland.html
Buckland, M., A. Chen, F. C. Gey, and R. R. Larson. 2006. Search across different media:
Numeric data sets and text files. Information Technology and Libraries 25: 181-189.
http://www.lita.org/ala/lita/litapublications/ital/252006/number4decembera/buckland.pdf
Buckland, M. A. Chen, F. C. Gey, R. R. Larson, R. Mostern & V. Petras. 2007. Geographic
search: Catalogs, gazetteers, and maps. College & Research Libraries 68: 376-387.
http://www.ala.org/ala/acrl/acrlpubs/crljournal/collegeresearch.cfm
Buckland, M., and L. Lancaster. 2004. Combining time, place, and topic: The Electronic Cultural
Atlas Initiative. D-Lib Magazine 10
http://www.dlib.org/dlib/may04/buckland/05buckland.html
DeRidder, J. L. 2007. The immediate prospects for the application of ontologies in digital
libraries. Knowledge Organization 34: 227-246.
Electronic Cultural Atlas Initiative. 2002. Going places in the catalog: Improved geographic
access. http://ecai.org/imls2002
Electronic Cultural Atlas Initiative. 2004. Support for the learner: What, where, when, and who.
http://ecai.org/imls2004
Electronic Cultural Atlas Initiative. 2006. Bringing lives to light: Biography in context.
http://ecai.org/imls2006
Electronic Cultural Atlas Initiative. 2007. Context and relationships: Ireland and Irish studies.
http://ecai.org/neh2007
Page 7
7
Gey, F., M. Buckland, A. Chen and R. Larson. 2001. Entry vocabulary: a technology to enhance
digital search. http://metadata.sims.berkeley.edu/papers/hlt01-final.pdf
Lancaster, L., M. Buckland, and R. Shaw. 2007. Event: Occurrence and agency in digital
resources. In: 6th International Semantic Web Conference and the 2nd Asian Semantic Web
Conference. Workshop 9: Cultural Heritage on the Semantic Web, Busan, Korea. (12 Nov
2007). Pp. 77-82. http://www.cs.vu.nl/~laroyo/CH-SW/ISWC-wp9-proceedings.pdf
Larson, R. L. and P. Frontiera. 2004. Spatial ranking methods for geographic information
retrieval (GIR) in digital libraries. Paper presented at the European Collaborative Digital
Library Conference, 2004. http://cheshire.lib.berkeley.edu/ECDL2004_preprint.pdf
McGrath, K. 2007. Facet-based search and navigation with LCSH : Problems and opportunities.
Code4Lib Journal 1. http://journal.code4lib.org/articles/23
Moretti, F. 1998. Atlas of the European novel. London : Verso.
Petras, V. 2006. Translating dialects in search: Mapping between specialized languages of
discourse and documentary languages. Doctoral dissertation, University of California,
Berkeley. http://www.sims.berkeley.edu/~vivienp/diss/vpetras-dissertation2006-official.pdf
Petras, V., R. R. Larson, and M. Buckland. 2006. Time period directories: A metadata
infrastructure for placing events in temporal and geographic context. In: Opening
Information Horizons; 6th ACM/IEEE-CS Joint Conference on Digital Libraries 2006. New
York: Association for Computing Machinery, pp.151-160
http://portal.acm.org/citation.cfm?id=1141782
Swank, R. C. Subject cataloing in the subject-departmentalized library. In: Bibliographic
organization, ed. by J. H. Shera & M. E. Egan. Chicago: University of Chicago Press, pp.
187-199. Repr. in Swank, R. C. 1981. A unifying influence. Metchen, NJ: Scarecrow Press,
pp.177-190.
Text Encoding Initiative. 2006. TEI: Personography Task Force.
http://www.tei-c.org/Activities/Workgroups/PERS/
Tylor, Edward Burnett. 1871. Primitive culture. London : Murray.
Gey, F., M. Buckland, A. Chen and R. Larson. 2001. Entry vocabulary: a technology to enhance
digital search. http://metadata.sims.berkeley.edu/papers/hlt01-final.pdf
Lancaster, L., M. Buckland, and R. Shaw. 2007. Event: Occurrence and agency in digital
resources. In: 6th International Semantic Web Conference and the 2nd Asian Semantic Web
Conference. Workshop 9: Cultural Heritage on the Semantic Web, Busan, Korea. (12 Nov
2007). Pp. 77-82. http://www.cs.vu.nl/~laroyo/CH-SW/ISWC-wp9-proceedings.pdf
Larson, R. L. and P. Frontiera. 2004. Spatial ranking methods for geographic information
retrieval (GIR) in digital libraries. Paper presented at the European Collaborative Digital
Library Conference, 2004. http://cheshire.lib.berkeley.edu/ECDL2004_preprint.pdf
McGrath, K. 2007. Facet-based search and navigation with LCSH : Problems and opportunities.
Code4Lib Journal 1. http://journal.code4lib.org/articles/23
Moretti, F. 1998. Atlas of the European novel. London : Verso.
Petras, V. 2006. Translating dialects in search: Mapping between specialized languages of
discourse and documentary languages. Doctoral dissertation, University of California,
Berkeley. http://www.sims.berkeley.edu/~vivienp/diss/vpetras-dissertation2006-official.pdf
Petras, V., R. R. Larson, and M. Buckland. 2006. Time period directories: A metadata
infrastructure for placing events in temporal and geographic context. In: Opening
Information Horizons; 6th ACM/IEEE-CS Joint Conference on Digital Libraries 2006. New
York: Association for Computing Machinery, pp.151-160
http://portal.acm.org/citation.cfm?id=1141782
Swank, R. C. Subject cataloing in the subject-departmentalized library. In: Bibliographic
organization, ed. by J. H. Shera & M. E. Egan. Chicago: University of Chicago Press, pp.
187-199. Repr. in Swank, R. C. 1981. A unifying influence. Metchen, NJ: Scarecrow Press,
pp.177-190.
Text Encoding Initiative. 2006. TEI: Personography Task Force.
http://www.tei-c.org/Activities/Workgroups/PERS/
Tylor, Edward Burnett. 1871. Primitive culture. London : Murray.
Sign up today - FREE
Mendeley saves you time finding and organizing research. Learn more
- All your research in one place
- Add and import papers easily
- Access it anywhere, anytime
Start using Mendeley in seconds!
Readership Statistics
1 Reader on Mendeley
by Discipline
by Academic Status
100% Assistant Professor
by Country
100% United States