From Bibliography to Timeline: Flexible Infrastructure Bears Fruit

  1. 1. Ian R. Johnson

    University Of Sydney

Work text
This plain text was ingested for the purpose of full-text search, not to preserve original formatting or readability. For the most complete copy, refer to the original conference program.

Heurist ( – introduced at last year’s
Digital Humanities conference – is an online system
which integrates internet bookmarking with standard
bibliographic functions, and provides strong social bookmarking
and workgroup management capabilities. These include
workgroup-centred tagging, rating and data sharing (private
and public wiki spaces, threaded discussions and automated
notifications) as well as publication of dynamically-generated
subsets of the database.
Heurist is aimed at an academic audience. Signup requires
authorisation by the system owner and, unlike generic systems
such as Delicious or CiteULike, depends on social constraints
to promote data quality (users are identifiable rather than
anonymous). Heurist is available as a free service to academics
(sensu lato) worldwide. We plan to release it as Open Source
software in the course of 2007, once we have developed
peer-to-peer communication between instances.
In developing the bibliographic functions of Heurist we did not
wish to follow the ‘easy way out’ solution adopted by almost
all existing bibliographic systems; that is, treating each
bibliographic entry as an independent record in a ‘flat file’
database, using ‘generic’ fields with different meanings in
different record types and redundantly repeating data such as
book and publisher details in every chapter record.
First, we developed a method of recording variant records using
metadata instances (a technique we pioneered for the Electronic
Cultural Atlas Initiative metadata clearinghouse –, allowing for a completely flexible data-defined
and instantly extensible structure which handles missing data
and repeating fields with ease. Fields and record types are
soft-coded within the database and can therefore be added
instantly through a web form; data entry forms are created
on-the-fly from the soft-coded definitions.
Secondly, we construct complete bibliographic records by
‘normalising’ data into separate records reflecting discrete
entities, thus reducing data redundancy and all its consequent
problems. For example, a book chapter will be composed of
four records - a chapter, a book, a book series and a publisher.
These four records are linked together by inter-record
relationships to form a complete bibliographic reference for
the chapter. Correcting the book information will thus correct
it for all chapters within the book.
There are a number of spin-off benefits in Heurist’s data
structure. In addition to all the normal field types –
variable-length text, integer, decimal, date, logical and
controlled lists – Heurist has three special field types which
reflect the needs of the data structure and the particular interests
of the research team:
• Geographic objects: Heurist stores point, point and direction,
circle, rectangle, (poly)line and polygon objects. Geographic
objects can be entered through a browser-based digitising
function embedded in the data entry forms. We believe
Heurist is the only spatially-enabled generic bibliographic
database in existence;
• File objects: Heurist implements a generic method for
uploading and managing files of various types – images,
text, spreadsheets, sound, video etc. Multiple files can be
attached to any record in the database and files can be
assigned specific roles (eg. logo, thumbnail, introductory
sound bite). Files can be embedded in output;
• Inter-record relationships: Heurist has rich functionality for
relating a record in the database to any other record(s) in
the database. To do this it defines a special type of field –
a relationship field – which is a pointer to another record
in the database. Relationship fields may be open or restricted
to point to a specific type of database record. During data
entry relationship fields pop up a search into the database,
filtered to the appropriate record type. Relationships
between records in the database are then just another type
of soft-coded database record containing two relationship
fields and a relationship type field. Some tricky code handles
addition of new relationship types and directionality, so that
a parent-child relationship, for instance, is reversed to
child-parent if viewed from the other side. Relationship
records also provide a datestamp (the range of time over
which the relationship applies) and all the annotation
capabilities of other Heurist records (including tagging,
personal notes, wiki pages, threaded discussion, personal
bookmarking and searchability).
We have been able to use the relationship functions of Heurist,
combined with its flexible record typing and geographic objects,
for some novel applications. These include mapping a network
of research centres, projects, grants, methods, sites, researchers
and publications (the 3DVISA network – and recording a set of
historical events and their relationships to create a browsable
view of history (Silk Road TimeLine - The underlying data for the Silk Road timeline is rendered as
a browsable list of interrelated events. When an event is
selected, all related events are listed, grouped by relationship
type – the web of relationships can be navigated by clicking
on one of the listed events to select it as the ‘root’ of a new list
of related events. The related events displayed in the list are
simultaneously shown on a TimeMap web map
( which is automatically updated whenever
the list changes. Links on the map and on the list allow
navigation between them and direct access to the backend
database records. The model can be extended to visualising a
different slice of history simply by copying the web page and
seeding it with a different starting record, or by linking a starting
record to the generic “Timeline Nodes” event which we have
created in the database (
Our next step in this project is to develop a visual timeline
which is generated on-the-fly from the current list of events
and also linked to the TimeMap map and the event listing.
In this paper I will show how Heurist and TimeMap have been
used to build this generic collaborative content creation model
and visualiser for historical events, illustrated through the
example of the Silk Road timeline project, and discuss the
underlying data structures. I will discuss planned developments
in the data model we use to describe historical events, and our
research on improved visual methods of entering and viewing
the web of relationships between historical events (or other
types of record), including timelie visualisations. I will also
discuss specific applications of the methodology adapted to
museums, visitor centres, online encyclopaedias and the
Heurist and TimeMap have been developed by the
Archaeological Computing Laboratory at the University of
Sydney, under my direction, as part of the Sydney Humanities
and Social Sciences e-Research Initiative (Heurist) and with
funding from the Australian Research Council, ECAI,
MacquarieNet and other user groups (TimeMap).

If this content appears in violation of your intellectual property rights, or you see errors or omissions, please reach out to Scott B. Weingart to discuss removing or amending the materials.

Conference Info


ADHO - 2007

Hosted at University of Illinois, Urbana-Champaign

Urbana-Champaign, Illinois, United States

June 2, 2007 - June 8, 2007

106 works by 213 authors indexed

Series: ADHO (2)

Organizers: ADHO

  • Keywords: None
  • Language: English
  • Topics: None