discover, reference, publish primary Archaeological data
http://www.opencontext.org: a site to discover, reference, and publish primary data data, images, reports and field documentation collected in archaeology
- data entry in archaeology is a complex discipline with its own science: Stratification (aka Stratigraphy).
- every piece of data, photo, artefact is bound to the exact place it was found, called context (aka Stratigraphic unit).
- Created by Alexandria Archive Institute, a not-for-profit organization
- spun off from Information and Service Design Program at the UC Berkeley School of Information
- main involved:
- Dr. Sarah Whitcher Kansa (Executive Director, AAI): archaeology
- Eric Kansa (ISD Executive Director, I-School Adjunct Professor): software development.
- Don't know if this is related to RDF in any way, but I marvel at the richness of information
Maria: study OpenContext. Does it use RDF in any way?
Open Context implements a subset of the Archaeological Markup Language (ArchaeoML), developed for the University of Chicago OCHRE project. ArchaeoML uses an item-based information model, where individual atomic units of observation are related to each other and their descriptive attributes. Each item does not belong to a predetermined class (pottery, bone, deposit, grave good, etc.), but is, instead, an abstract entity that can have multiple descriptive properties and different forms of linking relations with other items.
(From OCHRE Project: OCHRE does not impose a standardized terminology. Instead, it provides a core ontology within which each project’s terminology can be expressed. In database parlance, OCHRE has a “global schema” to which many different “local schemas” can be mapped. Unlike most databases, property names and values are not built into the structure of the database as table column headings and table cell values. In OCHRE, descriptive properties (called “attributes” in many database systems) are themselves treated as project-defined “data” within a more abstract structure. Each project can define its own descriptive properties and organize them into a meaningful hierarchy, yielding a project-specific taxonomy.)
There is plan for expression of ArchaeoML from machine-readbile XML format to RDF-triples using the CIDOC-CRM, though they say that any experts familiar with the CIDOC-CRM can describe relationships among Open Context content using the semantically richer framework of the CIDOC-CRM. See Open Context Concepts.
Vlado thinks that the same principles are used in openstreetmap.org.
- The application is pretty simple containing information about all Projects and Collections, without any collaboration and annotation tools.
- For each project is supported project overview, number of items per project grouped by categories (coins, bones, etc.)
- For each item is kept information about item properties (different for each category, and the same for all items from a category) and associated media files
- Here is example of CRM ontological model for archaeology information domain, which probably could be presented as a simpler one with Open Context.