Codex is being developed as part of the GEON and HERO projects, both funded by the National Science Foundation. For information, contact Bill Pike at wpike at psu dot edu.

A registration-free demo is available.

• Overview
• Selected papers
• People

Overview

Codex is a prototype portal that extends the notion of a network-accessible file sharing system with tools that provide a rich characterization of selected semantic relationships between resources. (It is called “Codex” after the style of manuscript notebooks such as Leonardo da Vinci’s). A central feature of Codex is a concept mapping utility through which users organize information resources into spatialized, visual representations of knowledge structures and share them with colleagues. Concepts are represented as nodes, and properties of those concepts that relate them to others are depicted as edges; a node’s location can also be mapped to its attributes. Nodes can be assigned roles corresponding to various elements of a problem, such as concepts, data resources, tools, people, and so on. The underlying structure of the resulting graph is represented in OWL, the Web Ontology Language.

As a user draws concept maps in the Codex Web client, the Codex server creates a representation of their structure in OWL. While Codex borrows many of the features of OWL, the knowledge structures produced in Codex are not formal ontologies. Unlike other collaborative tools that support shared understanding and concept graphing, Codex does not rely on fixed vocabularies for concepts and relations. Users can repurpose concepts and relations that others have created, modify them to fit their own needs, or define new ones. By representing information in a semi-formal manner we give researchers the flexibility to describe relationships that may be vague or inconsistent.

Codex maintains continuous records of a concept’s modification and use that allow one to take a temporal view onto a concept space; the provenance of a resource (be it an abstract concept or tangible data file) can be displayed visually. Similarly, one can view a concept space as a social network revealing relationships between researchers who used a resource in their own work. Each Codex user might rely on different data, methodologies or theories, and have different expertise and experience, but by translating visual depictions of knowledge into computational representations in OWL, it becomes possible to search for similarities across contexts, including researchers, locations, times, goals, methods, and so forth.

Codex will support several resource types:

(i) People – the individuals and groups who create or apply resources accessed through the Portal.

(ii) Concepts – descriptions of abstract ideas, including concept maps, domain and task ontologies and protocols or workflows.

(iii) Data resources (typically files) – including spreadsheets, text documents, curricula, learning and assessment activities, images, audio clips, maps, etc.

(iv) Analysis Tools – the methods used to analyze data and to construct concepts (categories) from data.

(v) Places and Times – geography is fundamental to integrative geoscience; places and times help researchers define the locations, scales and epochs under study. Place also accounts for large differences in epistemology between researchers, thus plays a major role in defining the nature of the defined concepts and the research outcomes produced. Thus these aspects must be logged when resources are used.

(vi) Tasks – people, concepts, files, tools, and places are linked together through tasks that might describe a workflow, an experimental procedure, or a problem-solving approach that link that connect observations or measurements (data) to the cognitive structures represented by concepts.

Selected Papers

The following papers describe aspects of Codex, the motivation behind it, and its role in a larger suite of computational aids to science work.

• Pike WA, Ahlqvist O, Gahegan M, Oswal S, 2003, Supporting collaborative science through a knowledge and data management portal. Workshop on Semantic Web Technologies for Searching and Retrieving Scientific Data, Second International Semantic Web Conference, Sanibel Island, FL, October 2003.
• Pike WA and Gahegan M, 2004, Visualizing concept relationships in a distributed knowledge sharing environment. GIScience 2004, Adelphi, MD, October 2004.
• MacEachren AM, Gahegan M, Pike W, 2004, Visualization for constructing and sharing geo-scientific concepts. Proceedings of the National Academy of Sciences 101(suppl. 1): 5279-5286.

[back to top]

People

Codex developers include:

• Bill Pike: codex architecture, concept mapping interface, and ontology management
• Sachin Oswal: Original standalone version of concept mapping applet
• Tawan Banchuen: Google search
• Gary Sheppard: User action capture
• Gary Liu: User action capture