Annotea: A Generic Annotation Environment Using RDF/XML
Art Barstow, Jose Kahan,
Marja-Riitta Koivunen, Ralph Swick
W3C
Introduction
Main goals:
- Support collaborative working groups with a general Web annotation
system
- Demonstrating the possibilities of Semantic Web architecture
- Implement a metadata infrastructure as part of W3C Semantic Web
Advanced Development work
History and statistics
- Basic version in Amaya November 2000
- Announcements and Annotea home page March 9 2001
- Some statistics from annotest.w3.org April 1 to 30, 2001, excluding W3C
staff
- 123 users
- 140 w3.org annotated documents
- 94 non w3.org annotated documents
- 93 new annotations by 22 users
- 1920 annotates request
- 730 body request
Underlying architecture
Uses RDF to support evolving forms of annotations
Annotea server protocol
- protocol uses http POST, GET, PUT, and DELETE
- RDF is sent to and returned from the server
- a custom query uses http GET to query any RDF triples using Algae
syntax
Annotation instance
Creating a reply by using a thread schema
Benefits of RDF based approach
- Use of RDF enhances flexibility
- New properties can be easily added e.g. dc:title
- Threads cause minimum changes to server - new built-in query or client
uses Algae directly
- Can decide later that replies are subtypes of annotations - no changes
to server
Future ideas
- Shared bookmarks
- External annotation bodies
- Adding other metadata using annotation mechanism; client-specific data,
chump irc annotations
- Automatic generation and collection of temporary annotations e.g. to
mark broken links or accessibility problems
More information
Demos: Annotea clients
- W3C's Amaya
- Art Barstow's Annotlet
- Mathew Wilson's Annozilla for Mozilla