Building: Windsor Hotel
Room: Oak Room
Date: 2015-09-30 03:00 PM – 03:15 PM
Last modified: 2015-08-29
Abstract
AnnoSys was tailored as a sustainable data quality tool to correct biodiversity data as well as allow specialists to enrich data in publicly available data portals. The current release of AnnoSys (https://annosys.bgbm.org/) establishes a number of workflows and methods enabling online annotations using the example of collection and observation data in the botanical and zoological domains as provided by the GBIF/BioCASE networks. It integrates information from a dozen biodiversity data portals and spurs a growing demand for further integration requests from additional data portals (e.g. the GBIF data portal). The latter requires the processing of additional data standards and the integration of a variety of additional use cases and new application ideas into the AnnoSys system. Further challenges include deriving data standards from within the involved communities as well as establishing and maintaining interoperability with other annotation systems.
The data portal integration includes the display of available annotations for a given collection data object as well as redirecting data portal users to AnnoSys' desktop-oriented web interface, which provides a comfortable means of editing, managing and publishing further annotations. Additionally, the web interface enables curators (on request) to communicate and justify decisions with regard to the acceptance or rejection of annotations referring to data objects in their collections. A subscription-based message system permits registered AnnoSys users (curators as well as researchers) to be informed by email about a self-defined subset of annotation processing events.
Based on a generic annotation context model, which is implemented using the W3C Open Annotation Data Model, the AnnoSys repository archives annotations and the related original collection data, which are publicly retrievable and integrable with other systems through Linked Data, REST (REpresentational State Transfer) and SPARQL-based web services. However, some limitations and aspects of inflexibility with regards to the context model, its backend implementation and the user interface are providing new challenges. Therefore in its second project phase, AnnoSys will identify these deficiencies and discuss, develop, and finally implement the most promising approaches.