On this page
Convener
- Paul J. Morris - Harvard University Herbaria/Museum of Comparative Zoology, MA, USA
Core Members
Motivation
This group serves as a forum for the discussion of mechanisms and concepts for the annotation of all kinds of biodiversity data.
The Annotations Group will discuss, evaluate, propose, and standardize mechanisms for the collection, distribution, and storage of annotations of biodiversity data. The scope of annotations is potentially broad, encompassing, among others, structured and unstructured textual data, multimedia, and data set metadata, with obvious interactions with other TDWG groups and with existing and proposed standards in the domain.
The group provides a forum for the advancement of existing and new mechanisms for the annotation of biodiversity data towards standards related to the structure, capture, transport, and management of annotations. The group will develop:
- Summary information about annotation standards and mechanisms for both data in general and scientific data in particular;
- Descriptions of annotation concepts and annotation management technologies in use and in development in the biodiversity informatics community;
- Descriptions of implementations of annotation capture, transport, and management mechanisms in use and in development in the biodiversity informatics community; and
- Applicability statements of annotation standards and concepts from other communities.
Becoming involved
The Annotations Group is open to all interested parties. You can add yourself to the mailing list. Full membership is available to individuals who intend to be materially active in the group by request to the Convener. We are currently seeking reference implementations of annotation mechanisms and case studies for the annotation of data in general. Biodiversity informatics datasets are seldom, if ever, finished. New assertions are continually being made about biodiversity data, which may change their interpretation or require update of resources based on them. With subsequent study, researchers may apply new determinations to specimens. The scope of taxonomic concepts and taxon names change over time, as do geopolitical boundaries and names. Biodiversity data are continually enriched, for example, by extractions to add molecular profiles or dissections to improve morphological knowledge. Each piece of data may be distributed and linked in ways in which an assertion concerning a part of one data set is highly relevant to data being used elsewhere. Images of a specimen may be held in one or more repositories that are different from where its physical specimen is housed and any of these repositories may be the sites for new assertions about the data that are relevant to records in more places than the sites where the assertions are collected. Assertions about scientific data can be viewed as annotations of those data, with associated motivations, actions, and decisions that document the history of proposed, accepted, and rejected changes to those data. The domain content of an annotation is likely to be well described by existing, proposed, or future standards in the domain (e.g. Darwin Core, SDD, TCS, MRTG), but the concern of the annotation is at a different level. For example, an annotation might carry a Darwin Core assertion about a new determination of a specimen, but also assert that the determination is based upon a set of characters observed in particular regions of interest in a set of images of that specimen as seen by the annotator in an image repository. These linkages amongst the who, where, why, what, and when of assertions about data are the particular concerns of annotations. Annotations span many scales from change tracking within an individual data set to transport and linking of assertions across heterogeneous distributed data.
The Annotations Interest Group will provide a forum for discussion about all aspects of annotations in biodiversity informatics, develop cross-linkages with other interest groups and standards, and generate Task Groups to develop specific standards and best practices related to the annotation of biodiversity data.