Wiki

TARDIS2

From TARDIS

Jump to: navigation, search

Contents

[edit] TARDIS 2.0

The next major iteration of TARDIS, version 2.0 will address shortcomings in the current implementation while extending the features available to the crystallographer.

[edit] Goals

  • Make deposition and harvest tools compatible with all major repository software
  • Refine/add to the existing crystallography-specific metadata
  • (Possibly) Incorporate all kinds of derived experimental data, not just raw diffraction images

[edit] Technical Information

This diagram was produced to clarify the changes being made. The current TARDIS workflow is represented by the red boxes.

Image:Tardis2diagram.png

  • A "Master" METS package will be produced that conforms with the Australian METS Profile. Information contained will include generic metadata (in MODS format), specific technical metadata relating to the crystallography experiment and datasets contained within, and a structure map describing the heirarchy of the files packaged for ingest into a repository.
  • A set of derived schemas, one per supported repository will be produced through XSLT on the master schema. This is due to repositories having specific requirements on the metadata it will handle on ingest of files. Each of these files should be extensions or modifications of the generic METS schema.
  • Software tools for deposition will share common code, through the implementation of SWORD.
  • Once data is in a repository, it must be harvested out. OAI-PMH, and possibly OAI-ORE will be used to harvest data out of repositories. Because different repository software provides access to their data in different ways, the harvester tool will have to support each of these access methods.
  • It is hoped that the need to split a large archived set of images will be removed, with only compression needing to be performed on a dataset.

Finally, TARDIS 2.0 will provide a streamlined interface to all harvested data.

[edit] Current Actions

  • Scott Yeadon (ANU) and Steve Androulakis (Monash Biochem.) are in the process of coming up with an example master METS xml package describing a crystallography dataset.
  • An extended meeting will be held, most likely at ANU to discuss the example METS package and modify it as needed.

This will form the base Australian METS for Crystallography Implementation profile that repository-specific packages will be derived.

[edit] METS Profiles and Examples

Profiles are by no means final, and serve as a base for discussion and feedback by the community. All profiles are based on the Australian METS Profile.

Scientific Datasets Profile

Crystallography METS Profile