Description
Project: Apache Taverna (incubating) http://taverna.incubator.apache.org/
Mentor: Robert Haines <hainesr@apache.org>, Stian Soiland-Reyes <stain@apache.org>, Ian Dunlop <ianwdunlop@apache.org>
Taverna Databundle is a file format returned by the Taverna Server containing the results of a Taverna workflow run and its intermediate values and provenance metadata.
The databundle is a structured ZIP file with metadata in JSON-LD.
https://github.com/apache/incubator-taverna-language/tree/master/taverna-databundle
The above is a Java API for working with the data bundle - but as it's just based on ZIP and JSON it should be processable in many languages without requiring this API.
This GSOC project proposes to create a web-based presentation (but not edit) of a workflow run, e.g. the following user story:
1) User uploads the databundle file for a workflow run (or provides the URI to one existing elsewhere) into the presentation system
2) The presentation system shows basic information about the workflow run
3) User navigates to inputs/outputs
4) The presentation system presents a list of inputs or outputs - embedding images etc. where possible, download links for large files, inline for small text, etc
5) User downloads a selected data file
Metadata provided can be presented in many different ways, for instance:
a) Which workflow was executed?
b) Which steps were executed?
c) What step produced this value?
d) What values were produced by this step?
e) What steps used this value?
This proposed project is separate from the existing Taverna Player - but could in theory be integratable within the Player.
https://github.com/myGrid/taverna-player-portal
This proposed project should on the other side also work without a Taverna Server, as it only requires upload of a wfbundle, that can have been produced with the Taverna Workbench, Taverna Command Line, Taverna Server or even non-Taverna tools that have used the Taverna Language APIs.
The student is free to choose programming language and web framework, as long as a free runtime is available (e.g. Ruby on Rails, Java CXF, Node.js, Bootstrap, Ember, Grails).
Your web viewer should aim to be added to the Apache Taverna as a release, so you would be a part of the Apache Taverna developer community http://taverna.incubator.apache.org/community/
which will be able to give feedback, testing and guidance for this GSOC project and beyond.
We can provide virtual servers for hosting the viewer prototypes.
Prospective students are encouraged to subscribe to dev@taverna mailing list and discuss their proposal in detail together with he Apache Taverna community - http://taverna.incubator.apache.org/community/lists