Usecase Publication Ready Archive

This is version 4. It is not the current version, and thus it cannot be edited.
[Back to current version]   [Restore this version]

Use Case Name: Archive a complete end to end run ("publication ready archive")

Version:
Summary:

Users would like the ability to create a "publication ready archive"--an archive that contains the Kepler workflow, all required inputs (data sources like database tables are included), and all produced output. For example, at the completion of a study, a user can produce this archive and submit it (or a link to it) alongside their paper to peers and journals.

Preconditions:

• User has created workflow that contains everything needed (eg data sources).
• User documents any external software the workflow requires

Triggers:

User exports workflow

Basic course of events:

User creates a workflow. User turns on "publication ready archive" mode. User runs workflow, and exports to an archive at completion.

Alternative paths:

User uses features from "Smart Rerun", "Manage a group of model runs", and "save intermediate derived data products" usecase (ie a special "publication ready archive" mode might not be required).

Postconditions:

Business rules:
Notes:

A specific "publication ready archive" option would not be required if requirements for usecases mentioned in Alt Paths are implemented.

Status: 07.14.09 - We've made good progress on this usecase. The basic course of events to get a kar archive that's becoming closer to being a 'pub ready archive' is currently:
User either launches kepler with the workflow-run-manager module or switches to this module from within Kepler using the module-manager (TODO, solve: the fact that they first have to do this is entirely unclear at this point). As provenance is on by default with WRM, any workflow run is recorded, and may be selected for export by right clicking on the run row in the WRM, and selecing "export archive" or "upload to repository".
The resultant kar does not yet contain everything needed however, and also needs to be improved to better support multiple runs. See bug #3668