PubFlow is a DFG funded (Scientific Library Services and Information Systems (LIS) - Geschäftszeichen HA 2038/3-1) research project. The goal of the project is to do research in the domain of (semi-) automated data publication. Therefore we are building a pilotapplication, which uses proven workflow technologie like JBoss JBPM to transfer research data  from the institutional repositories into the world data centers.

PubFlow can be integrated in different existing software systems, which are already used by many scientist and data managers. Usually PubFlow is started by creating a ticket in a ticket system like Jira. When the ticket is created, PubFlow starts the corresponding workflow and - in the case of a human task - interacts with the user through the ticket system. Finally the research data is uploaded to an archive.

During the execution of a workflow PubFlow collects provenance information for the processed research data. This information is collected by Kieker, a monitoring framework developed by our working group. The provenance information collection is independent from the PubFlow system. This assures that it is not affected by errors, which might occur in PubFlow. The collected data is stored in a repository build upon W3Cs Prov-O notation and neo4j. It will be available at this page very soon.