Participants from our group

• Anastasia Galkin (poster presentation here)
• Ole Streicher
• Michèle Sanguillon
• Mathieu Servillat

Topics discussed

ProvONE extension to PROV for workflows(simplify to our needs), instead of AbstractActivity/AbstractEntity

prov:Plan for the ActivityDescription (instead of prov:Bundle)

see Process/Program from ProvONE and UNICORE (poster for HPC workflows)

Subclass Parameter from Entity (see UNICORE poster)

Focus on Entities rather than on Activities (descriptions as a Plan, i.e. an Entity)

Triple Store, Neo4j...

SPADE: tool to capture provenance in an OS

Provenance Patterns (RDA)

python-prov may handle extensions (like ProvOne), if done correctly; we may derive from it ( not cut&paste) to include our extensions (votable handling, ProvOne etc.)

Contacts

• Luc Moreau (critics on our Bundle approach, directing to ProvOne)
• Dong Huynh (python-prov package maintainer)
• Paolo Missier (ProvOne)
• ??? (UNICORE, UniProv)
• Alessandro Spinuso ( s-provenance)
• Andreas Schreiber (DLR, reproducible science with python)
• Ashish Gehani (incremental re-computation, accountability)
• Bertram Ludäscher
• James Cheney, Edinburgh

Papers

A survey in provenance by Melanie Herschel et. al. 2017

Declarative Datalog Debugging for Mere Mortals by Köhler et. al. 2012 (?)

Yin & Yang: Demonstrating Complementary Provenance from noWorkflow & YesWorkflow by Pimental 2016

A Survey of Data Provenance in e-Science by Simmhan 2005

Provenance in Databases - Why, How and Where by J. Cheney

Links

https://openprovenance.org/