Enabling Biomedical Science with Common Workflow Language

Common Workflow Language (CWL) is an essential technology for describing biomedical analysis
workflows used by many organizations. Embracing CWL as a workflow standard enables reproducible
research. This ensures that scientific work can be understood, independently verified, and built upon in
future work. Improvements to CWL benefit multiple organizations and workflow systems that use the
CWL standard. For these organizations, there is an ongoing need to train new users and develop new
workflows with CWL. CWL faces barriers to wider use and adoption due to a perceived learning curve and
usability challenges. Our proposal outlines a plan to help reduce those barriers through community
engagement and improved usability. It covers documentation updates, community feedback aggregation
and response, specification development, and tool creation. We propose funding Peter Amstutz to lead
this effort alongside the PI Sarah Zaranek and a "community engineer." Peter is a leader in the
development of CWL, "cwltool", "cwl-ex" and Arvados. He has demonstrated the ability to serve the
needs of the community at large. He is familiar with "wearing many hats", including developer, technical
writer, community relations, and project evangelist. Our overarching goals is to make it easier for users
to create CWL workflows and run them at scale using any conformant CWL platform (drawing heavily on
examples from the open-source Arvados platform developed by Curii.) Throughout the effort we we will
use real world public domain (CC0) data and workflows from the global Personal Genome Projects where
we have contributed since 2005.