Dear Chris, I think providing a environment that others can use is an important part of the provenance question. It’s one thing to document dependencies, but in order to really make sure they are correct and complete you need to test them in a compute environment. https://mybinder.org/ is a great and free example of this setup. If you provide dependencies in standard text files, everyone can re-run your results (in python, R, others) in a virtual compute environment. Showing researchers the benefits of capturing provenance information is an important part of driving adoption.
Best, Tom Morrell Research Data Specialist, Caltech Library On Aug 13, 2018, at 6:36 AM, Chris Gates <[email protected]<mailto:[email protected]>> wrote: This is a wonderful thread and thanks to all. In bioinformatics, the web of dependencies is so dense and brittle, that beyond provenance there is increasing interest in sharing compute environments in a reproducible way using automated workflow frameworks and virtualization/containerization. In effect, the downstream users saying "please document your provenance - and (since I'm too impatient to read it) also please gather dependent data/programs/workflows into a an environment that I would actually use". This idea is, of course, distinct from provenance per se, so I don't want to muddy this thread - but it seems related, and worth mentioning. Best, cg The Carpentries<https://carpentries.topicbox.com/latest> / discuss / see discussions<https://carpentries.topicbox.com/groups/discuss> + participants<https://carpentries.topicbox.com/groups/discuss/members> + delivery options<https://carpentries.topicbox.com/groups/discuss/subscription> Permalink<https://carpentries.topicbox.com/groups/discuss/Te1cade367c0ab4ee-M533110c2162c11fb5ef71048> ------------------------------------------ The Carpentries: discuss Permalink: https://carpentries.topicbox.com/groups/discuss/Te1cade367c0ab4ee-M56d1de7821704659d5f162e5 Delivery options: https://carpentries.topicbox.com/groups/discuss/subscription
