Jon Zelner recently developed a neat Docker packaging of Stan, R, and knitr for fully reproducible research. The first in his series of posts (with links to the next parts) is here:
The post on making changes online and auto-updating results using GitLab’s continuous integration service is here:
* GitLab continuous integration
It updates via pushes to a Git repository hosted by GitLab.
Jon says, “This is very much a work-in-progress, so any feedback would be greatly appreciated!”. You can leave comments on the blog itself.
This seems like a sensible and practical approach. For those interested in Docker and containers, there’s a recent overview paper:
https://www.deepdyve.com/lp/institute-of-electrical-and-electronics-engineers/toward-a-standard-interface-for-cloud-providers-the-container-as-the-lGHrOl3w45
or
https://www.computer.org/csdl/mags/ic/2016/02/mic2016020066.pdf
Unfortunately, I cannot find an unlocked copy.
Also, the embedded link in the referenced piece doesn’t work for me (https://.about.gitlab.com/jzelner/reproducible-stan). However,
https://gitlab.com/jzelner/reproducible-stan does work.
Bob
Thanks for the links. I’m sure Jon will fix the broken one in his post.
Indeed – done and done. Thanks!
What would scientists think about a government mandate of this or similar technology? As a citizen, I find it pretty apalling that so much money is squandered on research that winds up inaccessible to the public.
After a one-time investment on retraining, it seems researchers would also reap a huge productivity dividend based on improved work flow. As a developer, I find it inappropriate that government doesn’t enforce code and data standards on grant recipients.
I kind of randomly ran into this post and find it funny that after 8 years, none of the links on a post about reproducible research is actually reproducible…