Supporting reproducible science on HydroShare

[ChristinaB]

We need documentation, workshops, and more communication for research software scientists developing tools- who are not software engineers with training.

• Publish/help page/newsletter on Best practices e.g. for linking to a resource ID dependent file location.

This code doesn’t work:

hs=hydroshare.hydroshare()
homedir = hs.getContentPath(os.environ["HS_RES_ID"])
os.chdir(homedir)
print('Data will be loaded from and save to:'+homedir)

• Publish/help page/newsletter/sample code by Tony on Best Practices for dependency checking and version control of Notebooks. TravisCI scheduling and commit ID. e.g. at top of Notebook add assertion to control for versions.

• Make utilities/hydroshare a package for local installations or different JupyterHub servers to access HydroShare (CyberGIS Team and Christina’s future proposals need this)

• Workshops/Help Videos on how to use Hydroshare with a paper tutorial Notebook, Docker, SciUnit, etc. Webinar once per semester training on using Notebooks in classrooms and workshops and publications.

• What is the role of CUASHI to support users in maintaining working resources given version changes? CUAHSI is responsible for education and communication of requirements.

• Add deprecation warnings to utilities/hydroshare. Communicate the bounds of what CUAHSI is responsible for and what the user is responsible for.

• Instructions for communicating on Slack and Github

• Developers who use CUAHSI JupyterHub server requires operational and maintaining for workshops and paper users. Developer is responsible.

• CUAHSI makes it easy for ‘research software scientist developers’ to collaborate and work with others. Set up a forum and listserv.

[ChristinaB]

Here are some useful links:

https://www.nature.com/articles/sdata201930

https://blog.zenodo.org/2017/05/30/doi-versioning-launched/

https://www.reviewnb.com/