Three steps to creating open datasets:

1) initial data build (choosing data to archive)

      1a) redudancy (lock-down original source)

      1b) alignment (create data matrix with consistent order)

      1c) test data quality (produce test plots)

2) annotation

      2a) label and metadata for each variable

      2b) basic descriptive statistics

3) verification step (before archiving)

      3a) sort/realignment (verify data matrix has a consistent order)

      3b) verify quality (reproduce plots)

4) archiving step

      4b) HDFView (https://www.hdfgroup.org/) for manipulating databases (HDF5 Format).

      4b) Open Science Framework (https://osf.io) or Github (https://www.github.com) for storage and providing stable URL. For Github repo storage, data can be converted to csv (comma-delimited) format, which will render in the form of a table in the repo.