The research community is rapidly moving toward facilitating access of raw “open data” as an integral part of publishing research papers with minimal data overhead to the researcher. Two excellent examples of these repositories are www.figshare.com and www.molssi.org, both of which already have HDF5 data available on their sites. MolSSI, although more domain specific than figshare, has a great list of use-cases at that are broadly applicable to the general task of enhancing raw data discovery. Also, they have a great paper available for download on the MolSSI platform.
As the developer of the H5s desktop scanner for HDF5, I’m evaluating and prioritizing directions and future work on the scanner that most benefit HDF5 and interested users. The quickest way to set the context for the question is a quick look at the H5s slideshow in the gallery of HDF5 data.
The question at hand is this:
Figshare already has an open API for publishing and searching datasets . Does building out to this API present a promising development direction for H5s to facilitate upload, download, and searching of published HDF5 data on figshare?
H5s is a modest application, so future development must be weighed carefully. Any comments, critiques, or recommendations much appreciated.
(Am posting with permission of the HDFGroup in order to get some feedback, and don’t want this to constitute a misuse of the forum.)