dataverse-client-python
dataverse-client-python copied to clipboard
IQSS
Create IPython Notebook for Dataverse APIs
Per discussion at http://irclog.iq.harvard.edu/dvn/2014-01-27 it would be nice to create an IPython Notebook for Dataverse APIs interesting enough to be listed at https://github.com/ipython/ipython/wiki/A-gallery-of-interesting-IPython-Notebooks
Fernando Perez had this wonderful idea: we could create a few notebooks, each one giving an example of how to use the Dataverse API from python for a different discipline (maybe we could choose 3-5 disciplines). What do you think?
Merce
Mercè Crosas, Ph.D. Director of Data Science, IQSS Harvard University http://iq.harvard.edu/merce-crosas
On Mon, Jan 27, 2014 at 1:25 PM, Philip Durbin [email protected]:
Per discussion at http://irclog.iq.harvard.edu/dvn/2014-01-27 it would be nice to create an IPython Notebook for Dataverse APIs interesting enough to be listed at https://github.com/ipython/ipython/wiki/A-gallery-of-interesting-IPython-Notebooks
— Reply to this email directly or view it on GitHubhttps://github.com/IQSS/dvn-client-python/issues/6 .
Just to recap some of our conversation with @mcrosas from last week: in addition to what she points out above, I also mentioned that in the longer term, it might be interesting to consider hosting a few notebooks alongside with datasets as official "computational companions" to the data. Think of these as "executable READMEs" to go along with the data, explain it better, summarize some of its key properties, etc.
The reason I mention this is because I think it would help enormously to bootstrap how people can use these datasets intelligently, if every time they grab one they can also get a document that explains it, provides additional references, and computes some quantities of interest from the data. Obviously people will want to explore further, but that can be a very useful starting point.
Such a notebook can also point out specific issues with the data, or provide any utilities that may be relevant to better analyze that particular dataset.
Such notebooks can be rendered statically, providing the web-based documentation about the data set, while also being available for local execution to users who download them.
@fperez you were great on http://podcastinit.com/episode-10-brian-granger-and-fernando-perez-of-the-ipython-project.html
I wanted to mention some interesting work by @4tikhonov where he's make an IPython notebook at https://github.com/4tikhonov/geo/blob/master/Clio%20Infra%20Historical%20boundaries.ipynb that operates not on Dataverse APIs themselves but rather APIs that he has built (using Python, I believe) that call into the PostgreSQL database behind a Dataverse installation. For details, see https://github.com/IQSS/dataverse/issues/2286