Hi all,
some threads here mention digital data capture using OpenDataKit.
My employer, the Department of Biodiversity Conservation and Attractions (Western Australia), are very happy ODK users - the amazing developer community (and the fact it’s FOSS as in we don’t waste taxpayers’ dollars) is a big plus to us. So far, we’ve collected ~20k records in ODK, and we’re just phasing out ODK Aggregate in favour of ODK Central.
ODK are progressing nicely with their new data clearinghouse ODK Central, which has a fantastic interactive API doco. (showerthought: ckanr, but for ODK Central)
ODK run an ODK Central sandbox, and the setup is pretty well documented (we run it in Kubernetes on Azure).
The online form builder provides a very useable drag-and-drop GUI, and at least among my colleagues there isn’t any real need to turn form building from a GUI with real-time validation into an R script. (Related: Field data collection workflow)
My real pain point / feature request / package idea is to automate the data ETL pipeline from ODK Central directly into reports / CKAN datasets (for smaller research campaigns), or into our downstream data warehouse (another RESTful API, for large monitoring campaigns), so I’ve started writing a wrapper in R, ruODK. While a few functions are still missing, I wanted to ask for your feedback - let me know what you think (or yell at me, the louder the better).
Follow-up: ruODK 0.6.0 is now up to date with ODK Central 0.6 (released 15 Aug 2019) and submitted for review.