May 20, 2013
We have been busy hacking away at code and our website. Here is an update on what we’ve been up to.
rplos/alm PLoS provides two different API services: the Search API and ALM API. As their names suggest, the search API lets you search and get text from their papers and associated metadata. The ALM API allows you to get article level metrics data on PLoS papers. Up until a few weeks ago, both APIs were accessible via functions inside the rplos package, but they really served two different purposes. Thus, we decided to make two packages: rplos, which wraps just the Search API, and alm, which wraps just the ALM API. It especially made sense to break off the ALM API into its own package as other publishers can use the ALM API for delivering their own article level metrics given that the PLoS ALM code is open source. Thus, down the road, you should be able to get altmetrics from XYZ publisher using the alm package just by changing the base URL (a.k.a. the API endpoint).
rWBclimate is a new package maintained by our very own Ted Hart. rWBclimate is an interface to download worldbank climate data directly into R, using their climate data API. Look for an upcoming blog post on the package.
rebi is a new package that lets you search and get data from over 1/2 million open access articles, citation counts, and more from the Europe PubMed Central RESTful web service. rebi is maintained by Najko Jahn. Check out the tutorial here.
rbefdata is a new package maintained by Claas-Thido Pfaff, that gives you access to data from the BEF (Biodiversity Ecosystem Functioning) data portal. The package was discussed in a recent post on this blog.
rbison is a new package that lets you search and get occurrence data for species in the USA via the USGS BISON API.
As always, ping us on Twitter (@ropensci) if you have any comments/questions, and let us know of any bugs or feature requests on the appropriate Issues pages of our repos.
We have a lofty goal of wrapping all science APIs. This is a mighty large task - impossible without a community of developers. A number of people have stepped up to contribute to some of our packages. The four core people, including myself, start a lot of R packages. However, we would have a hard time maintaining them all ourselves - the community is so important because they can extend the ability of rOpenSci to maintain quality R packages.
In addition, a few people now maintain their package as part of the rOpenSci suite. This is a great way to extend the breadth and number of packages in our suite. The four of us in the core team are all ecologists. Gasp! Thus, we could start R packages to wrap APIs for medical data, for example, but it makes much more sense for someone with domain knowledge to maintain that package.
At rOpenSci, we are trying to build tools for open science, in other words, get data from the web into R. We are of course not the only ones building packages in R for web data. We created a page on our site with a curated list of packages that access data from the web. This is a work in progress. Hit us up on Twitter @ropensci if you think we should add any packages to the page.