I’ve recently heard the idea of “offline first” via especially Hood.ie. We of course don’t do web development, but primarily build R interfaces to data on the web. Internet availablility is increasinghly ubiqutous, but there still are times and places where you don’t have internet, but need to get work done. In the R packages we write there are generally two steps to every workflow: Make a call to the web to request data and collect data Rearrange the result as some sort of R object (e.

Natural history museums have long been valuable repositories of data on species diversity. These data have been critical for fostering and shaping the development of fields such as biogeography and systematics. The importance of these data repositories is becoming increasingly important, especially in the context of climate change, where a strong understanding of how species responded to past climate is key to understanding their responses in the future. Leading the way in opening up such valuable data is a new effort by the Berkeley Initiative in Global Change Biology called the Ecoengine.

A number of the APIs we interact with (e.g., PLOS full text API, and USGS’s BISON API in rplos and rbison, respectively) expose Solr endpoints. Solr is an Apache hosted project - it is a powerful search server. Given that at least two, and possibly more in the future, of the data providers we interact with provide Solr endpoints, it made sense to create an R package to make robust functions to interact with Solr that work across any Solr endpoint.

rplos is an R package to facilitate easy search and full-text retrieval from all Public Library of Science (PLOS) articles, and we have a little feature which aren’t sure if is useful or not. I don’t actually do any text-mining for my research, so perhaps text-mining folks can give some feedback. You can quickly get a lot of results back using rplos, so perhaps it is useful to quickly browse what you got.

Upcoming Book on Open Science with R We’re pleased to announce that the rOpenSci core team has just signed a contract with CRC Press/Taylor and Francis R series to publish a new book on practical ways to implement open science into your own research using R. Given all the talk about the importance of open science, the discussion often lacks practical suggestions on how one might actually incorporate these practices into their day to day research workflow.