Expanding the R toolkit

19 February 2018
Associate Professor Paul Murrell, Department of Statistics
Associate Professor Paul Murrell

Associate Professor Paul Murrell’s work on graphics in R is a bit unusual – he mostly doesn't create graphs, but develops tools so others can.

If you want to, say, draw curves with variable width, or find out how to automate R demonstration videos, Paul has already come up with the solution. “For me, that's the most enjoyable sort of programming – a solution that generalises,” says Paul. “Why solve one problem when you can solve a whole lot of problems at once?”

Paul has an intimate knowledge of R, an open-source language and environment for computing and graphics used by tens of thousands of statisticians. The software was developed by Ross Ihaka and Robert Gentleman when both were in the Department of Statistics in the 1990s; Paul was also here, doing his PhD with Ross.

Paul’s doctoral research grew into the “grid” system in R, which allowed people to programme R graphics in a more sophisticated and flexible way than was previously possible; it has become the launch pad for a host of packages.

As an example, Dr Jonathan Godfrey of Massey University, who is blind, has combined a grid-based package with his own BrailleR package so users with visual impairments can turn R plots into text descriptions and generate tactile versions of plots. Another popular package built on grid is the data visualisation package ggplot2 by US-based R expert Hadley Wickham, a former Department of Statistics student.

Paul has long been an advocate of open-source software like R, but he also strongly believes that publicly-funded research should be freely available for anyone to view, share, and reuse, and that’s why he now self-publishes through the Department’s technical blog, Stat Tech.

“This not only avoids the expense and delays of publishing in journals that are controlled by huge publishing houses, but it allows me to use very open licences and experiment with modern technologies, which is especially fun for emphasising reproducibility.”   

In doing this, Paul is one of a growing number of academics resisting the stranglehold of traditional academic publishing by setting up other pathways.

“Once it becomes easy to publish, the number of publications is less meaningful,” he says. “The emphasis is more on quality, as measured by things like citations, or other measures of how much your work is used and valued by others.”  If you’re interested in how Paul disseminates his work, his 2016 talk Publish for Pleasure (additional resources here) outlines the tools and processes that researchers can use.

Paul is on sabbatical this year, and is looking forward to clearing a backlog of research – “lots of writing” – and completing the third edition of his best-selling book R Graphics – “lots more writing”.  

He adds, “I am hoping that because I will have fewer short-term deadlines, I will be able to work and think bigger.”