On CRAN, the official web home of all things R it says,
R is a free software environment for statistical computing and graphics.
Well, that sounds all hunky dory. But let’s take a close look at what this statement really is says.
- “free software” – what does this mean? Free as in gratis or as in Open Source free? As it turns out, R is not only priceless but is also Open Source, an aspect that is fundamentally required for anyone doing publicly funded research published in peer-reviewed journals. Without going into the details (because others have done it much more eloquently, e.g. here and this week in Nature) research has to be reproducible and only Open Source software can achieve this for computational research. Apologize to all you Mathematica (or whatever proprietary software you use for your research) users out there, I am sure your software looks nice but you are using a black box and as a result your research is irreproducible.
- “environment” – why settle for a piece of software when you can have an environment. Think software = swiss army knife (A jack of all trades and… you get my point) vs. environment = a tool box in the spirit of Unix philosophy (Do one thing and do it well. Write
programspackages to work together.) Bottom line, you don’t build a house using a swiss army knife.
- “statistical computing” – now come on people I know many of you guys (for some reason I can only find one woman on the official R contributor rooster) are into statistics, but you really are short-selling your baby here. I can’t remember the last time I used R for statistics (probably this morning, so yeah, I know) but R can do so much more, just look at the amazing diversity of packages available. There is a lot more than just statistics here. And yes, one can do mathematical biology with R (that’s biology without stats people).
- “graphics” – check!
So here’s what I think CRAN should be saying about R:
R is a free and open source environment for computational research and graphics.
Filed under: computing, open science, R