Revolution R and Fedora: Revisited

February 10, 2012
By

(This article was first published on Confounded by Confounding » R, and kindly contributed to R-bloggers)

A previous post of mine had suggested that, despite them being extremely similar operating systems, and really there being no clear reason why, Revolution R 5.0, which does support Red Hat Enterprise Linux, refused to work on Fedora 16. The installation failed, dependencies could not be installed, tech support was singularly unhelpful because I wasn’t using RHEL 5.0, and I essentially said “nuts to this” and went back to my trusty, working, free installation of R sitting in my beloved native OS X. But today, the plot thickened…

Since that time, Revolution has added support for RHEL 6.0. I figured what the heck, why not give it another go, see if things will be more cooperative this time. Lo and behold, they work! That’s right, it appears that the newly updated version of Revolution R – at least the academic version which I was using – will indeed install and work on the latest version of Fedora.

Now, some caveats:

  • I have no idea if it will work for everyone. I’m running a brand new download of Fedora 16 via Parallels on my Mac, installed with essentially every developer related package I could find in hopes of covering all my bases. And only the 64-bit version of Fedora works. Tried it on the 32-bit version, and it failed spectacularly.
  • It could break at any time. What with Fedora not being supported.
  • If it does break, you’re SOL. My experience with Revolution’s customer service/tech support folks to date suggests that if you’re not running RHEL (or maybe CentOS), you’re SOL if something stops working. So I’d still be a touch skittish about putting any major analysis into Revolution R running on a Fedora machine, or if you do and get it working, touch nothing.
  • I don’t actually know that it works. I mean, it installs. ‘$ Revo64′ works as a command. It comes up, and I can do my usual “R, are you working?” procedure, which involves nothing more than creating two vectors of random numbers, plotting them against each other, and running a quick and dirty lm(). All of that worked, but I sincerely doubt that really tested the guts of Revolution’s distribution of R. If someone wants to give me some heavier code to put it through its paces and see if it crashes, please feel free to drop me a line or comment.

But there you have it. Hope for the future.

Now Revolution, if you’re listening? My kingdom for an OS X native version.


Filed under: R

To leave a comment for the author, please follow the link and comment on his blog: Confounded by Confounding » R.

R-bloggers.com offers daily e-mail updates about R news and tutorials on topics such as: visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...



If you got this far, why not subscribe for updates from the site? Choose your flavor: e-mail, twitter, RSS, or facebook...

Comments are closed.