Introduction . . . .

In my Installation article, I mentioned that R does not provide a graphical user interface. By default, this is true. On Linux, the default R interface looks like this:

There's more.

Well, it's not much too look at, but I don't like the way pumpkin pie looks either. Unfortunately for my waist-line, I spend too much time working on R while eating pumpkin pie. In you use R on a Windows or Mac, the situation is better, but many of you make like to think about some alternatives. These versions of R do include a rudimentary application framework, but they are not complex interfaces. Fortunately, complex application interfaces are sometimes not very useful. The image below is a screenshot I have shamelessly appropriated from the R-Project's website:

R looks much nicer on OS X, and provides important tools such as syntax highlighting. In the coming months we will spend a lot of time talking about how to do the code part. Today, I would like to talk a little about the tools we should use to actually interface with R.

If you use Mac OS X, the default interface may be a good option for the time being. But, if you use Linux or Windows, the default frontends are just not good enough. Fortunately, all three platforms have some really excellent alternatives. Truthfully, all you really need is a solid text-editor that supports syntax highlighting for R and access to a terminal console. Everything beyond these two requirements is just icing on the cake. That's why I think the default interface on OS X is a realistic option. The default interfaces on Linux and Windows do not provide syntax highlighting, and have several other limitations as well. For me syntax highlighting is an absolute necessity. In short, my eyes like pretty colors schemes.

Disclaimer: Your friend Pluribus is proud to be a Linux geek. I have limited exprience with R on Windows. I don't own a Mac and have practically no experience with them. Therefore, please take my editor-choice opinions on these other platforms with an appropriately sized grain of salt. But, here are some options.

Pluribus' Number One Recommendation:

Regardless of platform, I think the best choice for interacting with R is emacs. I don't merely use emacs, I love emacs. I love it because it is so much more than a text editor. In this article, I don't have enough room to really explain what makes emacs so special, so I will refer you to a Wikipedia article and some lame humor. Installing emacs on Linux is easy. From the command line:

sudo apt-get install emacs ess

This will install the current version of emacs and a "major-mode" for R called ESS which is an acronym for Emacs Speaks Statistics which makes working with R much easier and more reproducible. Next week I will write a guide to using ESS. But, if you'd like to get ahead of me:

• Emacs Manual (Note: This comes installed with EMACS if you prefer to learn that way)
• ESS Manual

At this point it should be obvious that emacs comes with a steeper learning curve than some "modern" (lesser) text editors or IDEs. But, the flexibility and power of emacs is unrivaled. Once your fingers "know" emacs, other editors feel incredibly limited. Yes, I am biased.

If you want to install emacs on Windows, read this. If you want to use emacs on OS X, this looks a good option. But, I have read that Aquamacs is the real emacs alternative for OS X. But, I say those icons are clearly too pretty to belong to any realy real emacs clone.

Other Linux Options

The default text editor on Ubuntu Linux is Gedit ( See "Applications" -> "Accessories" -> "Text Editor"). In the past, gedit was . . . anemic . . . but recent versions of Gedit are very nice. If emacs looks overwhelming, this is a solid alternative. To make Gedit a little more effective, install the gedit-plugins package. This gives you an embedded terminal and a few other bells and whistles.

sudo apt-get install gedit-plugins

To use the plugins, you will need to configure them first. If you use Kubuntu or any other KDE based Linux, the default text editor Kate is even nicer than Gedit and also has syntax highlighting for R and an embedded console.

A quick Google search will show you several other options, including a plug-in for Eclipse. If you would like to go that route, you are on your own. I have no idea how to set it up, but I'm sure it's nice. If anyone else here is an Eclipse fan, I would appreciate the help. The last Linux option I am going to mention in JGR. JGR is a cross platform tool that works well on Linux, Windows and OS X. If you are going to use R on more than one platform, JGR is nice because your R programming environment can stay consistent. Installing JGR onto Linux is a little tricky, but if you follow these directions, you won't have any problems.

Windows

Yes, I do understand that a few of you may have legitimate reasons to use Windows. If you must, there are several nice options on this platform.  Tinn-R is a good IDE like environment for R on the Windows platform. And JGR, the Java front-end I mentioned above also works very well. It's also very easy to install on Windows.

OS X

If you use OS X and don't like Aquamacs or the default environment, I would look carefully at JGR. The Java runtime environment is well integrated on OS X and JGR is actually primarily developed for this platform, so installation should be real easy.

Goodbye

That's all for today. If you have any questions, fire away.

Free Tip: If you want to learn R quickly, join the mailing list. Lurking on this high-volume list is incredibly useful. I learn new tricks all the time. The list includes many developers of R and enough PhD's to start an independent Maths program. The "signal to noise ratio" is very good. In other words, the discussions are focused and there is very little BS. But, I do want to warn you, participating on this list isn't like using the Ubuntu Forums, or playing around on Facebook. If you are not used to participating in a programmer oriented list, I suggest joining the list and lurk for a couple of weeks before asking any questions. Doing so will help you learn how to ask questions appropriately and how to avoid problems. The list recommends asking direct questions and including relevant, commented code and many of the list participants enforce this ethic. They are not rude, but they are very concerned with ensuring that the list remains an excellent resource for users and developers.