General FAQ
I need help! Where can I ask?
In case you can’t find the relevant information in this wiki, see this page for the main ways to contact us.
Overview
What is RKWard?
RKWard is meant to become an easy to use, transparent frontend to the R-language. R is a very powerful, yet hard-to-get-into scripting-language with a strong focus on statistical functions. RKWard makes R programming easier and faster, by providing a graphical frontend that can be used by unexperienced users (thanks to easy-to-use dialog boxes) as well as experts (which are attracted by the integrated development environement proposed by RKWard).
RKWard then is (will be) something like a free replacement for commercial statistical packages. In addition to ease of use, three aspects are particularily important:
- It’s a transparent interface to the underlying R-language. That is, it does not hide the powerful syntax, but merely provides a convenient way, in which both newbies and R-experts can accomplish most of their tasks. A GUI can never provide an interface to the whole power of a language like R. In some cases users will want to tweak some functions to their particular needs and esp. to automate some tasks. By making the “inner workings” visible to the user, RKWard makes it easy for the user to see where and how to use R-syntax to accomplish their goals.
- For the output, RKWard strives to separate content and design to a high degree. It will not try to design its own tables/graphs, etc, which have to be converted to the style used in the rest of a publication by hand. Currently RKWard uses HTML for its output. Using appropriate style definitions reformatting this output to match the rest of the publication will be easily doable. In future releases RKWard will even seek stronger integration with existing office suites.
- it relies on a language, that is not only very powerful, but also extensible, and for which dozens of extensions already exist.
And of course, it’s free as in free speech!
Are there other R GUIs?
Yes, there are. Here’s the most complete list to date: http://www.sciviews.org/_rgui/ . The R Wiki has an overview, too (still lacking information at the time of this writing): http://wiki.r-project.org/rwiki/doku.php?id=guis:projects .
Some popular choices are:
- Sciviews (Windows, only).
- RCommander by John Fox might be mentioned as another versatile, but somewhat limited R-GUI, build upon TCL/TK and R language. RCommander is available for Windows and U-/Linux systems alike.
- Another option of a GUI-like approach is offered by Emacs/ESS which comes with powerful capabilities to master programming, debugging and profiling of self written GNU R functions. That said, don’t expect Emacs/ESS to be kind of a conventional GUI like SPSS or SAS (which by the way can also be run in Emacs/ESS!) rather than a terminal like application offering lots of R functionality.
Platforms
Linux / Unix
Is RKWard available for Linux / Unix?
Yes, it is!
How can I install RKWard on Linux / Unix?
Many Linux distributions include RKWard today. Check your distributions package manager, or see the list of known binaries and build scripts.
If no package exists, compiling from source is generally not too difficult on Linux systems. Detailed instructions are available: Building RKWard From Source
Windows
Is RKWard available for Windows?
Yes, it is! There are some platform-specific quirks, but RKWard runs on at least Windows 2000, XP, Vista, and Windows 7. Many users use RKWard productively on these platforms.
How can I install RKWard on Windows?
Several different options exist for installation, ranging from a bundled installation of everything you need to run RKWard, to compiling from source. See RKWard on Windows for details.
Can I install RKWard on a USB disk on Windows?
Yes, you can. Be sure to use the “complete installation bundle” (see RKWard on Windows). This installation can freely be moved to different paths.
Mac OSX
Is RKWard available for Mac OSX?
Yes, it is! This is still rather fresh, though, and it’s quite possible that you will run into a few quirks. We provide binaries for Mac OS X 10.7 or above. For older Macs, you will have to use MacPorts to compile from source. See RKWard on Mac for instructions on installing on the Mac, and please provide your feedback!
Troubleshooting
I get an “error while starting the R backend”. What does it mean?
“The ‘rkward’ R-library either could not be loaded at all, or not in the correct version”
This error is typically the result of mixing different methods of installation on a single system (e.g. from binary packages, and from source, or installation from source with different parameters). More specifically, R supports installing libraries to several different locations. Probably, in one of these locations, an older version of the rkward support library has been left over from a previous installation.
In general, you want to clear all leftovers of previous rkward installations. To do so, run R as root, and type:
for (libloc in .libPaths()) try (remove.packages ("rkward", libloc))
Note: On some systems, there are even several installations of R, e.g. one in /usr/local/bin/R, and one in /usr/bin/R. In this case, do the above for each installation of R.
You may want to apply the procedure mentioned below, to clean up any other left-overs of previous rkward installations, too, for good measure. When done, re-install RKWard.
“RKWard either could not find its resource files at all, or only an old version of those files”
Typically, this error occurs, if you installed from source, and did not use a suitable “-DCMAKE_INSTALL_PREFIX” option while running cmake. The error may also indicate that there are left-overs from a previous installation of RKWard on your system. Try running
find /usr /opt -name rkward
on the command line. Clear out all mentioned directories / files, then install again (using the correct “-DCMAKE_INSTALL_PREFIX” option).
I have upgraded to R, and now RKWard stopped working
While much of the time RKWard will simply continue to work with a new version of R (esp. minor releases), sometimes it is necessary to re-compile RKWard for the version of R you will be using.
Importantly this can be a problem when trying to use a compiled package of RKWard provided with your Linux distribution, while using an updated package of R, directly from CRAN. For some distributions, we, or third parties, provide appropriate updated packages. Please refer to the page on binaries and build scripts for more info. If no solution is listed for your distribution, you may have to compile from source, and / or ask your provider to create updated packages.
Design issues
Why is RKWard a KDE-application?
KDE offers a set of tools that allow us to make progress relatively quickly. The QT library on which it is based is one of the very best toolkits currently available.
Why is ECMAScript (Javascript) used for the plugin framework?
The reasons for using JS are:
- well known and stable (applies to python as well)
- not much functionality is needed in this context, and JS provides all that we need in a lightweight manner, with very little overhead compared to most other options.
See Converting_Plugins_from_PHP_to_JS for some background information.
Getting involved
What can I help with?
Please have a look at the Open Tasks for a list of good starting points. Or offer your help on the mailing list (see next question). We are looking for all sorts of skills and talents.
Donations
We develop RKWard, because we have use for a good statistics tool in our professional lives, but RKWard is not part of our jobs, and time we spend on the project is time that we can’t spend other activities such as advancing our careers. We enjoy working on the project for various reasons, but not all aspects of this work are fun.
Donations we receive will be split among the currently active developers (although many give the money back to the project). It helps us buy new hardware to develop RKWard on, and it helps us achieve the feeling that the time we spend working on RKWard is well invested.
To make a donation, please visit: http://sourceforge.net/donate/index.php?group_id=50231 .
Thanks!