Re: [R] Pros and Cons of R

From: GREGOR Brian J <>
Date: Fri, 23 May 2008 11:00:35 -0700

Monica, here are some other Pros to consider about R:

  1. IMHO, the most important reason for using R is that expressed by John Chambers as the aim of the S language: "to turn ideas into software, quickly and faithfully." The broad capabilities of R facilitate the integration of data maintenance and cleaning, exploratory data analysis, model estimation, model implementation, model calibration, model application, and the reporting and display of model outputs. This has tremendous productivity advantages. For example, I was able to meet a tight timeline for developing a regional land use model because R allowed me to easily move through the steps of model development from data analysis to implementation. I even found that I could do some geographical operations that could not be done using our GIS software. In addition, all of the model outputs (including a very many maps) were produced with R. Once you learn how to use R in this way, you will find it takes less time to program the outputs in R than to produce them in GIS. R yields productivity advantages in smaller ways too. We've developed a number of small applications to solve GIS or other problems that could not be solved as easily using other tools. Moreover, once they have been solved using R, the solutions are easily automated or recycled in other contexts.
  2. R facilitates documentation and replication. Previous to using R, we did our data analysis and implemented our models in a variety of platforms. For example, Access, Excel, SPSS and Stata were all previously used in household survey data processing and analysis. This was a documentation nightmare. All the steps can be done using R instead and documentation can be easily included in the scripts. If care is taken to use good naming conventions that emphasize readability, the scripts can be largely self documenting. This also facilitates group work.

Since we started using R in our work, we have been able to greatly increase our modeling capabilities and output with no increase in staffing.

Brian Gregor, P.E.
Senior Transportation Analyst
Oregon Department of Transportation
Transportation Planning Analysis Unit
555 13th Street NE
Salem, OR 97301

>Message: 22
>Date: Thu, 22 May 2008 16:00:10 +0000
>From: Monica Pisica <>
>Subject: [R] Pros and Cons of R
>To: <>
>Message-ID: <BAY104-W52F35E50A680C814F52417C3C60@phx.gbl>
>Content-Type: text/plain; charset="Windows-1252"
>I am doing a very informal presentation for my office about R
capabilities to deal with and analyze spatial data, display data and maps, and connections with GIS. I've used in my presentation info from the CRAN, the spatial Task view, and the more striking graphics examples from and NCEAS cs/OneMapProdWithRGraphics.html together with examples of my own work.
>I am finishing with pros and cons about R and I am wondering if you can
come up with other examples, or comments. Here they are:
>- R is a programming environment well suited for statistical analysis.
>- R is open source and cross platforms (Windows, Mac, Linux).
>- Fortran, C (C++), and Python wrappers are in place.
>- Deals well with spatial data, has a robust graphical interface and
has an active user group list / forum.
>- External packages for R are almost daily increasing, most of them
based on published up-to-date books and peer-reviewed articles.
>- R related books ? quite a few ?.
>- R has a very steep learning curve.
>- There is no perfect ?beginner? book.
>- Experience with other programming languages is a plus / minus.
>- You can save scripts, but not *.exe.
>- It is updated several times a year (good) but there are no up-grades.
>- It seems that it is hard to install correctly under Linux.
>- Everything you want to do is a command line, minimal GUI.
>- Memory management problems (depends on your OS), especially when
displaying big images at high resolution or working with huge matrices (hundreds of Mb).
>Also i am wondering if R works under 64 bit computers and if it takes
advantage of it.
>Monica mailing list PLEASE do read the posting guide and provide commented, minimal, self-contained, reproducible code. Received on Fri 23 May 2008 - 18:22:26 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Fri 23 May 2008 - 20:30:49 GMT.

Mailing list information is available at Please read the posting guide before posting to the list.

list of date sections of archive