[R] Pros and Cons of R - Summary

[Gabor Grothendieck]

On Mon, May 26, 2008 at 11:01 AM, Monica Pisica <pisicandru at hotmail.com> wrote:
>
>
> Hi,
>
> First of all thanks for all the answers .... all were very informative and helped me shape a better presentation. Although i received lots of emails saying that installation under Linux / Unix is easy .... still emails asking for help in installing R itself, or external packages and such crop more often than not on the list and for this reason i pressumed it is harder to install under Linux like OS ;-)) Myself i have no experience with it - i work under Windows with all the good and the bad and the ugly of it and anybody who can "push the install button" can install under Windows .... you even don't need to actually know how your Windows operating system works.... but this is digressing already.
>
> So .... new Pros (some new and some needed emphasis ...)
>
> - R is open source and free (emphasis)
> - very good support form the R help list (emphasis)
> - data maintenances and integration, exploratory analysis, testing hypothesis
> - model implementation, calibration, application and reproduction / replication
> - Automation, recycling and ..... batch
> - R facilitates documentation and replication
> - takes full advantage of 64 bit processor
>
> Cons ....
>
> - slow learning curve (yes, it depends on the person who learns it but still ....)
> - call by value, there are no references (e.g. Java)

I would not really say it lacks references since environments
are largely equivalent.  e.g.

f <- function(e = parent.frame()) with(e, x <- x+1)
x <- 3
f()
x # 4

f(.GlobalEnv) # explicitly pass global environment
x # 5

There are also a number of packages that facilitate
this by various approaches, e.g. proto package.

Other pros:
- large growing community
- large number of addons
- runs on PC, Mac, Linux/UNIX

Other cons:
- interpreted - can be slow
- datasets usually must reside in memory - constrains size
- no .Net on Windows (only the older COM interface via addons)
- PC, not web-centric

>
> Also, somebody mentioned 3d graphics .... yes/ no/ maybe .... I saw and know that there are 3d capabilities in R, but i never really needed them - maybe in the future ;-) .... 2D maps and graphs are quite sufficient for a very large amount of work, although i have to recognize that a 3d map / graph can give the "wow" factor ;-)
>
> Again thanks for all the answers,
>
> Monica
>
>
>
>
>
>
>
>
>
>
>
>
>
> _________________________________________________________________
> Give to a good cause with every e-mail. Join the i'm Initiative from Microsoft.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>