[R] Linux, UNIX, XP32, Vista X64 or ...? [SEC=UNCLASSIFIED]

[Patrick Connolly]

On Sun, 27-Jan-2008 at 07:38PM -0500, Gabor Grothendieck wrote:

|> On the PC there is a builtin GUI but not on UNIX and there are
|> some packages that are OS specific in which case you might
|> get more or less selection but probably more.  Also depending
|> on the specific system you may have greater difficulty installing
|> certain packages due to the need to compile them on UNIX
|> and the possibility exists that you don't quite have the right
|> libraries.  On Windows you get binaries so this is not a problem.
|> I have repeatedly found that common packages that I took
|> for granted on Windows had some problem with installation
|> on UNIX and I had to hunt around and figure out what the problem
|> was with my UNIDX libraries or possibly some other problem.
|> For all R packages this won't be a problem but for packages
|> that use C and FORTRAN this can be.  Although I am lumping
|> all UNIX systems together I think this varies quite a bit from
|> one particular type/distro of UNIX/Linux to another and I suspect if you
|> are careful in picking out the right one (if you have a choice) you
|> will actually have zero problems.

I've had a fair bit of experience with Redhat/Fedora/CentOS and with
Mepis distros.  I've had zero problems compiling R or packages.  

The only problem I've had is with *using* a package which uses R code
to hard-wire graphic options that don't suit X11, postscript or pdf
devices.  Virually every other package I've used doesn't hard-wire
graphic options, so that problem never arises.

YMMV

-- 
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.   
   ___    Patrick Connolly   
 {~._.~}          		 Great minds discuss ideas    
 _( Y )_  	  	        Middle minds discuss events 
(:_~*~_:) 	       		 Small minds discuss people  
 (_)-(_)  	                           ..... Anon
	  
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.