[R] R for various ports of linux
Ben Bolker
bolker at zoo.ufl.edu
Wed Nov 5 23:29:05 CET 2003
I've always built from source and almost never had to do anything beyond
"tar zxf sources.tgz; ./configure; make; make install" (on various Red Hat
versions). On the other hand ... I've been hoping to move in the
direction of an apt- or rpm-based solution to get a better handle on
tracking package versions etc. etc. across multiple machines ... It does
seem as though the Debian folks (Eddelbuettel [sp]?) are very quick about
packaging apt versions ...
Ben
On Thu, 6 Nov 2003, Jason Turner wrote:
> Nathan Leon Pace, MD, MStat wrote:
>
> > To all:
> >
> > I currently download the R binaries for Redhat 7.x Linux.
> >
> > There is considerable turmoil in the vendors of Linux. Redhat apparently
> > is changing it's business model to paid versions.
> >
> > This might motivate my department to use a different vendor of Linux.
> >
> > Is there anything predictable about which vendors/versions of Linux will
> > have R binaries in the future?
>
> Short answer: build from source. You won't regret it.
>
> Long answer:
> The "build from source" approach is remarkably painless under any Linux
> distribution I've tried (RH, SuSE, Slackware, et. al.). It's also
> painless under Solaris.
>
> The days of having to be a programmer to build R from source have been
> over for years. If you're computer literate enough to use R, you're
> probably over-qualified to build from sources.
>
> Kudos to R-core for their attention to detail in making what's
> complicated "under the hood" quite simple for the end user.
>
> Alternate answer:
> If you absolutely must have binaries, there will be binaries as long as
> there are users of your OS with time they wish to commit to building
> them. This may be where your sysadmin steps in :)
>
> Cheers
>
> Jason
>
--
620B Bartram Hall bolker at zoo.ufl.edu
Zoology Department, University of Florida http://www.zoo.ufl.edu/bolker
Box 118525 (ph) 352-392-5697
Gainesville, FL 32611-8525 (fax) 352-392-3704
More information about the R-help
mailing list