FOR IMMEDIATE RELEASE


Statistical Analysis Environment R turns 2.0.0
===============================================

Version 2.0.0 of the R environment for statistical computing and
graphics is released on Monday, October 4th (2004-10-04).

R, also known as "GNU S", is a language and environment for
statistical computing and graphics. R implements a dialect of S,
developed at Bell Laboratories by John Chambers et al. John Chambers
received the 1998 ACM Software System award for S and later joined the
R Core Team.

R has increased dramatically in popularity in recent years, in teaching
and academic research, notably in bioinformatics, and also in industry
and finance.

R is designed as a true computer language with control-flow
constructions for iteration and alternation.  It is easily extensible
with user code written in the R language or by dynamic loading of
compiled code. A package library system has been developed and more
than 400 contributed packages are available.

R is developed by an international team of 16 (the "R Core Team") with
many contributions from the user community. An R Foundation has been
established (incorporated in Austria) to underpin the R system. Highly
active mailing lists have been set up for general user interaction and
development issues as well as more specialised topics. The newsletter
R News has reached its 10th issue for a total of 69 high-quality
articles.

This new release marks more a coming of age than a radical change of
the product. Since the release of 1.0.0 on February 29, 2000, R has
developed steadily and settled on a release cycle with a "dot-release"
two times per year. 

Lots of things have happened over the last four and a half years: The
R language has acquired namespaces, exception handling constructs,
formal methods and classes, much improved garbage collection and we
now have generalized I/O via connection objects.

We have seen considerable improvements in the graphics area. The
"grid" subsystem offers increased flexibility compared to the older,
but still viable, ink-on-paper model. The "lattice" package builds on
this new foundation to reimplement and extend Bill Cleveland's
Trellis graphics for structured multiframe layouts.

The Windows GUI has been further refined, and we now also have a Mac
OSX port, using the Aqua interface. We also have a cross-platform
method for developing GUI interfaces via Tcl/Tk (similar to Python's
Tkinter). Work on alternate interfaces and use of other interface
toolkits is ongoing and in various stages of development. An R-SIG
(Special Interest Group) has been set up for discussion of these
matters. The issues of interfacing R to other languages and of
embedding R in other applications have been addressed.

The user workspace has been reorganized and so has the set of packages
that ship with R. We now bundle several "recommended packages",
maintained outside of the R core, but deemed indispensable in a
statistical system. This includes import/export functions for other
statistical packages.  The configuration scripts have been tuned to
make installation as smooth as possible and it is now much easier to
link against high-performance linear algebra libraries.

In addition, there has been a large number of more specific new
functions, tweaks, and bug fixes.


	   The R Core Team


Links:

The R Project homepage:  http://www.r-project.org

The Comprehensive R Archive Network: http://cran.r-project.org and
mirrors. This is where the software and add-on packages can be
fetched.