Many thanks to everyone that posted replies to this thread.  I used
some of the ideas from this thread and other sources to put together a
case for R and I just received formal approval from our IT department
today.  In case this can be useful to anyone in the future, here's a
summary of what was submitted (by the way, I love the slides found at
http://www.matthewckeller.com/Lecture1.ppt, the Harry Potter stuff is
brilliant):

Thanks again to everyone!
Dan Viar
Chesapeake, VA

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------


(head of IT name removed),

          Based on our conversation yesterday, below is some
documentation that might be useful in trying to evaluate the risk that
R might pose.  I also contacted one of the company's that is trying to
be a "Red Hat" for R and his response is included below.  The salient
points:

·         R is "the de-facto standard" for statistical computing and
(for example) appears in peer reviewed journals of Statistics

·         R is a quality open-source product, not some small piece of
freeware developed by an individual

o    For example there are currently over 19 individuals responsible
for the maintaining the source (see:
http://www.r-project.org/contributors.html).  These individuals are
arguably some of the most talented in the field of Statistical
Computing.

·         R is licensed under the GPL (e.g. see
http://www.groklaw.net/article.php?story=2008081313212422)

·         R requires no more support from IT than a similar stat
package (e.g. SAS).  In other words, we may want to put R on our
desktops (it doesn't write to the registry) or have it installed on a
server like "Bugsy" but R would not generate calls to the help desk
(unlike something like Excel).

·         Anecdotal evidence suggest that the technical support
offered by the R community (through forums, email lists, etc…) is
comparable if not better than that provided by commercial products.
For instance, we have been evaluating various commercial packages
(SAS, SPSS, S-Plus) and so far have had better responses to getting
technical questions answers on the R-help list than through the help
desks of company's trying to sell us their software.  As another
example, last night I posted "How do I get my IT department to 'bless'
R" and so far I have received 12 replies.

·         R benefits from external innovation that makes it able to
have quick reaction time to new statistical ideas.  It is not uncommon
for a cutting edge statistical technique to appear first in R and then
make its way into a commercial package.

R was recently featured in the NY Times:

http://www.nytimes.com/2009/01/07/technology/business-computing/07program.html

According to the article

"Companies as diverse as Google, Pfizer, Merck, Bank of America, the
InterContinental Hotels Group and Shell use it."



I recently came across the following links that show that some
commercial products like SAS and SPSS are providing functionality so
that their programs can call R (as a selling point).  Here's a link to
some SAS marketing:


http://support.sas.com/rnd/app/studio/Rinterface2.html


>From that site:


"R is a leading language for developing new statistical methods," said
Bob Rodriguez, Senior Director of Statistical Development at SAS. "Our
new PhD developers learned R in their graduate programs and are quite
versed in it."

It's open source software, and many add-on packages for R have
emerged, providing statisticians with convenient access to new
research. Many new statistical methods are first programmed in R.

While SAS is committed to providing the new statistical methodologies
that the marketplace demands and will deliver new work more quickly
with a recent decoupling of the analytical product releases from Base
SAS, a commercial software vendor can only put out new work so fast.
And never as  fast as a professor and a grad student writing an
academic implementation of brand-new methodology.

This sounds like a pretty strong endorsement for R, from one of its
commercial competitors.


I also found the following link which is a Power Point presentation
explaining why we would be interested in R (nice if you like Harry
Potter).  It does a good job of show casing the differences between
SAS, SPSS, and R.


http://www.matthewckeller.com/Lecture1.ppt


Let me know if there is something formally that I need to do (forms to
fill out, process, etc…).

Thanks,

Dan




R: Regulatory Compliance and Validation Issues A Guidance Document for
the Use of R in Regulated Clinical Trial Environments

http://www.r-project.org/doc/R-FDA.pdf

R installation and administration manual

http://cran.r-project.org/doc/manuals/R-admin.pdf



http://www.r-project.org/

http://www.revolution-computing.com/



>From Colin Magee [co...@revolution-computing.com]:



Hi Dan -



Well, we'd love to talk to your Head of IT/ Manager about this.



We can make them feel more than warm and fuzzy - our build of R is
built within a controlled and documented software development
lifecycle to GAMP5 standards, fully documented and supported.   We
have to be FDA compliant since REvolution R Enterprise is used in
clinical environments.   At the end of the day, it's got nothing to do
with the source code being open (which is a positive), it's all about
how the codebase is managed, tested, compiled and supported - from a
commercial perspective, that is.   This is what Red Hat did for Linux
and what we are doing for R.   We have a bunch of slides on our build
process, so happy to go through those.   And yes, whilst we of course
love R, we would always say it is better to use a supported
distribution within a commercial enterprise.

The other point I should have mentioned is of course that Microsoft
are one of our biggest supporters.... (supporting REvolution to
integrate 64 bit R on Windows, IDE currently in development, HPC
Server integration etc.)





Best,

Colin



Case Study: PFIZER using R

http://www.revolution-computing.com/downloads/case-studies/revo_case_study_pfizer_rpro.pdf



Example of R being used at Merck & Co., Inc (See page 44 of PDF):

http://www.r-project.org/doc/Rnews/Rnews_2005-1.pdf



Analyzing clinical trial data for FDA submissions with R:

http://blog.revolution-computing.com/2009/01/analyzing-clinical-trial-data-with-r.html



Supporting Institutions:

    * Astra Zeneca R&D Mölndal, Mölndal, Sweden
    * AT&T Labs, New Jersey, USA
    * Baxter AG, Vienna, Austria
    * Baxter Healthcare Corp., California, USA
    * BC Cancer Agency, Vancouver, Canada
    * Black Mesa Capital, Santa Fe, USA
    * Boehringer Ingelheim Austria GmbH, Vienna, Austria
    * Breast Center at Baylor College of Medicine, Houston, Texas, USA
    * Center für digitale Systeme, Freie Universität Berlin, Germany
    * Dana-Farber Cancer Institute, Boston, USA
    * Department of Biostatistics, Johns Hopkins University, Maryland, USA
    * Department of Biostatistics, Vanderbilt University School of Medicine, USA
    * Department of Economics, Stockholm University, Sweden
    * Department of Mathematics and Statistics, Utah State University, USA
    * Department of Statistics, University of California at Los Angeles, USA
    * Department of Statistics, University of Warwick, Coventry, UK
    * Department of Statistics, University of Wisconsin-Madison, Wisconsin, USA
    * Department of Statistics, Iowa State University, USA
    * Department of Statistics & Actuarial Science, University of Iowa, USA
    * Dipartimento di Statistica, Università Cà Foscari di Venezia, Italy
    * Division of Biostatistics, University of California, Berkeley, USA
    * Ef-prime Inc., Tokyo, Japan
    * European Bioinformatics Inst., UK
    * Hygeia Associates, California, USA
    * Lehrstuhl für Rechnerorientierte Statistik und Datenanalyse,
University of Augsburg, Germany
    * MPI for Demographic Research, Rostock, Germany
    * Massachusetts General Hospital Biostatistics Center, Boston, USA
    * National Public Health Institute, Helsinki, Finland
    * Norwegian Institute of Marine Research, Bergen, Norway
    * School of Economics and Finance, Victoria University of
Wellington, New Zealand
    * Spotfire, Massachusetts, USA
    * TERRA Lab, University of Regina - Department of Geography, Canada
    * ViaLactia Biosciences (NZ) Ltd, Auckland, New Zealand

Donors:

    * Adelchi Azzalini (Italy)
    * AT&T Research (USA)
    * Austrian Association for Statistical Computing (Austria)
    * BC Cancer Agency, Vancouver (Canada)
    * Fabian Barth (Germany)
    * Biostatistics and Research Decision Sciences, Merck Research
Laboratories (USA)
    * Brian Caffo (USA)
    * David W. Crawford (USA)
    * Dianne Cook (USA)
    * Yves DeVille (France)
    * Department of Economics, University of Milano (Italy)
    * Dipartimento di Scienze Statistiche e Matematiche di Palermo (Italy)
    * Emanuele De Rinaldis (Italy)
    * Zubin Dowlaty (USA)
    * Faculty of Economics, University of Groningen (Netherlands)
    * Jaimison Fargo (USA)
    * Peter L. Flom (USA)
    * David Freedman (USA)
    * Manfred Georg (USA)
    * Google Inc., Mountain View, California (USA)
    * Hort Research Institute (New Zealand)
    * Peeter Luikmel (Estland)
    * Vittorio de Martino (Italy)
    * Shigeru Mase (Japan)
    * Peter M Maksymuk (USA)
    * Merck and Co., Inc. (USA)
    * Telecom New Zealand (New Zealand)
    * Minato Nakazawa (Japan)
    * Network Theory Ltd, Bristol (United Kingdom)
    * Stavros Panidis (Greece)
    * Norm Phillips (Canada)
    * Bill Pikounis (USA)
    * James Robison-Cox (USA)
    * Saxo Bank (Denmark)
    * Julian Stander (United Kingdom)
    * Stanford University, California (USA)
    * Università Cà Foscari Venezia (Italy)
    * Université de Nantes (France)
    * Boris Vaillant (Germany)
    * Ivo Welch
    * Kevin Wright (USA)

______________________________________________
R-help@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.

Reply via email to