[R] Input appreciated: R teaching idea + a way to improve R-
pietr007 at mc.duke.edu
Wed Oct 24 05:14:11 CEST 2007
I agree with Frank - decentralization is key in a wiki environment
since it allows the system to dynamically correct itself. That said,
there are some ongoing experiments trying to change this -- check, for
example, http://scholarpedia.org/ , which seems to be somewhat along
the lines Ted described. You will notice that the content is of very
high quality, but apparently the site doesn't seem to expand its
content as fast as other environments that are completely
There is probably not a single answer to this issue, as different
degrees of centralization will serve different purposes to different
audiences in different environments and at different times ... such as
in ecological and evolutionary models
On 10/23/07, Frank E Harrell Jr <f.harrell at vanderbilt.edu> wrote:
> (Ted Harding) wrote:
> > On 23-Oct-07 16:11:12, Tony Plate wrote:
> >> [...]
> >> Is there any way on the R-Wiki for people to quickly and easily add an
> >> annotation indicating that they believe some particular advice is poor
> >> practice? Ideally, these annotations would be easily searchable so
> >> that other users could find and fix or respond to them.
> >> -- Tony Plate
> > I think the ideal medium for this kind of thing (and in my opinion
> > it can -- and in the future will -- expand to the general domain
> > of on-line publication) is on the following lines.
> > A. Someone puts up a document. This is "owned" by its author
> > and cannot be changed by anyone else. (There is also an argument
> > for stipulating that on such a medium the author cannot change
> > it either--the "back-trace" could be meaningful and important).
> > B. There is one exception to (A). Anyone can "mark" a place in the
> > document with a link to another contribution (which might be a
> > further contribution, a comment, a correction, a link to something
> > else altogether, ... ). All such links can also be followed in the
> > reverse direction.
> > C. Rules (A) and (B) appliy to all documents in the hierarchy.
> > D. At some stage, the original author or anyone else can "wrap up"
> > what has happened so far by creating a new "root" document. The
> > previous version can be archived.
> > E. There is a case for plain-text file format where the content
> > can be expressed in words. More generally, though (and, of course,
> > especially for content which includes graphics or mathematics),
> > a generally-readable file format with the necessary capacilities
> > should be used. This seems to me to imply PDF (and exclude such
> > proprietary formats as Word or Excel, and unfortunately even PS
> > which is not universally readable). Where data need to be included,
> > this whould be possible using CSV files.
> > Having said all that, I'm wondering what web format and software
> > can conveniently implement such a structure. I have very little
> > experience with Wikis (apart from reading them from time to time),
> > so I don't really know how well a Wiki would lend itself to this.
> > There are some other considerations which would be at least
> > desirable.
> > F. Searchability.
> > G. A user should be able to bring up a tree representation,
> > using edges to link nodes which, when clicked/hovered on,
> > would pop up a box giving a brief descrption of what the
> > link is about; and the user should be able to drop ("prune")
> > branches which are not of interest in order to simplify the
> > task.
> > I'd be very interested to see commments on these thoughts!
> > Best wishes to all,
> > Ted.
> The experience we're having with wikis has to my satisfaction shown that
> such levels of control, and keeping an initial draft intact, are not
> necessary and can be counter-productive. Refactoring and sparing
> readers from out-of-date thoughts is a key to productivity and knowledge
> > --------------------------------------------------------------------
> > E-Mail: (Ted Harding) <Ted.Harding at manchester.ac.uk>
> > Fax-to-email: +44 (0)870 094 0861
> > Date: 23-Oct-07 Time: 18:11:17
> > ------------------------------ XFMail ------------------------------
> > ______________________________________________
> > 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.
> Frank E Harrell Jr Professor and Chair School of Medicine
> Department of Biostatistics Vanderbilt University
> R-help at r-project.org mailing list
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
Ricardo Pietrobon, MD, PhD, MBA
Associate Vice Chair and Assistant Professor, Department of Surgery
Director of Biomedical Informatics, Duke Translational Medicine Institute
my calendar is available at
to view start and end times, click on the specific hyperlink and then
send me a time that best fits your schedule
More information about the R-help