[R] R Wiki and R-sig-wikii

Philippe Grosjean phgrosjean at sciviews.org
Wed Jan 18 13:12:21 CET 2006


Hello all,

This is to announce the creation of R-sig-wiki, a new R SIG (Special 
Interest Group) mailing list dedicated to the elaboration and 
maintenance of a R Wiki. You can subscribe at: 
https://stat.ethz.ch/mailman/listinfo/r-sig-wiki. There is currently a 
prototype for a new R Wiki at http://www.sciviews.org/_rgui/wiki 
(temporary address). The main idea is to offer a site where users could 
collaborate in writting various kind of documentation for R. It mainly 
targets R beginners, but is left open for more advanced sections. Any R 
user interested in the setting up of this Wiki is warmly welcome to 
participate and to subscribe to R-sig-wiki. For the others, we will send 
an announcement for the final R Wiki on this list when it will be ready. 
Hereunder is a (rather long) summary of the discussion we had so far on 
this topic.
Best,

Philippe Grosjean

======================================================================
There is a proposition to create a R Wiki, stimulated essentially by two 
facts:

1) The traffic in R-Help is very high, with many "trivial" questions 
asked repeatedly. Obviously, searching in R-Help archives is not so 
obvious for some R users. Perhaps another presentation, like plain HTML 
pages would be fine. Since the building of these HTML pages should be a 
collaborative work, a Wiki seems to be a possible solution (recall that 
a Wiki is essentially a simple way to collaborate on writting Web pages; 
see Wikipedia definition at http://en.wikipedia.org/wiki/Wiki).

2) Some threads in R-Help are very long and difficult to follow in the 
form of a succession of emails. Again, a more structured presentation, 
allowed by HTML / Wiki pages is suggested as a possible solution.

There is already one attempt to build a Wiki by Detlef Steuer at: 
http://fawn.unibw-hamburg.de/cgi-bin/Rwiki.pl. Despite not much effort 
was put in this wiki, several people collaborated to it with time, but 
it appears that it was below the minimum required to make it fly as it 
should.

Being unaware of that Wiki, I proposed recently (beginning of January 
2006) a prototype of another R Wiki, just to explore if and how it could 
answer those two problems on R-Help. The prototype is at 
http://www.sciviews.org/_rgui/wiki (temporary address).

There is a couple of other Wikis dedicated to R floating around, like: 
http://www.okada.jp.org/RWiki/ (in Japanese) and 
http://learnserver.csd.univie.ac.at/rcomwiki dedicated to R(D)COM and 
RExcel essentially.

After a discussion, the R-Core Team decided to give support to one or 
several intiatives to make a R Wiki, with a big concern about the 
quality of information in the Wiki and how to keep it in phase with the 
rapid development of R. Here is the mail send by Martin Maechler:

========================================================================
Martin Maechler wrote:
 >We've had a small "review time" within R-core on this topic,
 >amd would like to state the following:
 >
 >------------------------------------------------------------------------
 > The R-core team welcomes proposals to develop an R-wiki.
 >
 >- We would consider linking a very small number of Wikis (ideally one)
 >  from www.r-project.org and offering an address in the r-project.org
 >  domain (such as 'wiki.r-project.org').
 >
 >- The core team has no support time to offer, and would be looking for
 >  a medium-term commitment from a maintainer team for the Wiki(s).
 >
 >- Suggestions for the R documentation would best be filtered through
 >  the Wiki maintainers, who could e.g. supply suggested patches during
 >  the alpha phase of an R release.
 >--------------------------------------------------------------------------
 >
 >Our main concerns have been about ensuring the quality of such extra
 >documentation projects, hence the 2nd point above.
 >Several of our more general, not mainly R, experiences have been
 >of outdated web pages which are continued to be used as
 >reference when their advice has long been superseded.
 >I think it's very important to try ensuring that this won't
 >happen with an R Wiki.
===========================================================================

After that announcement, several people started to discuss the structure 
and content of the Wiki (Tony Plate and Ben Bolker took the initiative 
to propose one structure on the experimental R Wiki at: 
http://www.sciviews.org/_rgui/wiki/doku.php?id=varia:organization_discussion, 
and I propose to leave it open until the end of February. At that time, 
we will implement the proposed structure as a "proof-of-concept".

Another topic is about how to allow discussion on the R documentation in 
the Wiki. This topic was fed by Frank Harrell Jr, Gabor Grothendieck and 
others and it leads to a trial on: 
http://www.sciviews.org/_rgui/wiki/doku.php?id=varia:test_include. 
Basically, all CRAN and Bioconductor .Rd files would be converted 
automatically to read-only Wiki pages (regularly updated) that are 
themselves included in  fully editables "discussion" pages. A link to 
latest svn versions of .Rd files for base packages was also suggested. 
Finally, access to the documentation online in Wiki format should be 
included as naturally as possible in R (proposal of a helpWiki() 
function), so that users would be easily and naturally redirected to the 
most up-to-date information (latest .Rd man page followed by Wiki 
additions), if their computer is connected to the Internet.

Another problem is: which Wiki engine to choose and which features to 
propose. Indeed, there are more than 30 different wiki engines, and they 
range from simplistic one (like Tipiwiki, for instance), to very 
elaborate ones (like MediaWiki, the Wikipedia engine; Trac, a complete 
solution to follow software development with bug tracking and 
collaborative documentation writting in Wiki format; or TWiki, a very 
capable, but rather complex Wiki engine used in many big companies). The 
Web site http://www.wikimatrix.org/ is excellent to compare most popular 
Wiki engines. Using this tool, and after a couple of trials, I settled 
on DokuWiki: both simple enough (PHP only, plain text files) and 
powerful (and also completelly customizable/expandable through plugins).

Currently, Detlef Steuer decides to set its ancestor R Wiki as read-only 
to the benefit of the DokuWiki prototype (the idea is to move pages to 
the new one and fuse the two R Wikis in a single one). Although I am not 
sure everybody is completelly convinced that the customized DokuWiki 
engine I porpose (more than 10 plugins added, syntax coloring for R code 
added to GeSHI, MathML rendering of LaTeX equations added) is the one to 
use, there seems to be little alternative proposed (after disucssions, 
Trac and TWiki were considered as valid alternatives, but they do now 
overcompete DokuWiki as a good compromize between simplicity and 
performance for documentation writting).

To feed the Wiki with high-quality material (more than "just" the few 
thousands of .Rd man pages of R and additional packages in CRAN and 
Bioconductor, sic!), I proposed to several authors of Web sites 
dedicated to R documentation to move to the Wiki. This will certainly 
ease to keep the information up-to-date, since everybody could 
collaborate in maintaining this material. I already got a positive 
answer from Paul Johnson for Rtips 
(http://www.ku.edu/~pauljohn/R/Rtips.html#1.1) and from James Wettenhall 
for R Tcltk Examples 
(http://bioinf.wehi.edu.au/~wettenhall/RTclTkExamples/).

So, what next?
Well, I think we should not rush. We should take time to refine the Wiki 
engine, the Wiki server and the structure of the Wiki in order to 
optimize them for the intended use. So, I suggest we continue to 
experiment with the propotype for a month or two. Then, we will set up 
the definitve R Wiki site, hopefully with an address like 
http://wiki.r-project.org... (if possible)! So, currently, we are 
seeking for good ideas both about the customization of the wiki engine, 
about how to feed the .Rd files in it and about how to structure the site.

In a second phase, we will certainly need volunteers to help translate 
HTML pages from RTips, R Tcltk Examples, and possibly other Web sites to 
Wiki pages.

I hope I did not forgot something important in this summary?!
Best,

Philippe Grosjean
-- 
..............................................<°}))><........
  ) ) ) ) )
( ( ( ( (    Prof. Philippe Grosjean
  ) ) ) ) )
( ( ( ( (    Numerical Ecology of Aquatic Systems
  ) ) ) ) )   Mons-Hainaut University, Pentagone (3D08)
( ( ( ( (    Academie Universitaire Wallonie-Bruxelles
  ) ) ) ) )   8, av du Champ de Mars, 7000 Mons, Belgium
( ( ( ( (
  ) ) ) ) )   phone: + 32.65.37.34.97, fax: + 32.65.37.30.54
( ( ( ( (    email: Philippe.Grosjean at umh.ac.be
  ) ) ) ) )
( ( ( ( (    web:   http://www.umh.ac.be/~econum
  ) ) ) ) )          http://www.sciviews.org
( ( ( ( (
..............................................................




More information about the R-help mailing list