[R-wiki] R Wiki structure - level 1 & start page

Philippe Grosjean phgrosjean at sciviews.org
Thu Feb 2 13:50:42 CET 2006


Roger Bivand wrote:
> On Thu, 2 Feb 2006, Philippe Grosjean wrote:
> 
> 
>>OK, let's proceed step-by-step.
>>
>>Do we agree with the top level structure of the wiki:
>>
>>- beginners: collects together material for first time R users (and for 
>>curious people wanting to discover what R is). Subsections must still be 
>>defined.
> 
> 
> Is it at all possible to use the CRAN Task Views to "view" packages and 
> functionalities. I feel that Task Views are a potentially useful way of 
> reflecting emerging clustering, and do - even for beginners - give a view 
> of discipline or thematic groupings of functionality? Beginners in Finance 
> arguably can benefit from a different view than say Ecology?
> 
> Just a thought.
> 
> Roger

Such a structure (by subject-area, but indeed, it could reflect CRAN 
Task Views) is suggested for the 'tutorials' section. It is not 
implemented yet as proposed, but I will do this as soon as possible. I 
will make sure to respect current task views in this structure.

For 'beginners', we haven't discussed much what should go there. My own 
feeling is that it should be restricted to material useful only in very 
early learning, encouraging to switch to the other sections as soon as 
possible (in particular, 'tutorials'). If you propose too much there and 
if people feel they can do what they need with just those basic stuff, 
they tend to never move to the next step.

This is something I notice with R commander. R commander provides indeed 
a good set of statistical tests and a couple of models + some 
multivariate stat and a few data manipulation. This looks like enough 
for some biologists and many of my students clearly decide not to move 
to the scripts for that reason. If R Commander was less rich, they would 
be forced to do the jump!

So, for me, the 'beginners' section should contains what is required to 
install R and additional R packages, to start working with R, to import 
simple datasets, to produce basic analyses and graphs, and perhaps, to 
use a couple of GUIs, but that's it. Thus, not enough to justify a 
splitting of the material into subject-areas (this is pretty common 
stuff among all kinds of users).
Best,

Philippe Grosjean

>> [...]



More information about the R-sig-wiki mailing list