[R] New R website: R-Cookbook.com
phgrosjean at sciviews.org
Sat Sep 29 09:29:14 CEST 2007
You give a very long answer, and I don't want to comment each of you
points individually. The Wiki is flexible enough so that pages can be
arranged the way people want (they can add new sections too if they
want; the proposed structure is only a *suggestion* and by the way, the
structure is a consensus produced by a team of a couple of dozen R
users, which should give it a little bit of credit). There is a search
engine, and an RSS feed, too. Author that want to get authorship on one
page can ask me to lock their page, if they want, so that other people
cannot edit them. And finally, nothing forbid anyone to add other
solutions to a given problem, either on a different page, or on the same
page as the original answer. Searching for pages produced by a given
author is simple: just enter his name in the search box at top-right and
click search (try "Paul Johnson", for instance).
So, may be, you are not aware enough of the possibilities provided by
the Wiki, otherwise, you would know that most features provided by
R-Cookbook.com already exist in the R Wiki (in a slight different form,
for some of them). And the Wiki offers more, like unlimited version
checking, the possibility to revert back to any old version of the
document, to compare two versions, to include math equation in LaTeX, or
a simplified syntax called ASCIIMathML, etc, etc.
Furthermore, I really don't see an advantage to classify your recipes by
authors, when the kind of recipes you propose is more or less a copy and
paste from a discussion in the R mailing list, and the "author" is the
people that does this copy-paste, not the real author that formulated
the original answer in the R mailing list (sic)!
And YES, R-Cookbook.com does duplicate R-Wiki, indeed. Because the main
goal (a series of R recipes in addition to the official R documentation
provided by R users) is *exactly* the same one for both sites! In fact,
the two major sections of R-Cookbook.com, "recipes" and "guides" are
nothing else than a reincarnation in a different skin of the "tips" and
"guides" sections of the R-Wiki... and the third section, "forum", is
really duplicating the official R mailing lists.
Now, for just the little bit I discover from R-Cookbook.com, the
"non-linear" way of navigating through recipes is very nice (for
instance, the "similar recipes" section at the end... note that this
feature could be added as well in the wiki through a plugin).
I really believe that the community should join the force, not to spend
time to duplicate existing features. In this way, I would expect
comments and feedback on the existing tools (the R Wiki) to make it
better. What? You don't like the presentation? So, what could you
propose (it is Xhtml, so one can reskin the R Wiki site totally, if
needed). You don't like the "linear" presentation (calling a wiki
presentation as "linear", is absurd, but anyway)? So, why not to
contribute a dokuwiki plugin to provide a "similar pages" section at the
end of each wiki page, in order to navigate the same way as in your
Drupal site? It is PHP, and I believe you should be able to program this
plugin easily, if you were able to program a "HighlighteR" module on top
of Drupal for your R-Cookbook.com.
) ) ) ) )
( ( ( ( ( Prof. Philippe Grosjean
) ) ) ) )
( ( ( ( ( Numerical Ecology of Aquatic Systems
) ) ) ) ) Mons-Hainaut University, Belgium
( ( ( ( (
Jeff Spies wrote:
> Hi Philippe,
> I don't want to be competing in any way with any of the fantastic
> official R resources that exist; I only want to supplement them.
> Although this list is probably not the best place to discuss the proper
> use of wiki's for documentation/learning, I'll make a few comments on
> the role, I believe, R-Cookbook.com can play.
> First of all, R-Cookbook.com offers two organization schemes, dictated
> by the users: linear and (what's being call) non-linear. Recipes can be
> placed in guides in order to organize content linearly in a way more
> familiar to most users. Recipes can also be free-tagged, not only by
> the author, but by the community (much like a wiki--although you've
> actually implemented a linear scheme for tips and lack the non-linear
> Secondly, some people like to retain authorship of these sorts of
> things, and not only for selfish reasons, like self promotion. First,
> they get to control how their message is delivered--other's can't edit
> it. My guess is that you have the wiki configured as most wiki's
> are--community editing rights. To counter the issues associated with
> individual-control in a community-oriented site, at R-Cookbook.com,
> other users can comment on the recipe, or express their regards for the
> quality of the recipe via the rating system. Or they can create a
> recipe on the same topic, and the "similar recipe" engine should display
> it beneath the recipe in question. Some people also like to retain
> authorship merely so their friends/students/fans can keep track of their
> work. At R-Cookbook.com, users get a url specific to their recipes (in
> the format http://www.r-cookbook.com/recipes/[username]) along with a
> feed of all of their recipes. Also, this multi-level feed system is not
> available (I don't believe) via dokuwiki's framework. If a person were
> only interested in visualization, they could use their favorite RSS news
> reader, and only keep track of recipes related to visualization. Or if
> a person had a great deal of respect for a few authors, they could track
> their contributions the same way.
> In principle, a wiki may not be the best place to store "recipes"
> either. Because most wiki's like to have one article per topic, the
> community is forced to establish a "best" article via community
> editing. At R-Cookbook.com, there can be more than one recipe for a
> certain problem/issue, letting the community decide the preferred
> solution via ratings/comments. Sure, having more than one article can
> be done with a wiki's, but it almost defeats the purpose--I'm a purist,
> sorry. ;)
> More generally, recipes on R-Cookbook.com are free to go a different
> direction as what has been currently established as tips on the wiki (I
> know this can change, but I'm just saying...). "Recipes" can be very
> problem specific, related to many topics, or very specific to the
> authors data, or whatever the case may be--not necessarily related to
> one method of analysis or one function. Where would something like that
> go in your current linear organization of tips? At R-Cookbook.com, I
> would still appreciate and value the contribution because someone might
> find it useful, and they could discover it via tags, the search engine,
> a user feed, or the similar recommendation engine. Personally, I just
> don't see a place for that sort of entry in the wiki.
> With all of that said, if there is anything at R-cookbook.com that you
> believe would benefit the wiki more than perhaps a link to the recipe,
> the content is freely available to the public, and it is in your rights
> to put it there.
> Again, the site is purely to support the community, and I really don't
> believe it is in competition with what your wiki offers. I'd be glad to
> continue this discussion off-list and appreciate you bringing up the
> point though, it's a good question, that I hope I've answered.
> On Sep 28, 2007, at 4:11 AM, Philippe Grosjean wrote:
>> Hello Jeff,
>> Good initiative,... but why not to put this in the official R Wiki
>> (http://wiki.r-project.org)? There is a section named 'tips' dedicated
>> to such little recipes
>> (http://wiki.r-project.org/rwiki/doku.php?id=tips:tips). It should be
>> better to centralize all these little tips, don't you think so?
>> Should you have difficulties to use the Wiki, just tell me, and I will
>> Philippe Grosjean
>> ) ) ) ) )
>> ( ( ( ( ( Prof. Philippe Grosjean
>> ) ) ) ) )
>> ( ( ( ( ( Numerical Ecology of Aquatic Systems
>> ) ) ) ) ) Mons-Hainaut University, Belgium
>> ( ( ( ( (
>> Jeff wrote:
>>> R Community,
>>> I've put together a website that I thought this mailing list might be
>>> interested in: http://www.r-cookbook.com
>>> It's a (free) community-driven content management system for R
>>> "recipes", or working examples. Some of the features of the site are
>>> code highlighting, recipe ratings, recipe comments, personal "recipe
>>> boxes" to save your favorite recipes, community tagging, RSS feeds
>>> for each user and for each tag, and similar recipe recommendations.
>>> Although I imagine that many users will sort/search/find recipes by
>>> tags, I've implemented a linear organization for recipes as well:
>>> guides. These will be compilations of "recipes", organized in a
>>> logical fashion as to promote understanding of the topic of that
>>> particular guide and introduced with user-contributed pages. Over
>>> time, hopefully with the help of the community, more guides will be
>>> created and the ones I have will be filled-in to actually be useful.
>>> I have started several guides to give you an idea of the sort of
>>> thing I'm thinking: Introduction to R, Longitudinal Modeling in R,
>>> Exploratory Data Analysis in R, and more here, http://www.r-
>>> A couple of features that will be worked on in the near future are
>>> (1) the design of the site and (2) working on a more interactive code
>>> display (right now, functions are highlighted and linked to the r-
>>> docs, but that's it).
>>> I hope some of you might find the site useful and perhaps even
>>> consider contributing your own recipes. If you have any suggestions
>>> or feature requests, I'd be glad to hear them!
>>> R-help at r-project.org mailing list
>>> PLEASE do read the posting guide
>>> and provide commented, minimal, self-contained, reproducible code.
>> R-help at r-project.org mailing list
>> PLEASE do read the posting guide
>> and provide commented, minimal, self-contained, reproducible code.
More information about the R-help