[R-wiki] Beginner - Intermediate - Advanced... new attempt

Sun Jan 29 17:17:28 CET 2006

Overall these are my impressions:

- the technical aspects of the wiki are impressive

- I don't agree that a wiki is necessarily disorganized as the examples below
which I previously posted and repeat at the end show.  The current
confusing organization needs to be redone and using one of them as a model
would be one way since they are seem reasonably well done.

- the beginner/intermediate/expert are not only not useful, but actually
detract by providing clutter.  I don't think its adequate just to say not to
use them if you don't like them.  I don't think they should be an
organizing principile of the wiki.

  Tcl - http://mini.net/tcl/
       - this one has over 1000 pages

 Common Lisp - http://www.cliki.net/index

 PHP - http://www.php.net/manual/en/
        - not sure what to call this but users can add comments to end
of each page.

 Lua - http://lua-users.org/wiki/

On 1/29/06, Philippe Grosjean <phgrosjean at sciviews.org> wrote:
> Please, please, consider this:
>
> **Stop thinking you could ever control the structure of a Wiki!**
>
> By the way it is working (everybody can add pages everywhere), it is
> going to be unstructured, by nature!
>
> Also, stop thinking that the Wiki can be restricted to a given use:
> again, you don't control what people add to it. So, it is not reasonable
> to think that only material useful for beginners will be added, for
> instance.
>
> So, without a new idea about how to tag material regarding the skill of
> the user, the Wiki is going to suffer the same problem as does the
> current R online help: beginners have problems to find essential stuff,
> because it is not clearly separated from the rest.
>
> Currently, there are two mechanism I think about to help make a distinction:
>
> 1) the novice/user/expert tag. I think that, now, it is easy enough:
> everything is at user level, except:
>
>   - trivial things for someone that can install, start and run at least
> basic analyses in R. This material is clearly meant to help novices to
> start using R... and it should be tagged as such, so that other users
> can rapidly skip these pages (when they see the icon, thus, before
> reading anything).
>
>   - tricky stuff that a "normal" user does not really need to work with
> R, but that can be interesting for experts, or people willing to become
> experts (and thus, try to know a little bit more about the internals, or
> details about R). This material should be tagged as "expert", so that
> the other readers can skip these pages/sections and would not feel
> stupid just because they don't understand what is there.
>
> A final word: it is the author that is responsible for the tag on his
> page, and the tag is an optional feature (Gabor: if you don't like it,
> or still do not know how to rate material,... just don't use it, right?)
>
> 2) The second mechanism is the possibility to write custom,
> well-organized, table-of-content that sort material in the wiki. Someone
> could, for instance, decide to write a TOC for a biologist beginning
> using R from a Systat background (something really specific thus), and
> he could collect together the material he thinks is useful for such a
> reader (and the skill tag on the pages would help also to spot where
> material for beginners is located). Such TOCs are, indeed, similar to
> 'Task Views' on CRAN, in a way.
>
> Of course, I am open to any other *positive* and *realistic* proposition
> to help organize material in the Wiki, but I suggest you look at other
> Wikis, and perhaps also, you start making you own personal Wiki before
> commenting, because I feel that a couple of negative comments originate
> from people that do not fully understand what a Wiki really is.
> Best,
>
> Philippe Grosjean
>
> Damian Betebenner wrote:
> > I agree that the categorization into distinct categories is futile. I think that there are much
> > more meaningful characterizations of the content that can be made that will help the user
> > navigate its contents.
> >
> > The goal in designing the Wiki, I think, is to allow the user to zero in on what they're
> > interested in as quickly as possible. Someone mentioned recently on the list-serv
> > that the list-serv archives are not ideally structured for someone wishing to get
> > a question answered. Thus, the same questions are asked over and over.
> >
> > The R language already comes with a "dictonary" like structure for its commands. However,
> > when you aren't exactly sure what command you need, its tough to know what command
> > to look at.
> >
> > As a coder, often the most useful thing to me is a well done example. Some of the examples
> > in the R command help are wonderful, but I often discover them inadvertantly. What
> > I will use the Wiki primarily for it to examine really nice examples that people add. That
> > will lead me to examine the syntax more carefully which will likely in turn lead me to other
> > examples. It's this back and forth that is the power of hypertext. A Wiki strength is its
> > ability to present code alongside graphics and text annotation.
> >
> > Thus, my recommendation is that a navigation layout be thought up that (in addition to
> > other functionality) allows the user to quickly procced to examine certain types of
> > examples (e.g., string manipulation, categorical data analysis, etc.). I don't know if this could
> > be faciliated using some  sort of metadata, but being able to go back and forth between
> > examples and the commands that make up the examples seems most useful.
> >
> > Damian
> >
> >
> >
> >
> >  |  On Sat, 28 Jan 2006 10:12:39 -0500
> >  |  Gabor Grothendieck <ggrothendieck at gmail.com> wrote:
> >
> >>>>I must say I find this entire beginner/intermediate/expert baffling.
> >>>>I have no idea what goes where and to me its an categorization
> >>>>that is difficult to get right and probably not a good idea.  I would
> >>>>just omit the whole thing.  If someone can point to another
> >>>>language that has used such a categorization successfully
> >>>>I would be willing to modify this viewpoint but I know of none.
> >>>>I think more useful categorizations have already
> >>>>been discussed.
> >>>>
> >>>>On 1/28/06, Philippe Grosjean <phgrosjean at sciviews.org> wrote:
> >>>>
> >>>>>Hello,
> >>>>>
> >>>>>Following the previous discussion about the icons with three users (some
> >>>>>people did not like them and found them not very clear nor informative
> >>>>>enough), I make a second trial.
> >>>>>
> >>>>>This time, there are still three skill levels ("novice", "user" and
> >>>>>"expert"), but no possible mixture (like "novice" + "user", excluding
> >>>>>"expert", for instance).
> >>>>>
> >>>>>I also considered Tont Plate's suggestion of an explicit text, but I
> >>>>>integrate this text in the icon (I also took Ben Bolker's suggestion to
> >>>>>use road signs ;-).
> >>>>>
> >>>>>I designed three series of icons:
> >>>>>1) With text for top of page,
> >>>>>2) Big icon without text, for sections on the page and
> >>>>>3) small icons without text, to rate with discrete icons lists or table
> >>>>>of content entries.
> >>>>>
> >>>>>The Wiki site is modified to use these icons, so, you can make your own
> >>>>>idea more easily. Look for instance at:
> >>>>>- http://www.sciviews.org/_rgui/wiki/doku.php?id=start for the explanation,
> >>>>>
> >>>>>- http://www.sciviews.org/_rgui/wiki/doku.php?id=rtips:data:import for
> >>>>>the use in a page and,
> >>>>>
> >>>>>- http://www.sciviews.org/_rgui/wiki/doku.php?id=beginners:introduction
> >>>>>for usage in a table of content.
> >>>>>
> >>>>>Once again, I am waiting for your valuable comments!
> >>>>>Best,
> >>>>>
> >>>>>Philippe Grosjean
> >>>>>
> >>>>>_______________________________________________
> >>>>>R-sig-wiki mailing list
> >>>>>R-sig-wiki at r-project.org
> >>>>>https://stat.ethz.ch/mailman/listinfo/r-sig-wiki
> >>>>>
> >>>>
> >>>>_______________________________________________
> >>>>R-sig-wiki mailing list
> >>>>R-sig-wiki at r-project.org
> >>>>https://stat.ethz.ch/mailman/listinfo/r-sig-wiki
> >
> >
> > Damian Betebenner
> > Educational Research, Measurement & Evaluation
> > Lynch School of Education
> > Boston College
> > Chestnut Hill, MA 02467
> >
> > (617) 552 4491
> >
> > _______________________________________________
> > R-sig-wiki mailing list
> > R-sig-wiki at r-project.org
> > https://stat.ethz.ch/mailman/listinfo/r-sig-wiki
> >
> >
>
>