[R] A Few Suggestions to help out newbies

[Jonathan Baron]

Once again, I think a lot of this has been done, although not
quite the way you would like.

Another point is that the sort of things you are asking are
things that can be done by non-statisticians, e.g., anyone with a
web server and some extra disk space.

On 04/02/02 03:28, Zed Shaw wrote:
>Hey Folks,
>
>I may have some suggestions for you, based on my experience as a newbie
>with R.  Implementing these very simple resources would be fairly easy
>to do and would give volumes of help in return:
>
>1)  An R Cookbook section of the site where people can submit pieces of
>interesting code that satisfies a need.  This would be similar to the
>Perl/Python/Java Cookbook texts that O'Reilly puts out, but with a more
>dynamic activity.  The python folks have something like this, and people
>love it.  I learned a lot of python this way.

I like this, but see below.

>2)  A Series of Documents helping people translate from another package
>to R.  For example, "R for SPSS People", "R for SAS People", etc.

The "R for psychology" thing I wrote with Yuelin Li is a LITTLE
like this, as it makes a lot of comparisons with Systat and SAS.
(Systat, believe it or not, is the "scientific" package sold by
SPSS, and SPSS is the "business" package.)

>3)  A dynamic FAQ, placed prominently on the front page, ready for
>people to access and search.  The idea is that, as you encounter these
>dumb questions, you can slap up another faq question about it.  When it
>is asked again, don't bother replying, just *politely* say, "go to
>http://www.r-project.com/somefaqquestion/".  That saves everyone
>headaches, and encapsulates the knowledge on the list.  If it were setup
>right, it could be searchable through R.

There is a FAQ, and perhaps you can figure out how to make it
"dynamic".

>4)  Better web site layout.  It is hard to read the manuals if you can't
>find them.

They come with R itself, and you can point your browser to files
on your disk.  They are also on http://finzi.psych.upenn.edu, and
elsewhere.

>5)  Better search for the site.  It would be nice if you used google on
>the site, but even htdig setup properly would help.

I have htdig set up on http://finzi.psych.upenn.edu, along with
other things that might interest you and others.

>6)  Better layout of packages listed on CRAN.  This listing format will
>collapse under its own weight once it gets too large.
>
>7)  Create the "Encyclopedia of Statstics" online.  I would kill for a
>repository of all the "trade secrets" of statistics, related to R.  For
>example: a brief discussion of the merits of factor analysis,
>considering its heritage with IQ tests.  Or, "The History of Student".
>If this were organized right, it would even be possible to access it
>from R itself and provide people with help with the statistics part of
>using R (which is probably the most difficult portion).

I like this, and (1) above.  We have some of this in our "R for
psychology" piece, but more of this could be done, _if_ it were
properly searchable.

A problem is that a lot of uses of R (or any such program) are
discipline dependent.  Thus, an alternative is to develop things
like this for different disciplines, which is the approach we've
taken.  I'm not sure ANYONE, even the members of the core team,
could anticipate the needs of all the various users of R, well
enough to make a more detailed FAQ.  It is hard enough for me
just in the grab bag called "psychology," which includes
time-series analysis of neural impulses, fMRI data, questionnaire
studies, and treatment outcome studies.  But at least I know
about them because I sit through job talks!  Thus, I encourage
more of what we did, and I think you have provided some good
suggestions that we can use for revision.

>And, related to R:
>
>8) Command completion and contextual help in R.  The first one is
>probably fairly easy.  The second one is probably impossible.  It would
>involve giving out detailed help messages when things go wrong.  Not
>sure how to do that.

Doesn't ESS do this?

>9) Finally, my personal pet peeve of R:  tell me the line number of
>errors.  It's nearly impossible to fix a broken function when I don't
>know where it is broken.

If you run a script like source("myscript.R",echo=T) you see
exactly what it is doing and you can find where it bombs.
Moreover - and Systat never did this - all the variables created
up to the point where it bombs are there for you to examine.
Even when it bombs within a loop, the index of the loop is
available.  (This often happens to me when I have one subject
with no variance on the dependent variable in a within-subject
regression, and this is the easiest way to find such cases.)

>Anyway, those are just a few suggestions.  You'll notice that there is a
>common thread through all of them:  record the knowledge somewhere, make
>it easy to find.  I think doing at least some of these things would
>improve support for R, and make it fairly famous (especially if the
>statistics encyclopedia worked out). 
>
>Zed A. Shaw
>
>-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
>r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
>Send "info", "help", or "[un]subscribe"
>(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
>_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

-- 
Jonathan Baron, Professor of Psychology, University of Pennsylvania
Home page: http://www.sas.upenn.edu/~baron
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._