[R] Re: R-Gnome. I don't want abuse!

Paul Johnson pauljohn at ukans.edu
Thu Aug 31 18:45:32 CEST 2000


I'm getting some abuse for asking about R-Gnome and I don't think it is
fair. I did not create that project, I was just asking how it was going
and whether that other project would help out.

Now I'm getting some email that says some of you don't want a GUI
interface because you think interfaces like that are inconvenient.  OK.
I agree, in most cases.  I think GUIs for model fitting are mostly bad. 
You could carry the argument even further, actually. I have a colleague
who finds R to be an interference with his convenience, and he writes
programs to do factor analysis and t-tests in C.  

I don't propose to change the R experience in any way for people who
don't want a GUI.

I do think there are times when it would be nice to have a GUI, though,
for touching up graphs.

I would like a graph window where I can click on particular items and
adjust their properties, see how they look with tick marks in or out, or
spaced 5 units apart or 10, and so forth. I've been doing that with Axum
for about 7 years. If you have not tried  Axum (or perhaps even the S+
that assimilated Axum, a la the StarTrek Borg, although I don't use it
much), you should and you might revise your opinion about the value of a
good OO graphing package.  I can put a graduate student with Axum and
they can make graphs and easily tailor them to their needs. The
scripting is pretty easy, so one can graphs run over and over again or
adjust to particular datasets.  Their saved files work easily to
customize to new data and questions.  I expect, before students would be
as capable with R, it would take about 5 times as long.  Hell, think of
the time we could save for students if there were a File/save as menu in
R? (Doesn't every new user go berserk the first time they try to save a
graph or print it in R? no offense intended, but it drove me crazy and
every other week or so somebody else wanders through in the same
quandry.)

And, I do think SAS analyst is an example of how the GUI might be used
to generate the code. SPSS's journaling feature does that as well,
incidentally. And, at one time, I could do that with Axum, but I don't
recall about S+4.0.

Axum is not free, does not exist for Linux, it does not have the
programmabaility of R, and it does not have some of the statistical
modeling capabilities, so I'm not proposing to use Axum and I don't
expect R should try to immitate it.   If the powers that be say "we are
canceling support for R-gnome", I might be disappointed, but I will
still keep on trying to find my way around R.  I'm still assembling my
tip sheet, which you can see if you want to make sure that I'm actually
making an effort (http://lark.cc.ukans.edu/~pauljohn/R/statsRus.html). 
Pretty soon I won't need a GUI, but I might still wish I had one.

-- 
Paul E. Johnson                       email: pauljohn at ukans.edu
Dept. of Political Science            http://lark.cc.ukans.edu/~pauljohn
University of Kansas                  Office: (785) 864-9086
Lawrence, Kansas 66045                FAX: (785) 864-5700
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list