[R] What happenes with R-gnome?

[Yves Gauvreau]

Here is my penny worth on this. I think from the start GUI interfaces where
invented or created to simplify the existence of users with a more natural
or friendlier interaction with the application and/or system. As time went
by the systems and application became more and more complex to the point of
creating something like information overload graphically speaking as in the
example below. Thus now days, one still need a comprehensive knowledge of
the application and\or system in order to exploit them efficiently via
command line or GUI interfaces. Sometimes if not all the time it is more
efficient to use command lines interfaces when available.

This doesn't mean that applications like R should forget all about GUI or
should I say graphics. I would have no objection to leave out the "UI" part
of GUI but I would favor much more "intelligence" into the graphic model.
Things like automatic rescaling and redrawing in cases where for example the
addition of lines, points and\or text would fall outside the limits of the
first plot command. Things like zooming capabilities, like callbacks for
interactive purposes, comprehensive 3D and the works.

I dream of the day when systems and applications will isolate completely the
user from the specifics of the implementations. In other words the systems
and applications would be sufficiently "intelligent" to interpret what the
user ask verbally, ask for clarifications if necessary, query the pertinent
data, perform whatever task are necessary and produce the output of the
query in a form suitable for the problem at end and some.

I'm still young maybe in 20-30 years will have something à la StarTrek.

Regards

Yves Gauvreau


----- Original Message -----
From: "Douglas Bates" <bates at stat.wisc.edu>
To: "Peter Dalgaard BSA" <p.dalgaard at biostat.ku.dk>
Cc: "Paul E Johnson" <pauljohn at ukans.edu>; "R mailing list"
<r-help at stat.math.ethz.ch>
Sent: Thursday, August 31, 2000 8:04 AM
Subject: Re: [R] What happenes with R-gnome?


> Peter Dalgaard BSA <p.dalgaard at biostat.ku.dk> writes:
>
> > Paul E Johnson <pauljohn at ukans.edu> writes:
> >
> > > When I saw "SciGraphica":
> > > http://magnet.fsu.edu/~feiguin/scigraphica/screenshots.html
> > >
> > > I wished it could be a GUI frontend for R on Unix.
>
> ...
>
> > However, the choice of GUI toolkit is more or less orthogonal to the
> > issue of having a plotting interface like the one on those
> > screenshots. I do sympathize with the desire to have simple and
> > accessible interfaces like that, but one has to be aware of the
> > drawbacks. Consider the following R function (somewhat simplified from
> > the original)
> >
> > "trace.plot" <-
> >     function(x, y, g){
> >         g <- factor(g)
> >         dfr <- na.omit(data.frame(x=x,y=y,g=g))
> >         dfr <- dfr[order(dfr$g,dfr$x),]
> >         plot(x, y, type="n")
> >         for ( i in levels(g) )
> >             evalq(lines(x,y), subset(dfr,g==i))
> >     }
>
> I think the Peter's point here deserves to be emphasized.  GUI's are
> very helpful in performing a task for the first time but often get in
> the way when doing tasks repeatedly.  That is why I think there will
> continue to be a place for "command line" interfaces, like the S
> programming language, even in the age of Microsoft Office and even for
> things like creating data graphics.
>
> About a year ago Bill Venables and I were at a conference and happened
> to be sitting together during a presentation on state-of-the-art data
> graphics.  The presenter showed a series of about 10 slides to
> demonstrate how easy it way to produce a certain type of plot.  The
> commentary went like "You click on this gadget in the toolbar and
> (next slide) drag that pointer to the plotting area and (next slide)
> the plot changes to this form.  Then you go back to the data bin (next
> slide) and click on another variable and (next slide) drag that to the
> vertical axis and ..."  After the 10 slides Bill and I looked at each
> other and both remarked that you could do the whole thing in a 1-line
> trellis formula and furthermore it could be described in a fraction of
> the time required to describe the GUI approach.  (I should note that
> trellis is available in S-PLUS and not currenly available in R.)
>
> This is not to say that there is no place for GUI's.  I'm just saying
> that the usefulness of GUI's for tasks like creating data graphics is
> often overestimated.  Or, looking at things the other way, the
> usefulness of command line interfaces is often underestimated.  Bill
> has remarked that a common comment from students in his short courses
> is that they hadn't realized how quickly tasks could be accomplished
> when using the text-based interface to S-PLUS instead of the GUI.
>
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
-.-.-
> 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
>
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
_._

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._