Re: R-beta: mfg weirdness + future of graphics pars

Bill Venables (wvenable@attunga.stats.adelaide.edu.au)
Thu, 26 Mar 1998 23:19:30 +1030


Date: Thu, 26 Mar 1998 23:19:30 +1030
Message-Id: <9803261249.AA03145@attunga.stats.adelaide.edu.au>
From: Bill Venables <wvenable@attunga.stats.adelaide.edu.au>
To: Ross Ihaka <ihaka@stat.auckland.ac.nz>
Subject: Re: R-beta: mfg weirdness + future of graphics pars

Ross Ihaka writes:
 > 
 > I just checked my S manual and it appears that layout
 > parameters like can "mfg" only be given in par().

Yes.  pty is another, for example.

 > I think that in fact "mfg" is only meant to be queried.

No.  One use for setting mfg is to produce a page of plots in
portrait orientation with, say, two small plots on the top half
of the page and one long flat one along the bottom.  to wit

par(mfrow = c(2,2))
plot(... first ...)
plot(... second ...)
par(mfg = c(2,1,2,1)
tsplot(... long, lazy plot across bottom ...)

Although this is possible, I prefer to do this kind of thing
with the "fig" parameter.  In fact I use mfrow, mfcol and mfg as
little as possible since they waste so much of the space with
axes and labels.

 > 
 > Perhaps not unrelatedly ...
 > 
 > After some discussion with Paul Murrell I am inclined to think that
 > the whole graphics system needs a rewrite, even though this would
 > mean incompatibility with S.  A major problem is that the whole "par"
 > system does not work when devices can be resized....

....
 > Does anyone have any thoughts about how important it is to retain
 > compatibility with the S way of doing things?

I don't see much value in even trying to maintain compatibility
at this level.  The big corporate development at MathSoft Seattle
has been the acquisition of Axum, so the inventive energy and
experience are in place to give the graphics system (of S-PLUS at
any rate) the shake-up of its life.  S-PLUS 4.0 is effectively an
Axum dll, in fact, so the tail already wags the dog.  What next?
Compatibility is a worthwhile ideal in principle but not if it
means sacrificing flexibility or quality in graphics and
especially if it is going to be impossible to achieve, anyway.

If the announced MathSoft agenda is realistic (not all that
likely, I know, but if...) R and S will not even be approximately
compatible high level languages by the end of the year.

Bill.
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._