[R] R-help Digest V1 #34

R-help Digest owner-r-help-digest at stat.math.ethz.ch
Fri Dec 3 04:01:10 CET 1999


R-help Digest         Friday, December 3 1999         Volume 01 : Number 034



In this issue:

  [R] model.tables
  Re: [R] Installing R on Slackware Linux
  Re: [R] Installing R on Slackware Linux
  [R] nlmin
  Re: [R] nlmin
  Re: [R] nlmin
  Re: Summary: [R] Wanted: online Introduction to R
  [R] Applying a function of several variables to data 
  Re: [R] R and XML -- a near perfect combination?
  [R] problem installing R 0.90
  Re: [R] graphics
  Re: [R] graphics
  Re: [R] graphics
  Re: [R] graphics
  Re: [R] R and XML -- a near perfect combination?
  Re: [R] graphics
  [R] Plot Character Sizes in rw-0.90.0
  RE: [R] R and XML -- a near perfect combination?
  Re: [R] R and XML -- a near perfect combination?
  Re: [R] Plot Character Sizes in rw-0.90.0
  Re: [R] R and XML -- a near perfect combination?
  [R] Meaning?
  Re: [R] Installing R on Slackware Linux
  Re: [R] Meaning?
  Re: [R] graphics
  par(mkh) {was [R] Plot Character Sizes in rw-0.90.0}
  Re: [R] graphics
  [R] Large plot symbols (with pch = 1 etc)
  Re: [R] graphics
  Re: [R] graphics
  Re: [R] Large plot symbols (with pch = 1 etc)
  Re: [R] Installing R on Slackware Linux
  Re: [R] graphics
  RE: [R] Meaning?
  Re: par(mkh) {was [R] Plot Character Sizes in rw-0.90.0}
  [R] problem with par(fig=value)
  [R] Meaning?
  [R] Multi-threaded external routines with R?
  Re: [R] problem with par(fig=value)

See the end of the digest for information about r-help-digest

----------------------------------------------------------------------

Date: Wed, 1 Dec 1999 14:21:11 +1100 (EST)
From: John Maindonald <john.maindonald at anu.edu.au>
Subject: [R] model.tables

Berwin Turlach wrote -

> > >>>>> "JM" == John Maindonald <john.maindonald at anu.edu.au> writes:
> 
>   JM> At 08:02 30/11/99 +0000, Prof Brian D Ripley wrote:
>   >> On Tue, 30 Nov 1999, spoon <spoon at hilbert.maths.utas.edu.au> wrote:
>   >> 
>   >>> Hi,
>   >>> Is this a bug or do I just not understand model.tables?
>   >>> [...]
>   >>> Or am I just completely misinterpreting something basic?
>   >> 
>   >> Basically, yes. This is an incompletely replicated design, and d and e
>   >> occur on different litters.
>   >> 
>   >> The results are identical to the S-PLUS original. I think you are
>   >> probably looking for what dummy.coef gives you.
>   >> [...]
> 
>   JM> So what is it that model.tables() gives?
> Good question, in R (0.65.1) the documentation of model.tables says:
> 
>    WARNING:
>         The implementation is incomplete, and only the simpler
>         cases have been tested thoroughly.

Simon and I had noted that.  As S-PLUS has no such warning, perhaps
the discussion should be taking place on s-news.

>   JM> S-PLUS says they are estimates.  Of what?  These are not the
>   JM> marginal means of the fitted values, ignoring other factors.  Do
>   JM> they mean anything at all?  The issue of unbalance does not
>   JM> arise here.
> The documentation in R also states that
> 
>     Details:
> 
>         For `type = "effects"' give tables of the coefficients
>         for each term, optionally with standard errors.
> 
>         For `type = "means"' give tables of the mean response
>         for each combinations of levels of the factors in a
>         term.
> 
> Hence, my understanding would be that if `type="means"' is specified
> the marginal means of the observations ignoring other factors is
> given.  But this doesn't seem to be the case, at least not with an
> unbalanced design.  You can use either your dreamed up example or the
> data that spoon has posted.

On the usual definitions of `balance' both in my example and
Simon's example treatments are `balanced" over blocks (or litters).

...

John Maindonald               email : john.maindonald at anu.edu.au        
Statistical Consulting Unit,  phone : (6249)3998        
c/o CMA, SMS,                 fax   : (6249)5549  
John Dedman Mathematical Sciences Building
Australian National University
Canberra ACT 0200
Australia
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Wed, 1 Dec 1999 10:10:09 +0000 (GMT)
From: Bill Simpson <wsi at gcal.ac.uk>
Subject: Re: [R] Installing R on Slackware Linux

Become root: su -

Put the new *.tgz file the same place as the old one was (something
like /usr/local or /usr/local/src).

Unpack it: tar zxvf Rwhatever.tgz

Then read INSTALL for detailed instructions.

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 at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Wed, 1 Dec 1999 10:34:41 +0100 (MET)
From: Torsten Hothorn <hothorn at statistik.uni-dortmund.de>
Subject: Re: [R] Installing R on Slackware Linux

On Wed, 1 Dec 1999, Murray Jorgensen wrote:

> I mostly use Windows, but I am slowly learning to use Linux on my second
> machine. (It came in handy a couple of weeks ago when I had two files to
> sort, each of length about 2 million. Windows sort died after an hour, but
> the Linux sorts took 2-3 minutes each.)
> 
> Anyway the last time I tried installing R in Linux I found out from a
> helpful computer support person that with my distribution (Slackware) it is
> better to work from the source files than binaries. In fact he was so
> helpful that he did it all for me. This was great, except I didn't learn
> how to do it myself. 
> 
> Now I'd like to upgrade from 0.65.1 to 0.90.* and add some packages, but
> this time I'm all on my own. My linux skills are quite limited, but I have
> some books to help me. Is there anything written down somewhere that help
> someone compile and install R who has never compiled or installed anything
> on Linux before?

Well,

get R-0.90.0.tgz, copy it to /usr/src  and then type (as root)

$> tar -xzvf R-0.90.0.tgz	# untar the archive
$> cd R-0.90.0			# change to the R dir
$> ./configure			# create your Makefiles
$> make 			# if the output of configure is ok compile
$> make install			# and install the bin's

All those things are well documented in the FAQ and README's as in all GNU
projects ... :-)

Torsten




> 
> Regards,
> 
> Murray Jorgensen
> 
> Murray Jorgensen,  Department of Statistics,
> University of Waikato,  Hamilton,  New Zealand.  [maj at waikato.ac.nz]
> ______________________________________________________________________
> Hey! Look at this! 
> The survivors are the ones we had time to get the water to. 
> None of the eighteen other variables that we recorded were significant.
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> 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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Wed, 1 Dec 1999 09:33:56 +0000 (GMT)
From: JJ Dos Santos Ramalho <J.J.Ramalho at bristol.ac.uk>
Subject: [R] nlmin

I'm a very recent user of R. I have been adapting my Splus programmes
and I found only one (important) problem. There exists no function
"nlmin" in R and its substitute, "nlm", does not work well with my kind
of problems, sometimes no achieving convergence, other tines
"converging" to impossible values. My models are highly nonlinear and
are to be estimated by both GMM and empirical likelihood methods.

Is there any alternative nonlinear optimization command in R? In the
help file it is said that "nlm" is only a preliminary version, is it
expected to replace/improve it soon? Does anyone have any alternative
algorithm?

Thanks for your help,

Joaquim Ramalho
Department of Economics
University of Bristol

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

------------------------------

Date: Wed, 1 Dec 1999 10:59:24 +0100 (MET)
From: Jim Lindsey <jlindsey at alpha.luc.ac.be>
Subject: Re: [R] nlmin

> 
> 
> I'm a very recent user of R. I have been adapting my Splus programmes
> and I found only one (important) problem. There exists no function
> "nlmin" in R and its substitute, "nlm", does not work well with my kind
> of problems, sometimes no achieving convergence, other tines
> "converging" to impossible values. My models are highly nonlinear and
> are to be estimated by both GMM and empirical likelihood methods.

I have also been using nlm for the past 4-5 years in R and for another
5-10 years before that as a Fortran program. Most of my problems are
also highly nonlinear and I rarely have had a problem unless the model
is unsuitable for the data. nlm has several arguments that can be
modified to aid in convergence. If you obtain impossible values, then
you appear to have a constrained optimization problem. Often parameter
transformation can fix this (e.g. log for positive values, logit if
between 0 and 1). Constrained optimization is a difficult problem
for which really good algorithms are not available. All of the
numerical analysts that I have spoken to agree that the algorithm in
nlm for unconstrained optimization without supplying analytic
derivatives is about the best available. Note that, as of 0.90, you
can supply the derivatives to nlm, which may also solve your problem
(although my experience is that it makes little difference if the
model is suitable for the data). Jim

> 
> Is there any alternative nonlinear optimization command in R? In the
> help file it is said that "nlm" is only a preliminary version, is it
> expected to replace/improve it soon? Does anyone have any alternative
> algorithm?
> 
> Thanks for your help,
> 
> Joaquim Ramalho
> Department of Economics
> University of Bristol
> 
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> 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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Wed, 1 Dec 1999 10:16:33 +0000 (GMT)
From: Prof Brian D Ripley <ripley at stats.ox.ac.uk>
Subject: Re: [R] nlmin

On Wed, 1 Dec 1999, JJ Dos Santos Ramalho wrote:

> 
> I'm a very recent user of R. I have been adapting my Splus programmes
> and I found only one (important) problem. There exists no function
> "nlmin" in R and its substitute, "nlm", does not work well with my kind
> of problems, sometimes no achieving convergence, other tines
> "converging" to impossible values. My models are highly nonlinear and
> are to be estimated by both GMM and empirical likelihood methods.
> 
> Is there any alternative nonlinear optimization command in R? In the
> help file it is said that "nlm" is only a preliminary version, is it
> expected to replace/improve it soon? Does anyone have any alternative
> algorithm?

Yes, yes. Note, though that nlm was improved for 0.90.0 and can now
make use of gradient information.  Does your application have known
derivatives? (I know nlmin does not use them.)  The aim is to
allow alternative methods within the nlm function fairly soon (a couple
of months).  One issue to allow the methods to be used recursively,
which means basing them on C code not Fortran code.


- -- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

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

------------------------------

Date: Wed, 1 Dec 1999 08:11:08 -0500
From: partha_bagchi at hgsi.com
Subject: Re: Summary: [R] Wanted: online Introduction to R

One interesting page to consider would be John Pezzullo's interactive Stats
page:
http://members.aol.com/johnp71/javastat.html



                                                                                                              
                    Clive Jenkins                                                                             
                    <clive.jenkins at clara        To:     Peter Dalgaard BSA <p.dalgaard at biostat.ku.dk>         
                    .net>                       cc:     r-help at stat.math.ethz.ch                              
                    Sent by:                    Subject:     Re: Summary: [R] Wanted: online Introduction to  
                    owner-r-help at stat.ma        R                                                             
                    th.ethz.ch                                                                                
                                                                                                              
                                                                                                              
                    11/24/99 03:53 PM                                                                         
                                                                                                              
                                                                                                              




Peter Dalgaard BSA wrote:
>
> Clive Jenkins <clive.jenkins at clara.net> writes:
>
> > An excellent on-line book for the more advanced reader, especially
> > chapters 14 and 15 that deal with Statistics and Data Modeling:
> > "Numerical Recipes in C: the Art of Scientific Computing" (ISBN
> > 0-521-43108-5) Copyright (C) 1988-1992 by Cambridge University Press. A
> > Fortran version is also available, and both exist in PostScript and
PDF.
> > http://www.ulib.org/webRoot/Books/Numerical_Recipes/
>
> Um, not to put too fine a point on it: That is not a book I'd
> recommend, and particularly not for the statistics sections. It's a
> reasonable introductory text to numerical analysis subjects, but
> according to people who really know the stuff, it is not to be trusted
> in the details (there's an online reference to the criticism, but I
> can't remember it offhand), and their software licencing is.... (don't
> get me started!)
>
> --
>    O__  ---- Peter Dalgaard             Blegdamsvej 3
>   c/ /'_ --- Dept. of Biostatistics     2200 Cph. N
>  (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
> ~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907

OK, I stand corrected. What I wrote was not from the point of view of a
reviewer who has thoroughly researched the field. I am no specialist in
statistics and write with no authority. However, out of the limited
amount of material that I did find on the web, this book did provide me
with a lot of useful formulae for various distributions, t-tests,
F-tests, chi^2, regression ... and all the things that are probably
taken
for granted by those whose daily lives are dedicated to Statistics.

If any of you know of better on-line resources, then please spare a
minute and enlighten us.

Clive Jenkins.

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

------------------------------

Date: Wed, 1 Dec 1999 15:17:14 +0100 (MET)
From: Kjetil Kjernsmo <kjetil.kjernsmo at astro.uio.no>
Subject: [R] Applying a function of several variables to data 

Dear all,

After looking at several packages I decided I would use R for my thesis
work (we've even got a site license for S-plus), so I'm learning.

I have run into a problem, and nobody on the house has a solution, so I'm
wondering if anybody out there can help. 
I've got two functions of four variables, one outputs a scalar, the other
an array, and I want to be able to give an array as a parameter for any
(or none) of the variables, and get a multidimensional array of
corresponding dimensions out, with all combinations. In the case of two
variables it is easy to do with outer(), and I have looked if apply() 
could do this, but found only examples where there is a function of one
variable. Obviously, I could do it with a for-loop at some level, but
after reading so many bad things about for-loops, it itches in my
estheticle thinking about it... :-) Is there a nicer way to do this?

Best,

Kjetil
- -- 
Kjetil Kjernsmo
Graduate astronomy-student                    Problems worthy of attack
University of Oslo, Norway            Prove their worth by hitting back
E-mail: kjetikj at astro.uio.no                                - Piet Hein
Homepage <URL:http://www.astro.uio.no/~kjetikj/>
Webmaster at skepsis.no 

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

------------------------------

Date: Wed, 1 Dec 1999 09:55:41 -0500
From: Duncan Temple Lang <duncan at rice.research.bell-labs.com>
Subject: Re: [R] R and XML -- a near perfect combination?

The great thing about data exchange is that at least two systems have
to be involved in the exchange (for it to be non-trivial!)  John
Chambers, myself and others have been discussing XML _joint_
integration into several projects - Omegahat, R and S.  

The interesting thing from the general perspective (rather than
a particular project and pair of interacting applications)
is in defining some DTDs that people are comfortable using.
This applies not only to data frames but also model specification,
results, etc.

>From the implementation perspective of reading XML, Omegahat has it
automatically. It would be nice to have the one shared by both R and S
and this was on my list of things to do.  Irrespective of the choice
of C/C++ parsing system, one approach I was thinking of in R is to use
a closure that is associated with a DTD. Thinking out aloud, the idea
is to have functions in the closure that correspond to the different
elements in the DTD.  As the parser discovers each element instance,
it calls the associated function in the closure (or a default one)
with the attribute lists and potentially the "identifier" for the
parent node in the resulting tree (although R and S aren't exactly
designed for trees).

(If this doesn't make any sense, it could be attributed to too little
sleep!)

D.



> Cc: r-help at stat.math.ethz.ch
> References: <Pine.LNX.4.10.9911301721080.3817-100000 at www.approximity.com>
> From: rossini at biostat.washington.edu (A.J. Rossini)
> Date: 30 Nov 1999 15:59:59 -0800
> Lines: 33
> Sender: owner-r-help at stat.math.ethz.ch
> Precedence: bulk
> 
> 
> >>>>> "c" == cys  <cys at www.approximity.com> writes:
> 
>     c> Did anybody alreay write a XML parser for R?  XML, as we will
>     c> have tons of data-interchange with all sorts of other programs
>     c> and XML is good for giving meaning to raw data.
> 
>     c> Any pointers/comments would be highly appreciated.
> 
> It's a nice format, if you know what you are doing.  The main thought
> that I've been having for what you are proposing (data exchange of
> datasets) would be to write an converter from your XML format to a
> text representation of the corresponding data.frame.  
> 
> Reasonably simple, plus you are free to use whatever your choice of
> parser language is (C++, Java, Python, whatever).  Plus, you can grow
> it (a simple list is easy, adding row/col names isn't too hard,
> etc...  Do it using pipes, and you will be fine for Unix and NT.
> 
> The only problem with a generic parser is the necessity of doing XML
> to XML conversion, since you can't be sure that everyone wants to use
> the DTD (or style) that you particularly like.
> 
> best,
> -tony
> 
> -- 
> A.J. Rossini			Research Assistant Professor of Biostatistics 
> Center for AIDS Research/HMC	Biostatistics/Univ. of Washington
> Box 359931			Box 357232
> 206-731-3647 (3693=fax)		206-543-1044 (3286=fax)
> rossini at u.washington.edu	rossini at biostat.washington.edu
> http://www.biostat.washington.edu/~rossini/
> 
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> 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
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
> 


- -- 
_______________________________________________________________

Duncan Temple Lang                duncan at research.bell-labs.com
Bell Labs, Lucent Technologies    office: (908)582-3217
700 Mountain Avenue, Room 2C-259  fax:    (908)582-3340
Murray Hill, NJ  07974-2070       
         http://cm.bell-labs.com/stat/duncan

      "Languages shape the way we think, and determine what 
       we can think about."        
                                      Benjamin Whorf
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Wed, 01 Dec 1999 17:09:17 +0100
From: mathieu <mathieu.ros at free.fr>
Subject: [R] problem installing R 0.90

- --------------0959452E704C9F02EEB4171F
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

hello all,
when running ./configure in my [new] R-0.90 directory, the process stop
at the following step :

(...)
checking whether g77 and gcc agree on int and double...

can anybody tell me what's the problem (I believe it is due to g77 but
what can I do?)
thanks,
    Mathieu

I'm using Redhat 6.0 on a  Intel P90

- --
- ----------------------------------------------------------------------
 Mathieu Ros - 13 rue bévière - 38000 GRENOBLE - 04 76 491 370
 http://www.multimania.com/mathieuros/index2.html
 DESS ingénierie mathématique (biostatistiques)
 Universite Joseph Fourier, Grenoble
- ----------------------------------------------------------------------
La science détient la vérité, ne vous laissez pas avoir par les faits.



- --------------0959452E704C9F02EEB4171F
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
hello all,
<br>when running ./configure in my [new] R-0.90 directory, the process
stop at the following step :
<p>(...)
<br>checking whether g77 and gcc agree on int and double...
<p>can anybody tell me what's the problem (I believe it is due to g77 but
what can I do?)
<br>thanks,
<br>&nbsp;&nbsp;&nbsp; Mathieu
<p>I'm using Redhat 6.0 on a&nbsp; Intel P90
<pre>--&nbsp;
- ----------------------------------------------------------------------
&nbsp;Mathieu Ros - 13 rue b&eacute;vi&egrave;re - 38000 GRENOBLE - 04 76 491 370
&nbsp;<A HREF="http://www.multimania.com/mathieuros/index2.html">http://www.multimania.com/mathieuros/index2.html</A>
&nbsp;DESS ing&eacute;nierie math&eacute;matique (biostatistiques)
&nbsp;Universite Joseph Fourier, Grenoble
- ----------------------------------------------------------------------
La science d&eacute;tient la v&eacute;rit&eacute;, ne vous laissez pas avoir par les faits.</pre>
&nbsp;</html>

- --------------0959452E704C9F02EEB4171F--

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

------------------------------

Date: Wed, 1 Dec 1999 17:57:25 +0100 (MET)
From: Martin Maechler <maechler at stat.math.ethz.ch>
Subject: Re: [R] graphics

 >>>>> Peter Dalgaard BSA <p.dalgaard at biostat.ku.dk> adds:

    PD> Jim Lindsey <jlindsey at alpha.luc.ac.be> writes:
    >> I have been receiving complaints that my libraries no longer give
    >> reasonable graphics with R0.90. I have been replying that I have not
    >> changed anything. Among my travels (presently in Umea), I have
    >> finally had a chance to take a look. The best way to describe the
    >> current default graphics is HORRIBLY UGLY. The R core member who
    >> made those changes should buy glasses to see the screen better
    >> instead of imposing his near sightedness on the rest of us. The
    >> worst is that the most important options that have been modified are
    >> RO so we can do nothing. As PS graphics are also knackered, we
    >> currently have no way of viewing a reasonable graph produced by R.

    PD> Well, the font sizes are as I have had them for ages: 12pt on a
    PD> 100dpi assumed resolution. If that's too large, fine, complain and
    PD> if you gain a sufficient following we might reduce it to 10pt. The
    PD> old code would give you 100dpi or 75dpi fonts depending on which
    PD> one came first on your font path and was available at the desired
    PD> size, leading to general madness. If it bothers you enough, use
    PD> fix(X11) and change the default for the pointsize argument.

After a quite a bit discussion (incl. some own thought), I tend to vote
with Jim Lindsey: 
We *should* reduce the default font size:  Because the font size is 
relatively large, the margins (measured in characters via "mar") become
large as well, and (as Jim pointed out to some of us)
now use more than 50% of the device surface.

Fontsize = 10 (pt) looks more reasonable, to me, too.
(However, I propose to make it an option, see below).

    PD> The postscript is truly bug-ridden, but the latest snapshot should
    PD> be much better.

    >> Is there any way to set the default par options, say in Rprofile,
    >> without having a graphics window pop up at startup?

    PD> Nope, but we should probably have something like ps.options there,
    PD> the par() effect is also messing up example(): the code cannot set
    PD> ask=TRUE so multiple graphics in an example just fly by, unless the
    PD> user sets the parameter.
yes, definitely a reasonable feature request! -- see below

BUT, for a first "quick fix" : Why not change X11's argument
     pointsize = 12
to
     pointsize = .Options$X11pointsize
(we already have X11colortype !).

Now back to the feature request:

Do we want  par.defaults() for this?
It would have the same arguments as par()  plus additional
   device = .Options$device
with further possibility   
   device = "all"
i.e., by default,  par.defaults()  would set things only for the default
device, but there'd be a possibility to set these for all devices.

Note that
- -  This would lead to considerable side effects.
   [Starting a device and setting par()s would not be portable,
    for complete portability you'd have to first set (or reset) the
    par.defaults()]
- -  for  device = "postscript", 
   we could allow further arguments as those of ps.options() and postscript()
   and e.g. for device = "X11" ( = default on non-gnome Unix),
   one should allow arguments of X11/x11() such as pointsize, colortype, ...

- -----

When feature requesting: What I have wanted more than once is an
     par(reset = TRUE)
which would set all the par()s back to how they were when the device was
opened.   

     par.defaults(reset = TRUE)

would analogously remove all `site/user specific' par defaults
(effect: as if  par.defaults() was never set).
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Wed, 1 Dec 1999 18:39:54 +0100 (MET)
From: Jim Lindsey <jlindsey at alpha.luc.ac.be>
Subject: Re: [R] graphics

> Do we want  par.defaults() for this?
> It would have the same arguments as par()  plus additional
>    device = .Options$device
> with further possibility   
>    device = "all"
> i.e., by default,  par.defaults()  would set things only for the default
> device, but there'd be a possibility to set these for all devices.
> 
> Note that
> -  This would lead to considerable side effects.
>    [Starting a device and setting par()s would not be portable,
>     for complete portability you'd have to first set (or reset) the
>     par.defaults()]
> -  for  device = "postscript", 
>    we could allow further arguments as those of ps.options() and postscript()
>    and e.g. for device = "X11" ( = default on non-gnome Unix),
>    one should allow arguments of X11/x11() such as pointsize, colortype, ...

Is there a strong reason why a call to par with no device active must
open a window? If this 'feature' can easily be removed, then I would
be satisfied as I could then put my default par settings in Rprofile
without a window opening at startup. Of course, the above is a far
better idea but perhaps more difficult to implement.

(By the way, for those on the list who have not had access to all the
discussion about this in the last couple of days, Peter has pointed
out to me that X11 - note capital X - can be redefined in Rprofile so
that pointsize has any desired size. This is now working very well for
me.)

> 
> -----
> 
> When feature requesting: What I have wanted more than once is an
>      par(reset = TRUE)
> which would set all the par()s back to how they were when the device was
> opened.   
> 
>      par.defaults(reset = TRUE)
> 
> would analogously remove all `site/user specific' par defaults
> (effect: as if  par.defaults() was never set).

I'll certainly vote for those.
  Jim
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: 01 Dec 1999 19:28:24 +0100
From: Peter Dalgaard BSA <p.dalgaard at biostat.ku.dk>
Subject: Re: [R] graphics

Martin Maechler <maechler at stat.math.ethz.ch> writes:

> BUT, for a first "quick fix" : Why not change X11's argument
>      pointsize = 12
> to
>      pointsize = .Options$X11pointsize
> (we already have X11colortype !).

Which is a mess! We also have

.PostScript.Options$pointsize

So wouldn't it be more obvious to use

.X11.Options$pointsize

(and set it with X11.options(pointsize=12))?

If we go to 10pt default, we also need to discuss the multiframe
scaling issue. The current rules are that if there is two rows and/or
columns, reduce font size to 80%, if there are three or more reduce to
50% and a 5pt font @100dpi is a tad small. Note that this rule is
*device independent*, so one would have to check that it still looks
reasonable on postscript and windows. I think it works out OK if one sets the
rule to 60% giving a 6pt font, but I'd like someone elses eyes on
that. 

> Now back to the feature request:
> 
> Do we want  par.defaults() for this?
> It would have the same arguments as par()  plus additional
>    device = .Options$device
> with further possibility   
>    device = "all"
> i.e., by default,  par.defaults()  would set things only for the default
> device, but there'd be a possibility to set these for all devices.
> 
> Note that
> -  This would lead to considerable side effects.
>    [Starting a device and setting par()s would not be portable,
>     for complete portability you'd have to first set (or reset) the
>     par.defaults()]
> -  for  device = "postscript", 
>    we could allow further arguments as those of ps.options() and postscript()
>    and e.g. for device = "X11" ( = default on non-gnome Unix),
>    one should allow arguments of X11/x11() such as pointsize, colortype, ...
> 
> -----
> 
> When feature requesting: What I have wanted more than once is an
>      par(reset = TRUE)
> which would set all the par()s back to how they were when the device was
> opened.   
> 
>      par.defaults(reset = TRUE)
> 
> would analogously remove all `site/user specific' par defaults
> (effect: as if  par.defaults() was never set).

Hmm. I'm not so sure we're really gaining much from an elaborate
par.default() function. If one desires, one could do it via the
startup function, simply by a call to par based on the options).

For instance, try changing X11 to

X11<-function (display = "", width = 7, height = 7, pointsize = 12, 
    gamma = 1, colortype = options()$X11colortype, maxcubesize = 256) 
{
    .Internal(X11(display, width, height, pointsize, gamma, colortype, 
    maxcubesize))
    par(ask=T)
}

and then

example(barplot)
dev.off()
par(ask=F)
example(barplot)

- -- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: 01 Dec 1999 19:41:07 +0100
From: Peter Dalgaard BSA <p.dalgaard at biostat.ku.dk>
Subject: Re: [R] graphics

Jim Lindsey <jlindsey at alpha.luc.ac.be> writes:

> Is there a strong reason why a call to par with no device active must
> open a window? If this 'feature' can easily be removed, then I would
> be satisfied as I could then put my default par settings in Rprofile
> without a window opening at startup. Of course, the above is a far
> better idea but perhaps more difficult to implement.

I'm sure it can be worked around, but the issue is that the par()
settings must go to a device, so if none is open, we open the default
one, by the default method, which opens the window. The obvious fix,
not to open the window until something is actually drawn would have
the consequence of making a direct X11() call an apparent no-op... And
some of the things that par() can do involves requests to the window
system about the plotting window, I suspect.

- -- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 2 Dec 1999 08:36:27 +1300
From: Lyndon Drake <lyndon at stat.auckland.ac.nz>
Subject: Re: [R] R and XML -- a near perfect combination?

On Tue, Nov 30, 1999 at 05:25:17PM -0800, cys at www.approximity.com wrote:
> Did anybody alreay write a XML parser for R?
> XML, as we will have tons of data-interchange with
> all sorts of other programs and XML is good for giving
> meaning to raw data.

A good C XML library is libxml (wierd name, I know :-), at
http://rpmfind.net/veillard/XML/.  It's the one used in GNOME, but it is
supposed to be portable.  I don't know how well it would work on Win32,
but it would be worth looking at.  It is licensed under the LGPL.

- -- 
Lyndon Drake                       | Desktop:      http://www.gnome.org
isenguard                          | Mail client:  http://www.mutt.org
ICQ#: 12558803                     | Editor:       http://www.vim.org
http://stat.auckland.ac.nz/~lyndon | OS:           http://www.linux.com
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 2 Dec 1999 08:47:36 +1300
From: Ross Ihaka <ihaka at stat.auckland.ac.nz>
Subject: Re: [R] graphics

I think I mentioned this a while ago ...

The problem I have with the new X11 driver is that it appears to ignore
the actual DPI (dots per inch) of the X server and use 100 DPI.  My machine
is most definitely 75 DPI (I explicitly set it when I start the server).

The fonts I get from R are much too large and do not agree with those
displayed in xfontsel (which I think shows the official story).

	Ross
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 02 Dec 1999 08:51:47 +1100
From: John Maindonald <john.maindonald at anu.edu.au>
Subject: [R] Plot Character Sizes in rw-0.90.0

I have two supposedly identical installations, one under Windows 95,
and the other (at home) under Windows 98.  The puzzle is that under 
my Windows 98 installation, symbols plotted by default or with pch
come out huge, at about 2.2 times the height of the axis labels.
Setting cex=0.5 seems about what is needed to fix the heights.

On the Windows 95 installation, the symbols come out just a little
smaller than the height of the axis labels.

Unless someone has an immediate explanation, maybe I need to send a
bug report from my home machine?

A further point is that, contrary to the documentation of par(),
the
setting of mkh seems to have no effect on the height of symbols 
plotted with pch, e. g.
plot(1:4,1:4,pch=1,mkh=0.2)


John Maindonald               email : john.maindonald at anu.edu.au        
Statistical Consulting Unit,  phone : (6249)3998        
c/o CMA, SMS,                 fax   : (6249)5549  
John Dedman Mathematical Sciences Building
Australian National University
Canberra ACT 0200
Australia

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

------------------------------

Date: Wed, 1 Dec 1999 17:01:33 -0500 
From: "Carel, Roland" <rcarel at netgenics.com>
Subject: RE: [R] R and XML -- a near perfect combination?

>On Tue, Nov 30, 1999 at 05:25:17PM -0800, cys at www.approximity.com wrote:
>> Did anybody alreay write a XML parser for R?
>> XML, as we will have tons of data-interchange with
>> all sorts of other programs and XML is good for giving
>> meaning to raw data.

>A good C XML library is libxml (wierd name, I know :-), at
>http://rpmfind.net/veillard/XML/.  It's the one used in GNOME, but it is
>supposed to be portable.  I don't know how well it would work on Win32,
>but it would be worth looking at.  It is licensed under the LGPL.

Why not use the java parser from the IBM Alpha Works:
http://www.alphaworks.ibm.com/tech/dynamicxmlforjava
It should be easily portable to just about any platform.

- -Roland Carel
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: 01 Dec 1999 14:18:24 -0800
From: rossini at biostat.washington.edu (A.J. Rossini)
Subject: Re: [R] R and XML -- a near perfect combination?

>>>>> "CR" == Carel, Roland <rcarel at netgenics.com> writes:

    >> A good C XML library is libxml (wierd name, I know :-), at
    >> http://rpmfind.net/veillard/XML/.  It's the one used in GNOME,
    >> but it is supposed to be portable.  I don't know how well it
    >> would work on Win32, but it would be worth looking at.  It is
    >> licensed under the LGPL.

    CR> Why not use the java parser from the IBM Alpha Works:
    CR> http://www.alphaworks.ibm.com/tech/dynamicxmlforjava It should
    CR> be easily portable to just about any platform.

Would be fine with Omegahat, but probably not R.  Actually, maybe you
could get to it via RJava, but it would probably require a bit more
thought.  There might be licensing issues wrt to R's GPL, though,
depending which IBM license it's under...

best,
- -tony

- -- 
A.J. Rossini			Research Assistant Professor of Biostatistics 
Center for AIDS Research/HMC	Biostatistics/Univ. of Washington
Box 359931			Box 357232
206-731-3647 (3693=fax)		206-543-1044 (3286=fax)
rossini at u.washington.edu	rossini at biostat.washington.edu
http://www.biostat.washington.edu/~rossini/

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

------------------------------

Date: 01 Dec 1999 23:39:18 +0100
From: Peter Dalgaard BSA <p.dalgaard at biostat.ku.dk>
Subject: Re: [R] Plot Character Sizes in rw-0.90.0

John Maindonald <john.maindonald at anu.edu.au> writes:

> I have two supposedly identical installations, one under Windows 95,
> and the other (at home) under Windows 98.  The puzzle is that under 
> my Windows 98 installation, symbols plotted by default or with pch
> come out huge, at about 2.2 times the height of the axis labels.
> Setting cex=0.5 seems about what is needed to fix the heights.
> 
> On the Windows 95 installation, the symbols come out just a little
> smaller than the height of the axis labels.

Peculiar. They're supposed to be roughly the width of a "0" in the
base font. The Postscript driver had a bug in the size calculations,
in that it subtracted 18 pixels from the radius, but that wouldn't
seem to be your problem if setting cex fixes things up.

What does par("cra") and par("csi") say on your respective systems?
(par("cra")[2]/par("csi") should be the dpi of your device)

> Unless someone has an immediate explanation, maybe I need to send a
> bug report from my home machine?
> 
> A further point is that, contrary to the documentation of par(),
> the
> setting of mkh seems to have no effect on the height of symbols 
> plotted with pch, e. g.
> plot(1:4,1:4,pch=1,mkh=0.2)

Yes, I believe we already have a bug report to that effect.

- -- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Wed, 1 Dec 1999 22:46:14 +0000 (GMT)
From: Prof Brian D Ripley <ripley at stats.ox.ac.uk>
Subject: Re: [R] R and XML -- a near perfect combination?

On 1 Dec 1999, A.J. Rossini wrote:

> 
> >>>>> "CR" == Carel, Roland <rcarel at netgenics.com> writes:
> 
>     >> A good C XML library is libxml (wierd name, I know :-), at
>     >> http://rpmfind.net/veillard/XML/.  It's the one used in GNOME,
>     >> but it is supposed to be portable.  I don't know how well it
>     >> would work on Win32, but it would be worth looking at.  It is
>     >> licensed under the LGPL.
> 
>     CR> Why not use the java parser from the IBM Alpha Works:
>     CR> http://www.alphaworks.ibm.com/tech/dynamicxmlforjava It should
>     CR> be easily portable to just about any platform.
> 
> Would be fine with Omegahat, but probably not R.  Actually, maybe you
> could get to it via RJava, but it would probably require a bit more
> thought.  There might be licensing issues wrt to R's GPL, though,
> depending which IBM license it's under...

Many users of R do not run Java; for example no one porting R to Windows
does on their Windows machines AFAIK. We really do need something that
compiles on standard C compilers with minimal baggage, as total download
size is a significant issue.

- -- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

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

------------------------------

Date: Wed, 1 Dec 1999 19:37:28 -0500
From: "Yves Gauvreau" <cyg at sympatico.ca>
Subject: [R] Meaning?

Hi,

Sorry to ask this but what is the meaning of "AFAIK". From a darn Frenchman!

Regards.

Yves

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

------------------------------

Date: Thu, 02 Dec 1999 13:39:10 +1300
From: Murray Jorgensen <maj at waikato.ac.nz>
Subject: Re: [R] Installing R on Slackware Linux

I'd like to thank Markus Jantti, John Maindonald, Peter Malewski, Mathew
Wiener and Jonathan Yuen for their help. As most pointed out, the .INSTALL
file is quite good. Just the same, for a linux novice like me it did help
to have much the same thing said in several different ways.

I still don't know what to do with the old 0.65.1 directory tree. [ It was
a minor triumph of mine when I figured out how to use the "find" command
properly and by searching for config.status, which one of the replies let
me know was a file in the tree, actually succeeded in finding the R0.65.1
directory tree.]

Now what do I do?

* Delete the old directory tree before installing the new version?
  (and risk ending up with no R at all)
* Leave the old directory tree and install the new version?
   (and possibly still get the old version when I type "R")

Or maybe uninstalling is more complicated than deleting the tree - it
certainly is in Windows!

I truly do hope that there are lurking newbies out there that make my
asking such basic questions worthwhile!


Murray Jorgensen,  Department of Statistics,  U of Waikato, Hamilton, NZ
- -----[+64-7-838-4773]---------------------------[maj at waikato.ac.nz]-----
"Doubt everything or believe everything:these are two equally convenient
strategies. With either we dispense with the need to think."
http://www.cs.waikato.ac.nz/stats/Staff/maj.html       - Henri Poincare'

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

------------------------------

Date: 01 Dec 1999 21:15:42 -0600
From: Douglas Bates <bates at stat.wisc.edu>
Subject: Re: [R] Meaning?

"Yves Gauvreau" <cyg at sympatico.ca> writes:

> Sorry to ask this but what is the meaning of "AFAIK". From a darn Frenchman!

AFAIK => As far as I know

It is one of those acronyms that is sometimes adopted in e-mail to save typing.
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 2 Dec 1999 08:53:06 +0100 (CET)
From: Kurt Hornik <Kurt.Hornik at ci.tuwien.ac.at>
Subject: Re: [R] graphics

>>>>> Peter Dalgaard BSA writes:

> Martin Maechler <maechler at stat.math.ethz.ch> writes:
>> BUT, for a first "quick fix" : Why not change X11's argument
>> pointsize = 12
>> to
>> pointsize = .Options$X11pointsize
>> (we already have X11colortype !).

> Which is a mess! We also have

> .PostScript.Options$pointsize

> So wouldn't it be more obvious to use

> .X11.Options$pointsize

> (and set it with X11.options(pointsize=12))?

Why would we need a separate options variable anyway?  We might as well
do .Options$PostScript.foo instead ... or .Options$PostScript$foo ...

If we do the above we might also want an X11.options() in analogy to the
ps.options() which already exists, and perhaps also pictex.options()
etc.

???

> If we go to 10pt default, we also need to discuss the multiframe
> scaling issue. The current rules are that if there is two rows and/or
> columns, reduce font size to 80%, if there are three or more reduce to
> 50% and a 5pt font @100dpi is a tad small. Note that this rule is
> *device independent*, so one would have to check that it still looks
> reasonable on postscript and windows. I think it works out OK if one
> sets the rule to 60% giving a 6pt font, but I'd like someone elses
> eyes on that.

>> Now back to the feature request:
>> 
>> Do we want  par.defaults() for this?
>> It would have the same arguments as par()  plus additional
>> device = .Options$device
>> with further possibility   
>> device = "all"
>> i.e., by default,  par.defaults()  would set things only for the default
>> device, but there'd be a possibility to set these for all devices.
>> 
>> Note that
>> -  This would lead to considerable side effects.
>> [Starting a device and setting par()s would not be portable,
>> for complete portability you'd have to first set (or reset) the
>> par.defaults()]
>> -  for  device = "postscript", 
>> we could allow further arguments as those of ps.options() and postscript()
>> and e.g. for device = "X11" ( = default on non-gnome Unix),
>> one should allow arguments of X11/x11() such as pointsize, colortype, ...
>> 
>> -----
>> 
>> When feature requesting: What I have wanted more than once is an
>> par(reset = TRUE)
>> which would set all the par()s back to how they were when the device was
>> opened.   
>> 
>> par.defaults(reset = TRUE)
>> 
>> would analogously remove all `site/user specific' par defaults
>> (effect: as if  par.defaults() was never set).

> Hmm. I'm not so sure we're really gaining much from an elaborate
> par.default() function. If one desires, one could do it via the
> startup function, simply by a call to par based on the options).

> For instance, try changing X11 to

> X11<-function (display = "", width = 7, height = 7, pointsize = 12, 
>     gamma = 1, colortype = options()$X11colortype, maxcubesize = 256) 
> {
>     .Internal(X11(display, width, height, pointsize, gamma, colortype, 
>     maxcubesize))
>     par(ask=T)
> }

> and then

> example(barplot)
> dev.off()
> par(ask=F)
> example(barplot)

Agreed.

- -k
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 2 Dec 1999 09:38:30 +0100 (MET)
From: Martin Maechler <maechler at stat.math.ethz.ch>
Subject: par(mkh) {was [R] Plot Character Sizes in rw-0.90.0}

    PD> John Maindonald <john.maindonald at anu.edu.au> writes:

      JM>      <....>

      JM> A further point is that, contrary to the documentation of par(),
      JM> the setting of mkh seems to have no effect on the height of symbols
      JM> plotted with pch, e. g.  plot(1:4,1:4,pch=1,mkh=0.2)

    PD> Yes, I believe we already have a bug report to that effect.

True; question is : What should be fixed?  Documentation or implementation?

Here the Status:

   Contrary to S, in R, we have always been using  par("cex") and the
   <highlevel plot> cex argument  *both* for text and all graphical symbols.
   S uses mkh for  pch=<integer>   symbols and
	  cex for  pch=<character> ones and text
   which is really not what we (R&R, initially) wanted to copy.

   In R, mkh settings have always been completely ignored.  
   In spite of this, par() (and the internal code in  par.c & graphics.c) have
   been set up to allow setting and querying mkh.

Now, there are several possibilities of which I'd advocate mainly 1) or 2) 
[I'm talking about implementation; documentation must be updated in any case]:

1) Implementation unchanged + using  mkh gives a warning saying 
   that mkh is practically ignored.
   
2) setting par(mkh) or  <highlevelplot>(* , mkh = ..)
    
    gives a warning (``mkh : setting cex'')
    and then behaves as if "cex" was used.

3a) Even more S compatibility:
   Setting par(mkh) or  <highlevelplot>(* , mkh = ..)
   changes "cex" accordingly *IF* pch=<numeric> .

3b) Similar but more logical; need some "notation" : 
    The symbol size used in the core function plot.xy()  is
           cexbase * Pcex   (--> plot.c, l.1089)

    where "cexbase" is par("cex") and  Pcex is the "cex" argument of 
    plot(),points(),...

    Now we could change this to
           cexbase * mkh * Pcex
    for the case of numeric pch (internally pch <= 31 or something) where
    mkh is *either* par("mkh") or the mkh arg. of plot.xy().


4) complete S compatibility (cex *not* working for pch=<numeric>)

   This is completely out of question for me!


Martin Maechler <maechler at stat.math.ethz.ch>	http://stat.ethz.ch/~maechler/
Seminar fuer Statistik, ETH-Zentrum  LEO D10	Leonhardstr. 27
ETH (Federal Inst. Technology)	8092 Zurich	SWITZERLAND
phone: x-41-1-632-3408		fax: ...-1228			<><
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 2 Dec 1999 09:48:27 +0000 (GMT)
From: Bill Simpson <wsi at gcal.ac.uk>
Subject: Re: [R] graphics

As I understand it, all this discussion is referring to how the graphics
look on the screen. Maybe it is pointless to mention the following, but
here I go:
- - what is on the screen (X11) should agree with what is printed
(dev.print)
- - the default sizes of letters and symbols are too small in
ouput produced by dev.print. Should be double the size. I always use
cex=2, but numerous things get buggered up. In my opinion, the current
behaviour is a bad legacy from S. Default S plots use labels and symbols
that are too small. Remember that plots submitted to journals
and book publishers always get scaled down, so in the original version
they have to be quite large. Take a look at V&R, pp16-17 for plots whose
axis labels are smaller than surrounding main text.

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 at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 2 Dec 1999 20:23:20 +1100 (EST)
From: John Maindonald <john.maindonald at anu.edu.au>
Subject: [R] Large plot symbols (with pch = 1 etc)

On my home (Windows 98) machine I find:
> par()$cra
[1] 42 16
> par()$csi
[1] 0.1666667
> 

On my work machine I believe I had
par()$cra
[1] 17 16

So it looks as though R is reading the width wrongly,
and that it is the width that is somehow used to
determine the plot character size.

Incidentally I noted Brian's comment that R had misread
his character dimension information wrongly.  I have a
Matrox MM2 board on my Windows 98 (SE) machine.  
I understand he will be using a Matrox G200.  Is there a 
Matrox connection?

Regards

John Maindonald               email : john.maindonald at anu.edu.au        
Statistical Consulting Unit,  phone : (6249)3998        
c/o CMA, SMS,                 fax   : (6249)5549  
John Dedman Mathematical Sciences Building
Australian National University
Canberra ACT 0200
Australia
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 2 Dec 1999 09:39:55 +0000 (GMT)
From: Prof Brian D Ripley <ripley at stats.ox.ac.uk>
Subject: Re: [R] graphics

On Thu, 2 Dec 1999, Bill Simpson wrote:

> they have to be quite large. Take a look at V&R, pp16-17 for plots whose
> axis labels are smaller than surrounding main text.

There are no plots on those pages! Presumably you have an old edition.

It is quite common in journals to have minor labels smaller than the main
text: after all the footnotes and index are smaller than the main text in
our book. Those plots have been checked and approved (three times) by
professional copy editors.

What you get with dev.print is entirely up to you: it depends on how you
set pointsize and in particular if you like looking at plots with your head
turned through 90^o. (I get very frustrated with students who produce
landscape plots that way: the default pointsize is relative to a much
smaller plot if it is portrait not the default (for some reason that has
always escaped me) landscape.)

- -- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

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

------------------------------

Date: Thu, 2 Dec 1999 11:03:09 +0000 (GMT)
From: Bill Simpson <wsi at gcal.ac.uk>
Subject: Re: [R] graphics

On Thu, 2 Dec 1999, Prof Brian D Ripley wrote:

> On Thu, 2 Dec 1999, Bill Simpson wrote:
> 
> > they have to be quite large. Take a look at V&R, pp16-17 for plots whose
> > axis labels are smaller than surrounding main text.
> 
> There are no plots on those pages! Presumably you have an old edition.
2nd edition, figs 1.6 and 1.7

> It is quite common in journals to have minor labels smaller than the main
> text: after all the footnotes and index are smaller than the main text in
> our book. Those plots have been checked and approved (three times) by
> professional copy editors.
I know it amounts to taste. My taste is: figure labels should be larger
than surrounding text.

> What you get with dev.print is entirely up to you: it depends on how you
> set pointsize and in particular if you like looking at plots with your head
> turned through 90^o.
I tried setting pointsize in the past without success. Maybe it works now
and I should try again. In the past the only solution was cex=2

> (I get very frustrated with students who produce
> landscape plots that way: the default pointsize is relative to a much
> smaller plot if it is portrait not the default (for some reason that has
> always escaped me) landscape.)
I agree that portrait should be the default. I always set
horizontal=FALSE.

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 at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 2 Dec 1999 10:33:11 +0000 (GMT)
From: Prof Brian Ripley <ripley at stats.ox.ac.uk>
Subject: Re: [R] Large plot symbols (with pch = 1 etc)

> Date: Thu, 2 Dec 1999 20:23:20 +1100 (EST)
> From: John Maindonald <john.maindonald at anu.edu.au>
> 
> On my home (Windows 98) machine I find:
> > par()$cra
> [1] 42 16
> > par()$csi
> [1] 0.1666667
> > 
> 
> On my work machine I believe I had
> par()$cra
> [1] 17 16
> 
> So it looks as though R is reading the width wrongly,
> and that it is the width that is somehow used to
> determine the plot character size.
> 
> Incidentally I noted Brian's comment that R had misread
> his character dimension information wrongly.  I have a
> Matrox MM2 board on my Windows 98 (SE) machine.  
> I understand he will be using a Matrox G200.  Is there a 
> Matrox connection?

It was my laptop (our G200's are primarily Linux), and it is a NeoMagic
256. The problem was with Windows, not R: R acts on what it is told by
Windows, and that was nonsense.  I am fairly sure that problem resides
in the graphics drivers, but we need to work around it.  Not long
before the release of 0.90.0 he and I were getting very different
font and symbol sizes on our laptops from identical code.

As far as I can see in the Nov 19th change Guido changed the cra
setting to come from the character metrics.  That is, cra[1] is set to
what Windows reports as the maximum width of a character in the font.
It looks as if that is unreliable.  I think we should ask it for "O".
Incidentally, I am beginning to learn that Window 98SE is not just
a minor upgrade, so we are going to have to test on that too.

Brian

- -- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

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

------------------------------

Date: 02 Dec 1999 12:44:06 +0100
From: Peter Dalgaard BSA <p.dalgaard at biostat.ku.dk>
Subject: Re: [R] Installing R on Slackware Linux

Murray Jorgensen <maj at waikato.ac.nz> writes:

> I'd like to thank Markus Jantti, John Maindonald, Peter Malewski, Mathew
> Wiener and Jonathan Yuen for their help. As most pointed out, the .INSTALL
> file is quite good. Just the same, for a linux novice like me it did help
> to have much the same thing said in several different ways.
> 
> I still don't know what to do with the old 0.65.1 directory tree. [ It was
> a minor triumph of mine when I figured out how to use the "find" command
> properly and by searching for config.status, which one of the replies let
> me know was a file in the tree, actually succeeded in finding the R0.65.1
> directory tree.]
> 
> Now what do I do?
> 
> * Delete the old directory tree before installing the new version?
>   (and risk ending up with no R at all)
> * Leave the old directory tree and install the new version?
>    (and possibly still get the old version when I type "R")
> 
> Or maybe uninstalling is more complicated than deleting the tree - it
> certainly is in Windows!
> 
> I truly do hope that there are lurking newbies out there that make my
> asking such basic questions worthwhile!

Well, it will unpack in a separate directory and you can build and
test the new version there, it is only when you enter "make install"
that you'll be overwriting files in system locations. The R-0.65.1
directory can be kept if you wish.

Normally, no harm comes from installing on top of a previous
installation, but if you want to be sure delete /usr/local/bin/R
and  /usr/local/bin/*Rd* (hoping that nothing else matches the
wildcards...) and "rm -rf /usr/local/lib/R" befor installing.

If something goes haywire, it should still be possible to go to
R-0.65.1, type "make install" and have the old version reinstalled.

All assuming that you used "make install" with the standard locations
the last time round.

- -- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: 02 Dec 1999 14:48:33 +0100
From: Peter Dalgaard BSA <p.dalgaard at biostat.ku.dk>
Subject: Re: [R] graphics

Bill Simpson <wsi at gcal.ac.uk> writes:

> As I understand it, all this discussion is referring to how the graphics
> look on the screen. Maybe it is pointless to mention the following, but
> here I go:
> - what is on the screen (X11) should agree with what is printed
> (dev.print)

This is a bit problematic. I'd say that it should be readable before
anything else. If one wants exact preview, use ghostview & friends.
Also for a pleasant working environment (at least with X11), I like to
use unscaled fonts as far as possible.

> - the default sizes of letters and symbols are too small in
> ouput produced by dev.print. Should be double the size. I always use
> cex=2, but numerous things get buggered up. In my opinion, the current
> behaviour is a bad legacy from S. Default S plots use labels and symbols
> that are too small. Remember that plots submitted to journals
> and book publishers always get scaled down, so in the original version
> they have to be quite large. Take a look at V&R, pp16-17 for plots whose
> axis labels are smaller than surrounding main text.

(As Brian pointed out, pointsize is better than cex for this purpose.
Pardon the pun...)

I do agree that the S defaults are a bit small, but it all depends on
what they are to be used for: 

 - as an intermediate tool in model development &c
 - full-page illustration in book/journal
 - half-page, in-text illustration
 - slide

(with increasing text-to-plot ratios). And it also depends on personal
tastes - e.g. Jim Lindsey pretty squarely want to maximize the plot
area, i.e. minimum legible fonts, if I interpret him correctly.

I think I'm with Brian in saying that the plots in V&R are certainly
not seriously bad, and there's not really any cause for assuming that
the sizes of text on a plot are directly related to those in the
running text (unless one has been foolish enough to use the same
font!) They're readable enough and the whole thing is in reasonable
balance. The labels are a bit too close to the axes and tick marks
though....

Incidentally, I reached out for the nearest stats book from the pre-S
era (Breslow&Day, 1980) and found some pretty ugly clashes between
plots and the surrounding text. It isn't easy...

- -- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 2 Dec 1999 16:36:16 +0100
From: "Karl M. Syring" <syring at email.com>
Subject: RE: [R] Meaning?

Yves,
there are ancronym finders (of course) on the web. Try
http://www.ucc.ie/cgi-bin/acronym?rotfl

Regards
Karl


> -----Original Message-----
> From: owner-r-help at stat.math.ethz.ch
> [mailto:owner-r-help at stat.math.ethz.ch]On Behalf Of Yves Gauvreau
> Sent: Donnerstag, 2. Dezember 1999 01:37
> To: r-help at stat.math.ethz.ch
> Subject: [R] Meaning?
> 
> 
> Hi,
> 
> Sorry to ask this but what is the meaning of "AFAIK". From a darn 
> Frenchman!
> 
> Regards.
> 
> Yves
> 
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
> -.-.-.-.-.-.-
> 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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: 02 Dec 1999 17:08:06 +0100
From: Peter Dalgaard BSA <p.dalgaard at biostat.ku.dk>
Subject: Re: par(mkh) {was [R] Plot Character Sizes in rw-0.90.0}

Martin Maechler <maechler at stat.math.ethz.ch> writes:

> 1) Implementation unchanged + using  mkh gives a warning saying 
>    that mkh is practically ignored.
>    
> 2) setting par(mkh) or  <highlevelplot>(* , mkh = ..)
>     
>     gives a warning (``mkh : setting cex'')
>     and then behaves as if "cex" was used.
> 
> 3a) Even more S compatibility:
>    Setting par(mkh) or  <highlevelplot>(* , mkh = ..)
>    changes "cex" accordingly *IF* pch=<numeric> .
> 
> 3b) Similar but more logical; need some "notation" : 
>     The symbol size used in the core function plot.xy()  is
>            cexbase * Pcex   (--> plot.c, l.1089)
> 
>     where "cexbase" is par("cex") and  Pcex is the "cex" argument of 
>     plot(),points(),...
> 
>     Now we could change this to
>            cexbase * mkh * Pcex
>     for the case of numeric pch (internally pch <= 31 or something) where
>     mkh is *either* par("mkh") or the mkh arg. of plot.xy().
> 
> 
> 4) complete S compatibility (cex *not* working for pch=<numeric>)
> 
>    This is completely out of question for me!

Um, as I read my S docs, cex works for numeric pch, but only if mkh=0
which is the default, and mkh works for pch=1 only. That mkh doesn't
work for non-numeric pch is a bit odd, and we could fix that if we
could decide on a reasonable definition of the height of such symbols
(probably, it should depend on the font, not the character). 

But is the real issue not that one will sometimes want to set the size
in "screen-inches", rather than by scaling the defaults? Looks wrong
to me to have it scaling with anything except the pixel resolution.


- -- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 02 Dec 1999 19:35:24 +0100
From: mathieu <mathieu.ros at free.fr>
Subject: [R] problem with par(fig=value)

- --------------1E8E8E7AB4B0CCCC413712D2
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit

hello all,
I want to draw a figure with multiple plot on the same page using the
par(fig=value) parameter but

> par(fig = c(0, 50, 60, 95)/100, adj = 5/10)
> eboulis(iris.acp)
> par(fig = c(45, 100, 60, 95)/100, mgp = c(3, 1/2, 0))
> boites(iris.acp)

draw the graphics on 2 different pages.
what am I doing wrong ?
thanks for your help.
    Mathieu

[using R 0.65 under Linux Redhat 6.0]

- --
- ----------------------------------------------------------------------
 Mathieu Ros - 13 rue bévière - 38000 GRENOBLE - 04 76 491 370
 http://www.multimania.com/mathieuros/index2.html
 DESS ingénierie mathématique (biostatistiques)
 Universite Joseph Fourier, Grenoble
- ----------------------------------------------------------------------
La science détient la vérité, ne vous laissez pas avoir par les faits.



- --------------1E8E8E7AB4B0CCCC413712D2
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
hello all,
<br>I want to draw a figure with multiple plot on the same page using the
par(fig=value) parameter but
<p>> par(fig = c(0, 50, 60, 95)/100, adj = 5/10)
<br>> eboulis(iris.acp)
<br>> par(fig = c(45, 100, 60, 95)/100, mgp = c(3, 1/2, 0))
<br>> boites(iris.acp)
<p>draw the graphics on 2 different pages.
<br>what am I doing wrong ?
<br>thanks for your help.
<br>&nbsp;&nbsp;&nbsp; Mathieu
<p>[using R 0.65 under Linux Redhat 6.0]
<pre>--&nbsp;
- ----------------------------------------------------------------------
&nbsp;Mathieu Ros - 13 rue b&eacute;vi&egrave;re - 38000 GRENOBLE - 04 76 491 370
&nbsp;<A HREF="http://www.multimania.com/mathieuros/index2.html">http://www.multimania.com/mathieuros/index2.html</A>
&nbsp;DESS ing&eacute;nierie math&eacute;matique (biostatistiques)
&nbsp;Universite Joseph Fourier, Grenoble
- ----------------------------------------------------------------------
La science d&eacute;tient la v&eacute;rit&eacute;, ne vous laissez pas avoir par les faits.</pre>
&nbsp;</html>

- --------------1E8E8E7AB4B0CCCC413712D2--

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

------------------------------

Date: Thu, 2 Dec 1999 12:47:09 -0500
From: "Yves Gauvreau" <cyg at sympatico.ca>
Subject: [R] Meaning?

This is a multi-part message in MIME format.

- ------=_NextPart_000_035E_01BF3CC3.58F31C40
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi,

Thank you all for your quit an aboundant responses. HAND all of you.

Thanks again.

Yves



- ------=_NextPart_000_035E_01BF3CC3.58F31C40
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Diso-8859-1" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.2314.1000" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT size=3D2>Hi,</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=3D2>Thank you all for your quit an aboundant responses. =
HAND all=20
of you.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=3D2>Thanks again.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=3D2>Yves</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV>&nbsp;</DIV></BODY></HTML>

- ------=_NextPart_000_035E_01BF3CC3.58F31C40--

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

------------------------------

Date: Thu, 02 Dec 1999 10:30:50 -0800
From: Harry Mangalam <mangalam at home.com>
Subject: [R] Multi-threaded external routines with R?

Hi All,

While R is not multi-threaded, it supports dynamic loading of external
routines writ in C or Fortran.  It seems that it should be possible to link
in multi-threaded code as easily as serial code, as long as the external
routine syncs correctly and returns valid R objects.  

Is this true or ar there R internals that would prevent this?  

If it is true, can anyone point me to examples of this kind of usage?

Thanks very much,
Cheers
Harry
- -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

------------------------------

Date: Thu, 2 Dec 1999 14:12:54 -0500 (EST)
From: Ben Bolker <ben at zoo.ufl.edu>
Subject: Re: [R] problem with par(fig=value)

  I think you need to look at par(new) ...

On Thu, 2 Dec 1999, mathieu wrote:

> hello all,
> I want to draw a figure with multiple plot on the same page using the
> par(fig=value) parameter but
> 
> > par(fig = c(0, 50, 60, 95)/100, adj = 5/10)
> > eboulis(iris.acp)
> > par(fig = c(45, 100, 60, 95)/100, mgp = c(3, 1/2, 0))
> > boites(iris.acp)
> 
> draw the graphics on 2 different pages.
> what am I doing wrong ?
> thanks for your help.
>     Mathieu
> 
> [using R 0.65 under Linux Redhat 6.0]
> 
> --
> ----------------------------------------------------------------------
>  Mathieu Ros - 13 rue bévière - 38000 GRENOBLE - 04 76 491 370
>  http://www.multimania.com/mathieuros/index2.html
>  DESS ingénierie mathématique (biostatistiques)
>  Universite Joseph Fourier, Grenoble
> ----------------------------------------------------------------------
> La science détient la vérité, ne vous laissez pas avoir par les faits.
> 
> 
> 

- -- 
- --------------------------------------------
Ben Bolker                                  bolker at zoo.ufl.edu
Zoology Department, University of Florida   http://www.zoo.ufl.edu/bolker
318 Carr Hall/Box 118525                    tel: (352) 392-5697
Gainesville, FL 32611-8525                  fax: (352) 392-3704

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

------------------------------

End of R-help Digest V1 #34
***************************

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
Digest of R-help mailing list; About R, see  http://www.ci.tuwien.ac.at/R/
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-digest-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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list