[R] problem with labeling plots, possibly in font defaults

[Prof Brian Ripley]

See 
http://cran.r-project.org/bin/macosx/RMacOSX-FAQ.html#I-see-no-text-in-a-Quartz-plot_0021

And the default font is not serif ... that FAQ says it is Arial, but I 
do not know if that is current.

Mac-specific questions to R-sig-mac please.  (This must be Mac-specific 
as the default device, quartz(), is.)

On 07/08/2014 23:24, David Winsemius wrote:
>
> On Aug 7, 2014, at 12:59 PM, Tim Blass wrote:
>
>> Hello,
>>
>> I am using R 3.1.1 on a (four year old) MacBook, running OSX 10.9.4.
>>
>> I just tried making and labeling a plot as follows:
>>
>>> x<-rnorm(10)
>>> y<-rnorm(10)
>>> plot(x,y)
>>> title(main="random points")
>>
>> which produces a scatter plot of the random points, but without the title
>> and without any numbers along the axes. If I then run
>>
>>> par(family="sans")
>>> plot(x,y,main="plot title")
>>
>> the plot has the title and the numbers on the axes (also and 'x' and 'y'
>> appear as default labels for the axes).
>>
>> I do not know what is going on, but maybe there is some problem in the
>> default font settings (I don't know if that could be an R issue or an issue
>> specific to my Mac)?
>
> It hasn't happened to me recently (since updating from Leopard and SnowLeapard to Lion)  but it used to happen pretty frequently that I would get a duplicate font that printed empty square boxes. (I wasn't the only one. It got reported several times on R-SIG-Mac.)  One could fix that sort of problem by deleting the offending duplicate entry using Font Book.app
>
> Since this is happening with the default serif font,  you would probably find the duplicate in Times. (It used to be happening to me with Symbol.)
>
>> quartzFonts()$serif
> [1] "Times-Roman"      "Times-Bold"       "Times-Italic"     "Times-BoldItalic"
>>
>
>
>>
>> This is clearly not a big problem (at least for now) since I can put labels
>> on plots by running par(), but if it is indicative of a larger underlying
>> problem (or if there is a simple fix) I would like to know.
>>
>> Thank you!
>
> David Winsemius
> Alameda, CA, USA


-- 
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 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595