[Rd] plot.lm: "Cook's distance" label can overplot point labels

[Michael Dewey]

At 13:05 20/02/2009, John Fox wrote:

>Dear Thomas,
>
>Though far from an expert on the matter, it's my understanding that
>red-green confusion is the most common form of colour-blindness. I guess
>that the best way to put it is that it would be desirable to choose colours
>for the standard palette that minimize the probability of perceptual
>problems.

I wonder whether there are two separate issues: what is the best 
standard palette and whether mainstream plots should use colour to 
carry essential information. For instance there seems little problem 
in biplot having red arrows and black points because the colour is redundant.

I am not an expert on colour vision either but there certainly are 
people who report difficulty at scientific meetings with interpreting 
the slides.


>Regards,
>  John
>
>
> > -----Original Message-----
> > From: Thomas Lumley [mailto:tlumley at u.washington.edu]
> > Sent: February-20-09 4:32 AM
> > To: John Fox
> > Cc: 'John Maindonald'; 'Prof Brian Ripley'; r-devel at r-project.org; 'Martin
> > Maechler'
> > Subject: Re: [Rd] plot.lm: "Cook's distance" label can overplot point
>labels
> >
> > On Thu, 19 Feb 2009, John Fox wrote:
> >
> > > Dear John and Brian,
> > >
> > > My point about colour-blindness was partly tongue-in-cheek, but I think
> > that
> > > it's a bad choice to have the second and third colours in the default
> > > palette as red and green.
> > >
> >
> > Looking at the standard palette with dichromat::dichromat() it seems that
>it
> > depends on which flavour of red-green anomaly you have. For deuteranopia
>the
> > red and green are quite close. For protanopia they are pretty distinct and
> > the confusion is between colours 3 and 7 (yellow vs green) and between 4
>and
> > 6 (blue and magenta).
> >
> > I agree that the standard palette isn't ideal, though.
> >
> >       -thomas
> >
> > Thomas Lumley                 Assoc. Professor, Biostatistics
> > tlumley at u.washington.edu      University of Washington, Seattle
> >
>
>______________________________________________
>R-devel at r-project.org mailing list
>https://stat.ethz.ch/mailman/listinfo/r-devel
>
>
>No virus found in this incoming message.
>Checked by AVG - www.avg.com
>Version: 8.0.237 / Virus Database: 270.11.1/1961 - Release Date: 
>02/19/09 18:45:00

Michael Dewey
http://www.aghmed.fsnet.co.uk