Re: default colormap

[Carlo de Falco]

Il 04/nov/2015 13:43, "Jordi Gutiérrez Hermoso" <address@hidden> ha scritto:
>
>Moreoever, if they have, what viridis is doing by changing it
> slightly but still looking almost exactly like parula does not make it
> any less derivative (in the case of copyrights) nor less functional
> (in the case of patents). In the absurd case that they trademarked the
> colours (and we should investigate), maybe you can argue that changing
> their name from parula to viridis can avoid their trademark.
>

Viridis (and the other new perceptually uniform sequential colormaps magma, inferno, and plasma in matplotlib) are NOT just a slightly modified version of parula:

https://bids.github.io/colormap/

It seems that also R is adopting these colormaps (I'm not sure whether they'll make them default though):

https://cran.r-project.org/web/packages/viridis/index.html

In this case I personally care more about consistency between octave python and R rather than exactly reproducing Matlab.

I do see why you consider claims from TMW inconsistent but IANAL so my opinion on this point does not count much. They do not want us to use parula and I see no reason to insist on using it.

c.