Re: default colormap

[Ben Abbott]

> On Nov 4, 2015, at 5:54 AM, Pantxo <address@hidden> wrote:
> 
> Carlo de Falco-2 wrote
>> On 4 Nov 2015, at 00:02, Jordi Gutiérrez Hermoso &lt;
> 
>> jordigh@
> 
>> &gt; wrote:
>> 
>>> On Sun, 2015-11-01 at 23:57 +0000, Carlo De Falco wrote:
>>> 
>>>> The purpose of this changeset:
>>>> http://hg.savannah.gnu.org/hgweb/octave/rev/17e507df10e3 was to
>>>> change the default colormap for the 4.2 release.
>>> 
>>> Btw, why can't we just set it to parula? I can't see why this colour
>>> map is any different from our existing ones that match Matlab's. It
>>> doesn't matter how the Mathworks created parula as long as we
>>> replicate parula without consulting Matlab's source code.
>>> 
>>> How did we replicate the other colour maps?
>>> 
>>> - Jordi G. H.
>> 
>> Jordi,
>> 
>> The same discussion has gone on over at the matplotlib mailing list, 
>> I don't have the link at hand right now but I will try to find it.
>> 
>> IIRC the point is that it seems TMW are claiming "intellectual property" 
>> rights over the colormap itself, not the code that implements it.
>> 
>> Whether such a claim makes sense or not is unclear to me, but the reaction
>> of the matplotlib guys was: "well, if they do not wat us to use parula,
>> why 
>> not try to do something better?" I like their approach ...
>> 
>> After all, do we really need compatibility at this level? 
>> There's good reasons to not use jet as the default colormap, but
>> do we really need to care that the default colormap be exactly the same as
>> Matlab?
>> 
>> 
>> c.
> 
> Seconded, I very much like viridis which fixes all the issues raised by Jet,
> and is not so different from ML's "parula". Why try to mimic parula (and
> eventually get into troubles) when we already have a very good replacement?
> As for compatibility we could make "colormap ('parula')" return a warning
> such as "parula is proprietary, you may use viridis as free replacement".  
> 
> Pantxo

Mathworks recenlty changed a lot of things when it introduced its new graphics 
system.

http://blogs.mathworks.com/loren/2014/11/05/matlab-r2014b-graphics-part-3-compatibility-considerations-in-the-new-graphics-system/

I’m not sure what “compatible” means at the moment. The pre-2014b Matlab and 
post-2014b Matlab are not compatible with each other. The default colormap is a 
trivial example.

Ben