[Tig] color table: system gives name?

[Dominic Case]

> There are some good natural standards of colour - bananas , lemons,
oranges, tomatoes, peppers.
> There are some obvious colours such as grey (if you have defined white)
and black. There are the
> printing primaries (if you distinguish SWOP in the US and FOGRA in Europe.
However, there is a
> large gap of reliable color words in the blue-violet-pink range. 

Berlin & Kay's work that I alluded to before puts colours in the following
order:

Stage I: Black and white
Stage II: Red
Stage III: Either green or yellow
Stage IV: Both green and yellow
Stage V: Blue
Stage VI: Brown
Stage VII: Purple, pink, orange or grey

- primitive languages at stage 1 of linguistic development only have terms
for black and white. Languages that have more colour terms can be arranged
reliably according to this list.  It's interesting that blue is the last of
the primaries to be added. It was originally claimed that the reasons for
these language differences were physiological - "primitive" people were
effectively colour blind (though not in the "defect" sense that we
understand it from the Isahara tests - deuteranomalia etc, but a universal
non-perception of the difference between blue and grey, for example, in a
community).Berlin & Kay's theory is now challenged, but those who support it
do so mainly on linguistic grounds - as languages develop in
sophisitication, speakers coin terms for ideas that are important to them.
Thus red, the colour of blood, is the most alarming non-neutral colour of
all, and the one most likely to be named. Green is important as it
demonstrates life and health in plant crops. Blue? Not only are we short of
referents apart from the sky (not always reliable!), it's hard to see blue
as a word that is crucial to our survival. Especially the dark blue that we
know as primary blue.

>I have often thought that we ought to have voice control for color editing.

This prompts such alarming visions of coloring sessions that I can't think
of where to start commenting. Shouting matches come to mind. However, the
technology is all available (and only needs to be assembled) to be able to
point (with your finger at the screen) at an object in the image, and say
"warmer" or "more chroma" etc to have that object change. Presumably
youcould then point at one face, say "match" then point at another face, say
"this" and have the system recognise the shape, and correct one shape to
match the other sample area. You could totally screw up an image without
touching a button.

>but it does mean I do not always see colours as other people do
Vale Vincent van Gogh, Turner, Monet et al.

_________________________________
Dominic Case
Atlab Australia
________________________________