Color Representations for Brown and Violet

[Mickey Segal]

Check out http://mindprod.com/jgloss/wavelength.html
that converts a frequency of the rainbow to a Java colour.

This sounded very enticing but I couldn't get it to run. I downloaded the
wavelength11.zip file, expanded it and I see a "just.bat" file that attempts
to load on an E drive. Is there some simple way to launch this in a way
obvious to someone who never fiddles with Java applications?

 

[Andrew Thompson]

..you shouldn't have to put text notes below an applet.

For *end* *users* I would agree, but you might hope that
developers would have the sense to RTFM!

(Maybe if you moved the message *above* the applet?
Some people are a bit 'scroll challenged'.)

 

[Andrew Thompson]

Check out http://mindprod.com/jgloss/wavelength.html
that converts a frequency of the rainbow to a Java colour.

It also states..
"The wavelength of an electron is much shorter ¡X 2.42 picometers."

De-Broglie's wavelength fromula would suggest that figure is
only correct for one particular electron *velocity*. I would
guess the velocity in this case would be the orbital velocity
of the electron when located in the 1st (innermost) electron shell.

BTW - the spectum bars.. pretty.

And to Mickey. 'Spectrum' has a main(). I might have
written a .bat to launch it, but didn't bother since
TextPad launched it for me, using the class files that
were in the .zip distributable.

 

[Thomas Fritsch]

or quoted :

Except you are mixing light not pigments. so
red+green = yellow
green + blue = turquoise
blue + red = violet

the color mixing rules for pigments and light have little to do with
physics and at a lot to with anatomy and the variable sensitivity of
different cells on your retina to different frequencies of light.

Agreed.
Light mixing (or additive mixing) is physically correct for video screens.
But it is not that intuitive, and sometimes surprising,
for example: red(255,0,0) + green(0,255,0) = yellow(255,255,0)
Taking the average resembles more closely the human intuition, which we got
from our paint-box at the age of 6.
for example: [ red(255,0,0) + green(0,255,0) ] / 2 =
brownish_green(128,128,0)
Other examples can be constructed at will.

 

[Roedy Green]

For *end* *users* I would agree, but you might hope that
developers would have the sense to RTFM!

Please read this book. It may change your life:

The Design of Everyday Things
0-385-26774-6
Donald A. Norman
Though this is not particularly about designing GUIs, I love this
book. It was like meeting somebody else who understood my frustration
with the ineptness in the design of user interfaces for computer
programs, household appliances and just about anything mechanical. He
grabbed ill-formed ideas out of my mind and laid them clearly on
paper. A classic.

http://www.amazon.com/exec/obidos/ASIN/0385267746/canadianmindprod

 

[Roedy Green]

Light mixing (or additive mixing) is physically correct for video screens.
But it is not that intuitive, and sometimes surprising,
for example: red(255,0,0) + green(0,255,0) = yellow(255,255,0)
Taking the average resembles more closely the human intuition, which we got
from our paint-box at the age of 6.
for example: [ red(255,0,0) + green(0,255,0) ] / 2 =
brownish_green(128,128,0)

A have composed 18 colour tables and 2 colour picking Applets you can
get at from http://mindprod.com/jglossl/colour.html

By looking at the colours sorted all sorts of ways, and looking at the
colours and numbers, you can begin to get to the point where you can
compose colours out of your head with numbers, or know what they would
look like just from the number.

Most of my intuition in the f0f0f0 -- ff ff ff range.

I long time ago I read that Edwin Land, the guy who invented the
Polaroid discovered the eye can be tricked into seeing colours with
only two colours of light mixed. I have never seen any mention of it
since.

I remember my bafflement learning about the light spectrum and at the
same time the strange colour mixing rules from my second grade art
teacher. The two things just did not jibe at all in my mind. It was
not until very late in life did it dawn on me that colour mixing is an
anatomical phenomenon, not anything inherent in the spectrum, and
further than each type of retinal cell had its own response curve to
ALL frequencies of light. They were not narrow band detectors.
However, the brain can make sense of all this and provide the illusion
we have explicit frequency detector pixels in our eyeballs.

 

[Roedy Green]

Given the range of colors, any definition of color by name
seems ..limiting and pointless.

I can bring to mind exactly what PapayaWhip looks like, but I have
only a vague idea what #ffefd5 looks like.

If the names were known and standard, it would be easier to proofread
a program with the names, at least glaring errors would stand out and
also tiny differences in colour are more obvious by the names.

Perhaps some future SCID will show colour selections three ways -- by
name, sample and number, configure for the current need.

 

[HGA03630]

Here's Japanese standard color names and their approximate RGB values.
This list does include X11 colors for reference purpose.
http://homepage1.nifty.com/algafield/JaColors.html
(beware, charset is Shift_JIS)

 

[Andrew Thompson]

Please read this book. It may change your life:

Big claim. 'Jonathon Livingstone Seagull' did nothing for me.
[ Mostly because it got the aerodynamics completely wrong. ;-) ]

The Design of Everyday Things

...
Nice tea-pot on the cover.

It probably has some good tips, I'll add it to the
(long) list of books I intend to buy (one day ...am
not too flush at the moment).

 

[Andrew Thompson]

I can bring to mind exactly what PapayaWhip looks like,

?!? You, but not me..

..but I have
only a vague idea what #ffefd5 looks like.

If the names were known and standard,

You mean, all.. 16,777,216 of them?

..it would be easier to proofread a program with the names,
ditto?

..at least glaring errors would stand out

Amongst those 16,777,216?

..and
also tiny differences in colour are more obvious by the names.

...to you maybe. Tell me. What is the difference between
Fushia, Fucia and Fucsia? Or rather, can you immediately
picture in your mind the colors I am 'referrring to'?

Try this

<html>
<body>
<table>
<tr><td bgcolor='fushia'>Fushia</td></tr>
<tr><td bgcolor='fucia'>Fucia</td></tr>
<tr><td bgcolor='fucsia'>Fucsia</td></tr>
</table>
</body>
</html>

All three colors are rendered differently to each other,
yet are consistent across late model IE, Opera and Mozilla.

I do not regard that useage of color as being either
intuitve *or* practical. I would have to see them rendered
to know what they looked like, then I would want *one* of those
three - with just a little more green, or less blue.

That is where the 'name for color' system falls apart.

Perhaps some future SCID will show colour selections three ways -- by
name, sample and number, configure for the current need.

I recognise that some people are incapable of dealing with
colors as spearate RGB. For them, a small palette of
consistently rendered colors might be handy, but it becomes
cumbersome to define names for a great number of colors.

 

[Joan]

Class <Color> includes constants for a lot of useful colors,
but not-
able by their absence are brown and violet. Can anyone give me
rgb
values for those two colors so I can create them using the
standard
constructor? Thanks in advance.

---Kevin Simonson

/usr/lib/X11/rgb.txt

 

[Chris Uppal]

Try this

<html>
<body>
<table>
<tr><td bgcolor='fushia'>Fushia</td></tr>
<tr><td bgcolor='fucia'>Fucia</td></tr>
<tr><td bgcolor='fucsia'>Fucsia</td></tr>
</table>
</body>
</html>

All three colors are rendered differently to each other,
yet are consistent across late model IE, Opera and Mozilla.

I do not regard that useage of color as being either
intuitve *or* practical. I would have to see them rendered
to know what they looked like, then I would want *one* of those
three - with just a little more green, or less blue.

It's off-topic I know, but maybe a good way to name colours, attempting to
maximimse both precision, visual/intuitive clarity would be to use a fairly
small number of "ordinary" color names (i.e not "paypaya whip", "taup", "puce",
or "white with a hint of marketing") plus a smalling delta expressed in RGB (or
other colourspace). Gray + 2r0b0g.

-- chris

 

[Mickey Segal]

A have composed 18 colour tables and 2 colour picking Applets you can
get at from http://mindprod.com/jglossl/colour.html

Roedy mis-typed; he meant:
http://mindprod.com/jgloss/colour.html

I long time ago I read that Edwin Land, the guy who invented the
Polaroid discovered the eye can be tricked into seeing colours with
only two colours of light mixed. I have never seen any mention of it
since.

Land gave an excellent lecture when he dedicated a university building that
he donated, which some claim is in the shape of a Polaroid Land camera:
http://en.wikipedia.org/wiki/Harvard_science_center

The lecture (and a similar Scientific American article) were mostly about
the effect of nearby colors on perception of other colors, and the effects
are presumably due to neuronal properties and circuits in the retina, the
lateral geniculate or the occipital cortex. There was also some stuff about
using only two colors of light, and they are presumably related by the
ability to get neurons to make decisions based on limited information. I
never figured out whether this had anything to do with how the Polaroid Land
camera worked, but I always wondered about that because the colors in the
resulting instant photos always looked a bit off. The only relation of this
work to computer displays that I can think of is the need to consider the
color of one component in relation to its nearby components, which, alas,
led to the confusing display in the applet that I made to choose component
colors (www.segal.org/java/color/).

 

[Mickey Segal]

The lecture (and a similar Scientific American article) were mostly about
the effect of nearby colors on perception of other colors, and the effects
are presumably due to neuronal properties and circuits in the retina, the
lateral geniculate or the occipital cortex. There was also some stuff
about using only two colors of light, and they are presumably related by
the ability to get neurons to make decisions based on limited information.
I never figured out whether this had anything to do with how the Polaroid
Land camera worked, but I always wondered about that because the colors in
the resulting instant photos always looked a bit off.

It seems the Polaroid camera did not use Land's intriguing two color
results:
http://www.greatreality.com/Color2Color.htm
"Land never found a practical way to apply the phenomenon to instant
photography, and he ultimately developed Polaroid color film using a
conventional three-color system."

 

[Oliver Wong]

I can bring to mind exactly what PapayaWhip looks like, but I have
only a vague idea what #ffefd5 looks like.

The opposite is true for me. I know Papaya is a fruit, but no idea what
it looks like (or what colour it would be), and no idea how "Whip" would
modify a color. But looking at the RGB values, I have a very good idea of
what the colour looks.

- Oliver