Talk:Color

It has been appropriate for the past several years for this page to be titled 'Color' as most Wikipedia users in the early days of the 'net were from North America. But Wikipedia is getting enormous respect and recognition world-wide, now, and the Internet is far wider, too.

So, perhaps 'Colour' would be a better name. It would be the usual spelling for most English speakers, and hence make the article easier to read. And that will make it a better article.

If that's too big a change for some (though it is hardly a 'big issue'), then at least we should have both spellings in the title?

quota

We already have both spellings, the way it's always done in Wikipedia: Colour is a redirect to Color. We don't need to have this debate again. --jpgordon ∇∆∇∆ 19:18, 10 Dec 2004 (UTC)
Obviously we do, as newcomers are still expecting the more common spelling and are making corrections and edits that way. Let's just switch the redirect and article and have done with it. quota

WE DO NEED TO HAVE THIS DEBATE AGAIN.

The spelling "color" only exist in the USA, Thus the spelling "color" must be on the disambiguation page, and the main article retitled with the original, standard, and most universal English spelling, i.e.: "COLOUR". The prevalence of the American dialect of English in this international encyclopaedia is an example of cultural imperialism. The standard English should be standard English as in England, with American-specific articles in the American dialect. The process of "de-Americanisation" must be granted official sanction, so that as edits take place to correct the spelling, there is no wrecking or obstruction by any Americans who want to resist the de-Americanisation and internationalisation of Wikipedia. At the moment there are even British-focussed articles written in the American dialect! The Isle of Man article was the last one I saw like that.

The alternative, is that the Wikipedia hierarchy just be open about it, and say that this is an American resource which is designed to promulgate an US-centric world view.

MACDADDY

The US dialect is twice as popular as all the others put together. If you say language can not change, then shall it be Olde English or what? What century is OK with you? The common root ancestor from somewhere around 1600? Which town?

Metameric colors

I removed this part:

"cyan is a pure spectral color whose wavelength is located just between the responsitivity peaks of the "green" and "blue" cones. A cyan color experience can thus also be generated by an equal mixture of those two peak wavelengths, as long as these don't stimulate the red receptor."

One can easily see by regarding the CIE chromaticity diagram that that no pure spectral color can be reproduced by a combination of other spectral colors.

Hankwang 19:43, 13 Feb 2004 (UTC)

This is not correct. Key is "as long as these don't stimulate the red receptor", which is a very artificial condition. In other words, if you disconnect (poison?) the red receptors in your eye, then you can make cyan from green and blue. AlbertCahalan 05:13, 19 Feb 2005 (UTC)

PS this part was on the top of the talk page earlier - sorry, newbie mistake. - Hankwang 20:08, 18 Feb 2004 (UTC)

What are the primary colors??

Please read this slowly and carefully!

A while ago, notes about RGB color coordinates on certain colors at Wikipedia were added, trying to make what they are talking about based on the psychological primary colors, not the traditional primary colors or the light primary colors. (Look up primary color at http://dictionary.reference.com for what the psychological primary colors are.) Psychologically, color definitions are based on 3 scales, one of which is a black-white scale, one of which is a blue-yellow scale, and one of which is a red-green scale. However, I looked at the histories of the entries of Yellow, Red and, Green at Wikipedia and it appears that people are trying to change the color entries so that they are using the light primary colors as the standards. Somewhere on the Internet, there needs to be a poll that talks about what set of primary colors sounds most natural to you, with the choices being traditional (red, yellow, and blue,) light (red, green, and blue,) and psychological (red, yellow, green, blue, black, and white.) My vote is psychological.

[It is Jerzy who notes that the above is the result of the following edit: 01:25, 2004 Feb 18 66.32.118.38 (What are the primary colors??)]

I think you're missing the point: the point is not to try to decide "what the primary colours are", but to point out that the term "primary colour" is used to mean very different things in different contexts. In the realm of painting and printing, it refers to the basic pigments used to obtain paints of almost any colour appearance. The first such set of basic pigments was developed by painters, who settled on red, yellow and blue pigments. Later, this selection was refined into magenta, yellow and cyan, and a fourth black pigment was added. This CYMK system is now the standard in printing. But since there is still an amount of perceptual colours which are not reproducible using this set, further sets such as Hexachrome (which adds two other basic pigments) have been devised. Any of these different sets may be called a set of "primary colours", but "primary pigments" would be clearer and more proper. In the realm of screens and displays, "primary colours" refers to the set of pure spectral lights with which it is possible to prompt the perception of most colours, a set of three light wavelengths which only when in isolation appear as red, green and blue. As in the case of the primary pigments, there's still an area of the perceptual space not covered by the RGB, and again, a term like "primary lights" or "primary wavelengths" would be clearer and more proper than "primary colours". Finally, in the realm of psychological perception, "primary colours" are those colour perceptions that appear to be pure, unmixed, and are indescribable in terms of others. These are six: white, black, red, yellow, green and blue, and a clearer way to refer to them is "elementary colours" or "elementary/basic/primary colour percepts". Uaxuctum 00:30, 2 Oct 2004 (UTC)

FWIW:

I feel that cyan/magenta/yellow are the most natural primary colors, because that's what I used to experience when mixing paints at school. It feels natural to create new colors by substractive mixing (plus white paint). (We students would call the colors blue, red, and yellow, although the cans actually did read C, M, Y instead of B, R, Y.)
As a physicist, I find RGB the most logical set of primary colors, but it is not very intuitive to me how new colors are created by mixing them additively (e.g. when I need to pick RGB values for web pages).

You've hit the spot: If RGB is so "logical", why is it not intuitive? Why can't I predict offhand what the RGB composition of any given colour is, while on the other hand I can easily describe lilac as blue+red+white, pink as red+white, orange as red+yellow, maroon as red+black, etc.? The answer is simple: RGB deals with what happens at the level of retinal cone stimulation, which is fine to know in order to find a way to "fool the eye" for colour reproduction techniques, but which is only the very beginning of the story of how our colour vision works. The raw tri-stimulus data produced by the cones is then much processed through a series of neuronal networks which quickly turn it into a colour signal that is no longer based on the three parameters originating from the cones, but on the parameters of an achromatic scale and of two dimensions of opponent pairs of hues, that is, based on the six elementary colour percepts which in isolation appear as the only six colours (white, black, red, yellow, green and blue) that to our mind look completely pure and indescribable in terms of others. That's why it doesn't feel intuitive to work with RGB values: when the colour information reaches our mind, it simply is no longer dealing with such kind of data, and thus we are unaware of what the pre-processed tri-stimulus signal originating at the cones consisted of, so when presented with that kind of information it doesn't look intuitive at all. The precise details of the functioning of that very complex neuronal network that turns the raw cone signals into the final processed colour data haven't been fully understood yet and involve many kinds of cells and structures, such as ganglions, amacrine cells, double-opponent cells, cortical blobs, etc., although much investigation has already been carried out which has clarified at least parts of it. Here's a very quick summary of some of the processes at the retinal level: http://webexhibits.org/colorart/ganglion.html, and here you can find some much more in-depth descriptions of what is already known about the inner workings of our visual system: http://webvision.med.utah.edu/index.html. Uaxuctum 00:30, 2 Oct 2004 (UTC)

I had never heard of psychological primary colors.

Sadly, this situation is very common. Sad, because most people have been taught something about the primary colours of printing and TV, even though for the most part that information seems unintuitive and is rather useless for everyday life, while nothing is usually said about the basics of the human perceptual colour space, which is the one that describes how colours actually appear to our mind, an information that is both intuitive to understand and the one most useful in order to realize of the perceptual relations and harmonies between colours, and so the one really helpful in such everyday tasks as combining the colours of clothes in pleasing ways. Uaxuctum 00:30, 2 Oct 2004 (UTC)

Yes, a red can be bluish or yellowish, and likewise for the other color combinations except blue/yellow and red/green. But to me, it seems like an arbitrary way to divide the hue circle into four quadrants. It does not deal very well with intensities. To me, brown does not feel as a mixture between black and yellow/orange. Moreover, by introducing white in this list, one crosses the border between perception of colored light and the tricky area of the perception of material colors, where the brain does some processing to identify the color of the material decoupled from the ambient lighting conditions.

The hue circle doesn't deal with intensities, it merely shows the mutual relationships between the saturated hues as for chromaticity (e.g. that orange "belongs" between red and yellow), not as for the inherent lightness value of each hue (e.g. that pure yellow is per se lighter than any other hue). The most logical way to show the mutual relationships as for chromaticity between the hues in the colour circle is to define four quadrants signalled by the four pure, unitary hues (red, yellow, green and blue) and to place each scale of binary hues in the quadrant between the two hues it is composed of (the oranges in the quadrant between red and yellow, the purples in the quadrant between red and blue, the bluegreens in the quadrant between blue and green, and the yellowgreens in the quadrant between yellow and green). Colours such as brown, pink, maroon, navy, etc. are unsaturated, that's to say, they are composite perceptions consisting of a hue component (an amount of an element from the hue circle) and an achromatic component (an amount of white/grey/black), and are the ones filling the area of the perceptual colour space between the hue circle and the achromatic scale. Nothing stops you from representing the tridimensional perceptual colour space taking into account the inherent lightness value of hues, for example making the value at the white-to-black axis reflect it, thus making the representation of the hue circle ondulate according to the inherent lightness value of each hue so that yellow would be placed at a "height" close to that of white, while red would appear at the height of a certain rather darkish shade of grey, and the quadrant of the hue circle with the scale of oranges would rise from the height of red to the height of yellow, etc. The resulting representation of the perceptual colour space would be halfway between those of the NCS and Munsell systems. OTOH, can you find a "pure" brown, that is, one that is impossible to define in terms of other colours just like white or red are by all means impossible to define in terms of other colours? I cannot. Any shade of brown that I can see and think of has at least a tinge of yellow and/or red and a tinge of black and/or white, even if that tinge may happen to be very very subtle in a certain shade of brown and even if the combination of those several percepts into one composite perception makes the result look distinctly different from how those percepts look on their own. This relationship is clarified for example if you place a brown square next to squares in other colours and check which ones the brown one appears to have an "afinity" with: if you put blue or green next to brown, you won't feel any affinity, but you will if you put red or yellow next to brown—it may be a very subtle affinity, one of merely a very slight, barely perceptible trace of yellowishness and reddishness in the case of the darker kinds of browns, but there is one which is completely abstent when comparing brown to blue or green (well, except between olive brown and green, in case you consider dark olive = yellow+green+black(+white) as a kind of brown). Another experiment: put a brown square next to a square in navy blue and another in forest green, the three of them in shades of similar darkness; the fact that there is something yellowish/reddish in brown should become more obvious. Yet another way to make clear the relationship between brown, red and yellow: take orange (= red+yellow, very clearly) and increasingly add black; the result is that the orange becomes increasingly darker and less saturated until you will no longer comfortably call it orange but brown—the hue component hasn't changed (the proportion of reddishness to yellowishness has remained the same, and their presence hasn't been diluted completely), but the addition of increasingly more black to the composite perception makes it cross the fuzzy frontier between the realm of the linguistic label "orange" into that of the linguistic label "brown". On a different issue, "material" colors like gold (which are patterns of interaction between dots of several different colours distributed over a surface) are a in very different category from that where white belongs; gold is rather a chromatic texture instead of a true colour. Uaxuctum 00:30, 2 Oct 2004 (UTC)

- Contrary to your comments, psychological primary colors were not invented by this Wikipedia user; they are mentioned at the "primary color" entry of http://dictionary.reference.com and several Internet sites, as you will find by doing a Google search. 66.32.126.20 21:39, 3 Apr 2004 (UTC)

Hankwang 21:00, 18 Feb 2004 (UTC) (clarified 23 Feb 2004)

I belive RGB is the standard set of primary colours, although I believe both RGB and cyan/magenta/yellow should be represented. However, the standard should (IMO) be RGB.

I too had never heard of psychological primary colours. I believe this is a very grey area (no joke intended) - if you're colourblind, surely your psychological primary colours will be different to other peoples?

Nick04 21:19, 20 Feb 2004 (UTC)

Documenting recent color-related anon 66.32... contributions

[This begins a seven-section contribution by Jerzy, which ends with a similar note.]

Red

The text

The RGB coordinates of red are 255 0 128. Note that the color most people call "red" on computers is in fact red with a little yellow in it.

was added to the article Red in the following edit:

18:37, 2004 Feb 14 . . 66.32.124.157

(with no summary)

(To avoid confusion, note that a "(last)" of article Red (http://en.wikipedia.org/w/wiki.phtml?title=Red&diff=2422410&oldid=2422400) shows that the edit

11:37, 2004 Feb 17 129.169.8.153 (RGB co-ordinates of 'red' are clearly 255,0,0. Not 255,0,128 which is Magenta-like.)

consisted of changing ",128" to ",0".)

Green

The edit

18:44, 2004 Feb 14 . . 66.32.124.157

with no summary inserted the text

The RGB coordinates of green are 0 255 128. The color that most people call green on a computer is actually green with a little yellow in it. Many people think of green as a mixture of blue and yellow, but this is because the blue, the yellow, or both already contains green in it.

(I, Jerzy, reverted that edit with

22:40, 2004 Feb 14 Jerzy (rv to Arj (killing confusion abt color theory))

and perhaps the note at Talk:Green is addressed to me. FWIW, i'm not convinced, but also not prepared to convince others that 66.32... is wrong.)

Various color-named articles

The contributions (http://en.wikipedia.org/w/wiki.phtml?title=Special:Contributions&target=66.32.124.157) of the above IP are 10 articles whose names are colors, all around 18:40, 2004 Feb 14.

This page (Talk:Color)

As noted above:

01:25, 2004 Feb 18 66.32.118.38 (What are the primary colors??)

Talk:Red

The edit

01:33, 2004 Feb 18 Talk:Red (RGB coordinates for Red)

adds the section == RGB coordinates for Red ==; it has three 'graphs and close to 200 words, and describes a straightforward-sounding experiment using PowerPoint.

Talk:Green

The edit

01:37, 2004 Feb 18 66.32.118.38 (RGB coordinates for Green)

put this text at the end of the talk page, under a heading == RGB coordinates for Green ==:

See the message in the talk about red titled "RGB coordinates for Red", the paragraph about replacing red-words with green-words.

Various color-named and color-system-named talk pages

The contributions (http://en.wikipedia.org/w/wiki.phtml?title=Special:Contributions&target=66.32.118.38) of the immediately above IP are 1 edit each of

the 3 talk pages pages above,

01:39, 2004 Feb 18 Talk:Primary color (Psychological Primary Colors), and

01:54, 2004 Feb 18 Talk:RGB color space (Explains the RGB coordinates for the psychological primary colors)

, all around 01:40, 2004 Feb 18

User:Michael Hardy has responded at Talk:RGB color space, apparently after finding some of the further experiments described there disconfirming of the IP contentions.

Do you mean Talk:RGB color model ? Kim Bruning 22:51, 3 Apr 2004 (UTC)

Perhaps the above catalog will be of some help to others who carry the real burden of this discussion.

[This ends a seven-section contribution by Jerzy, which begins with a similar note.] --Jerzy 04:21, 2004 Feb 18 (UTC)

Comment: Colour is actually used almost everywhere apart from America (inc. Canada and Australia) - it's the US that uses the odd spelling. -- anonymous

Color wavelengths

I changed the wavelength ranges in the table on the top to be in agreement with the CIE diagram (meanwhile I discovered that I put some of the wavelength labels in the diagram at wrong positions). I've seen my share of laser beams with well-defined wavelengths and a number of the cited wavelengths were outright wrong. A 514-nm argon laser generates green light, not cyan. I would call 590 nm (low-pressure sodium street lights) really orange and not yellow/orange. The boundary between yellow and orange is around 580-585 nm in my perception. What to call "orange" is probably very subjective, so I didn't touch those wavelengths. Since I don't have a reference table of what people generally call "blue", I used the CIE diagram to roughly define what I call green, blue, or cyan.

I think that indigo should be removed from the table. The difference between 430 and 450 nm in the CIE diagram is extremely small and it is agreed upon that Newton introduced this color just to get 7 "elementary colors" instead of 6.

Furthermore, I adjusted some of the HTML colors a bit to my feeling and my monitor gamma. I'm not sure what to make of blue and violet. 400 nm looks vastly different from whatever blue/red mixture you can create on a CRT and does not give a "purple" feeling to me. Pure CRT blue (#0000FF, around 465 nm) is probably closest. An other CIE diagram on the web (Google for it) lists 483 nm as "unique blue", which would be best approximated by a green/blue mixture on a CRT.

Hankwang 19:51, 25 Feb 2004 (UTC)

Color Wheel

Based on the discussion of HSV color space, I think it would be beneficial to add a subsection on "Color Wheel" just above the "HSV" subsection. This subsection could explain in a little more detail the reason for adding the mixtures of red and blue to the pure spectral colors in order to form a continuous wheel.

I think this would be a good separation to make in any case, but I am particularly motivated by my understanding of the history of HSV. To me, the idea a "color wheel" as a representation of hue, and the wheel's relation to the pure spectral colors might be something known to Newton, but I thought HSV was just a modern invention and formally a non-linear rearrangement of RGB color space. That's why I think the HSV color wheel should be discussed as a combination of RBG, not as spectral colors plus the purple line. "Color wheel"s in general are something that could be profitably related to CIE, while the HSV hue wheel is a derivative of this underlying theory that is designed with the parameters of RGB in mind.

Maybe I'm being too much of a stickler for the scope of this article, but I'll see if I can find time to make this change without messing up the structure of the article. -- Chinasaur

The HSV/RGB issue sounds plausible. My background is that I once took an introductory course about the physics of color vision. I don't know much about (historical) conventions and terminology. If you know about the history of HSV, then edit ahead! (As an aside, can you, in the future, sign your contributions on discussion pages? Just type four tildes in a row) -- Hankwang 19:56, 28 Feb 2004 (UTC)

Should "cyan" be "00FFFF" rather than "00E0E0"?

color	wavelength interval	frequency interval
red	~ 625-740 nm	~ 480-405 THz
orange	~ 590-625 nm	~ 510-480 THz
yellow	~ 565-590 nm	~ 530-510 THz
green	~ 500-565 nm	~ 600-530 THz
cyan	~ 485-500 nm	~ 620-600 THz
blue	~ 450-485 nm	~ 670-620 THz
indigo	~ 430-450 nm	~ 700-670 THz
violet	~ 380-430 nm	~ 790-700 THz
Continuous spectrum Not drawn to scale.

color	wavelength interval	frequency interval
red	~ 625-740 nm	~ 480-405 THz
orange	~ 590-625 nm	~ 510-480 THz
yellow	~ 565-590 nm	~ 530-510 THz
green	~ 500-565 nm	~ 600-530 THz
cyan	~ 485-500 nm	~ 620-600 THz
blue	~ 450-485 nm	~ 670-620 THz
indigo	~ 430-450 nm	~ 700-670 THz
violet	~ 380-430 nm	~ 790-700 THz
Continuous spectrum Not drawn to scale.

I would have expected "cyan" to be "00FFFF" rather than "00E0E0". Here I contrast the two with each other; the two tables appearing side-by-side have somewhat different "cyans". Should this be changed in the article? (I am pleased with what "orange" looks like in this table; I've seen other "oranges" on the web that didn't seem to land right in the middle of what "orange" should look like. But maybe that's just my browser?) Michael Hardy 22:37, 2 Mar 2004 (UTC)

In principle, both are cyan. The problem is that 00FFFF (maximum brightness cyan) has a much higher intensity/luminosity than 0000FF (maximum brightness blue). I feel that the current RGB values result in a smoothly varying intensity, without cyan sticking out as being brighter than both blue and green.

Eh, Cyan IS more brighter than blue or green...

MWAK--217.122.44.226 09:10, 22 Jul 2004 (UTC)

As far as orange is concerned, it might be a trick your eyes play on you because of the contrast with the surrounding colors. How about this one? It is the same shade.

Do you feel that indigo belongs in the table?

-- Hankwang 09:22, 4 Mar 2004 (UTC)

Should violet really have a quoted frequency? It doesn't appear in the continuous spectrum.

Suggested changes

I edited the article in order to deal with most of your comments -- Hankwang 00:21, 6 Mar 2004 (UTC)

I went through and corrected some typos and grammar. In the process I also noticed a few things that conceptually don't seem quite right to me, but I thought I should run it by the rest of y'all.

"White has no definite position in this diagram since any light spectrum that stimulates all three receptors, whether it is the warm light from an incandescent light bulb (more red) or the cooler daylight (more blue)".

This isn't a complete sentence, not totally sure what it's trying to say but I changed it to this: "White has no definite position in this diagram; rather it is defined differently depending on the application, mostly depending on the overall color temperature desired."

"The human color space is a infinite horse-shoe-shaped cone"

Maybe I'm just way off base, but this doesn't sound right to me. How can the cone be infinite? The color receptors do saturate, so it is not infinite in scope. Further, color perception has significant difference thresholds, so perceptual color space isn't really infinite in degree either. Right???

"No mixture of colors, though, can produce a fully pure color perceived as identical to a spectral color."

I'm not so sure this is true. Along the longer wavelength end of the curve (from around 550-700nm), things get very linear, implying that if I mixed primaries of 550 and 700nm, I should be able to reproduce the perception of an intermediate pure spectral color pretty much perfectly... In fact, now that I think about it this also agrees with something I read on the biology, which is that it actually is possible to stimulate almost entirely just L photoreceptors using light towards the 700nm. This is because the tuning curves drop off more sharply towards long wavelengths than short ones, so to the long side of the L cone peak you can get pretty good single cone stim, thus pretty good mixture recreation of spectral ("monochromatic') colors in that range as long as your primaries are spectral themselves. Maybe nitpicky, but I think it's worth it to be correct.

You are right. By the way, there are some tetrachromacy theories i. e. about rods having moderate influence on color vision in some conditions, but I have not read much about them. Artjt 6 Mar 2004.

"The HSV color space was already used by 19th century physiologist Ewald Hering."

I think this is another instance of mixing up HSV in the modern sense with the general idea of a color wheel, saturation, lightness model. If you check HSV color space it is asserted there that HSV (as the modern system based on modern RGB) was invented in the 1970s).

"The values for S, M, and L are obtained by integrating the spectrum of a light beam with the published sensitivity curves of the three receptors. The z coordinate represents luminosity and is obtained by integrating the spectrum with a fourth "overall sensitivity" curve that peaks at green wavelengths."

This terminology is not familiar to me: "integrating this with that", what does that mean? I think the proper way to say it is "integrating over the product between the light spectrum and the sensitivity curve."

Finally, seems like one thing missing here is at least a brief discussion of opponent process theory.

Go ahead! I've no clue what that is. -- Hankwang 00:21, 6 Mar 2004 (UTC)

Anyway, if people agree with me on these things, change them and then you can delete it from my list here too. Speaking of which, what's the policy for deleting old posts here?

I'd prefer to archive discussions instead of removing them. -- Hankwang 00:21, 6 Mar 2004 (UTC)

Chinasaur 23:17, Mar 5, 2004 (UTC)

I am not sure about the paragraph about people having two types of cones in Color reproduction section. I added it, because I know these people can see some things more easily. i. e. look at the first picture at http://webexhibits.org/causesofcolor/2D.html, but is not it related rather to discerning shapes or textures? I am not sure this is about colors, so I deleted the sentence. In contrast to dogs, cone types of the people are usually a subset of these found in people with normal color vision. Perhaps it should be moved to some other page and talk about shapes? Or is this something about neural signals? Anyone knows these topics? Artjt 6 Mar 2004.

I wasn't too sure about that statement either; I'm glad you changed it; the new version we can be more confident is right.

It is true that for someone with anomalous sensitivities you might get better difference threshold detection in some ranges that normal people don't. The way people usually generate plots of difference thresholds is based on assumed "standard" sensitivity curves. We can probably find one based on an anomalous set of cones that will predict anomalous difference detection.

In terms of color spot tests, I know some protanopsia (missing the red cone entirely) tests where the protanope misses the number normals see, but instead is able to see a number pattern that normals miss. But I don't think this is better color detection, just pattern detection. I dunno about spot tests designed for anomalous cone tuning, but I doubt it's any different. The page you sent either doesn't appear right on my monitor or is designed for deuteranopes or anomalous cone tunings; it doesn't seem to work for me (protanope).

Chinasaur 19:24, Mar 6, 2004 (UTC)

Requesting peer review of dominant wavelength. --Chinasaur

Will someone clarify the bit about web-safe colors that reads "In earlier years of the Internet browser card computer screens"? I have no idea what that means, and I remember the days of HTML 3.2.

Disturbing new edits by User:Information-Ecologist

I did not revert, and won't have time to make carefully considered edits in the near future, but I wanted to draw the community's attention to the latest edits and explain why I find them disturbing.

My main source of concern is that many of the edits (here and to related articles, see RGB color model, or to cut to the chase just see the contribution list (http://en.wikipedia.org/w/wiki.phtml?title=Special:Contributions&target=Information-Ecologist) for Information-ecologist or for his earlier login Ecology2001 (http://en.wikipedia.org/w/wiki.phtml?title=Special:Contributions&target=Ecology2001)) are drastically self-promotional. It almost seems to me that the main goal of the edits is to create as much interlinkage as possible to a fairly small set of external websites that are all fairly obviously owned by the same person. If you don't see where I'm going with this, please refer to Google bomb. Along this vein, please note the page on Collective_intelligence_agency, in particular the objectives section.

At the same time, this editor has put a considerable amount of work into edits unrelated to his personal websites, so it's not clear to me whether he is simply overzealous for my taste, or intentionally manipulating the system to generate higher page ranks. I don't want to jump to any conclusions about the editor's intentions, and I fully hope that he will appear here shortly to answer these concerns.

I don't mind also mentioning that I find most of the discussion of "peace cubes", etc. to be gibberish. However, this is not a value judgement on the content itself; perhaps it is simply my ignorance which makes some of the latest additions nonsensical to me. Nevertheless, it may be a value judgement on the way the edits were approached: I believe it would be better not to add information on the subjects of "peace cubes", and related topics to this article until the meaning of these concepts is more clearly laid out elsewhere. At present, when I find these terms in the color article, and try to understand them by following available links (including to external pages) I am still rather at a loss. Without intending to offend, I would go so far as to say that the whole thing seems almost like a bad joke. If the intent is in fact serious, then it should be clear from my experience why the over-hasty addition of all these linkages before the concepts are well defined in their own articles is not desirable.

At any rate, I congratulate Information-Ecologist on discovering Wikipedia and making a whirlwind of successful first edits. I hope also that he will understand the reasons for my concerns, and that he will be willing to argue his case for these latest edits, since at present I remain skeptical.

--Chinasaur 10:19, Mar 23, 2004 (UTC)

This is starting to make sense now. This guy has been over at en.wikibooks.org too. We were all sort of wondering what to do with his "books"... but now it strongly appears that they are computer-generated via a Novell DataPerfect relational database. In other words, he's a damn tough wiki spammer. His website needs to go into the banned URL list. AlbertCahalan 05:06, 19 Feb 2005 (UTC)

Edits 23 March 2004

This section by: Hankwang 10:43, 23 Mar 2004 (UTC)

CMYRGB color table

I removed the CMYRGB color table. I don't see what it adds to the article. I moved the original spectrum table to the section "physics of colors", where it is discussed.

Web colors, Netscape

There are pages about web colors and such. I feel that it dives too much into implementation details (i.e., how do you define a color on a web page in the year 2004.

Virtual light & color cubes

I removed this section:

The RGB color model defines a virtual three-dimensional space for color that can be made visible as twin virtual light & color cubes - each a mirror image and photographic negative of each other.

Within each virtual light & color cubes, the color at any point is defined as the sum of its red, green & blue coordinates, where the expression #rrggbb is understood as a three-dimensional sum of the red, green and blue coordinates, ordinal

I cannot make sense out of this text. What is virtual light? or is it virtual (light&color) cube? "Virtual" seems to be a synonym for "mathematical" or something like that.

Color tables with grey background

I originally put in the example of RGB colors on a black background to illustrate the difference between coloring power and brightness wrt the CIE color model. A discussion about visibility on grey and white backgrounds breaks the discussion. I moved it to a section "Contrast", but actually I think it really belongs in an article about ergonomy.

To do: part of the remark about arbitraryness in CIE applies to the tristimulus space as well.

Color Coordinates

In the next couple of days (when I'm feeling bored), I'll add color coordinates in RGB, CMYK and HSV to each of the color articles.

RGB values I know off the top of my head. In cases where I'm uncertain I look at the nice color bar someone has already added to each page. :-)

HSV I get with the help of the GIMP color picker , which calculates the values for me, not that that's so hard to do. There are also formulas on the [[HSV] page you can use to check me.

Many CMYK values are really obvious, and I can do those from the top of my head too. If I'm uncertain, I can always use the formulas at CMYK.

I'm less familiar with other color systems. Maybe at some point the lists can be replaced with tables containing coordinates in all kinds of different color systems.

Note that currently I'm specifying colors using as integers in the range <math> [0,2^8-1] <math> This is because computers often express colors with 8 bits per channel (page hasn't been created (yet), so see bits per pixel for now). This is the most common way to express colors on computers, so I'm using that.

For completeness, there are several other methods to express the same colorspace: namely proportion, percentage, different # of bits per channel etc...

I hope this is helpful! Kim Bruning 22:48, 3 Apr 2004 (UTC)

I would have scaled all numbers to the range [0,1], except maybe the hex triplets. As you say yourself, computers often express colors in 8 bits, but high-quality digital images (e.g. directly scanned from a photo negative) may very well be expressed in a higher resolution. In five years from now, you may have to update all your numbers to 12 bits, with a remark about older computers, and another ten years later to 16 bits... -- Hankwang 13:24, 26 Apr 2004 (UTC)

Some code

I got fed up calculating everything by hand, so I've used my (as yet terrible) python talent to write some code to help me.

If you want to use it: Run the programme , enter the hex triplet (you can look it up on the color page itself) and cut-and-paste the output into a color page.

Note that the leading spaces on the lines are significant.

Feel free to wiki-edit code and colors and soforth :-)

see: User:Kim Bruning/colorspace.py

modified all color pages

I've now managed to modify all the color pages to show coordinates in 3 colorspaces: RGB, HSV, CMYK. I noticed some irregularities in the pages while doing that, so please check to see if they're correct. :-)

I got a lot of help from Dysprosia for CMYK, and an amazing amount of help from Chinasaur sorting out the HSV oddness.

Have a nice day! Kim Bruning 14:32, 7 Apr 2004 (UTC)

Eeeew, you should remove those. CYMK and HSV are not well defined. CYMK is device-specific. The CYMK for one printer is different from the CYMK for another printer. With RGB, at least you can rely on sRGB to be well-defined. HSV is just plain nuts. If you want something reasonable other than RGB, choose CIE-XYZ or CIE-L,a*,b* instead. This is in fact how the USA specifies highway sign color paints. AlbertCahalan 01:13, 24 Apr 2005 (UTC)

I'm working on another approach to this, so that the CMYK and HSV can be supported with qualifications. (Of the coordinates presented, only the hex triplet can really be said to be device independent. Even RGB is potentially subject to different ranges on different machines.) See my proposed article explaining some of the differences. The idea is that the various coordinate lists would point into this article (which would be at Wikipedia:Normalized Color Coordinates) rather than into the individual articles. Comments are welcome at my talk page. CoyneT ^talk 00:31, 26 Apr 2005 (UTC)

Violet

I think it would be useful to have some explanation as to why the violet area of the spectrum is perceived to be a similar color to that of purple (a mixture of red and blue). The reason being that the red cones in the eye are sensitive to the wavelengths in the violet region as well as the longer wavelengths in the red region of the spectrum. Violet is therefore perceived as a stimulus to both red and blue cones. [anonymous, 25 apr 2004]

added a paragraph to violet -- Hankwang 12:36, 26 Apr 2004 (UTC)

CIE coordinates

I think the stuff about the LMS and XYZ coordinates is based on misconceptions (my misconceptions, since I wrote most of it). See CVRL Color & Vision database (http://cvrl.ioo.ucl.ac.uk/index.htm).

Phase perception

I just added this sentence: "Humans cannot perceive phase effects of light except in special cases of interference (e.g. see thin-film optics) where phase effects lead to perceivable amplitude changes." But actually I'm fairly certain that this is true of all biological color vision systems. I just can't think of a good way for any biological system to use phase information constructively (except in very special cases), so I would assume none do it. Thoughts? --Chinasaur 18:00, 11 Jun 2004 (UTC)

Squid can handle phase but not frequency (wavelength). I can imagine that this might have some use in detecting directional orientation toward the ocean surface. AlbertCahalan 01:09, 24 Apr 2005 (UTC)

Excuse me, that would be polarization. AlbertCahalan 00:19, 11 Jun 2005 (UTC)

Proposing a move

I think this article should be moved to Colour and all the instances of "color" in it changed to "colour". As the first sentence of the article states (Color (American English) or colour (most other forms of English, including Australian English, British English, Canadian English and New Zealand English)), "colour" is the spelling used in the larger portion of the world. blankfaze | (беседа!) 01:45, 26 Jul 2004 (UTC)

If you think it's worth the bother, I have no problem using the British spelling although I may slip up occasionally. Other considerations:

I suspect the majority of the readers of the English wiki are American (any numbers anywhere on that?).
Do the other frequently contributing editors have any preference?

—Chinasaur 22:40, 26 Jul 2004 (UTC)

International standards (e.g., ISO, ECMA, usually use Oxford University Press guidelines (which in this case would suggest 'colour', but in other cases such as -ize/-ise endings use the American spellings)). A quick Google shows 'color' has about 20% more hits than 'colour', but that's probably an artifact of the Internet having started in the USA. As the rest of the world catches up, the USA numbers will soon be in a minority, so it would make sense to make the article as easy to read as possible for the majority. Many uses of the word in the article could be replaced by 'hue', for example, which is more neutral. quota

Regarding the Google search, I think it's a tricky issue; if you look at the hits for "colour" you'll see that many of those pages don't use that spelling anywhere on them, or in the keywords, or anywhere in the source. There are various reasons this might be, but anyway my impression is there's actually more of a bias towards "Color" than you observe above. The international standards issue is intriguing, although I don't understand why Oxford would recommend a hybrid standard.

Google hits: yes, a vague guide. But it is certainly true that Brit English is used by more people that USA English. Consider India... [Oxford not really hybrid, more a simplification—separate discussion.] quota

In the end I don't really see what the big deal is. There's thousands of other articles on wikipedia where one convention or the other has been adopted (although perhaps not so many where the title word itself is in dispute). Are you going to Briticiz(s)e all of them?

It's the same issue as date formats -- in fact maybe that's the solution. Readers should see the format (spelling) they want. quota

I don't even think the readership is the chief concern.

Ouch. What's Wikipedia for if it isn't to be read?

The only really compelling reason I could see to make a change is if editors were constantly saving the wrong spelling accidentally and having to reedit to bring things back to conformity. This isn't really a problem currently. --Chinasaur 17:28, 19 Aug 2004 (UTC)

It certainly is for any editor more used to the 'colour' spelling. Or maybe they don't count?

Probably better not to interleave your comments. Since I feel like I should be defending myself at this point... First, I "don't think the readership is the chief concern" because I don't think they really care about this issue. Not enough to outweigh the inconvenience to editors. If editors are inconvenienced, the potential for damage to the article and thus to the reader is much more serious than making them read "color" or "colour" IMO. Second, I didn't even begin to suggest that British spelling editors don't count, I said they don't seem to be having a problem remembering the English spelling. Anyway, that's the end of my contribution to this immaterial controversy. --Chinasaur 06:22, 20 Aug 2004 (UTC)

Given that there are only 402 million english first speakers, and there are 290 million english speakers in the United States, I would suggest that more people spell "color" than "colour." Based on that fact alone I would suggest that "color" be used. It's not that the other people don't count, but simply that there are 'more' people that spell it color. While I understand the majority of people that speak english as a second language would use the british spelling (i 'think', don't quote me on that), those people can usually use their specific version of the wikipedia.Ctachme 21:38, 19 Aug 2004 (UTC)

Well, as I said above .. seems like the right answer here is to have the word marked up, like dates, so it can be seen by the reader in the appropriate format. It could then, in fact, be spelled either way in the source, for the convenience of whoever is editing, but be presented consistently in one or other way depending on the locale of the viewer. (In fact Wiki could do that without markup being needed for the relatively small set of differently-spelled words: colour/color traveling/travelling, etc.) quota

I object on the following grounds:

The majority of citizens in the English (http://en.wikipedia.org) Wikipedia are probably American, and...
The Wikipedia servers are housed in Florida, which is in the United States. Because this is a subject which is not specific to any country, American should be the default spelling.
If it is moved, it is likely to be moved back anyway by an American. (Why? WMD hidden among the wording? :-))

—Mr. Grinch (Talk) 17:46, 18 Aug 2004 (UTC)

I suspect most of the readers of the English Wikipedia will not be American. Because of its much larger size than most other language Wikipedias it will attract plenty of people from non-English speaking countries who speak English. There may be a majority of Americans amongst first-language English-speakers, but why does it matter whether it's people's native tongue or not? There are also millions upon millions of English-speakers in India who would spell this word "colour". — Trilobite (Talk) 04:07, 7 Sep 2004 (UTC)

Moreover, there may be an artificial increase in hits for "color" since it is used in HTML and other markup languages. It would be difficult to separate this from the real use of the two spellings. Stemonitis 12:32, 4 Mar 2005 (UTC)

Truth about the above 7-section color discussion

The truth about the discussion above is at Talk:Complementary color. 66.32.241.40 02:34, 8 Aug 2004 (UTC)

Disambig on specific color pages NEEDED!

White has sections 'People whose surname is white' and a bizzare 'Other definitions' with game fictional character. Most of red article is taken by Red 'Usage, symbolism, colloquial expressions'. Green and black...let's not even talk about that. I am not even going to look at pink.

What all those pages need is wikifying with (or into) a disambiguation. I'd personally leave only references to color in physics (spectrum and stuff) on those pages and move all other info to other pages, leaving a disambig on the top. Definetly all cultural/language/biology/etc. references (not to mentions surnames, towns and stuff) should be moved to a diffrent page(s). --Piotr Konieczny aka Prokonsul Piotrus 17:53, 10 Aug 2004 (UTC)

Tint, hue, shade, value

Wow, a whole scientific piece with not a word about these terms as used by artists--! Who volunteers? Elf | Talk 00:15, 2 Sep 2004 (UTC)

Color constancy

I just reverted this quote out of the 1st paragraph of "Color reproduction"--but it looks like it might be a useful term that someone might want to define? Elf | Talk 16:05, 24 Sep 2004 (UTC)

Template

Please make feature for copper in the template below. 1 Oct 2004 (UTC), Heegoop

Page move (not done)

(from WP:RM))

Color → Colour

Many editors have attempted to use the more common international spelling of this word for this article, but this has been blocked by the minority who say 'please use the spelling in the title of the article'. But the spelling in the title cannot be changed by a Move because of the redirect. Can we please switch the redirect and the article?

[Obviously cut-and-paste can be used, but that's messy for a big article like this.]

In general .. surely Move should be able to switch a redirect? Thanks! quota

Rejection. While (as an American) I do prefer British as opposed to American spellings, there is no reason to waste time on such bickering between the two versions of English spelling. There is no institutional attempt to internationalise spellings. Some people write in British English, some in American English. That does not indicate an organized movement to adopt one or the other. Someone wrote the article, titled it "color"...let is stand as written. The other spelling "colour" is referenced in the first line and the redirect suffices, so there's no confusion that will be born from leaving it as it stands. There is no reason to bicker over it, because color is not an exclusively British phenomenon (as was an earlier requested move, when someone asked that Euroscepticism be moved to the Americanized spelling Euroskepticism—which was overwhelmingly rejected). The only reason there are two spellings in English is because the American English form is closer to the Latin root of the word, color, while the British English form is still stuck with words like colour, honour, valour, etc, as influenced by the spellings of the Isles' Norman French invaders—the bastard offspring of the French -eur ending.

You should check your history. The Norman invasion was in 1066. The spelling color was in use in England through Shakespearean times, and the 'u' spelling didn't become common until later (after the USA was colonized) - though it was in use as early as 1350.
- Actually I know my history and my linguistics well. The Middle English period, where the language was heavily impacted by the French the Normans (who were actually Norsemen) brought with them, started after the invasion, in 1066 to roughly 1500 AD. Shortly after, the Elizabethean period brought forward a shift in vowel pronounciation, but the -our endings (based from the French) had been cemented into English spellings centuries before. Also, sign your comments. —ExplorerCDT 17:56, 11 Dec 2004 (UTC)

There are important things to waste our time with, quota...this is not one of them.

There is constant churn in the 'Color' article due to this problem. That is the waste of time.
- Agreed, thank God I have nothing to do with that. And Wikipedia has a policy about bickering over British vs. American English spellings. Follow it, and you'd save yourself a lot of grief. Sign your damn comments. —ExplorerCDT 17:56, 11 Dec 2004 (UTC)

Lastly, I take considerable umbrage at your pre-emptive characterization of the people who would potentially reject this as a "minority." I doubt 280 million people in the United States, and the myriad millions of people who learn the American English spellings

10,000,000,000? You exaggerate.
- Again, sign your damn comments.
  - apologies; thought that would be in the history [and there was no need to swear] quota
- Never said 10 billion.
  - Huh? A myriad is 10,000. 10,000 millions is 10 billions. quota
    - Myriad as 10,000 is an archaic usage not used since the 19th century. Today, the adjectival form takes precedence, meaning a vast number or composed of numerous diverse elements or facets. —ExplorerCDT 19:39, 11 Dec 2004 (UTC)

- It is estimated that just over 400,000,000 people speak English as a first language, worldwide. 68% of these live in the United States. Also estimated that 350,000,000 to 1,000,000,000 more speak it as an additional language. So, at least 750,000,000; at most, 1.4 billion. I said...280,000,000 + a myriad millions + India, and others...not enough numbers there to make 10 billion, but enough of a difference to recommend you repeat kindergarten and learn how to add again. —ExplorerCDT 17:56, 11 Dec 2004 (UTC)
  - Since you seem to have descended to ad hominem 'arguments'; I'm unwatching this discussion. There's no point in attempting a discussion with the irrational. quota
    - Okay Thrasymachus, pick up your toys and go home because of a tongue-in-cheek comment. —ExplorerCDT 19:39, 11 Dec 2004 (UTC)

as a second tongue would be a "minority"...considering even India (with their billion or so folks) tries to officially favour American English over the version which a small island nation of 70,000,000 people tried to enforce over them.

You seem to be forgetting the European Community, which uses International English as its second language. And China. And Australasia. Exactly where, outside the USA and its colonies, is American spelling taught?
- I didn't forget them, just didn't mention them individually (hence, and others abovestated). International English, which doesn't exist as a single solitary "language" (see the article) and is (according to the MLA) mostly limited to international academia...Most of the impact of English on foreign non-English populations is a combination of the influence of Ogden's Basic English and Voice of America's Special English on the non-English speaking world...and those are dependent mostly on American English spellings, the former being endorsed by Winston Churchill as an international language. Notice, they both spell it "color." Also, Europe (and Japan) after World War II saw a rise in non-English speaking people's learning English. However, the level of impact has been linked to which army controlled which sector (the British, Canadian, and American) and most recently the invasion of American commercial culture throughout the world. —ExplorerCDT 17:56, 11 Dec 2004 (UTC)

Like it or not (I, for one, don't), American English is the lingua franca of the world today.—ExplorerCDT 21:52, 10 Dec 2004 (UTC)

object. Wikipedia:Manual of Style says keep the status quo - no reason to go through the difficulty of a move for an unimportant change. And what Explorer said. --[[User:Whosyourjudas|Whosyourjudas\^talk]] 23:52, 10 Dec 2004 (UTC)
Object, see above. [[User:Neutrality|Neutrality/^talk]] 02:56, Dec 11, 2004 (UTC)
Object. Stare decisis. Jonathunder 20:01, 2004 Dec 11 (UTC)
Object. Both Google.com and Google.co.uk get significantly more hits for "color" than for "colour." --LostLeviathan 01:32, 12 Dec 2004 (UTC)

subject ('cos I subjectively object, and I'm feeling whimsical) - UtherSRG 03:29, 12 Dec 2004 (UTC)

Agree. sure, why not. The original word was colour, and is still used by many people people in the world. Canada uses it, the UK uses it. Anyone can understand either. I say agree, but if it stays where it is, it wont make a tiny bit of difference. SECProto 21:11, Dec 12, 2004 (UTC)
- The original word was color from the Latin. The French screwed up the system almost 1500-2000 years later. —ExplorerCDT 21:51, 12 Dec 2004 (UTC)
  - That might be. But when the word was first adopted into the English language, it was spelt colour... Unless the English adopted color from latin, and then changed to colour around the time of William the Conqueror? SECProto 00:30, Dec 13, 2004 (UTC)
Object. We already have a good policy about these spellings. Sometimes you have to work nicely with others, Quota. --Yath 22:58, 12 Dec 2004 (UTC)
Object. Since Wikipedia is an international project, it may make sense to standardize on international English spellings—but there is well-established policy not to attempt to standardize either U.S. or international spellings. Unless this policy changes, the proposed move is inappropriate. FWIW, OED gives both spellings (with preference, of course, to the spelling with a u); the chronologically first spelling given is colur (13th–16th c.), follwed by colure and coulur (14th c.), coloure (14th–17th c.), colour (13th c.–present), and color (15th c.–present). —Tkinias 02:29, 13 Dec 2004 (UTC)
Object: DCEdwards1966 04:31, Dec 13, 2004 (UTC)
Object to the move. We have a policy on which spelling to use, and in this case, it says to stay at the spelling which the article was originally written in. Try to gain consensus to change the policy, if you like. Tuf-Kat 00:10, Dec 30, 2004 (UTC)
Object The academic standard is to use the American spelling. While it makes sense to use metre or litre since those spellings are probably used more often, this is not worth the trouble. Drunkasian 00:46, 1 Jan 2005 (UTC)

Footnote: today there was a rash of unexplained changes from color to colour in this article, in Pantone and Cyan. It was a new IP address, so I think it was someone who hadn't seen this discussion.

Editing the section "Color vision"

I have done some careful tidying in this section, and added a couple of things I thought were needed. I have added a "See also" section at the article Color vision, and put this comment in Talk there:

While the first word of the article provides a link to the article Color, I felt it desirable to mark that link in a "See also" section, because there is substantial information on color vision at that article to assist those wanting an accurate summary in the general context of theory of color. Is there any principled objection to this? It seems to me, as a relative newcomer to this whole color-and-vision domain in Wikipedia, that the domain needs considerable careful work, and disciplined re-organization.

I think much of this article is still in need of fine-tuning for clarity and strict accuracy, but looking through the history and the discussion here I am not sure how this can be achieved. I hope it can be achieved, because people have obviously put in a great deal of fine work already. We'll see. --Noetica 23:57, 5 Mar 2005 (UTC)

Indigo

At what point in history did the spectrum of colours change from Red, Orange, Yellow, Green, Blue, Indigo, Violet to Red, Orange, Yellow, Green, Cyan, Blue, Violet, or is this just some personal preference which differs from Newton's? Arcturus 21:22, 31 Mar 2005 (UTC)

An edit in Oct. (http://en.wikipedia.org/w/index.php?title=Template:EMSpectrum&diff=7159784&oldid=6667483) on Template:EMSpectrum implied that substitution. I also saw it claimed somewhere that "cyan is a much more significant color today", IIRC defending an equivalent change. --Jerzy (t) 00:46, 2005 Apr 1 (UTC)

Newton was an 18th-centurian and at that time (IMO) the word "cyan" didn't exist in the English language. Today we have more words and use them wherever they are sensible. Which version makes more sense to you?? (Please explain your answer in whatever detail you can.) Georgia guy 22:14, 31 Mar 2005 (UTC)

Neither version makes any sense. If you want sense, read on:

The 7-color spectrum including indigo is two things:

Newton's forcing onto his observations a naming scheme that has nothing to do with physics, and is at odds with human vision, to satisfy his metaphysical notions (alchemy etc.) whose coffin his scientific work was driving major nails into.
Physicist's nostalgia for the details of how Newton expressed his observations, which is totally unjustified by the deserved respect they hold for his experimentation and mathematics.

The 7-color spectrum including cyan is simple foolishness; as noted implicitly above, cyan doesn't substitute for indigo, it's closer to cyan renaming Newton's "blue", and "blue" renaming Newton's indigo. The only smidgen of sense it makes is if you accept the stupid, magical idea of 7 colors.

This is not to say that cyan has no role, but this particular role is utterly nonsensical.

The only two natural, human-vision-related spectra of interest are:

Red Orange Yellow Green Blue Violet
Magenta mumblemumble Cyan mumblemumble mumblemumble mumblemumble

They are called the subtractive (pigment) and additive (light) schemes respectively, and the reason i mumbled about the additive one is that i and most of us have little intuitive grasp of the additive version, tho it is perfectly valid. The bold colors are the primary colors of the respective spectra, and the other colors are secondary colors resulting from combining the primaries in pairs, subtractively or additively.

One six-color spectrum would do the job in the EMSpectrum template well; IMO two six-color spectra would be somewhat better. Most important to say is that color is less about a linear spectrum as at Color#The physics of color than about what would probably work reasonably well in the template: showing the relationship among the colors as primaries and secondaries. Ideally, i'd like to see the template include two star-of-David patterns with six color patches each:

Subtractive:
- Primary triangle has R,Y,B with big labels
- Secondary triangle has O,G,V with small labels
Additive:
- Primary triangle has M,G, and whatever ("Blue-violet"?) with big labels
- Secondary triangle has the secondaries with small labels

As to the linear spectrum of Color, 7 colors including indigo serves no purpose but nostalgia; 7 including cyan serves no rational purpose whatsoever. 6 colors is still a poor illustration of the continuous spectrum, far worse than a graphic that approximates a continuous spectrum. (You can't display a true continuous spectrum on modern monitors, because they are pixel-addressed and the colors are specified digitally for each pixel.)

I'm at a loss as to why anyone wants to label specific colors on that spectrum anyway. (And how crucial can an illustration be, that is garbage for the colorblind 5% of the population?) Blue is not primarily a color anyway, but a range of hues between yellow and green; why promote arguments about which point in the blue range is the real blue by taggiing specific hues with such vague labels?
--Jerzy (t) 00:46, 2005 Apr 1 (UTC)

The "change" is really more practicle than real. Indigo, which is described in dictionaries as a "shade of blue-violet", is really pretty much all that we can see in the visible spectrum. The effect occurs because the response of the blue cones in our eyes is relatively poor, compared to the other cones, and the red cones show an increasing response to wavelengths above blue. (See this article (http://www.everything2.com/index.pl?node_id=1707879&displaytype=linkview&lastnode_id=1707879).)

The result is that, as the visible spectrum moves to wavelengths shorter than 440 nm the blue cone response (which is tailing off rapidly) begins to approach the increasing red cone response. The result is equal or nearly equal response from both sets of cones, which our brains interpret as violet - almost, because if you are looking at a real spectrum, the mixed response never really lets you reach a violet unless the background is extremely dark. This image from the main page is a fairly accurate rendition of what you would see from a prism in sunlight:

The increasing prevalence of naming violet as the "7th" color derives from the behavior of computers. In the continuous spectrum, yellow light corresponds to a discrete wavelength, such as 577 nm. In computers, we have no easy means of producing light at just any wavelength we like: Instead, we are limited to producing 3 wavelengths which more-or-less correspond to the human cone-response wavelengths of 420 nm (blue), 534 nm (green) and 564 nm (red). To get white, we mix proportions of these colors based upon cone response, in order to emulate the white which would be seen from the continuous spectrum of the sun. If we delete, say, blue from this mix, our eyes are fooled into thinking the resulting color is yellow, because they are seeing the same response from the red and green cones as would be produced by a light of wavelength 577 nm. This is a rendition of the computer color spectrum, showing the relative intensities of each of the three colors which are combined:

Missing image
Computerspectrum.png
Image:Computerspectrum.png

If we then take the same white and delete the green wavelength, our brains interpret the resulting strong response of the red and blue cones - and absence of green cone response - as the violet which we can almost see in the real spectrum. As a result, the computer color spectrum, reduced to various percentage mixes of the three colors, results in a "color wheel" which starts at red, moves to yellow then green, cyan, blue, violet and back to red. (If we used 4 colors in a monitor, the response would not be nearly so simple to represent.)

So, following the full-strength model, red full-strength yields red; red+green yields yellow, green full-strength yields green; green+blue produces a strange color called cyan; full-strength blue produces blue; and blue+red yields magenta (violet). So the computer color spectrum could be said to be red-yellow-green-cyan-blue-violet. Orange isn't listed, because on the continuous-mapping spectrum, it is really quite narrow: It is hard to produce a good orange on a computer. Likewise, indigo is quite narrow.

Humans being what they are, what we see on the computer has begun to enter the community conscience as the "proper" spectrum; hence the addition of the colors cyan and violet.

But, most people who distinguish electromagnetic (as produced by a prism from sunlight) response from computer emulation, though, would still describe the electromagnetic spectrum as red-orange-yellow-green-blue-indigo-violet, with violet being very weak. This is because, in this spectrum, these are the colors most easily distinguished. Cyan is not easily distinguished in a rainbow, for example, because of the extremely strong response of the green cones, which tends to "wash out" the blue cone response; while, again, the strong green response tends to create a very prominent orange color distinct from the yellow.

The "red/yellow/blue" list comes from early versions of subtractive color (http://en.wikipedia.org/wiki/subtractive_color) paints: These were traditional artist colors before (absolutely no offence intended) the color model of the eyes was well understood. Say you mix red and yellow paints: The red absorbs green light and (weakly) blue light, and reflects red light. The the yellow absorbs blue light strongly, reflecting red and green. Therefore the mixture, when we look at it, appears orange because of the reflected red and green light (again, this will not be a true continuous spectrum).

Now, we know that better subtractive colors are magenta (not red), yellow, and cyan (not blue). The problem with red/yellow/blue is that it is really not possible to mix them and get a true black or even anything close, because the absorption/reflection doesn't correctly match the wavelengths in our eyes. Magenta/yellow/cyan produces a much better (if not perfect) response and therefore can accurately represent more colors.

As for why we name segments of the spectrum: It is much easier to say things like, "I want a red car;" as opposed to, "I want a car with a spectral response peak around 670 nm." We name everything else, why not colors? CoyneT 22:07, 1 Apr 2005 (UTC)

Color naming does appear to follow some patterns ([1] (http://www.pnas.org/cgi/reprint/1532837100v1.pdf)). Newton, however, chose seven colors because it was a good number (seven notes in the diatonic scale, etc), two of those names where determined by social factors: Indigo was an important product (used to dye things blue-purple) and Newton was good friends with the Duke of Orange. Hyacinth 22:50, 1 Apr 2005 (UTC)

It seems from this debate that any division of the optical spectrum is bound to be arbitrary. I don't see why Wikipedia should come along and change 400 years of custom (seven colours including Indigo). Schools all over the world teach ROYGBIV. What are the children to make of it if they come to Wikipedia and find ROYGCBV? If there are no serious objections I'll change the template accordingly. Arcturus 10:54, 9 Apr 2005 (UTC)

I, for one, do have a serious objection and request the template being fixed to what was suggested above:

Red Orange Yellow Green Blue Violet

The whole point of wikipedia is to reflect the way that humans interpret and know the world, as such since that is the most widely used spectrum, it is the most logical thing to use in wikipedia. In my entire education (be that art class or physics) class, this was the spectrum used, and so-called Indigo was never used once. I agree that the decision is arbitrary, but wikipedia is not "prescriptive" it's "descriptive." Since the 6-color spectrum is what is used my the majority of people, that's what we should describe. --Ctachme 00:53, 19 Apr 2005 (UTC)

Move the Color Model section

I suggest that we move the color model section to it's own page, it is a separate concept enough and this article is pretty long as it is. --Ctachme 02:32, 16 May 2005 (UTC)

I agree - and I think that it and the color space article should be merged, but whether under the title "color space" or "color model" I don't know. PAR 02:51, 16 May 2005 (UTC)

Text running over onto image

in the Physics of Color section. I dont know how to fix it... Jaberwocky6669 20:14, Jun 10, 2005 (UTC)

(Broken color page.png) That's a picture of what it looked like to me (Firefox, 1024x768) Cuahl 20:51, 11 Jun 2005 (UTC)

I used to see that too. Then I added a 1em left margin to the table and it fixed the problem. Is anyone still seeing it like that? — Chameleon 21:02, 11 Jun 2005 (UTC)

Missing image
Broken_color_page_(fixed).png
Image:Broken color page (fixed).png

Looks great mate, cheers, Chameleon! Cuahl 21:27, 11 Jun 2005 (UTC)

Color Patents

Is it possible to patent a color? 68.40.169.172 03:10, 14 Jun 2005 (UTC)

Probably not a single color, as it is easy to demonstrate prior art if existing technology can make the color. (On the other hand, how could parts of genomes ever get patented?) However a novel use or arrangement of colors might be patentable if it is an invention. Also, a color's name could be protected by trademark legislation, and lists of colors perhaps even by copyright. (See Pantone). Any design made using colors is potentially protected by copyright, and might be registered as a trademark. Even the use of a single color in specific contexts can be trademarked. Trademarks are often better than patents as they do not have a set expiry date. (I am not a lawyer, just speculating) Notinasnaid 12:39, 14 Jun 2005 (UTC)

Retrieved from "https://academickids.com:443/encyclopedia/index.php/Talk:Color"

Talk:Color

2004

Featured Candidate

It's December, a good time to reconsider the name of this page

Metameric colors

What are the primary colors??

Documenting recent color-related anon 66.32... contributions

Red

Green

Various color-named articles

This page (Talk:Color)

Talk:Red

Talk:Green

Various color-named and color-system-named talk pages

Color wavelengths

Color Wheel

Should "cyan" be "00FFFF" rather than "00E0E0"?

Suggested changes

Disturbing new edits by User:Information-Ecologist

Edits 23 March 2004

CMYRGB color table

Web colors, Netscape

Virtual light & color cubes

Color tables with grey background

Color Coordinates

Some code

modified all color pages

Violet

CIE coordinates

Phase perception

Proposing a move

Truth about the above 7-section color discussion

Disambig on specific color pages NEEDED!

Tint, hue, shade, value

Color constancy

Template

Page move (not done)

Color → Colour

Editing the section "Color vision"

Indigo

Move the Color Model section

Text running over onto image

Color Patents

Navigation

Search

Toolbox

Personal tools