Himba color naming

[zompist]

Taking a loot at the Wikipedia page, subtractive primary/printing cyan is clearly blue, additive secondary/display cyan is more blue than green, [...] .

Er, those are exactly the same color. You do realize that the diagram of 'printing cyan' is appearing on your monitor, where it is displayed using 'display cyan'?

Maybe I'm misunderstanding, but then why do they have different hex codes and different places in sRBG, CMYK, and HSV color spaces?

I was looking at the diagrams, where they are exactly the same color, but I see what you're referring to now.

But, read the explanation under "process cyan". There is no hex code for printed cyan; printing doesn't use hex codes or RGB. They just darkened the cyan to look more like print. (Why they added some magenta is unfathomable. Cyan ink is the color of cyan ink, not cyan + a little bit of magenta.)

If you're looking at a printed page, the cyan will look plenty bright. If you put the printed page next to your monitor, it will look dark in comparison, because the monitor is basically a bright light. Our color perception is based on relative colors rather than absolute colors anyway. RGB #00FFFF and printed cyan are attempts to get the exact same color in different media.

(Also, #00FFFF just means 'turn on the green and blue bits full blast'; it doesn't specify a color in some absolute way. You can, after all, turn your screen brightness higher or lower! Turn it down low enough and it will match the printed page.)

[mèþru]

@Zaarin
I perceive teal to be more blue than green, and turquoise to be equally green and blue. However, what I call turquoise does not correlate to what most other people call turquoise.
@Chengjiang
I find it really hard to distinguish some types of oranges from yellows or reds, but "plain" orange is pretty distinctive for me. With "cool" purple, you get the whole dark blue/navy blue/indigo/bluish purple fiasco (not with warm purples however).

[Pole, the]

this thread is now about whether 'cyan' is a basic color term

Considering I see "cyan" as blue and everyone else I know insists it's green, I argue that it should be...

Cyan is a shade of grue. End of topic.

[JeremyHussell]

But, read the explanation under "process cyan". There is no hex code for printed cyan; printing doesn't use hex codes or RGB. They just darkened the cyan to look more like print. (Why they added some magenta is unfathomable. Cyan ink is the color of cyan ink, not cyan + a little bit of magenta.)

I think this might be because the standard reference-white paper is slightly yellowish compared to computer-monitor white. The extra bit of magenta ink thus balances the slight yellowness of the paper, making the printed cyan appear truly cyan instead of slightly yellow (or rather, slightly more green than computer-monitor cyan).

[Salmoneus]

(This apparently includes many pre-modern users of the term, since it used to be common to say cyan-blue, and Isaac Newton, whose blue in "red, orange, yellow, green, blue, indigo, violet" appears to refer to cyan with indigo referring more or less to RGB primary blue.)

But there is no cyan in the spectrum. Rather, cyan is a colour commonly used on computers to represent part of the spectrum (because our eyes interpret the two colours in a similar way), but imperfectly - and apparently this happens to be the area of the spectrum that is hardest to represent additively.
Spectral colours are monochromatic - they have a single frequency - whereas cyan is formed by the superposition of two different colours (green and blue).

[zompist]

(This apparently includes many pre-modern users of the term, since it used to be common to say cyan-blue, and Isaac Newton, whose blue in "red, orange, yellow, green, blue, indigo, violet" appears to refer to cyan with indigo referring more or less to RGB primary blue.)

But there is no cyan in the spectrum. Rather, cyan is a colour commonly used on computers to represent part of the spectrum (because our eyes interpret the two colours in a similar way), but imperfectly - and apparently this happens to be the area of the spectrum that is hardest to represent additively.
Spectral colours are monochromatic - they have a single frequency - whereas cyan is formed by the superposition of two different colours (green and blue).

I think you're not quite understanding metamerism here. When you perceive a particular color, it may be produced by light of a single frequency (as by a laser), or by two or three frequencies mixed together (as by a computer screen), or a messy set of many different frequencies (as by sunlight). There is an infinite set of frequency combinations that will produce any given color; these are metamers.

Of course cyan is on the spectrum. You'll find it at 480 nm, give or take 10 nm or so. You can produce it just fine with monochromatic light.

You can also produce it with a combination of unique blue and green. And by an infinite set of other metamers. (Look around the room; anything cyan you see, except from your computer monitor, is actually a very complex set of frequencies, produced either by sunlight or a light bulb and modified by the absorbance of the 'cyan' surface.)

(There are colors that aren't on the spectrum— violet, for instance. You cannot have violet monochromatic light. More strangely, you generally can't have red either. At most illumination levels, any monochromatic light you see has a bit of yellow in it.)

[din]

Our color perception is based on relative colors rather than absolute colors anyway.

God, yes.

Whenever I edit photos and want to change the saturation or brightness of something that looks totally green to me, I usually end up having to edit the yellows instead. Just a second ago a couch that appeared 100% teal to me turned out to extend through the greens and far into the yellows (to the point that there was very little actual teal), just because of the color of my flash and the lighting in the room. I mean, I know the couch appears teal in daylight, but it also appeared teal to me in that photo, but apparently it wasn't.

[Tropylium]

I learned turkoosi early on and pretty much accepted it as a relatively primary color (at least on the same level as orange and definitely more so than pink); and I've also already years ago moved towards considering cerulean a distinct color from blue; but I never seem to have settled on which of these "cyan" falls under. For example #00FFFF fits usually best within cerulean, but already #40FFFF bridges over to turquoise.

"Teal" is a mystery to me and many people's usage often strikes me as not even turquoise, just slightly greyish green.

[gach]

Of course cyan is on the spectrum. You'll find it at 480 nm, give or take 10 nm or so. You can produce it just fine with monochromatic light.

Such as this narrowband filter at 485 nm, though the image is only an approximate rendition of the true colour of the filter when you look through it. There's also this one at 501 nm, even if its colour looks more greenish to me when I handle these filters. Honestly though, you have to press me to care enough to call these anything more than shades of blue or green.

Colour perception is, of course, very subjective, and I don't see a reason to try to define colours independently from how we see them. In the wild and unruly world beyond our perception, the only thing you have are spectra. You'll catch people talking about things like the "blue" or "green" part of the spectrum, but these are simply handy colloquialisms and people will struggle to give robust definitions of them. The human eye tries to make sense out of the spectra by combining information from three types of sensors with different wavelength bands. This is obviously massively lossy as vastly different spectra can produce the same stimulus, i.e. look equally bright in all the three bands. You might want to define colours using more than these three bands, but that's a questionable choice since you'll end up straight away with colours that are distinct from each other but look indistinguishable to your eyes. On the other hand, you might want to separate "monochromatic colours" from "composite colours", but it's not clear to me how these would be perceived fundamentally differently from each other. And isn't talking about the wavelengths of light instead of "monochromatic colours" anyway much clearer?

[zompist]

On the other hand, you might want to separate "monochromatic colours" from "composite colours", but it's not clear to me how these would be perceived fundamentally differently from each other.

They aren't, that was the main point of my post. Your eyes can't tell a single-frequency light source from a mixture. (But a spectrograph can.)

[gach]

Yet another point is, how narrow would the wavelength band really have to be to count as monochromatic and not just a narrow composite band? Remember that truly single-frequency light is a physical impossibility, there's always some spread in there.

[zompist]

Yet another point is, how narrow would the wavelength band really have to be to count as monochromatic and not just a narrow composite band? Remember that truly single-frequency light is a physical impossibility, there's always some spread in there.

Hmm, I had thought that laser light really was monochromatic, but you're right, there's some spread. For a He-Ne laser this turns out to be about 0.01 nm.

The eye can distinguish between frequencies 2 nm apart, so from our point of few that's plenty monochromatic.

(Diode lasers have a much wider spread, more like 1 nm.)

[Zaarin]

I learned turkoosi early on and pretty much accepted it as a relatively primary color (at least on the same level as orange and definitely more so than pink); and I've also already years ago moved towards considering cerulean a distinct color from blue; but I never seem to have settled on which of these "cyan" falls under. For example #00FFFF fits usually best within cerulean, but already #40FFFF bridges over to turquoise.

"Teal" is a mystery to me and many people's usage often strikes me as not even turquoise, just slightly greyish green.

In my usage, "teal" is just dark turquoise. On the other hand, for me "cerulean" is high value low saturation blue--what Wikipedia describes as Pantone Cerulean or Cerulean Frost. To me, #00FFFF is waaaaay too green to be cerulean; I'd call it turquoise or cyan.

[Pole, the]

Trigger warning: anegdotical evidence

A few days ago I was in a shop and I saw a little girl with her mother near the cardboards full of oranges. The girl said Jakie żółte! „How yellow!”, after which the mother corrected her Pomarańczowe „Orange” (as an adjective). The girl saw the colour as yellow while it was orange for her mother.

So, even now there are some people not having orange as a basic colour, even if they are just children.