Okay, I think I understand where the crux of this disagreement is. Let me try to sum it up to make sure we're talking about the same things.
The initial point of contention was PatrThom saying that the rainbow has only three colours in it. To Pez, this cannot be correct because, if you take those three colours as being representative of three exact, specific wavelengths of light (or points on a colour wheel, or whatever system you prefer to use to discuss colour theory), there are obvious gaps between those three specific points that would then have to be missing from the rainbow. If you light a room with lights of those three specific wavelengths, there would be swaths missing from the full spectrum of colour that humans perceive. Yes?
If I haven't fundamentally misunderstood the discussion, the hangup to all of this is in that initial assumption that "red," for example, is only representing a singular wavelength of light (which I guess is where the discussion of pixel-created colour came in). The issue with that assumption, and the reason that there have been repeated examples of cultures that view colour differently, is that any of our colour words can be referencing much broader sections of the colour wheel than one specific point on it. "Red" could, given the right language and cultural context, be reasonably applied to a large section of it, say a third of the whole. Then "Green" could cover another third, and "Blue" the last third, leading to a rainbow of three "colours" that still contatins every wavelength of human-visible light, or every point on a colour wheel. It feels horribly imprecise to a first-language-English speaker who was raised in modern Western culture, but that's an effect of our biases, not any inherent trait of the light or of colours. It doesn't require genetic variation to see and describe a three-colour rainbow, even if it contains what we would describe as other visually distinct colours in it.
The initial point of contention was PatrThom saying that the rainbow has only three colours in it. To Pez, this cannot be correct because, if you take those three colours as being representative of three exact, specific wavelengths of light (or points on a colour wheel, or whatever system you prefer to use to discuss colour theory), there are obvious gaps between those three specific points that would then have to be missing from the rainbow. If you light a room with lights of those three specific wavelengths, there would be swaths missing from the full spectrum of colour that humans perceive. Yes?
If I haven't fundamentally misunderstood the discussion, the hangup to all of this is in that initial assumption that "red," for example, is only representing a singular wavelength of light (which I guess is where the discussion of pixel-created colour came in). The issue with that assumption, and the reason that there have been repeated examples of cultures that view colour differently, is that any of our colour words can be referencing much broader sections of the colour wheel than one specific point on it. "Red" could, given the right language and cultural context, be reasonably applied to a large section of it, say a third of the whole. Then "Green" could cover another third, and "Blue" the last third, leading to a rainbow of three "colours" that still contatins every wavelength of human-visible light, or every point on a colour wheel. It feels horribly imprecise to a first-language-English speaker who was raised in modern Western culture, but that's an effect of our biases, not any inherent trait of the light or of colours. It doesn't require genetic variation to see and describe a three-colour rainbow, even if it contains what we would describe as other visually distinct colours in it.