Alright I'm really starting to miss something with what you're saying here. I need to ask questions and stop making assertions it seems.
If I get an apple, and that sucker has the pigments (whatever conjugated chemistry they are) it's going to absorb and reflect parts of visible light. In this case I would say it reflects green light (~510nm), a piece of all visible light. Now, what am I missing by stating many times that this piece of light travels and hits your photoreceptors which are sensitive to this wavelength* and transduce it to your brain did I miss?
*By the way, I don't mean that this wavelength is singular, I just haven't willed myself to go find a good lambda to copy/paste.
Well, first off, you're speaking very, very loosely--way too loosely to even describe the situation properly. Let's just start with a problem statement; imagine it's phrased in the form of a question. "What color is the apple?" Now let's suppose you're building a robot to try to figure out the answer to this question. Your robot is equipped with a detector, and that detector registers 510nm photons coming from the apple. Does your robot then say that the apple must be green?
If so, here's the problem with your robot. Just about every object in nature reflects 510nm photons. Imagine a poor photographer who tries to take a film picture of a scene, but he has the camera set for a much dimmer environment. Well, predictably, his camera will be overexposed; when he develops his film, he's going to find that everything in the photo is white.
Now process that for a second...
every thing in the photo is white... why? Because everything is reflecting photons all across the spectrum. That's why your robot will fail. There's a lot more you need to do to your robot than to add a 510nm photon detector.
As an alternate example, suppose you're about to watch a movie using a projector. The screen is white. Now note something significant--projectors cannot erase light... they can only add light. And yet, while the movie plays, you see black objects in it. Think about that as well--somehow, by only
adding light, we made a spot on a white projector look black.
When you build your robot, make sure it can figure out what objects are black in a movie projected on a white screen as well.
Sort of like how a color-blind person is consistent in the correlation of their color association (I hope that word works, we haven't used it yet) for red, even if they're missing their red cones so red is really just dark colors, but not red. It's consistent even if the corresponding photons for red aren't perceived.
That's a fair example. Let's suppose Mary is one of those rare tetrachromats rumored to exist. Both of us I presume are trichromats. George is a protanope.
We have objects A, B, C, D, E, and F, all with different spectra. Let's say that A, B, and C are red; D is green; and E and F are blue. Since trichromats like us define what red, green, and blue means, this would imply that you and I can tell which of those six objects are red, which are green, and which are blue by looking at them. Mary can do this just as well as we can. However, even though A, B, and C are all red (and look identical to us), Mary can see that they are in fact different; A and B may look the same to her, but C looks like "the other kind of red" as she might describe it (or "rich red" or some other invented term meant to convey the distinct percept she gets from it).
George cannot tell red from green, but he can play us off each other and figure out that we're not making up the category green. He does this by asking you which of A, B, C, and D are green (he knows enough to know E and F are definitely not green), get your answer, and then independently ask me without telling me you already told him. Since I'm able to independently produce the same answers you produce, then he knows there's a real category of objects called green.
We can do the same test with Mary, though because there's only one of her, we have to settle for a different format. We can shuffle the A, B, and C objects randomly, while having C labeled on the bottom out of sight somehow, to where neither of us knows which is C is but either could figure it out by looking at the bottom label. We can then ask Mary which is the odd one out, and verify her choice. (We can use this same kind of testing to figure out if someone has a psychic sense, speaking of the forum at large).
The ability to distinguish these categories of objects is all perception has to do to be useful; we want to be able to tell that this is the green banana and that that is the yellow one. We learn that green bananas are hard to peel, tough to eat, and not so sweet; and that yellow bananas are easy to peel, soft in the mouth, and nice and sweet. By divining the color of the banana we can get the rewarding properties every time.
Mary might similarly be able to pick out the good avacados.
That's where the real meanings and utility lie. How exactly it's perceived doesn't matter. Now remember this... we have a real percept, and it correlates to some real thing, and that correlation is the measure of the utility of the percept. We still have to figure out what it correlates to and how to exploit it, as well as be able to.
Remember, this all started because Bodhi said (paraphrased and possibly out of context, I can't remember and honestly can't be bothered to check it anymore, it doesn't matter if he said it or not) that the guy on acid who was "seeing" things was "just as real" but I said something to the effect of "No".
I don't have an immediate objection to Bodhi. It is, in fact, true that no matter what we perceive, there is something in reality that correlates to that perception. A different way of phrasing this is to say that all perception is the result of some arrangement of electrochemical processes or some similar real thing. In this sense, all perception is "real".
I don't think Bodhi is saying that dafydd's shoes were made of glass. I think he's just saying that if dafydd perceived his shoes turned into glass, then something in reality must have occurred correlating to that. It could be "all in his head", but even if it is, it might be a useful sort of thing, or might not. It might, for example, be an indicator of a particular psychological trait, or problem. Or it could simply mean that particular neurons somewhere in his head have more receptors that are blocked by the drug. Whatever it is, his perception has some real world correlate, because only the real world causes perception.
It's only when you're stuck on what the perception is "supposed" to correlate to that you run into these sort of issues; if seeing glass shoes is "supposed to" mean that my shoes actually were glass, then the perception doesn't reflect that my shoes actually were glass. This doesn't mean the
perception doesn't correlate to reality though; it only means the
interpretation doesn't. The perception may still mean that particular neurons in dafydd's head have more receptors to be blocked by the drug, however useful that factoid may be.
I understand that actually. Now what I've been trying to say is that this to me doesn't really matter. It's consistent and perceived consistently, but not accurate.
Accuracy in this case is part definition. If we can always identify that A, B, and C are the red objects, then we have a pretty accurate sense of red. But the concept of "red objects" relies on our particular form of trichromaticity. Mary the tetrachromat thinks B and C are different colors. Maybe you can say her perceptions are more accurate, but how would you compare if it so happened that Mary's two additional L clones did not in fact line up to our L cones, and she couldn't reliably pick out the red objects? She can still pick out more colors than we can, and maybe spot useful properties, but would her perceptions then be "less accurate"? (Maybe she cannot tell red from green, meaning she has difficulty finding berries--but she can still pick out ripe bananas just as easy as we can, and can pick out the good avacados from their color unlike us).
As I said, you need to have a criteria here.
Alright so color (shapes, and all the rest) is just a gestalt of what reality is and this is what we call perception (stop me if I am wrong)
Hopefully I answered that above. A perception is "good" if we can correlate the perception in the same way that some external property correlates; if we can pick out the sweet and soft bananas from the sour and hard ones, or if Mary can pick out the ripe avacados from the bad ones.
normally they are for any believers in "the supernatural", I just find it funny that the people using them (most of the times) are obtuse and fanatical in a way that reminds me religious fundamentalists. 
