File talk:US states by date of statehood.PNG

This is an interesting image, but since there's no "intuitive" ordering of colors, wouldn't a greyscale image be more informative? Everyone can understand differences in lightness of gray, but only those with a lot of experience with frequencies of light (i.e., "rainbow order") will see at a glance the chronological information the colors are trying to reflect. - dcljr 07:41, 7 June 2006 (UTC)

I suppose, although with greys its harder to see what an individual place is - you might get a general idea better, but if you look at a specific state and want to know what range it is, its a little harder to tell the difference in shades --Astrokey44 11:32, 7 June 2006 (UTC)

I've uploaded this attempt. I compromised between a single color (e.g., shades of gray) and a "rainbow" of colors by using a gradient between (more or less) two different colors (red and blue). Actually, I had to extend the spectrum toward a third color (green) on one end of the spectrum and go back towards pink on the other. I just couldn't reliably tell which color was which when I stuck purely with two colors. A couple more attempts and I might be able to get it, though. - dcljr 05:31, 19 June 2006 (UTC)

Nice! It does seem more stylish using less colors --Astrokey44 11:05, 19 June 2006 (UTC)

Okay, you liked that one so much, how do you feel about this one? It uses only shades of red (and pink) and is restricted to "web-safe" colors. I know the red is a bit strong, but it was the one that was the easiest (for me) to reliably distinguish the shades of (as opposed to blue and green, the other two colors I tried). Unfortunately, Wikipedia's Color blindness article says some people have trouble distinguishing dark shades of red, which might be a problem with this latest version. - dcljr 21:22, 19 June 2006 (UTC)

hmm I like this one too! I cant decide.. The grey inset looks good. Maybe the 1840-59 color should be a little less saturated or lighter (same with the 1860-79 blue in the last one?) By the way there is a collection of scales here Image:Scales mapping.PNG although these are mostly spectrum ones rather than one color ones. Anyway I have posted a question about it at w:Talk:List of U.S. states by date of statehood which more people are likely to see than this talk page. --Astrokey44 23:24, 19 June 2006 (UTC)

And I've asked for opinions in the context of color blindness at w:Talk:Color blindness#Need opinions from CB folks on two images. - dcljr 04:17, 21 June 2006 (UTC)

Okay, I think I might have come up with the final version... finally (hence my momentous decision to return to the left margin): one with a red, white and blue color scheme (well, okay... red, gray and blue — and actually, the blues look more periwinkle on my monitor... but they're supposedly blue according to the RGB web-color model). Again, only "web-safe" colors were used. - dcljr 05:26, 21 June 2006 (UTC)

Hello, I am color blind. I have weak green perception. Some greens look greyish to me and some browns are hard to distinguish from dark reds. The majority of color blind individuals are like me, and the vast majority have red-green disorders. The key to helping color blind people is not by adjusting the hue.. (I can tell a lighter red from a very slightly darker red) but by using only one color, or very different colors. So using a dark green and a grey would be similar in hue to us. A map that went from green to brown would also be difficult because a dark brown may be like a dark green, but a map that was increasingly darkening green would be easier.

Your map with the blues is not good because although I can tell the difference between the tealish color at the bottom and the pinkish at the top, they look somewhat similar (our perception of greens are bad.) The other map with the increasing red is good because we can distinguish hues accurately. However people who cannot see red at all (rare) will have probs. papile 05:26, 25 June 2006 (UTC)

I see you commented on Image:US states by date of statehood gradient.PNG and Image:US states by date of statehood red.PNG, but what about Image:US states by date of statehood RWB.PNG? Is that last one okay? - dcljr 05:06, 3 July 2006 (UTC)

Perhaps an animation which does not show hue changes would be a good way to show this information? --Astrokey44 10:37, 6 July 2006 (UTC)

(The following comment was copied from w:Talk:Color_blindness#Need_opinions_from_CB_folks_on_two_images. - dcljr)

Actually, this one is the best. Never use blue and purple together as a protanope will not distinguish these. With the white-black-red scheme, the worst you will have is that a protanope will see the dark red as very dark grey, which should still be distinguishable from black, so it basically just turns into a grayscale gradient for him/her. Respectfully, if you are concerned about colorblind people, it's not clera to me why you didn't just make this a greyscale graphic in the first place; it only has one dimension of information. The newest one is confusing and the middle ranges aren't distinct enough. --Chinasaur 07:36, 7 August 2006 (UTC)

As seen above (in the first comment on this page), doing it in greyscale was actually my original idea, but the problem is picking the shades so they can not only be distinguished from one another, but also associated with the appropriate years (as pointed out by Astrokey44). It doesn't help to be able to tell the difference between two shades on the map if you can't reliably tell which year ranges they correspond to. I thought maybe using more colors (not just shades of one color) would help the year-identification problem. If anyone thinks they can do better, please try. The best solution, it seems to me, would be to use fill patterns, along with color, for the later years (since those generally correspond to larger states), but I don't know enough about The GIMP yet to be able do that. <g> - dcljr 15:41, 7 August 2006 (UTC)

I see, so just grayscale is not so great. Fill patterns does seem like a good solution, although probably it won't be beautiful. At that point, just putting the years on the individual states may be just as good. Your red version seems to work pretty well, although maybe you should use a blue gradient combined instead to avoid the protanope problem. Tritanope is much rarer, so blue channels are less likely to be lost. You might think that purple is better as purple gradient should be visible by everyone (except monochromats), but I tend to find purple confusing. Thanks for your efforts! --70.91.135.229 04:51, 10 August 2006 (UTC)