The final result is:
* R = 25%
* G = 40%
* B = 36%
As expected (and described in the thread), the gamut is pretty low (ie. not an immense possibility of colours, due to fact that masto polls options sum is 100)
0..100 to 0..255 (in gimp) give:
* R = 63.8 (64)
* G = 102
* B = 91.8 (92)
It's a soft, dark, bluish green!
It's complimentary color is #66404A (102,64,74)
We could call it "Fédi-vert"?
@jz beware that using #RGB is not enough to specify a color:
> A "#ColorModel" is an abstract mathematical model describing the way #colors can be represented as tuples of numbers (e.g. triples in #RGB or quadruples in #CMYK); however, a #ColorModel with no associated #MappingFunction to an absolute #ColorSpace is a more or less arbitrary color system with no connection to any globally understood system of color interpretation.
@julm Thank you! I was waiting to see how long before the first colour nerd would bring that up! ;) I am NOT disappointed! ;)))
Which colour space do you suggest to use? a non-proprietary one? What would be your suggested recipe to turn %ages into values?
@jz I'm no expert, but AFAIK it depends on where you want to put that color. The #sRGB #ColorSpace is standardized and widely supported by monitors (though some may not actually meet that standard, or need to be color calibrated). The percentages' accuracy may overflow the accuracy of the widely supported #TrueColor #ColorDepth (8-bit/channel) but that accuracy overflows the human eye's anyway. Whose perception depends on the #AmbientLighting on top of that. But again, I'm no expert.
@jz We are, if I'm not mistaken, at #426657 (26%, 40%, 34%), a kind of dark verdigris. Can do better 😋
Fortunately, there are 6 days left....
@jz how will you convert the percentage result from the poll into the 255 scaled RGB mixer ?
My concern is that percentages scored in a poll by each option are interdependant and can't be all zero% nor all 100%, but RGB channels are independant from each other and can be all 0 or all 255.
A result of 33% for each RGB channel would result in 255 x 0,33 ~ 84, which is grey.
@jz next time allow multiple choice, because now the gamut is extremely limited (R+G+B == always 100%)
@jz we have a color. In CSS:
background: rgb(calc(255*0.25), calc(255*0.4), calc(255*0.36));
Yes, there are other ways of converting these results into RGB, but this seems the simplest.
@jz another way could be:
1. assume 100% = 255×3 (R+G+B) = 765
2. for each color, multiply that by the result:
R: 765×0.25 = 191.25
G: 765×0.4 = 306
B: 765×0.36 = 275.4
3. Obviously we are over the max of 255 for G and B, so normalize to 255 based on the largest value (G):
R: 191.25×(255/306) = 159.375
B: 275.4×(255/306) = 229.5
CSS: background: rgb(159.375, 255, 229.5);
Oooh, I like the gradient idea, but I would use it a tiny bit differently: if we have historical data on poll values, the gradient could be the color-through-time, from start (0, 0, 0), to whatever the end result is, using poll values from different moments in time!
@jz You can see the fuller gamut if you reach into Masto's json response:
R = 470 / 1918
G = 762 / 1918
B = 686 / 1918
Mamot.fr est une serveur Mastodon francophone, géré par La Quadrature du Net.