I'm not generally colourweak and I have trouble decoding it too, J1 and J2 in particular look pretty much the same to me.

J1 and J2 are the same colour. The key used for the slices is on the left of the image. The labels on top show more detail, but several haplogroups are grouped into the same colour.

For distinguishing between red and green hues for some people, http://colortest.it/enhance-r:http://brilliantmaps.com/the-g... might be useful (or the "Enhance green" setting, or switching between the two).

colour coding is generally really bad, and especially when there are more than 4 or 5 things to be labelled..

would it be so hard to put the short one/two/three letter haplogroup names from the keys inside of the largest slices? or to show the pies consistently with the same groups in the same order?

Pretty bold to call the site "brilliant maps" if they rely on crappy colour coding like this. Even with helper tools I can't make head or tails of the map.

Wow. You guys really dislike colour coding maps. Colour coding is really okay for most of us, even if its not easy to distinguish between all colours. What alternative do you prefer?

About one in twelve men has some kind of colour weakness. If you're relying on the ability to distinguish between colours on hue or saturation alone, you're making it difficult or impossible for one in twelve men to use your product.

It's not that hard: if you need to distinguish a small number of colours, make them sufficiently different in luminosity: if you wouldn't be able to tell the colours apart on a greyscale photo copy, someone will be unable to use it. If you need to distinguish between a large number of colours, use additional clues (e.g. labels, patterns, etc) other than the colour itself.

It's not "not easy to distinguish between all colours". It's impossible to distinguish between some colours, making the map itself meaningless. This is akin to having a heatmap where the "hottest" and "coldest" areas are indicated with the same colour: the result is useless unless you're specifically interested in "average" vs "extremes" (with no regard for direction).

I just expressed my wish that someone would try to come up with a better way: different colours combined with hatching maybe?

I'm not even colourblind, just colorweak and I am fairly sure there must be some set of colours that are both differentiable and looks ok at the same time.

Alternatively here's a feature idea for someone who creates charts / chart toolkits: make several palettes, including one with safe colours and/or hatching.