Adjust darkness for various elements

[alexsb]

Here's an image with the new proposed ticks:

Two things stand out to me: the dots and ticks aren't as prominent as the bars. And the NAN dashes are as prominent as the dots and tick. Both aren't great. Hence:

• We should increase the darkness of the dots and ticks. They’re the same color as the bars, but have much less body. When ticks are shown in an aggregated cell, they should be lighter/more transparent though, so that we can see overplotting.

• We should reduce the darkness of the NAN dashes and/or reduce the "height" of the dashes so that they are thinner.

[sgratzl]

note: besides the N/A indicators, the colors are driven by the color of the column

[alexsb]

Good point; we need a systematic solution for this.

In our style guide we distinguish between lark mark and small mark colors:

So the marks should use the hue given by the column, but adjust the darkness according to the mark size.

[alexsb]

We agreed on the following solution:

• A cell has a given color, for simplicity assume it is given a saturation value of 0.5
• That value is either based on a color for all marks in the column, or is driven by a color mapping.
• For a large mark, the saturation value is decreased by a value N, which we will have to experimentally determine, but a good starting point would be to decrease saturation by 20%
• For a small mark (dot, dash), the saturation value is increased by a value M, again, try increasing by 20% to start.
• The NaN/None dash should be thinner and more desaturated, maybe at the same level as the large mark.

We could also experiment with a bit of changing saturation for bars so that smaller bars are more saturated and longer bars are less saturated. But that change should be small and I'm not sure whether it'll work.

[sgratzl]

note: since gray values are in the form hsl(0,0,76%) this saturation change has no effect on pure gray columns.

[alexsb]

Good point. Then use lightness or alpha instead. Might be better anyways.

…

On Thu, Dec 23, 2021, 15:23 Samuel Gratzl ***@***.***> wrote: note: since gray values are in the form hsl(0,0,76%) this saturation change has no effect on pure gray columns. — Reply to this email directly, view it on GitHub <#525 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAPR4TZCTHAPGJCUDGUX2Z3USMWEVANCNFSM5J6B4BXQ> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you authored the thread.Message ID: ***@***.***>