It is just a tiny extremely annoying thing that may be too hard too fix:
ratcheting has a sad side-effect of sometimes recolouring my previously green/blue (success!) dots on previous days.
ratcheting may turn them yellow or even red what is misrepresenting situation and makes graph uglier
Thanks so much for this bug report! Are you referring to the smaller dots without the black outline?
(For those who don’t know: Those are intermediate datapoints, not the official aggregated ones for the day. Like if you’re beeminding pushups, say, and you enter 5 in the morning and another 5 at night then your total for the day is 10 but the intermediate 5 is shown as a smaller, un-outlined dot.)
So my first question is to see if this only affects the intermediate dots. (If you could add both a before-and-after shot, maybe next time you ratchet a goal, that’d be especially helpful, though of course we’re very grateful for any kind of bug reports, including just “this seemed confusing at first”.)
If so (and just to help establish the severity, not saying this is a satisfying resolution) what do you think of the workaround of unchecking the “plot all” checkbox so only the official datapoint each day is plotted?
I actually think even the aggregated dots can change color when the yellow brick road changes, maybe not to orange/red but in your example there are dots that were presumably green that are now blue.
I think it’s going to be easier said than done to fix this because the dots are colored according to whatever the current yellow brick road is, not what it used to be. I think I agree though, we should remember every dot’s original color and never alter it when the yellow brick road is changed.
Thanks again for pointing this out!
It affects both small intermediate ones and big official ones.
I will do this, but at this moment I have no graph where I want to ratchet and would reproduce this.
It affects also main ones, but it will help a bit.
I thought that for many cases it would by defining color of rendered dots not based on “how many days in future derail at this level would happen” but “how many days with this number, at slope as of that day would trigger derail”. But this may break other things…
