search

OpenDRR / riskprofiler

Web Application to Support Disaster Resilience / Application web pour soutenir la résilience aux catastrophes
MIT License
10 stars 7 forks source link

Redundant legends for collapsed buildings #124

Open tieganh opened 7 months ago

tieganh commented 7 months ago

When exploring the M4.9 Georgia Strait scenario, I was looking at the number of collapsed buildings at the CSD level. The legend has 20 different colours to represent 0-1 and 1-1 casualties. This is confusing because it's obviously way too many legend colours, such that I can't tell the difference between several of them anyways, and 1-1 is meaningless.

anthonyfok commented 7 months ago

Thanks Tiegan! I can see it too: image

I wonder if we could try fixing it by reducing the rounding, i.e. increase the number of significant digits? Even though a building is either collapsed or not collapsed (i.e. whole number), I guess the user would understand it is some kind of probability and thus would accept seeing decimal numbers.

That said, I agree that 20 different shades of colours are too many, especially in this case. Perhaps something like 5 to 10 different colours?

tieganh commented 4 months ago

Hey @anthonyfok - thanks for looking into this!

My initial thought is that I don't think we want to change the rounding because we wanted to avoid showing people fractions (hard to understand and probably too precise), although I agree that using red to represent 1 collapsed building is going to be misleading and it's tricky to set it up as a binary (<1, >1). Since we are using fractions in the y-axes of our bar charts, I think it can be appropriate to add a more detailed legend to the map. So let's increase significant figures for the legend to 1 (ex: 0.1 collapsed buildings) and reduce to 5 shades, if that works.

Will this impact only this scenario or should we think about how to apply this to others?