Category colour blindness
TL;DR Viridis is a better palette than Rainbow for identifying a subtle feature in data. How does a plot’s colour palette affect our ability to find a subtle feature in data? Friend and colleague of Better Figures Tim Graham came up with a fun way to find out. Tim sent some Python code to plot a nice […]
How are we doing in getting rid of the dreaded rainbow colour scheme in scientific visualisations? On a whim, I asked for people to count presentations or posters with Rainbow palettes at this year’s meeting of European Earth sciences mega-meeting, the EGU general assembly. Many thanks to those that stepped up. This was clearly a […]
What an achievement! Yesterday, the lovely perceptually accurate open source colour palette, Viridis (we discussed it here) turned up in a major paper. The paper itself was on some gravity discovery or something, but we need to focus on what’s important here. Well done everyone.
That’s Royal Meterological Society at the European Centre for Medium Range Weather Forecasting. We had an afternoon of Visualisation of Meterological data, part of the ECMWF visualisation week, and Better Figures was lucky enough to be invited to present (download the slides). ECMWF was great, in that they have preserved the 1970’s decor beautifully, alongside […]
At the time of writing, the US media is tracking Hurricane Joaquin, and trying to predict whether it will make landfall in the US over the next few days. There are lots of great visualisations, but one thing that this event does highlight is just how many different types of rainbow colour palette there are. […]
Here is a great video, describing viridis, the new default colourmap for matplotlib in the programming language Python. This has some great features – It’s perceptually uniform across its range, it’s pretty, and it works for the colourblind. Oh, and you’re free to use it. The last five minutes or so describe the process and thinking […]
Here are some recommendations for making scientific graphics which help your audience understand your data as easily as possible. Your graphics should be striking, readily understandable, should avoid distorting the data (unless you really mean to), and be safe for those who are colourblind. Remember, there are no really “right” or “wrong” palettes (OK, maybe […]