One dimension that I would add is inclusivity. Effective visualization is inherently subjective - what works perceptually varies widely across the consumers of these visualization efforts. Color vision deficiencies are the obvious example, but not the only one. In reading this post, what hit me here is the opportunity to make subjectivity visible rather than abstract. An interactive explorer that lets the consumers of these visualizations toggle color vision deficiency or background luminance can quickly expose how fragile many good visualization designs actually are, particularly for CAE and CFD data sources. This kind of experience is more powerful than following rigid best practices. This type of direct experience could help to shift effective visualization education away from prescriptive practices toward more empathetic, audience aware design thinking.
You nailed it Matthew! I agree 100%. The experiential aspect is everything here, and this is why I am convinced this is a big deal. Thanks for suggesting color deficiency; it is definitely something I should add to my list. If you happen to create one, please do share it with me!
Enrico, thanks for the kind words. There are some good resources for converting color images to match what people with color vision deficiencies would see. (e.g. https://www.color-blindness.com/coblis-color-blindness-simulator/). This has influenced my choice of colormap for CFD simulation results.
I had not thought of putting something together myself that would take any colorbar spectrum, from an existing image or animation, and convert that to be accessible to those with either protanopia (cannot perceive 'red' light), deuteranopia (cannot perceive 'green' light) and tritanopia (cannot perceive 'blue' light). Having this utility as part of a CAE/CFD workflow would add a lot of value.
Ah! This is awesome and exactly what I was hoping to see! I love the explainer. @vctorpascual103545, did you consider making the effect even more extreme (like extending the y-axis minimum to even higher values)? Very cool. Thanks so much for sharing. If you make more, please do share them with me!
Yes! That's what I got in less than an hour. To be fair, then it takes a while to fix the little issues you start to see when you play with the tool. For instance, at the beginning it created a linechart from categories, which does not make sense at all.
I'll definitely try to build more explainers. Thank you very much for the inspiration!
This is a great idea! Have you published the tool somewhere? I think it would be a good idea to create an explainer about the zero-axis issue in bar charts and line charts. I will try to find some time to do it and I will share it.
Oh, this is so interesting. It hadn't occurred to me to gamify the visual variables like this, but it's such a neat, demonstrative concept. And yes, since iteration costs so little, we get to make ourselves lots of small, useful tools like this. More intentional iteration = better results. You're sparking lots of ideas in my head. For instance, I spend a lot of time on viz.wtf looking for bad charts to use as examples, but now I can just use AI to make one. Correlation != causation examples.
I'm also thinking I could make a tool explaining the principles of animation. There are some ease visualisations for gsap already, but I think marrying easing functions with the 12 principles of animation could be really cool.
Yes, yes, yes! I think you are having the same epiphany I had a few weeks ago! đŸ¤¯đŸ¤¯đŸ¤¯ Isn't it amazing!!! If you create something, can you please share it with me? I'd love to see what you come up with. You already have fantastic ideas in your comments.
One dimension that I would add is inclusivity. Effective visualization is inherently subjective - what works perceptually varies widely across the consumers of these visualization efforts. Color vision deficiencies are the obvious example, but not the only one. In reading this post, what hit me here is the opportunity to make subjectivity visible rather than abstract. An interactive explorer that lets the consumers of these visualizations toggle color vision deficiency or background luminance can quickly expose how fragile many good visualization designs actually are, particularly for CAE and CFD data sources. This kind of experience is more powerful than following rigid best practices. This type of direct experience could help to shift effective visualization education away from prescriptive practices toward more empathetic, audience aware design thinking.
... also, what did you use to create your explainer? I'd love to know the details. Thanks!
You nailed it Matthew! I agree 100%. The experiential aspect is everything here, and this is why I am convinced this is a big deal. Thanks for suggesting color deficiency; it is definitely something I should add to my list. If you happen to create one, please do share it with me!
Enrico, thanks for the kind words. There are some good resources for converting color images to match what people with color vision deficiencies would see. (e.g. https://www.color-blindness.com/coblis-color-blindness-simulator/). This has influenced my choice of colormap for CFD simulation results.
I had not thought of putting something together myself that would take any colorbar spectrum, from an existing image or animation, and convert that to be accessible to those with either protanopia (cannot perceive 'red' light), deuteranopia (cannot perceive 'green' light) and tritanopia (cannot perceive 'blue' light). Having this utility as part of a CAE/CFD workflow would add a lot of value.
I couldn't resist! It took me 40 mins to develop the 0-axis explainer that I'm planning to use in my next lectures! https://vpascual.github.io/explainer-zeroaxis/
Ah! This is awesome and exactly what I was hoping to see! I love the explainer. @vctorpascual103545, did you consider making the effect even more extreme (like extending the y-axis minimum to even higher values)? Very cool. Thanks so much for sharing. If you make more, please do share them with me!
Yes! That's what I got in less than an hour. To be fair, then it takes a while to fix the little issues you start to see when you play with the tool. For instance, at the beginning it created a linechart from categories, which does not make sense at all.
I'll definitely try to build more explainers. Thank you very much for the inspiration!
This is a great idea! Have you published the tool somewhere? I think it would be a good idea to create an explainer about the zero-axis issue in bar charts and line charts. I will try to find some time to do it and I will share it.
Oh, this is so interesting. It hadn't occurred to me to gamify the visual variables like this, but it's such a neat, demonstrative concept. And yes, since iteration costs so little, we get to make ourselves lots of small, useful tools like this. More intentional iteration = better results. You're sparking lots of ideas in my head. For instance, I spend a lot of time on viz.wtf looking for bad charts to use as examples, but now I can just use AI to make one. Correlation != causation examples.
I'm also thinking I could make a tool explaining the principles of animation. There are some ease visualisations for gsap already, but I think marrying easing functions with the 12 principles of animation could be really cool.
Yes, yes, yes! I think you are having the same epiphany I had a few weeks ago! đŸ¤¯đŸ¤¯đŸ¤¯ Isn't it amazing!!! If you create something, can you please share it with me? I'd love to see what you come up with. You already have fantastic ideas in your comments.