Anjney Midha, founder of AMP PBC, recommends AI labs require teams to read Edward R. Tufte’s “The Visual Display of Quantitative Information” before releasing charts

Earlier this year I was getting frustrated with Claude's charts, fed this book to claude and had it generate a Tufte skill. Instantly got simpler/more beautiful visualizations.

https://gist.github.com/aparente/e48c353755958621b3c0004593105a90

need this as a skill

@AnjneyMidha just make sure the models have read it atp

claude’s take

@draparente @AnjneyMidha Tufte is ok but I’d really recommend trying to feed the Semiology of Graphics by Bertin instead. A bit more dry for the human reader but way more rigorous and clear.

@AnjneyMidha Even better with the whole bundle

Tufte is the right starting point. You see the same failure mode at the SME implementation layer — one mega-chart that buries the two or three signals executives actually need. Small multiples, consistent scales, and a ruthless data-ink ratio fix most of it in an afternoon. In practice that usually turns three-meeting alignment into a single pass.

@willccbb this and "Clean Code" ;)

@AnjneyMidha Ch 9

@AnjneyMidha It’s like we are synchronized to the same reference materials

@draparente Interesting. Do you have any examples of the output?

@AnjneyMidha Tufte, Stephen Few, Dan Roam, and Amanda Cox and all people to follow and learn

@tszzl @AnjneyMidha i’d like to see gpt front end brockmann pilled, i bet some RL on the classics would go far

@willccbb

@willccbb here you go

@jeevish @draparente https://github.com/HugoVeltorai/publicskills

@AnjneyMidha My strategy plan.

⤵️

@AnjneyMidha ah yes i should put it on the merch section of the museum of chart crimes

@draparente Has someone already fed this and generated an md file for wider users to benefit?

@tszzl @AnjneyMidha tbph i found him to be full of himself in a way that made much of the emperor’s new clothes look sad and disappointing,.,