Still one of the 🐐 troves of viz knowledge around!
— Mara Averick (@dataandme) June 6, 2019
A veritable #dataviz education:
"Visualization Analysis & Design" w/ all @tamaramunzner's talks 🎉! https://t.co/CxTLoO9u8W #infovis pic.twitter.com/2sW14IsgkY
Just released version 3.1 of Altair. Lots of usability enhancements and minor bug fixes: https://t.co/rWP8fQMtjI
— Jake VanderPlas (@jakevdp) June 6, 2019
See the full change log here: https://t.co/ZhNGIZczK6 pic.twitter.com/DdGEa1jtj5
This is what the diversity tax looks like 👇 https://t.co/AOiRynASYU
— Kate Crawford (@katecrawford) June 6, 2019
Matplotlib tip: if your saved figure files are too large, pass `rasterized=True` to the plot element that's causing the issue: that element will be saved as a more compact raster (i.e. pixel) representation: pic.twitter.com/Pezq0VTtHg
— Jake VanderPlas (@jakevdp) June 4, 2019
D3-annotation by @DataToViz is brilliant and perfect and small and fast has been my lifeline for labelling things in d3.
— Monica Dinculescu (@notwaldorf) June 3, 2019
🏷 https://t.co/8wrFhHbBJC
Riding in a car is probably the riskiest activity we engage in on a regular basis. #datavizhttps://t.co/upIi8y2B9S pic.twitter.com/fehWL9qZox
— Randy Olson (@randal_olson) June 2, 2019
Top 10 most viewed #YouTube videos, 2006-2019. #datavizhttps://t.co/lCT7jo7xsR pic.twitter.com/5Y2X79QFXl
— Randy Olson (@randal_olson) June 2, 2019
Looking forward to playing with @thomasp85's patchwork package. I've recommended cowplot in the past, but this looks magic! #SDSS2019 https://t.co/g6VdMptjlj
— Amelia McNamara (@AmeliaMN) June 1, 2019
Top 10 most valuable companies in the world, 1997-2019. #datavizhttps://t.co/6UKIE5olf3 pic.twitter.com/qu1TODdg28
— Randy Olson (@randal_olson) May 31, 2019
Ethnic composition of the #USA. #datavizhttps://t.co/dXTGayCROV pic.twitter.com/FvGklTSRb3
— Randy Olson (@randal_olson) May 31, 2019
Wow - the most common causes of US death, vs the amount of media attention they get.
— Liv Boeree (@Liv_Boeree) May 30, 2019
Unsurprisingly, violence is hugely overblown. Terrorism, for example, is over-represented by a factor of 4000.
Imagine what this does to the public’s psyche?
via @OurWorldInData pic.twitter.com/mCScYhwpwa
A+ like spectrum lines , when sparklines show unbinned distributions, this is a fine display, a good example of journalism using technical data directly without restyling into the house style. But do data look too good to be true? https://t.co/amIIVdDdJc
— Edward Tufte (@EdwardTufte) May 28, 2019
