Daniel Zvinca

Sharing the current stage of an interactive violin visualization project. The concept is already there, even if the implementation is not yet as clean as I’d like. iOS rendering quirks forced heavier drawing logic than expected. danz68.github.io/visualprojec...

Desktop or mobile, with rich interactivity, experimental ideas, and a few easter eggs: center in-out size sorting, light border separation, adaptive/pinned tooltip, focus and zooming concept, unusual stack tracking, adaptive scale, shape morphing, and a nod to the law of large numbers.

Joe Rogan and Donald Trump talking over a connected scatter plot. Didn't expect that. Anyway, triggered by a this thread www.linkedin.com/posts/merod_... I decided to reinterpret the same data without using the time connected scatterplot designed by Max Roser ourworldindata.org/us-life-expe....

Each curve is an estimation based on a polynomial model (any decent model or smoother will equally serve the purpose). Doing so, there is a more clear distinction between curves, helped also by the white borders and proper z-order painting.

Joe Rogan and Donald Trump talking over a connected scatter plot. Didn't expect that. Anyway, triggered by a this thread www.linkedin.com/posts/merod_... I decided to reinterpret the same data without using the time connected scatterplot designed by Max Roser ourworldindata.org/us-life-expe....

I spent quite a bit of time developing the density dots plot graphics. Originally intended as a replacement for the beeswarms, it soon evolved into a complex graphical encoding technique. A bridge between the descriptive and inferential aspects of distributions, a quantile and a density estimator.

For comparison, the same dataset without tolerance adjustment (max error 2.15 dots, one dot=20).

I spent quite a bit of time developing the density dots plot graphics. Originally intended as a replacement for the beeswarms, it soon evolved into a complex graphical encoding technique. A bridge between the descriptive and inferential aspects of distributions, a quantile and a density estimator.

Each dot encodes a price with a tolerance of 500 rupees (the data set contains rental prices in one region of India). A seriously vertically squeezed Wilkinson dot plot (unit histogram) for comparison (the dot piles are much higher).

Triggered by Carlos Barboza LinkedIn post, www.linkedin.com/feed/update/... this is how a density dots plot design looks using the very same data (yearly cycle perspective).

Besides the density shape faithfulness (the main goal of the density dots plot, here using the Epanechnikov kernel BW = 10x dot size), the dot encoding accuracy is way better than the Wilkinson dot plot (max error 0.25x dot size vs. Wilkinson's 0.5x dot size, trading-off dots total visibility).