Ask HN: What is your preferred data visualization library and why?

For geographic data https://github.com/mapbox/tippecanoe (examples https://github.com/mapbox/tippecanoe/blob/master/MADE_WITH.m...), if I had enough time to learn a new one http://kepler.gl/

For bar charts http://www.chartjs.org/ and when requirements are more complex http://nvd3.org/ or https://d3js.org/

For time series it used to be https://omnipotent.net/jquery.sparkline/ but last release was 5 years ago.

My interest in the different paradigms and approaches to data visualization stems from some problems I'm working on having a lot of parallels with how plotting library APIs work.

Personally, my main encounter with plotting was in Python. I'm not a big fan of matplotlib, I got the impression that with increasing plot complexity, code complexity grew exponentially. Then there's bokeh [0], which I preferred to matplotlib, due to it being more declarative. HoloViews [1] is more declarative than both matplotlib and bokeh, and boasts that "usually [you can] express what you want to do in very few lines of code, letting you focus on what you are trying to explore and convey, not on the process of plotting". I've not used HoloViews yet.

Then I've heard of R's ggplot [2], which is based on (or inspired by?) The Grammar of Graphics [3]. This books is definitely something I want to check out.

Vega [4], an “assembly language” for visualization, is neither here nor there as far as this discussion goes, but nonetheless I just stumbled upon it and I'm quite optimistic about the initiative. Maybe someone will not have heard of it.

[0] https://bokeh.pydata.org/en/latest/docs/user_guide/concepts.... [1] http://holoviews.org/ [2] http://r4ds.had.co.nz/data-visualisation.htm [3] https://www.amazon.com/Grammar-Graphics-Statistics-Computing... [4] https://vega.github.io/vega/about/