Data Visualization and the Modern Imagination | Hacker News Reader

Data Visualization and the Modern Imagination

exhibits.stanford.edu

227 pointsrasmi20 comments

I would like to add one of my finds (if it's already mentioned in the original post, sorry for the double) - the National Atlas of Japan (1977) [1].

Take a look at the railway traffic statistics [2]. The visualization there must have been painstaking to make.

[1] https://www.gsi.go.jp/atlas/atlas-e-etsuran.html

[2] https://www.gsi.go.jp/atlas/archive/j-atlas-d_e_49.pdf

edit:

The earthquake epicenter visualization is also worth mentioning:

[3] https://www.gsi.go.jp/atlas/archive/j-atlas-d_e_11.pdf

Those are some beautiful visualizations. How did they do it in 1977, considering modern software like Tableau don't exist? The amount of expertise that goes into this must've been substantial.

In that timeframe, most visualisation was hand-drawn (especially on maps which were all hand drawn).

On the subject of great-looking maps I highly recommend the second Austro-Hungarian Military Survey [1] from the 1860s-1870s, some of the details in those maps are truly exceptional (one of my personal projects consists in mapping the forested area of Romania's territory from nowadays using those maps).

A little less accurate but more beautiful is the first Hapsburg Military Survey, from the late 1700s [2], when and if I'll ever get a bigger house I'll definitely hang some prints of those maps on the walls.

[1] https://mapire.eu/en/map/europe-19century-secondsurvey/?laye...

[2] https://mapire.eu/en/map/europe-18century-firstsurvey/?layer...

I have long been fascinated by timeline designs. So this is right up my street. So impressive what these artists of old managed to create without access to computers and design software.

For more modern timeline visualisations, you might be interested in this list of timeline designs I compiled:

https://www.tiki-toki.com/blog/entry/ten-amazing-online-time...

For timelines, the book "Cartographies of Time" [0] has lots of historical examples.

There is also a recent presentation of the design space of timelines in "Timelines Revisited: A Design Space and Considerations for Expressive Storytelling" [1, 2]. This then developed into the Timeline Storyteller tool [3].

[0]: https://papress.com/products/cartographies-of-time

[1]: https://timelinesrevisited.github.io/preprint.pdf

[2]: https://timelinesrevisited.github.io/

[3]: https://timelinestoryteller.com/

What I have learned about data viz over the years is that it is a very powerful solution to problems, just as Nightingale's diagrams demonstrated the effect of hygiene, which changed how societies and governments respond to epidemics. What I have also learned is that a solution obviates a dynamic someone thinks it is their job to manage, and a visualization that solves a problem is its own problem.

The "best," diagrams show change over time, and provide the presenter with a way to demonstrate how they are the important pivot point that optimizes and drives that change. The "worst," diagrams are the ones that illuminate the problem in such a way that it is no longer difficult, which humiliates the people it was presented to and designed to help.

I recommend using data viz privately, to reason through and solve problems quickly, and then use the time you save for self investment. The real value I think is to use data viz not as a product for productivity, but as an arbitrage tool for leverage.

Do you have any examples of these "worst" diagrams?

WebOWL Ontology tools, most Sankey diagrams, digraphs, radial dendrograms in d3, and pretty much anything that can represent more than a couple of degrees of freedom.

I had no idea W. E. B. Du Bois was a data and info graphic guru. Wow.

There don't seem to be many (any?) network diagrams (or what mathematicians strictly refer to by "graphs"). Are they a very recent invention?

Not that recent at non-mathematician timescales, a few hundred years at least.

As diagrams? I couldn't find anything, who used them then?

Yes. If Knuth wasn't making stuff up then you can find some historic accounts (1300s-1500s) in "Two Thousand Years of Combinatorics."

More recently, Euler's paper in 1736 didn't have what you'd recognize as a modern graph diagram, Konig's textbook in 1936 did, and the papers developing the subject between those dates eventually used the modern notion of lines connecting dots as a way to represent edges and nodes.

Yes, your second paragraph was basically what I thought was the case, it does seem quite recent. I don't have access to the Knuth article, I'd love to know what he means. Ramon Llull maybe?

The "Packet Boat and Steamship Connections between Europe and Overseas Ports" is a network diagram: https://exhibits.stanford.edu/dataviz/feature/exploring-time

The closest thing commonly seen in modern visualisation is the "Radial Tidy Tree": https://observablehq.com/@d3/radial-tidy-tree

The Data Visualization Society is a great community for discussing historical visualization.

How does that work?

I've flicked through this and it's honestly wonderful and beautifully presented, but why go to all that effort and then just render the headings in images without even providing any alt text?

I hope this is an ongoing project and that will be addressed.

There are real-text headings above the button bank at the top and the images have the role="presentation" attribute, meaning they're meant to be ignored by assistive tech. What is there to address...

Reply on news.ycombinator.com