[temporaryvector]

For me, it's a pretty interesting article on a topic I very much enjoy and they did some very impressive work, but it's ruined by some pretty bad UX.

Reading this felt like a chore. I would greatly prefer a PDF file or flat website with the text and a separate viewer.

[LargoLasskhyfv]

Hm, the site seemed a little slow at times, but i wasn't sure if it was my old laptop. The UI was a little bit unusal at times, especially that << backarrow-thing, which looked like it would be for going back, instead of moving on along some path. But it was obvious from context for me. I really enjoyed the 3D-inserts where you could zoom, pan, rotate, and the changes, rising and disappearing buildings across the timeline. Though i had to click on reload sometimes to have them appear at all. Besides that you could always click on the drop down menu in the upper left and use the TOC to jump around, or skip something.

[temporaryvector]

>Besides that you could always click on the drop down menu in the upper left and use the TOC to jump around, or skip something.

I think that was kinda the main problem for me, actually. When you have a PDF article or book or a plain webpage, you can jump around without much penalty, and you can skip ahead then go back and so on, you can skim parts and can check if something relevant appears ahead. At least, that's how I read scientific articles: read the introduction, skim through the text, go back to read the more interesting parts and jump around until you grasp it and so on. Any supplementary data or figures provided I usually have open in another window for quick reference as I'm reading the text.

The best UX for scientific articles for how I read them is still paper for the text and a computer for anything else. Second best would be a good PDF reader.

Thus, for this kind of interactive visualization, I would have preferred it if they had made a good visualizer and an accompanying article, instead of mixing the two into a single thing. Feels like they added unnecessary friction to both sides of the presentation and overall the effect would be better if split. For example, an apparent problem is if I want to jump back to a piece of text I read before I have to remember where in the TOC it was (can't just scroll back up) and then wait for all the 3D inserts to load when what I want is just the text.

A project that I think did this mostly right is GeaCron (geacron.com), at least the visualization part. It's not really comparable since GeaCron is a world map, but with it I can visualize a particular point of interest and then read up the information at my own leisure. I've wasted hours with that map this way.

I don't want to diminish the effort of these authors, for it is impressive and the 3D-inserts are great and I still enjoyed it despite the UX problems, but as someone who is a bit passionate about visualization in academic environments, I often see projects like this that I feel could be so much better if they didn't try to reinvent certain wheels.