Data Vis Dispatch, August 18: Data art, low water levels, and 😂 | Datawrapper Blog

Justy The tree rings for city growth are gorgeous, Cody. And annoyingly, they make me care about urban change before I’ve even read the labels.

Cody That’s the whole trick, though. The dispatch is quietly arguing that a visualization’s form is part of its reasoning, not a decorative shell glued onto a dataset.

Justy Yeah. This one jumps from Giorgia Lupi’s library installation to Nadieh Bremer’s Growing Cities project, then into eclipse maps and low-water reporting. It’s a pretty persuasive little collage of different jobs a chart can do.

Cody Right.

Justy Also, my week has been a lot of tabs pretending to be research. So a dispatch that just says, “look at these careful things people made,” felt weirdly restorative.

Cody That is not research avoidance. That’s visual literature review with better colors.

Justy Exactly. We’ve somehow made episode eight hundred eighty-three into a defense of staring at charts.

Cody A noble use of the machinery, clearly.

Cody Bremer’s piece is the most interesting technical example here. She worked with E P F L’s URBES Lab and rendered the evolution of thirteen Chinese cities as radial growth, like tree rings or bacterial spread. A radial encoding gives time a physical feeling of accumulation. You can see expansion as something that leaves a trace, rather than a sequence of disconnected annual snapshots.

Justy Mm-hm.

Cody But I’d hold the bacteria analogy loosely. Cities don’t grow through one clean outward mechanism. Boundaries shift, development jumps over empty space, policy changes the shape. The visual is honest if it’s an invitation to inspect the history, not a claim that urbanization literally behaves like biology.

Justy That caveat actually makes me like it more. The product value isn’t “now planners can make zoning decisions from a lovely circle.” It’s that someone who would never open a spreadsheet gets a durable mental model, then maybe asks a sharper question about why one city’s rings change differently from another’s.

Cody Sure.

Justy And Lupi’s SUNY Binghamton library installation is another version of that. It’s data-driven and interactive, but it lives on a wall, in a shared place. That changes the relationship. It’s less dashboard, more institutional memory with a pulse.

Cody I see it. Though this is where I become the boring person at the table. An installation can earn attention through atmosphere, but it still needs legible provenance somewhere nearby. What was measured, over what period, who made the choices. Otherwise the visual language gets to borrow the authority of data without offering its receipts.

Justy Come on, Cody, you can’t demand a tooltip from a wall.

Cody I absolutely can. A plaque is just a tooltip that gave up on hover.

Justy That is such a bleak little interface requirement.

Cody I’m right, though.

Justy You are. And it’s the same split we keep finding: the visible layer gets people in the room, then the boring layer has to support the claim. We’ve been saying the interesting work is one layer below the recommendation for months. Apparently that applies to beautiful walls too.

Cody Unfortunately, yes.

Cody The eclipse coverage shows the other end of the spectrum. El País used satellite imagery for the event, then published a Spain visibility map organized by place and hour. That map has a real decision attached to it: where the eclipse can be seen and when. The dispatch also notes reporting that tracked effects like an electricity-consumption drop and searches for eye pain. That’s a useful move from spectacle toward measurable consequences.

Justy Right, right.

Justy That’s who should care about this collection, honestly: teams making something people need under a deadline, not only people making gallery work. A visibility map is a public-service interface. The question isn’t whether it looks sophisticated. It’s whether timing, location, and uncertainty are clear enough that somebody can act without decoding a graduate thesis.

Cody And maps can fail exactly there. Smooth gradients and polished animation can imply accuracy the underlying data doesn’t have. I don’t know enough about El País’s inputs to audit that map, so I’m not calling it out. I’m just saying the visual grammar needs to match the measurement resolution.

Justy Fair. You found the error bars in an eclipse.

Cody I find error bars in everything. It’s not a personality, it’s a rendering bug.

Justy Okay, that one stays.

Justy I like that the dispatch doesn’t crown one chart type as the winner. It puts an expressive city-history piece beside practical reporting on water levels and heat, and lets the contrast do the work. Different stakes, different visual contracts.

Cody Yeah. The only overreach would be treating a curated set of great examples as proof that visual novelty always improves understanding. It doesn’t. But it does show that restraint and imagination aren’t opposites. The right form can make evidence easier to enter, provided the underlying structure is there when someone looks closer.

Justy That’s enough chart-staring for one Tuesday, Cody. Go make your plaque-tooltip before I turn it into a feature request.