The facades on the buildings built over a hundred years ago are amazing to me because so much art was involved in the architectural designs. It would have been fun to be in the construction business back then. #architecture #art #buildings #barcelona #photography #streetphotography

Paused MAD MEN while Ted Chaough’s partner was in the hospital and the HBO Max interstitial ad made it look like his dying wish was for a Whopper.

I suspect it’s probably related to why reading in a physical book results in better retention: there is a physical/spacial aspect to information storage. Where something was located on a page, where on a page we wrote something. We are embodied creatures, not brains in an abstract environment.

Happy #WorldWaterDay! Fetch yourself a big glass of the good stuff and take a look at the watery wonder which is Hamonshu (1903), a Japanese book of wave and ripple designs: https://publicdomainreview.org/collection/hamonshu-a-japanese-book-of-wave-and-ripple-designs-1903

reading [good] documentation and trying to get into the head of someone (or some people) who made a library or designed a language is interesting. figuring out how they did something - and why - is rewarding. learning out their way of thinking about a problem can be applied to my problems is great.

oh look at that www.pew.org/en/research-...

Good news!

As we head into another wildfire season: "We find that the observed increase in extreme fire weather bears a clear externally forced signal, detectable at 99% confidence above natural variability and attributable to human-induced climate change." www.science.org/doi/epdf/10....

In all the foofaraw surrounding AI, many people seem to have forgotten that product development is not all about coding, and that coding is maybe 10% of the actual work. 1/7

it's one of my core frames. context is how we make meaning, and the whole pitch of LLMs is that you don't need an embodied context to make sense of the world

I learned about a brilliant project yesterday to support British Sign Language (BSL) users, creating new signs for scientific terms in chemistry, oceanography, data science & more. And they're fabulously descriptive. Do have a look and share! www.ssc.education.ed.ac.uk/BSL/index.ht...

This is a fantastic interactive essay about clothes sizes for women (and how they so often poorly fit adults - hello mid-size gap!). Finding professional clothes as a roughly average sized woman is a terrible experience. pudding.cool/2026/02/wome...

where i come down on this is that you either care about what Amanda is talking about or you do not. and if you don't that's your preference. i think these tools can be impressive. and i read to try to get closer to other people...to let them occupy real estate in my brain. these things are different

Currently reading "Disabling Intelligences" by Rua M. Williams @fractalecho.bsky.social in the UCGIS Humans and Geospatial AI Book Club. I should have read this one sooner!

I do see posts from LLM enthusiasts that boil down to “critics are just mad the computer is better than them at trivia now” and idk, I think *most* critics are just genuinely terrified they’ll never find work they don’t hate that pays a living wage again in the current batshit climate

I can write good abstracts not because I have a natural talent for it but because I have spent twenty years consciously practicing it. The only way to get good at something is to keep working through the part where you're not good at it … yet. cf. james.grimmelmann.net/files/advice...

Short circuiting the learning process is the whole point of #AI. Users offload the "simple" or "mundane" thinking that nurtures growth. Students borrow its "understanding" and think that understanding is thiers. Use at the risk of stupidity. #academicsky phys.org/news/2026-02...

🌐 Are you attending #AAG2026 with @geographers.bsky.social in San Francisco? 🚨 Sspecial events, timely plenaries, and sessions are held, recognizing geography leaders and trailblazers. Be sure to add these must-see events to your schedule: 🌟 www.aag.org/events/aag20... #AmericanGeosciences

What coders lose by relying on AI. From our event with the University of Washington Office of Public Lectures. (with @emilymbender.bsky.social)

"Incorporating evidence from 26 studies on AVs’ impacts on the flow of car traffic, researchers find that a future where driverless cars become widespread is likely to increase the total number of miles traveled by vehicles in the US by around 5.95 percent." www.vox.com/future-perfe...

I've seen a lot of people describe AI tooling as less a way to reduce or implement work directly, but to hide moving work from one person to another person. A developer can use AI tools to generate a ton of sloppy code, but then it's going to put a much bigger burden on the reviewer

mapping power is so hot right now longhaulmag.com/2026/02/27/a...

"Screw Covid, I'm going to Sturgis!" The Sturgis Motorcycle Rally, held in August 2020, was the epitome of the individual choice mindset. Of course it spread cases outward and caused infections, illness, and death. A new research paper allows us to watch the Sturgis Covid wave spread across the US.

Have you seen bethdeconinck.com/2026/03/04/o... ? I thought Beth gave a wonderfully empathetic yet sharp analysis.

There are lots of intersections that are dangerous for cyclists, but what made Ipsley Cross so lethal was a kind of eldritch geometry that let cyclist and driver see each other a *long* time before the collision, while providing the illusion that they were *not* going to collide, until they did. 3

Of the many, many issues with how quickly "AI" became a core part of things is illustrated here: There just isn't a way to defend against prompt injection. Imagine if SQL was constructed in such a way that injection is impossible to mitigate; It would be insane to rely it.

Saw a thoughtful thread about AI, don't want to QT or argue. But. The biggest rage factor with LLMs is the people who, because genAI is transformative for coding, think it's transformative for everything else, because they devalue every other form of work and labor and knowledge.

New post: Can AI Replace Social Science Researchers? (No. No it can't. Come on, now.) davekarpf.beehiiv.com/p/can-ai-rep...

Michael A. Duprey, Georgiy V. Bobashev: Enhancing Computational Efficiency in NetLogo: Best Practices for Running Large-Scale Agent-Based Models on AWS and Cloud Infrastructures https://arxiv.org/abs/2602.15317 https://arxiv.org/pdf/2602.15317 https://arxiv.org/html/2602.15317

Philip Koopman, William Widen Redefining Safety for Autonomous Vehicles https://arxiv.org/abs/2404.16768

yes, you have to understand the lay of the land of the field you are attempting to contribute to I don't understand how people came to conflate sheer output with the transformative process of internalizing and analyzing information and creating something new from it

Every word of this post (and the original posters, too). The point of undertaking a literature review is to engage in the slow, cumulative cognitive process of learning as you read and write. It’s not “processing thousands of references to get a statistical summary of what the LLM was fed”

i don't think i can say this any more directly: if you want to get better at something, you have to do it a lot there are other requirements but that one is non negotiable

Every day I become more and more convinced that the main thing we need to do to improve literacy is just make kids read a lot of books, both on their own and as read-alouds

We are going to need to strengthen policies for identity use after death...

The answer to problems in sociotechnical systems is not always "more software", "better software" or "open source software". A lot of the time the best solution is _less software_ and sometimes even _no software_.

And this is why you should ALWAYS check when reference managers import metadata (as you can tell, these two articles, if the second one exists, are related in topic but definitely NOT the same): (I call this process "Cleaning References")

In my experience, the vast majority of waste in a software shop is time spent building things nobody wants on top of overly complex architectures that solve problems we don't have. Focus on that. Put solving your customer's actual (not imagined) problems first. 1/3

People tend to think of functional illiteracy as just being unable to read words aloud or such. What it *really* means is such as not being able to perceive subtext, context, understand who the audience of something may be, etc. Media illiteracy is fundamentally part of functional illiteracy.

Why? 1. Because some people care about ethical issues. Devs are using a tool (Claude) that is being used to kill people. And that's only one aspect. 2. Because some devs want to feel they are coding. They don't want to be passive observers.

Saving this one for reasons.

She told me that she believes that she (and other Tesla drivers) get too dependent on the screen, and that the screen doesn’t really see cyclists. Ok WOW. So, please, Tesla folks, look out for cyclists. I don’t think you’re evil or anything, just please don’t kill me please! Please!

****majors are not the only metric of a department or program’s value****

I have said this before, but eliminating programs that teach a lot of students but don’t have many majors is like a restaurant not buying flour in its grocery delivery because people aren’t ordering flour on the menu.

Yeah a lot of this has to do with the muscle memory that we develop when regularly manipulating physical buttons / levers / etc, most of which have a unique position and purpose in the cabin. All of that is lost with a touch screen which uses the same 12x12" (or whatever) area to control everything.

1. Kevin Gross and I have a new paper out today PLOS Biology. We used economic models based around screening games and the market for unpaid labor to highlight a meltdown cycle threatening peer review.

Writing helps you organize and clarify your own thoughts. When others read, it helps them do the same, whether they agree with you or not. Both of us probably need the reminder a year from now. This is why it’s valuable to write, even if your ideas aren’t unique.

Companies aren't hiring UXers much anymore, but the ones they do (and the ones that are already there) are no longer user centered. Deception and dark patterns rule the day. Just another sign of the rotten timeline.