Music Video Telephone

Telephone is a game in which participants whisper a phrase person-to-person, and see how it evolves as people guess at words they mishear.

The following music video for True Thrush takes this a step further, giving participants one shot to view and memorise a short video, before asking them to recreate it.

Telephone is entertaining because people's natural automatic error correction (tendency to recognise and reproduce actual words) fights with the noisy communication channel of a quiet whisper. The True Thrush video is more about the unreliability of memory and creativity, and what details seem salient.

Newton's Fractal

Grant Sanderson aka 3blue1brown has a wonderful channel on Youtube where he creates accessible yet deep educational maths videos. The are literally all very good, but the last two are a really great introduction to a topic in fractal geometry, and a demystification of the Mandelbrot set, presented with supreme clarity and fantastic visualisations. Probably requires high-school or first-year undergrad maths to really understand the technical content, but if you have interest in the topic I think you would get a lot out of this even if you have no formal education.

Why Is NFT Art So Ugly?

I came across this entertaining essay about NFTs (digital certificates of authenticity traded in a speculative asset bubble, aka "non-fungible tokens"), not from a financial, technological or environmental perspective, but instead from an art-criticism perspective.

As soon as you actually try to talk about this art as art the whole thing sort of falls apart, it just absolutely cannot stand up to the scrutiny. Doing so is about as cringy for the writer and the reader as it is for the viewer of the art itself, which I have to think is why the entire art world seems committed to talking only about the technology, the transmission mechanism, the great great value, all the swirling bullshit AROUND NFTs rather than, god forbid, the amateurish nonsense itself.

Plenty of social and environmental issues with NFTs too, of course, but those have been discussed to death by people more knowledgeable than me.

Edit: the follow-up essay "Why is NFT art so lazy?" is also good, focussing more on procgen in and out of NFTs.

My Universe

I turned on Radio 1 today for the first time in years. While I know essentially nothing about the current pop music scene, the unique combination of my decades-old pop music knowledge and a tiny slice of contemporary pop music knowledge (courtesy of my far-more worldly sister) allowed me to identify this song as a Coldplay/BTS colab purely from the singers’ voices.

Unfortunately I don’t really like it :/

New preprint: "Sensorimotor distance: A fully grounded measure of semantic similarity for 800 million concept pairs"

Together with my colleague and lab PILouise Connell, I have developed a new measure of semantic distance between concepts. It is based on the senses and body parts involved in experiencing those concepts — in other words it is fully grounded in sensorimotor experience. This sets it aside from other measures of semantic distance, such as those based on distributions of words in language, on encyclopaedic databases, or on lists of properties or features. It also is fairly comprehensive (thanks to the expansive norms collected by colleagues), with distances available for nearly 800,000,000 pairs of concepts.

Two panels. Both show arrangement of dots labelled with concepts. Left panel: select nouns for tools, emotions, fruit and celestial objects. Right panel: select verbs for leg, hand, mouth and cognitive actions
Left panel: select nouns for tools, emotions, fruit and celestial objects. Right panel: select verbs for leg, hand, mouth and cognitive actions. Within each panel, positions are based on sensorimotor distances between concepts, transformed into two dimensions using Sammon mapping.

The measure is described in a new preprint, and you can search, visualise and play around with the distances (e.g. the above image) using an online app I also developed.

Let us know if you do anything cool with it!

"Have one fewer child"

I’ve been looking for suggestions of individual actions one can take to reduce CO2 emissions. In particular, ones which actually make a big difference (unlike, e.g., switching to LED lightbulbs). Everywhere I look, all I see is "have one fewer child" dwarfing all other actions by effectiveness. It's all anyone talks about.

Graph showing comparative impacts of different climate actions. "Have one fewer child" so overshadows others that the y-axis has been clipped to accommodate it.
Wynes & Nicholas (2017, Environmental Research)

The Web is full of figures like this one. It shows the one-fewer-child recommendation being equivalent to about 60 t/y, where the next-best options like “sell your cars”, “stop flying” or "be vegan" are mostly in the 1–2 t/y range. So high, they had to clip and compress the range on the graph so you could even see anything else. That looks pretty stark. That figure is so high, it seems as if nobody could ever hope live sustainably if they had even one child. They might as well be taking a long-haul flight every 10 days for the rest of their life! Now, we definitely shouldn't discount conclusions just because they are surprising or uncomfortable, but we should scrutinise them. That feeling of "…really?" is the first hint something might not be right.


Every now and then when I find something online which I want to remember or show to somebody — usually an image or a video — I save it in a text file; one per person. Literally kilobytes of the stuff. And even more occasionally, I look back through those text files, and post something I find there on this site. (At this rate it'll be the year 3888 before I get through it all.)

This is one such thing: