#9: Wearable iridescence | Glitch in the machine | Algorithms and power
art | science | digital culture | design | creativity | words | tech
art | technology | science
Iridescence
Inspired by the Anna’s hummingbird’s gorget, IRIDESCENCE is an interactive collar equipped with a facial tracking camera and an array of 200 flipping quills. The Custom-made quills flip their colors and start to make patterns, in response to the movement of onlookers and their facial expressions.
As I prepare to teach a course in Wearable Technology for second-year undergraduate students, I’m digging around interesting projects and came across Iridescence by Behnaz Farahi.
It’s a remarkable work that melds sophisticated technological capabilities with natural bird behaviours with fashion in service of a broader concept of remaking relations between humans. The idea of wearable technology as a vehicle for creating and modifying relations is a key theoretical concept I explore with my students. Those relations could be interpersonal, between a person and their own body, or even autobiographical (a remaking of the stories we tell ourselves about ourselves).
There is a lot of interesting detail about this project on Farahi’s site but I really enjoyed seeing the Making of Iridescence video to get a sense of the technological details and fabrication methods.
I’m not sure my undergrad students can quite get to this level in a mere 12 weeks, but it’s surely inspirational!
art | digital culture
Glitch in the machine
I’m fascinated whenever algorithms go wrong. It mostly happens at the extreme limits of what they were designed for but really show how automated technology can break down. It’s amusing but also a warning for us as we rely on algorithms more and more, with the wide adoption of machine learning and neural nets making the inner workings even more obscured.
What got me thinking about the topic was a news story last week you might have seen that showed this on the US National Weather Service radar around San Diego:
It’s not a huge rainstorm on the way. It’s a “bloom” of ladybugs, taken to the sky. There is a report on it in the LA Times which describes the bloom as being 80 miles (~130 km) across! You can see two versions of the radar animated image on Twitter:
and click through to here. You could say this isn’t really a glitch—just something unexpected, but when we have algorithms making decisions, such as to redirect flights coming into an airport that had a huge storm on the way, we start to have more serious consequences.
There are lots of different kinds of glitches from automated systems but I quite like the ones to do with maps. So here are a few.
Artist Clement Valla has put together a collection he calls Postcards from Google Earth. It includes great images like these:
Peder Norrby has collected a set of glitch images from Apple Maps on iOS. Here are a few that come from the app trying to make 3D versions of locations using automated processes:
And here are some glitched images taken by the cameras on top of Google’s cars that drive around for their Street View service:
Glitches are more common than we realise, I think, and they are a sign that the technologies we use are right on the verge of failing. But they are also a useful tool for artists who exploit glitches to test the limits of systems—to break them, modify them, misuse them, and potentially reinvent them.
technology | digital culture
Algorithms and power
I came across a tweet by accident that showed the following slide from a talk by Catherine D'Ignazio (@kanarinka):
It was posted by Moritz Stefaner who photographed it onscreen during D’Ignazio’s talk at EYEO2019. The table comes from a book by D’Ignazio and Lauren Klein called “Data Feminism”. You can read a full draft of the book free but it will come out officially in 2020.
The last line of the table really caught my eye. There is a huge push in some sectors for more people to understand computation and algorithms but this is the first time I’ve seen the idea challenged in this particular framing.
Last week I read an essay this reminded me of. Bill Thompson, an insightful writer on tech and society, wrote on his blog about how the division between science and humanities decried by CP Snow in his famous Rede Lectures in 1959 has been somewhat replaced:
…today’s fundamental division is not between science and humanities but between those who understand computational thinking and those who don’t.
At first glance, this statement fits in the left-hand column of D’Ignazio and Klein’s chart above. But I know Bill has much more sophisticated views than that quote taken without the context of the essay. Indeed, he points out the human aspect of code:
Today the code running on the machines was developed, written and tested by other human beings, and the worldview embedded in that code comes at least from human bias and prejudgement.
He then points out that the new generation of algorithms, driven by neural networks and machine learning are completely obscured from human understanding. We don’t know and we can’t know how these algorithms work beneath the surface. He poetically puts the consequences this way:
Something more than mere computation has been loosed upon the world, but this time we know what rough beast is slouching towards the silicon foundry to be born: it is the soul of the new machine; the source of the lines of non-light arranged in the mind that will one day constitute cyberspace; the emergent force that through the fibreoptic fuse drives the algorithm; the dawn that will break behind the CCD sensors as agency emerges from our craft and silicon art.
I think it is this observation that ties the perspectives together, but a nuance that is lost in all the hype about getting people to learn to code and a nuance lost when people talk about the importance of understanding algorithms. Learning to code matters because it is a path to deeper understanding not so much of the skill but of how computers and algorithms are embedded in our current cultural circumstance and how they transform our history into our future.
It’s always good to hear from readers so let me know what you found interesting here. You can reach me directly at physicsdavid@gmail.com.