Review
Leave a comment

What Algorithms Want – Reflecting on (Human) Agency in an Age of Automation

One of the biggest questions the Silicon Valley regime poses is ‘how much should we let algorithms shape our lives?’ While algorithms filter the search results and headlines we see and match us with prospective mates and jobs, most popular analysis of them is a mile wide and an inch deep. What Algorithms Want: Imagination in the Age of Computing is dedicated to remedying this situation. Written by Ed Finn (an Assistant Professor at Arizona State University’s Center for Science and the Imagination) and published by the MIT Press this past spring, the book considers the impact of central digital age protagonists like Uber and Netflix while articulating how a constellation of devices, apps, and services have changed how we experience the world.

↑ Putting theory into practice: Norbert Wiener and his moth are one of many vintage precedents Finn contexualizes

Google PageRank, Amazon recommendations, Uber dispatch and routing – Finn explains how many of today’s leading cultural companies are basically “cultural wrappers” for algorithms. From Leibniz and Descartes right through Palantir the degree to which which computation has become a universal solvent in our quest to find and manage knowledge is clearly mapped out. Acknowledging our enduring embrace of cybernetic thinking, Finn describes how an algorithmic object of study is a system-in-motion and the most fertile spot for investigation is the gap between an algorithm’s intended functionality and its effects – that is where epistemological bias, problematic data or training, or unchecked cyberlibertarianism will reveal itself. Finn posits we need to look beyond ‘the glitch’ or ‘the crash’ (an approach gleaned from Paul Virilio’s notion of ‘the accident’) as “they are only windows into the broader opening between computation and reality. We construct the gap, or create space for it, on both sides.” We need to mind it, basically.

To cultivate this space between intent and effects, Finn conducts several thematic scans. These discussions often start with a discussion of a service, platform, or product and evolve into freewheeling, multifaceted analyses. We learn how the weird DARPA-funded (of course) pre-history of Siri colours her affable incompetence at processing our simplest requests, or what the popularity of Ian Bogost’s Cow Clicker tells us about contemporary attention and modes of engagement, and how the friction between Uber’s seamless UI and its wrecking ball-like disruption of labour and legislation paradigms is intense. While these discussions have all happened in myriad places before Finn really owns each service he tackles with nimble, thought-provoking commentary that deftly draws on both commercial and cultural references. “In refining ‘how fans are made’ … Netflix has constructed a distinct temporal aesthetic, a kind of eternal present or network time” he writes about the degree to which the wildly popular video streaming service has turned time, taste, and dissemination in television and film inside out.

All conversations around disruption must culminate with cryptocurrency and the blockchain! This are the rule these days apparently, and the penultimate chapter of What Algorithms Want is dedicated to value. While Finn starts out by hitting some expected notes on the strange post- high-frequency trading algorithms market – the flash crash, network latency, meditations on microseconds – he takes the discussion somewhere unique. In discussing the decentralization intrinsic to cryptocurrency he builds a bridge to the attention economy, and incisively describes how the blockchain’s inversion of the traditional relationship of privacy and transparency is the opposite of what PageRank does. This is all bundled in nuanced conversation about value and labour that somehow builds a bridge between Gawker and mining rigs – it’s great!

What Algorithms Want diligently observes how “cultural machines” have ascended to reshape everyday experience. From its charming opening consideration of Silicon Valley slogans (“to make the world more open and connected,” “evolving the way the world moves,” etc.) Finn clearly demonstrates that while operationalizing a business that harnesses computation is one thing, the way that its internalized logic impacts people, places, and flows is another. The book is exceedingly accessible – so much so that the world would be a better place if it was handed out on the first day within computer science and MBA programs. Beyond that, it’s a ‘must read’ for artists and thinkers addressing algorithms, apps, and web services as material within their practice; that latter endorsement is particularly important as, as Finn notes in his conclusion: if there’s one thing deploying algorithms widely necessitates, it’s voices and perspectives that make sense of the rapid change brought about by these pervasive and enmeshed systems-in-motion.

Get it from Amazon UK / Amazon US | What Algorithms Want | Ed Finn