It is this time of the year again—(drum roll)it is the SUMMER OF CODE. Quite a few programming related conferences took place this summer. And, thanks to the age of Zoom, I could finally afford to attend a couple of wildly different conferences. I guess you could say that most of them were industry-focused. The organisers probably would disagree with me and would argue that it was for the open-source community. But hey, it was rather hard to look past the fact that most hard-skill workshops were sponsored by various companies. And don’t get me to start with that false promise of the full-stack web development ‘speedrun demo’ (i.e. how to build a fancy website way faster than average)—which turned out to be only achievable through subscribing to a ‘digital product’ (a meaningless term that was invented for exacerbating a false sense of ownership by subscribing to a digital service IMO.) However, I also came across The Future of Code Politics, which focused on the cultural critique of the Tech industry, with speakers primarily from an artistic and/or activist background. True, who would want to be a party pooper at an industry event, while thousands of computer science graduates were here for business networking and ultimately, recruitment opportunities. Critiques always come easier from the outside.
Of course, I was pleasantly surprised to see quite a few talks on AI and Machine Learning Ethics in the industry conferences too. Some ‘got it’ more than the others. What I meant by ‘not getting it’ is seeing Ethics as a computational problem, which is solvable with mathematical means, such as using a ‘de-bias’ function in your code.
‘Rather, the bias lies in the setup as such…which therefore involves a priori choices and decisions as well as unacknowledged biases. Interpretation hence constitutes the setup, while at the same time being disclaimed by the analysts. Hermeneutics, in other words, is always at work in analytics, though it is rarely acknowledged as such,’ as commented by media theorist Florian Cramer, responding to the idea of a ‘de-skewing algorithm’ featured at the MIT Technology Review.
In my early days as a Statistics student, pretty much all lecturers recommended students to read How to Lie with Statistics (1954) at some point, just to remind us that statistics are pretty vulnerable to manipulation. If I were a statistics teacher now, I would probably throw in additional material, such as Signal and Noise (2013) by Nate Silver, who became a celeb statistician (Yes, this exists) for his work in the area of polling predictions, and Weapon of Math Destruction (2016) by Cathy O’Neil, a former-maths-professor-turn-hedge-fund-analyst. Despite their political differences (Silver refers to himself as a libertarian/liberal and O’Neil ran Alternative Banking Working Group for OWS,) their work both demonstrated how post-capitalism is the driving force of the statistical panopticon. This led us to my favourite question from this summer of code, which is posed by Dorothy Gordon in her keynote: ‘In the age of artificial intelligence, it (the stereotype of a programmer) has evolved into the evil programmer …Most good programmers are simply working to the task. They are given a set of parameters and they work within those parameters…Can they be expected to take those issues into account? Because we know that most programmers are working to earn a salary. What is the reaction of the people that are in control?’