Beginning with understanding algorithms
Algorithms influence every big and small decisions that we take during our everyday life. Whether it is about choosing what book I should read next, which tv series to binge watch or what stock should I buy to increase my capital income, and so on. Ed finn describes Algorithms as cultural machines that shape human culture by acting as filters to the information that reaches us. They do so through a series of “naturalised” process that considers machines as an integral part of living and working environment. We now think, read (and reflect) through a computer. They are shaping our thoughts, our visions and thus the kind of culture we develop. And when capitalised on such a culture that shapes us in a direct and individual way, those who create algorithms can produce and even manipulate our values.
As machines get more advanced, they will feel still feel primitive to us until they mimic human ‘intelligence’ to the best accuracy. Super AI as discussed by Ed Finn will be more sophisticated than human intelligence. It is the time to think how we treat such a machine and intelligence? To think of it as a parent-child relationship, we have to continue to encourage the growth of the system. But to what extent will the child who grows into an adult, takeover? Will some routines and jobs remain “machine proof”? Ed fins talks about creativity which has been expanded and multiplied through algorithms. Algorithms work at a different level than human thinking. There are multiple dimensions of data whereas human struggle to keep track of even 4 dimensions. Which is why algorithms surprise us, they produce unintended consequences while still trying to adapt to human thinking. Google’s DeepMinds is investing to accelerate growth in machine learning and big data. We are seeing new things being introduced such as complex games, self-driving cars and other such technologies that people once thought would take decades for computers to produce. A hybrid between design and physics-mathematics-computer science can define a design problem into smaller parts with a purpose of systemising the process and defining logical pattern that can be interconnected with each other. As designer’s how might we abstract and multiply outcomes produced by these algorithms to solve for multiple contexts OR (another simple question) what are the things we want to solve and that algorithms can’t do such as the concept of plural navigation? (2)
Oribaka — A one way sound language to teach origami
The challenge was How Might We create a sound only language which could stand alone instruct users to follow and make origami structures. My individual responsibility was to learn all the fold of origami as per http://origami.me/ and iterate a vocabulary of the language. This required to simplify the instructions given in the website and bring out only the common words used to instruct. After we framed the vocabulary, Apoorva and I framed the instructions for each fold. I participated in the initial testing and recording of the lessons. The exercise to devise a new language introduced us to the notion of deconstruction wherein we created new “terminologies”. We simplified logic of instructing origami and found a pattern for our new language that can be applied to give any instructions for origami. The effort we put into making this happen made me realise that it’s not going to be easy, But I was quite amazed and contended with the result and the group effort.
References :
- http://mitpress.podbean.com/e/the-serendipity-of-semiautonomous-systems/
- ApoorvaAvadhana. 2019. Plural Navigation: Navigation for a Wholesome Experience. In 31ST AUSTRALIAN CONFERENCE ON HUMANCOMPUTER-INTERACTION (OZCHI’19), December 2–5, 2019, Fremantle, WA, Australia. ACM, New York, NY, USA, 4 pages. https://doi.org/10.1145/3369457.3369545
