Friday, December 14, 2018

Final Project link

Monday, December 3, 2018

Progress Report #2

I have thought more about how I want to make my project less static, and I think there are three ways to go. I can make a video showing typing, I can make it interactive, or I can make it a performance. This last one is a new idea and I think it could be cool. The idea is to have two speakers; the first one represents a writer who starts to speak a line, the second represents autocomplete and the this second speaker will finish the line based on what autocomplete might say, perhaps by interrupting the first speaker. I think this is pretty feasible and wouldn’t require more technical expertise than I possess.

Monday, November 26, 2018

Project progress

After discussing my project, I think it is too ‘static.’ My proposal was essentially just static text on a page, so I think I might want to make it a bit more dynamic. I think I want to show someone typing in things and seeing how autocomplete behaves in that situation. The other option that came to mind was to make the project interactive and actually make my own version of autocomplete that is drastically different from the standard in nature and purpose. This idea is exciting, but I need to consider how practical it is in terms of implementation.

Week 11

The idea that glass in cathedrals is thicker on the bottom because of the pull of gravity over time is a common misconception. The true reason for this is an effect of the glass manufacturing process. It is true that we can’t really cleanly divide the concepts of solid and liquid, but it is not true because of glass or gravity or anything real at all. In the real world we might be able to have solids in liquids, but we as humans, do not live in the real world; we live in our minds. We can talk about empirical or pragmatic knowledge but ultimately when we say ‘solid’ and ‘liquid’ we are just saying words, muddled and obfuscated attempts to express ideas through the messy development of human language over thousands of years. We can talk about the platonic world of forms, but we can’t live there. The author states that we can acknowledge the work of an actor or the criticism of a critic may be in part of fractionally ‘perfect,’ but fundamentally any notion of ‘perfection’ is essentially and absolutely inaccessible. I agree with the author that ‘all human activity, [is] absurd.’ I don’t think criticism is about trying to judge absolutes or even relativity; I think it is about movement towards an understanding of how we can perceive meaning from what we experience.

Monday, November 19, 2018

Project Proposal

Title: autocompleteness Summary: I’m going to write a poem exploring how autocomplete and similar bots affect human expression. Autocomplete is based on statistics. Based on what you have written already it calculates what you are most likely to write next. Of course not everyone lives at the center of this statistical distribution. By deferring to what is most common it implicitly disadvantages someone who is writing something atypical. Furthermore, as humans rely on bots more and more to write, this statistical software does more cause inconvenience; it makes it actually harder to express thoughts that aren’t typical. Well at least it could, in theory. Keywords: Bots, Poetry, Generative Details: The idea is to write a poem where part of each line is written and the rest is generated by autocomplete. The story told by the written part will diverge from what the autocomplete tells if taken at face value.

Sunday, November 18, 2018

Translation Experiment

I made a markov generator mixing two text, the original german of Kant's Critique of Pure Reason and a machine translation from an english version back into german. I call it 'Critique of Pure Translation.' It can be viewed here:

Monday, November 12, 2018

Week 9

I find the piece “Machine Translation and Global English” quite fascinating. I think indeed English has come to be viewed by many people as a universal language and many people look to English as the piece puts it: “the master code and origin of all other languages.” Obviously, this sort of view presents problems and can frankly be viewed as colonialist. However I think there is an interesting additional point that can be developed based on this dilemma. By virtue of how translation works, it is a function. That is to say given some input it produces an output, hopefully coherent in a different language for example translating Spanish to English. However the interesting point here is that a function can only produce one output to every input but this is somewhat incongruous with how language functions in actuality. There is more than one way to say a certain thing and it seems this may be lost in translation. Furthermore, as translation softwares must select a dialect and register to translate into, e.g. “Basic English” as referred to in the piece, so do other dialects and registers fall by the wayside. This too may be fraught with problems. Different people have different ways of speaking the same language and by selecting a standard form of a language, the other forms are ignored and obscured. As translation by machines becomes more significant in society, so too does the magnitude of this potentially harmful effect. For example, certain minority communities have developed their own dialects of English here in the US, e.g. Ebonics. If machine translators ignore this and only translate into Standard English, this may have the effect of furthering the divide between different communities.

Final Project link