Tag: machine learning
My political theory of poetry is fairly straightforward: poetry is the most primitive form of communication. Poetry is the creative use of a limited resource. So the nonsense that comes out of mouths of anyone learning the vocabulary & syntax of a new language is also, clearly, poetry. As the limitations of language lessen, the need for creativity lessens, people opt (typically) for the most comprehensive method of communication.
What a marvel the first bird must have been!
I’m going to test this theory by teaching a machine to write poetry. I hypothesize this will be much easier than teaching it the intricacies of typical speech. This is a fairly easy supposition for me to make, as most machine-generated texts already read like they are a new language. I have no idea how long this is going to take, but I’m starting here: Machine Learning Mastery. Once I can swing the hammer correctly, I’m going to start my digital child pounding on the 1000 most common words in English.
If I raise it up as it should go, maybe it’ll even take over Poetry 4 Free duties.