ilteris kaplan blog

Week 1:Computational Poetry and Mac Low

January 24, 2006

My code is not working as I want it to be, but then again if it were what would be the reason of taking this class :) So I am going to post it in the end of this post. By the way I am curious how Shiffman is using #IDs to skip on the page through titles since I haven’t figured that out in WordPress.

Whatever today I was thinking when walking from the school, can an algorithm write/ read a poem? Hmm. To answer that question we should first address what is a poem imho. What makes poem a poem? Could it be the verse, metrics, or context or how intense it is, it is a difficult question to answer. I find similarities between poem and music as I dig deeper, what makes a music, music? So for me the answer is somewhere between the saturation and our past lives experience which builds how you perceive your surroundings thus giving you your taste. So with this approach, yes an algorithm might write and read a poem.

Another approach though is purely mathematical, can an algorithm write and read a poem, well my answer to that would be, no way! I think I am becoming more holistic lately (after I had this operation maybe :) That algorithm could give you the perfect output in linguistics but still that kind of reductionism couldn’t be compared with worst poems I know. maybe. I borrowed Virtual Muse from the library last week , Hartman gives some insight both conceptually and technically how to approach a computer poetry with examples. I got the impression of his words that the best generative poetry is the one that finds its way, like evolve itself, as opposed to imitating the way human approach poetry.

[page 72]
“…
AutoPoet embodied an innappropriate idea of poetry. As long as the goal was the imitation of the human poet ’ or as long as the poem’s reader was encouraged to think that was the goal ’ I wasn’t likelt ro get any farther. What’s wrong with the AutoPoetry I’ve quoted here (an all the other reams of it the machine would produce until it was turned off) is exactly that it’s imitation poetry. All our habits of reading are called upon, all the old expectations, and then let down. ‘Monologues of Soul and Body’ had worked because its ‘body’ sections were so different from human poetry, It had successfully demanded its own way of reading…”

I cannot say he succeeds in finding that working formula of the perfect poem but you should read the book if you are curious about his projects[since I don’t have neither the english skills nor the memory to remember those], it is rich in content and he mentions about Mac Low throughout the book with some of his projects which was the main reason I got this book. But yet, there is some good potential on the book about the projects that might pop for the final.

So if we come to assignment part, I was really curious of what Mac Low wanted to achieve through his diastic readings and thought it might be a good exercise to re-write this algorithm. Only with errors :) I am posting the code below, we can talk more about it in the class, the main problem is, it screws with multiple words of phrases. Actually I have realized it tonight… So I am thinking to improve this for next week, with couple of new optimizations hopefully.


To Download


Written by Ilteris Kaplan who still lives and works in New York. Twitter