Kurzweil Gets A Patent For Poetic Software

theodp writes "Ray Kurzweil, inventor of the Kurzweil Reading Machine for the blind, has developed what he calls a cybernetic poet, software that allows a computer to create poetry by imitating but not plagiarizing the styles and vocabularies of human poets. A sample: 'Sashay down the page...through the lioness...nestled in my soul.' Impressed? The USPTO, who sponsored the Independent Inventors Conference Mr. Kurzweil spoke at on Nov. 17, seems to be. On Nov. 11, Ray Kurzweil received U.S. Patent No. 6,647,395 for Poet Personalities."

Comments

As a human being

[Anonymous Coward]

As a human being, I think I am qualified to judge, and that isn't poetry. Even Frost could tell that.

After looking at this closer...

[clifgriffin]

I'm unimpressed.

It's AI seems only capable of duplicating style...but it turns out peoms that make no sense. It seems to have no concept of word relationships, outside of simple grammar and organization.

Like I said, gimme Robert Frost or Emily Dickinson...who needs this?

Clif

Re:After looking at this closer...

[Pike]

The sad thing is that most modern poetry really isn't any different from the stuff this program produces. Randomness and Hip Vagueness have pretty well killed any popular taste for poetry. After all, why read poetry when most of it appears to have no meaning and have required no talent?

This is where modern art has led us. The end result of trashing common sense is the heat death of the literary world. Everyone is a poet, therefore no one is a poet.

This person said it rather well. I have this only to add: the question is not whether art should change, but whether art should become intrinsically worthless.

-JD

Re:After looking at this closer...

[HiThere]

Most poetry during ANY period is trash. The difference now is that literally anything can be published.

Nearly all great works of art that we know of were panned severely when they first appeared. A great work of art creates something that is unexpected, and which we are unprepared for. (Not that I'm claiming that THIS was great art...most, as I said, really *IS* garbage. But don't judge based on initial reactions.)

Most good works of art are appreciated... and performed on commission. They are refinements of prior works and ideas. This doesn't make them less powerful, but it makes it easier for people to appreciate them.

Many schools of art don't really have room inside them for many great, or even very good, pieces. So people who keep trying for great novelty are continually trying to create totally novel ways of expressing themselves...ways outside of any extant school. Unfortunately(?), there appear to be limits as to what people can, even over time, learn to appreciate.

If one is willing to be satisfied with good, and very good, however, there are many classic schools that appear to be deep enough that any one person can never plumb their depths. The saga is one such form. It's not popular now because it DOES require a prolonged attention span to appreciate it. And it's difficult, requiring much craftsmanship. in it's place we have positioned the novel. A form that is at least as deep, somewhat wider, and which doesn't require as much skill to produce acceptable works. And which also can require less attention on the part of the audience. (This last *isn't* guaranteed. Many very good novels require, or at least reward, the same degree of attention that any epic poem can require.)

OTOH, even quite restricted formulae, e.g. the Haiku, can be quite expressive over a wide variety of issues. (Here I mean the strict form of Haiku, including the strictures of seasonal references as well as length and stress patterns.) For that matter, if it weren't for historical context (e.g., it's popularisation by Edward Lear), the Limerick might well be an equally expressive form. I've done a bit of experimenting, and I don't find it intrinsically any less or more confining than the Haiku. But the audience expectations mean that it can be difficult to deal with serious topics (unless the wry twist is a part of the point).

As to "modern poetry". Perhaps you should choose a different selection of poets. Julia Winograd, e.g., is a noted modern poet, and her works are quite accessible. They aren't, however, light. She lives among the poor, and reveals the darkness that they dwell in, without being maudlin. I know that you can purchase her works at Codys Books in Berkeley, although I don't find them in the on-line store (apache internal error). And Google doesn't seem to know her. But she has many collections published...self published, actually, but they've been on sale for years.

P.S.: This may partially explain why you think modern poetry is bad. I hadn't realized how difficult it was to find her works. Perhaps the publishers won't publish anything that they find offensive. After all, poetry isn't a moneymaker except on a very small scale. I do know that even recognized authors have difficulty getting poetry published. You may be able to find Logan by Paul Edwin Zimmer (or possibly Zimmer-Bradley). It was published once that I know of, and deals with classic american themes. In this case how the Iroquois nation was destroyed, and by who. And is in a classic form. And it was only included because 1) his sister was a best selling author, and 2) the editor was determined to include it. Yet it is a poem so moving that I had great trouble reading it. It should be a part of every history cirriculum, as it covers the facts of an important period of early american history. And it explores the nature of political action. In it's way it is similar to "Advise and Consent", but it is more factual. (Well, possibly not. I don't really know the background of "Advise and Consent".)

But it's poetry, so nobody pays attention to it.

Re:Says more about modern poetry then Kurzweil

[Pike]

Poe's poems, like any good poems, have meat because they were vested with real thought, effort and genius by their author. As such they have intrinsic merit.

You can't really think that anything on that program's page is just as good as

Ah distinctly I remember it was in the bleak December,
And each seperate dying ember wrought its ghost upon the floor

or Frost, for example:

Some say the world will end in fire
Some say in ice;
From what I've tasted of desire,
I hold with those who favor fire
But if it had to perish twice,
I think I know enough of hate
To say that, for destruction, ice
Is also great,
And would suffice.

Artists have lost the idea nowadays that real art has intrinsic value proportional to the real talent and effort that goes into it. What is this idea that words are just generic symbols, devoid of any of their own meaning? Words have well-defined meanings as well as emotional value, and this is why we use them and what makes them powerful in art.

syntax vs semantics

[incal]

Simulation is just not enough. simulated fire dont burn my flesh. simulated poetry dont burn my mind.

Let Kurzweil simulate Hoelderlin. or ee cunnings. how far he can go with them?

(anyone has read: poietic software?)

Computer chips on a wet black bough

[Txiasaeia]

I can't believe nobody has asked this question: what's the point? Humans write poetry in order to express a thought, idea or emotion; it's just as hackneyed if I read a bunch of Frost or Empson poems and slice and dice 'em together to create my own masterpiece, so why is it better that a computer can do so?

When I read poetry, I like to have the illusion that what I am reading was carefully thought about and created; trying to find meaning in a computer generated poem is as pointless as trying to find meaning in a book from Borges' Library of Babel.