Thursday, August 31, 2006
I've goofed yet again
I owe Jeffrey Nelson an apology. I spanked his Poetry Cube in Monday's post. I hadn't realized that it was a teaching tool, not an independent art work. I deeply regret the mistake.
Jeffrey's real work exemplifies why computational art will eventually prevail. Take a peek here--you'll see. Amazing stuff!
And Jeffrey, Thanks for the gracious response.
Wednesday, August 30, 2006
My code can beat up your kid's code
This afternoon, I coded this java statement:
infinitive = Verb.class.getMethod(stemMethodName,String.class).invoke(null,verbFormConstructedFrom).toString();
I'd like to say that I coded it spontaneously, that I'm a master of Java reflection, but it took an hour or so to get right. The problem was that I didn't realize that any method so invoked must be public. I'm used to C++ where you can assign private functions to pointers to members.
But that's not what this post is about. It's about needlessly complicated lines of code--like the one above. We programmers really do like to show off and one way to do that is to string together long sequences of object references without ever declaring a variable to assign them to. The code above would be much more readable like this:
Method method = Verb.class.getMethod(stemMethodName,String.class);Much easier to understand, but a lot less fun as well. That said, during my consulting years, I often noticed that hard-core computer science programmers tend to write in succinct but obscure styles while business programmers tend to write in a more verbose, but simpler style. One style privileging self and the other, the Other .
infinitive = (String)method.invoke(null,verbFormConstructedFrom);
I am more and more convinced that computer programming as writing and therefore as communication is constrained by socially constructed value systems, that the values of the programmer are evident in his or her code--indeed that it is impossible for it to be otherwise as the code is the programmer.
Monday, August 28, 2006
Sigh.....
An example of why the literary world tends not to take us very seriously: The Poetry Cube. Though I'm sure the authors worked very hard on this, it's neither original, nor particularly interesting, or even clever. Paraody perhaps? Oh well....
Friday, August 25, 2006
The originality problem
As the "easier" problems with ALG (grammaticality, surface realization, form) come closer to being solved, we're left with the fundamental problem, the essential problem, of how to make new texts seem original. Poetry generators produce boring work, not the first time, but somewhere between the 5th and 100th time. Simply put, an artificial author's poems start to all look alike. Jump over to my ETC site and see. Erica's work is full of participials and personal pronoun subjects and objects--so many that, even though ETC is not template based, it starts to appear to be.
Many electronic text products rely on randomization or weights to decide what trope or form they should use at a given point in the poem. ETC's abstraction index is an example. That index is really a probability, not 0-100, but 0-1. For lines other than questions, the software simply does a weighted selection on the index and if it comes back true, writes a fragment. Over time an abstraction index of 75 will result in 3/4 of the lines being fragments. That's just way too predictable.
If you read a collection of Wallace Steven's poems, you find way too many references to snow. It works in "The Snow Man," but after a very brief while, becomes tedious. Exaggerate the effect via the machine and the results pretty much suck.
I'm beginning to think that ALG systems, to be effective need another component in addition to the standard document planner, micro-planner, and surface realizer: a trope and word choice memory. A way for the machine to say to itself, Damn! That sounds really cool, but I used it in another poem--I can't use it here. But thinking that and designing it (let alone implementing it) are quite different things.
Wednesday, August 23, 2006
I'm baaackk!
I'm back home--urgent personal business forced an early return. But we still did over 7500 miles and 16 states in 20 days. Since we were just around the corner (Flagstaff), we made it back to New Jersey in four days (rain slowed us down a bit).
I could use the trip as source for a few posts on sights seen, roads ridden, nuggets of wisdom gleaned, motorcycle touring tips and tricks, and friendships deepened--but there's enough of that kind of pap around already. So just this one bit of pap: If you're ever ordering lunch in a roadside cafe in New Mexico and your waitress asks if you want green chiles on your burger, say No.
Wednesday, August 02, 2006
Slip sliding away...
I've been nearly silent the last few weeks. I've been finishing up a major project and preparing for the fall semester so that I can get away for a while, which means I'll be completely silent for the next few weeks. I'm off to Sturgis with a few of my motorcycle buddies--Harley riders all. (We admit to the rights of other folks to ride Hondas and other unnatural machines; but though we are polite to them in public, we don't associate with them socially, nor do we permit our children to form intimate relationships with them.) After Sturgis, we're off to make the great Western tour, we figure about 8,000 miles in four weeks, which if you don't know motorcycles is a real lot.
