March 4, 2006
Lebowitz’s Universe, part 1
As mentioned in the first post of this series (1 2 3), the primary designer of Universe is Michael Lebowitz (also, according to the acknowledgments in Lebowitz’s 1984 and 1987 papers, work by Paula Langer and Doron Shalmon made significant contributions to the project and Susan Rachel Burstein helped develop many of the ideas). The Universe system shares a certain intellectual heritage with Minstrel and Tale-Spin, and it also has another unusual shared feature in common with Tale-Spin. As we see with Tale-Spin‘s “mis-spun tales,” the most famous story attributed to Universe has a somewhat more tenuous connection to the project’s output than one might assume. Here is the story:
Liz was married to Tony. Neither loved the other, and, indeed, Liz was in love with Neil. However, unknown to either Tony or Neil, Stephano, Tony’s father, who wanted Liz to produce a grandson for him, threatened Liz that if she left Tony, he would kill Neil. Liz told Neil that she did not love him, that she was still in love with Tony, and that he should forget about her. Eventually, Neil was convinced and he married Marie. Later, when Liz was finally free from Tony (because Stephano had died), Neil was not free to marry her and their trouble went on.
Though a number of well-known books provide this as an example of Universe‘s output, in fact this is a summarization of a plot from Days of Our Lives. It appears in a paper about Universe as “an illustration of the kind of plot outlines we would like to generate” (1985, 172). Unfortunately, Universe has never been able to display the level of mastery that the authors of Days of Our Lives — a remarkably popular and long-running daytime television melodrama, or “soap opera” — have achieved. However, this story does nonetheless point to a number of important ways in which the goals of Universe are significantly different from those of systems such as Tale-Spin and Minstrel.
First, Universe is designed to to generate continuing serials: stories that never end. Second, Universe is working in one of the world’s currently most popular story forms (television melodrama) rather than the somewhat archaic (and more difficult for contemporary audiences to judge) forms of Aesop-style fables and tales of Arthur’s knights. Third, Universe‘s goals are defined in terms of what kinds of story and character structures it will generate, rather than in terms of the model of human cognition that the system’s operations will simulate.
The last of these is, perhaps, the most significant. While ideas such as “memory organization points” are important in the conception of Universe, the system is not presented as a simulation of a model of human cognition. Rather, it is presented as a means of generating a universe of characters and an ongoing plot that interconnects them. In fact, in lieu of any cognitive science theory, Lebowitz writes: “Our methods are based on analysis of a television melodrama” (1985, 483). This allows Universe to be designed specifically for the generation of stories, and of a particular style of stories, rather than for the simulation of the behavior believed to generate stories.
As with Minstrel and Tale-Spin, story generation in Universe is organized via plans. In fact, Lebowitz goes so far as to claim, “Story generation can best be viewed as a planning task” (1987, 234). But the goal of Universe‘s stories is to create complex interweavings of events between a core group of interrelated characters. Given this, before planning begins, Universe must create a cast of characters.
Universe‘s characters
A Universe story begins with the creation of characters, much as happens in Tale-Spin. But rather than a small number of characters who seem to come into existence at the moment the story begins, Universe creates an interconnected group of characters with somewhat detailed histories. This is accomplished through a character creation cycle. As outlined in Lebowitz’s 1984 paper, the cycle begins with a few characters in a queue, soon to become the universe’s (m/p)atriarchs, who need the details of their lives fleshed out. One character at a time is removed from the queue, and a simple simulation of that character’s life is carried out — focusing on the gaining and losing of spouses, the birth of children, and the possibility of death — until the present is reached. Any new characters created through this process are added to the queue. New characters aren’t created for each marriage, however, because the system may select an already existing eligible character (defined as single at the time of the marriage, of appropriate age and sex, and not directly related to the character). This begins to create interconnections between the families.
Once this basic framework of marriage, birth, and death is filled in, each character is further fleshed out. This begins by giving each character a set of traits, some of which are inherited from their parents (if known), and selecting a set of stereotypes that work well to explain those traits. Stereotypes include many familiar character elements (e.g., lawyer, doctor, gangster, big-eater, swinger, video-game-player) and have normal values for some traits and not others (e.g., “lawyer” has normal values for intelligence, guile, and self-confidence, but not for religion, promiscuity, or moodiness). Following this, the system adds further detail to the characters’ pasts by creating simplified versions of the sorts of events it will create in the present (once story generation begins).
And I’ll take up Universe‘s story generation further in my next post. For now, let me note that these are the publications of Lebowitz’s to which I’m referring in the post above:
Lebowitz, Michael. “Creating Characters in a Story-Telling Universe.” Poetics 13: (1984) 171–194.
Lebowitz, Michael. “Story-telling as planning and learning.” Poetics 14: (1985) 483–502.
Lebowitz, Michael. “Planning Stories.” In Proceedings of the Ninth Annual Conference of the Cognitive Science Society, Seattle WA. 1987, 234–242.
March 5th, 2006 at 11:36 pm
Noah,
It is clear that these (and previously described) systems are not quite complete for the English story generations. But what about reusing them for something slightly off-the-wall: language learning.
For (foreign) language learning, it is important to have a story with many sentences based on a simplified grammar and with a large number of shallow variations. An automatically generated story with adjustable grammar level (e.g. start with everything in present tense, then extend to new vocabulary, past tense, etc) would be useful for that purpose.
Does this make any sense or is it beyond such systems as you describe?
Regards,
Alex.
March 6th, 2006 at 12:37 am
Alex, it sounds like an intriguing thought. I know that branching story systems have been used for language learning (e.g., as part of MIT’s Project Athena, for example) but I don’t know if these sorts of story generation systems have been. Does anyone else reading this know?
March 9th, 2006 at 6:59 pm
Although I’m not familiar with the mechanics of Universe, if there’s any chance that the system could run into problems such as those in Minstrel, using it to teach might be a bad idea.
“The knight…did something…with the princess…hmm…ate? That can’t be right.”
If we’re going to try to teach comprehension with these generated stories, we’d better be sure that the stories they generate are comprehendable.
March 13th, 2006 at 10:57 pm
Yes,
Knight eating the princess would be bad even for translation purposes. However, the point would be to make mini-stories with adjustable grammar rather than long extremely exciting ones.
So, you could make a story about buying travel ticket or shopping for dinner prepartion and it would be different every time.
I think the language/scenario subset is a key. These systems already do this, but for language learning purposes, it might be possible to restrict the scenarios even further with even better results.
Regards,
Alex.
October 30th, 2007 at 5:46 pm
[…] sted a series of thoughts about two story generation systems: Minstrel and Universe (1 2 3 4 5). I had some critical things to say about the Minstrel system, but they were b […]
November 19th, 2007 at 10:31 am
[…] the approach taken by Universe last year, in a series of posts contrasting it with Minstrel (1 2 3 4 5). But, as with my writing about other systems, this was entirely based on reading publications […]