August 7, 2007
“Narrative Variation in IF” Dissertation Online
My dissertation, “Generating Narrative Variation in Interactive Fiction,” is now online.
If my slides and defense summary interested you, you can check out this long form writeup of my work. I guess there will be an “official” version coming from ProQuest/UMI before too long, too, via their open access option. I think that “official” in this case will mean that the papers I printed out will be scanned in again, imbuing the result with the authority of the page.
August 7th, 2007 at 6:10 pm
congrats again, I look forward to reading it soon.
August 8th, 2007 at 3:28 pm
Yeah, UMI is no longer free. I think (can’t recall, though) that at participating institution you can
get the first 24 pages. And yeah, I think the first page of that might even be some junk they stick on
there to make it “official” as you say. But the general public at home you can’t get more than the titles
these days from their site, I’m afraid. I’m looking forward to reading this through, when I get a chance.
August 8th, 2007 at 7:03 pm
Ah, thanks for this. Saves me coming hat in hand to ask for an electronic copy. Looking forward to the read.
August 9th, 2007 at 10:38 am
[…] th, 2007 | Links | alexjc | Share This
Nick Montfort of Grand Text Auto has just posted his dissertation online. It’s entitled &# […]
August 9th, 2007 at 12:23 pm
I signed up to have this appear in PQDT Open, so the whole thing should be available through ProQuest for free, even though other UMI material isn’t. This option was only recently made available, and you have to jump through a few hoops (open access wasn’t advertised as being available at Penn) and a pay an extra fee.
August 13th, 2007 at 2:35 am
Excellent work, Nick. This is one of those thesis I wish I wrote.
Have you given any thought to how ‘nn’ could be adapted to a MUD environment? I’ve been toying for a while with the thought of making a new MUD with up to date parsing and NLG algorithms.
August 13th, 2007 at 8:42 am
[…] ps’ review of Planetfall. I’ve never gotten around to playing this one myself. Nick Montfort’s dissertation on nn, an IF develop […]
August 13th, 2007 at 1:38 pm
Having begun to read the thesis (not finished it), I am happy to see an attempt to divorce
the worldmodel from agency. It would be correct to say there are multiple agents inhabiting
the IF game: narrator (and facilitator- for save/restore/help/breaking the fourth wall/etc),
PC, and NPCs. I think they all deserve their own agents, and I am happy to see the most important
division being made- that between the PC and the worldmodel itself. All the other objects that
inhabit a game- brass laterns, containers, chairs, doors, keys, etc. are something other than agents.
What is clear to me from a CS perspective is that while a message passing interface will suffice
to allow all the worldmodel stuff and the non-narrator agents (PC, NPCs) to get along, there is
a lot of work still yet to be done to allow the narrator to work optimally. The narrator “container”
contains a lot of attributes and some of those include the plot arc, in order to tell the story
in different orders, from different perspectives, etc. But in order to take the results of all
the message passing, collate that into attribute switches that reside predominantly in the narrator
object, and then determine what “summary text” gets displayed on any arbitrary game move, it seems like
it requires a lot of nonlinear processing to get it right. Without an *actual* demonstration game
to show the ability (or inability?) of the agency model to cope with typical IF-game level stresses,
it’s hard to evaluate whether nn is a “useful abstraction” or whether it is the difference between
trying to compute armed with a knowledge of Roman numerals and an abacus versus with a Cray and a fortran
compiler.
> STILL AWAITING INPUT
August 14th, 2007 at 10:30 am
Nick– thanks for your generosity in making this easily available. So far, it looks quite solid.
August 18th, 2007 at 1:59 pm
Thank you for making this available, I look forward to reading it soon, Nick!
August 24th, 2007 at 12:36 pm
Nick, thanks for letting everyone read it,
am I right to think that nn is currently implemented in Python?
August 24th, 2007 at 9:41 pm
Ash, that’s right. I am planning to make the Python system available, too, as soon as it expresses the ideas of the project a bit more usably and coherently.
September 29th, 2007 at 3:10 am
[…] arget=”_blank”>Familiarity breeds contempt: Buidling game stories that flow, and of course Nick’s recent dissertation on a new IF system foc […]
October 2nd, 2007 at 3:43 pm
[…] ction) So, I promised last month to Nick Montfort I would be posting on his dissertation. Then I got stalled for various reasons, due […]
October 18th, 2007 at 9:55 am
[…] 55 pm (Interactive Fiction, Narrative Variation) Continuing with the dissertation, there’s an interesting section in 3.3 […]