Yeah, I’ve tried not to be too much of an annoying fanboy. I’ve stopped and thanked several people briefly, but have so far resisted the temptation to add to the endless string of people wanting to talk to Graham Nelson and Emily Short. 
I missed most of the morning yesterday. I’m staying in an AirBnB about seven miles out, so I brought my unicycle and thought I’d ride in. And I’m a full-time small farmer, so I’m an early bird – I thought I’d have time to stop by the Arnold Arboretum for a bit, and…yeah, by the time I got there it was about 11:40. So I hung out in the expo room with Oxygen Trail for a bit. Dunno what I was thinking bringing that: it’s not really a narrative game, it’s a graphical resource management game told primarily through text. And…yeah, written by three programmers. I did have a couple people play it, and Lorri Hopping had some very well-considered thoughts on how it could be more of a narrative game. Sounds like she’s directing (or moving towards directing) a company/project on her own dime and I’ll be very interested to see how that turns out.
The talks I did see:
I apparently didn’t take any notes at Cat Manning’s talk? All the usual things about scope creep and how it happens and how to combat it, with verve and style and examples from her experience. Very well done.
I was debating whether to go to Graham Nelson’s talk or Aaron Reed’s: I was tipping towards Inform since I’m not an analog storygame sort of person. I can squint at them and go, “yeah, that would be a lot of fun for a certain kind of person…who isn’t me.” But then a group of people walked by whose work I respect and said something to the effect of, “well, it looks like Aaron’s talk is the one to go to” which tipped me back the other way.
So Aaron’s talk was great. Packed to the gills: I think there were people standing all along the back wall. He was talking about rules or mechanics to set players up for success in collaboratively telling good stories, and asking how we might adapt that to the digital space. Being a DM is a learned skill: how can we make rules that help people tell good stories and don’t require quite as much from them in terms of structure?
He started with a game that Andrew Plotkin apparently designed around 2007 on the Myst forums: two players post alternately on a single forum thread: one is from the present-day (archaeology-ish) and one from the past of a culture. Each starts with 10 fact points (my words here: they were just “white stones” in forum): they can spend one to mark an element of their post as narratively important. And one…maybe you’d call them resolution points? You started with one, and could earn extras by combining two elements and resolve them so that one (or both?) are no longer important. You could spend one to re-open a closed element. Or spend two to contradict an element, and then the other player has to explain why they were mistaken.
So the mechanics limit story length (limited number of fact points), allow collaboration (you mark things that the other player has to respond to, but there is no out-of-band collaboration). The resolution points are somewhat rare, so you can spend them to have upsets and surprises in the story line, but not too many.
Then he roughly divided actions into four categories: Generation, Storywrighting, Negotiation, and Administration.
Generation - generating ideas, etc. Even just the setting of the game is a big influence: the associations of a crystal skull are much different in D&D vs. Call of Cthulu. The original D&D was apparently notorious for having tables of percentages of penalties for all kinds of things with percentages that were too small to bother with (1-2%) – old age, illness, lots of more imaginative things – but they’re idea generation helpers: you might want to put these sorts of things in your scenario.
He mentioned Jason Morningstar’s 2015 game Juggernaut, where you draw predictions about the future from a deck of cards representing an infallible computer oracle, and it’s a game about rationalizing how those predictions come true (as a player, your job is to make sure they come true in the resulting story).
And Annals of the Parrigues as an example of generation in collaboration with a computer.
Negotiation – He mentioned Drachen & Smith’s 2008 paper with a title that starts Player Talk and appeared in (Computers in Entertainment) about negotiating with children when playing with blocks? And Polaris (Ben Lehman, 2005) which has 12 “performative utterances” which are formalized language for the narrative actions. It being a game with two opposing players, the most important is “but only if” (and he read an over-the-top dramatic example from the manual?).
Administration – I didn’t take note much here: I think he just said it may sound boring but there is interesting work to be done in the space of how and what you track. He mentioned Apocalypse World by Baker & Baker in 2010, which lays out what the Moves are, and is a system successful enough to spawn a number of other games.
Then a numbered (?) list of things, briefly: Expressive Input, Expressive Output, Common Ground (can we find a world/language/etc. that lets computer/human agents act on the same representation? Brenda Laurel Computers as Theatre), Supportive Play Partners, A Broader Range of Stories (Dream Askew/Dream Apart, Alder/Rosenbaum 2019), Understanding Structure (story structure: Robin Laws Hamlet’s Hit Points and Hillfolk).
Katherine Morayati’s talk on Writing IF Like a Pop Song used pop songs as a source of metaphor, drawing parallels between structure in songs vs. structure in narrative, starting by comparing intro/verse/chorus/…/outro to branch-and-bottleneck, where you can go off to a variety of different verses, but then come back to the branch point (chorus). She was asking what we can learn and apply to IF from this: polish the intro first because it’s what everybody sees first. If the chorus comes again and again, polish that more than the verses which will only be seen once.
Pacing & contrast in volume, syllables, emotional tenor (cheerful songs about gruesome subjects). Time cave -> songs with seemingly no related pieces or normal structure. Minimalist songs -> limited parser work. Songs with a twist. Bowie’s Verbasizer.
The core message of the talk was how to find where something sags and where it works well, and that the structure in songs is more visible than that of movies, so if we draw parallels to music, it might give us new insights into tightening up our stories.
Lightning talks:
Ian Michael Waddell talked about failure being fun, and it was everything you’d expect from the author of Animalia. I don’t think there was anything particularly new, but it was a very fun talk.
Toiya Kristen Finley talked about adding sensory details: they add subtext and encourage player imagination. “OK” writing often lacks description or sensory details and simply descripes the action in such detail as to stifle imagination. Don’t over-focus on sight: describe the other senses too! Verbs & adverbs are descriptive too (examples from Nine Hours, Nine Persons, Nine Doors). Don’t go too overboard unless you want purple prose. Very well-presented.
Mark Baumann gave a talk on their work towards a multiplayer IF tool. It was about what you’d expect from a computational scientist and made all the usual mistakes that people pitching new IF systems usually make. And he ran over 5+ minutes on a fifteen-minute (?) talk. There were a couple of interesting ideas buried in there, but nothing earthshaking.
Ben Schneider talked about writing short: coming from his initial experience writing response “barks” for units in real-time-strategy games. Some good thoughts, and some interesting numbers. He said he had linked his slide deck from his twitter (DrOctothorpe).
Seven miles? On a unicycle???