News

StoryNexus Developer Diary #2: fewer spreadsheets, less swearing

By Alexis Kennedy, August 5, 2012 · Failbetter Tagged with ,

Long past due: some discussion of why we chose the approach we have for StoryNexus and, indeed, most of our other work. That approach is QBN, ‘quality-based narrative’, which is to say, little bundles of interactive story (storylets) whose appearance controls and is controlled by changeable state (qualities).

Very broadly, the two approaches to building interactive stories are branching narratives and world-models. Choose-your-own-adventure stories, and the narrative underpinnings of most CRPGs, are branching narratives. Parser-based interactive fiction and Dwarf Fortress are world models.

Building branching narrative is conceptually very simple, the barrier to entry is really low, and you can produce something very quickly – but it’s terribly easy to make broken stories or get yourself into content debt, and the player experience is limited and inflexible. World-models make for more deeper and more satisfyingly complex experiences, but the barrier to entry is much higher and it’s harder to finish.

It’s a continuum, of course – you can get world models with branchy elements, and branching stories which track state. QBN is a compromise between the two.

It’s very easy to put together a storylet or ten – as easy as building a CYOA story. But you can’t get lost in big, unfinished trees, because – by design – after every choice you have to describe the effects of that choice with quality changes. This naturally brings the story back to the same place. It helps creators be disciplined about interactive effects, and it gives the player more freedom in what they choose to do next.

(That said, we may ultimately allow creators to stitch two or more storylets together to make a bigger tree – especially after seeing what people have managed with Varytale. It is hard to manage longer dramatic scenes with such a staccato rhythm. But the simple rhythm of setup-choice-result, setup-choice result is the StoryNexus idiom.)

We also wanted, from the start, to allow collaborative world creation. You can build branching narratives collaboratively, but it means spreadsheets and swearing. But if a team agrees some basic rules – we won’t give out more than x of a quality at a time, we won’t advance this storyline quality without checking – then it’s easy to add one or a dozen storylets into an existing narrative, to grow it like coral. (This is, of course, what we’ve been doing with Fallen London for years.)

The coral metaphor also works like this: QBN stories don’t have to be complete to be playable. If a branching narrative is incomplete, you hit a dead end. If a QBN story (or a world model) is incomplete, you can play and experiment. You might run up against the buffers or an under-construction sign – but you can go back and try something else. This allows creators to build worlds in perpetual beta, and to bring players on to the team over time.

And a pragmatic point. We wanted creators – including ourselves – to be able to charge a little entry fee for bits of story and particular choices, while leaving as much of the game free-to-play as we could. QBN is a really great fit for this – it’s a natural, unintrusive and frankly tempting way of selling content, without the fourth-wall-breaking effect of something like DLC.

Finally, we wanted to make stories that could be consumed in small chunks, like baklava or dim sum…. but over days or weeks. A table full of baklava or dim sum that you pass by every day and pick a couple of bits out, anyway. I love big, sprawling, immersive, lose-a-week-of-your-life stories – Planescape, Deus Ex, Morrowind, Psychonauts, Baldur’s Gate, all the greats – but I have less and less time to devote to that kind of completely immersive experience. By doling out story in smaller chunks, we allow the sense of scope and pacing that you get in larger narratives, without necessarily needing to build a twenty-hour time sink. I appreciate Fallen London does have a certain amount of pacing filler in it… which is one reason we never intend to build one game as big as FL, ever again.

So: pieces of story like mosaic tiles, not pipes or complex machinery. Put them together. Build something wonderful.

7 Comments

Leave a reply


Standard Patterns in Choice-Based Games | These Heterogenous Tasks Jan 27, 1:56am

[…] they generally impose some more linear-progression structure on it. (Alexis Kennedy uses the term quality-based narrative to describe the general approach: ‘pieces of story like mosaic tiles, not pipes or complex […]

Alexis Kennedy Aug 9, 2:45pm

Yes, but not in the very first phase of the beta. I go into slightly more detail at the bottom of this page: http://community.failbettergames.com/topic734-how-storynexus-will-work.aspx?Page=0

kate Aug 9, 2:42pm

Will we be able to use our own artwork in the games we create? I'd really like to make something with my sister who is primarily an artist!

Alexis Kennedy Aug 9, 12:54pm

It's still tucked away somewhere private - but it'll be out very soon (probably ahead of the creators' tools.)

Genny Aug 7, 12:03pm

Is Cabinet Noir somewhere we can all look at it? Or is it perhaps having a private moment with your beta StoryNexers?

Alexis Aug 6, 2:35pm

Vael - we'll never build a single world as big as FL. But as per dev diary #1, we expect universes to consist of many worlds. If players love a franchise, we can always build more related worlds, but it doesn't need to be a megalith.

Vael Victus Aug 6, 4:50am

It's interesting what you've said about your intention to never build games as large as Fallen London again. I'm wondering what you think of the "end game" of smaller projects, then. How would you deal with the player really getting invested into the story, but soon after that, it ends? Let's say they've played 80 hours worth of actions and there's 20 hours left. They then complete the story, and now - plah. Would you create an end-game where they could put in 10 more hours of actions to get a few more storylets, until finally that is gone too? And as a creator on a platform where these games can be described as in "perpetual beta", how can you know when to finally say "This is the end, there is NO more content!"?