two-star says, “Well, novels are capable of conveying widely different scales of passing time. Parser IF considers time at the pace of individual actions. Choice IF can do time at different scales pretty well though.”
— from the last meeting of the IF discussion group.
This is the contention that we’re going to be discussing at our next meeting (April 5).
Parser IF pieces, for the most part, involve actions that take roughly uniform amounts of time. There are a few exceptions to this approach, for instance games that make EXAMINE or REMEMBER or THINK ABOUT instant, or that assign different durations to all of the game’s possible actions. Eric Eve wrote an Inform extension, Variable Time Control, to provide this functionality (though I don’t know whether there are any games that actually make use of it). And some parser games do include significant scene breaks where the action jumps to a new location in time and space, sometimes (as in Photopia) moving backward as well as forward to narrate scenes out of temporal order.
But with vanishingly few exceptions, the player’s actions are usually small in scope (moving objects, choosing individual lines of dialogue, traveling short distances) rather than large (pursuing a job, having a full-length conversation with someone, moving house, etc.).
To what degree is that an inevitable feature of parser games? How much does it constrain their expressive ability? Are there stories that aren’t possible to tell this way? Are there aspects of current time modeling methods in IF that could be extended? Conversely, what (if anything) is lost in CYOA pieces when actions don’t represent consistent amounts of time?
One of the things we concluded at the end of our last meeting is that it would help to have some common ground for discussion set up in advance. This isn’t mandatory, so please don’t feel like you can’t join in if you haven’t looked at the reading, but on the club page, I’ve linked some things I think it might be relevant to know about before the discussion. I reproduce them here:
- Thread started by Jan Thorsby on pacing and the way that common parser game behavior interferes with it.
- (Added late) Hoist Sail for the Heliopause and Home, a parser-based game that takes place over vast expanses of space and time.
- Choice of Broadsides, a Choice of Games piece in which the scope of player action varies significantly from one node to the next. Even playing the first fifteen or twenty minutes of the game will demonstrate this variety, though you may enjoy trying out the whole thing.
- David Welbourn’s collection of time modeling examples, which provide a number of useful samples of how parser IF games currently communicate the passage of time to the player.
- If you really want to dig deep: Past RAIF threads on the passage of time in IF.
Many thanks to David Welbourn for collating the examples for us.
Emily Short's Interactive Storytelling 
Just for the record, I don’t know if any other games make use of my Variable Time Control extension, but Nightfall does (which is what I originally wrote it for).
Aha, that makes sense. Thanks!
Just to expand on that for anyone who’s interested, the purpose of Variable Time Control in Nightfall was mainly to prolong the game. I wanted to create the impression of passing time (partly to give a sense of urgency) but the way the game/story was set up meant that it had to end before dawn. If every turn took a notional one minute, there was a danger that the following dawn would be reached too soon (in terms of the number of turns), not allowing players time to explore the game fully and reach a more satisfactory ending. Having a means to make many actions take only a fraction of a minute was my solution to that particular problem.
I used Variable Time Control in Six and I’m using it in the thing I’m doing now. I’ll probably use it in every thing I ever do forever due to its ability to ‘take no time’!
Yeah, ‘take no time’ is the main part of Variable Time Control that I use.
I know of six games that use Variable Time Control by Eric Eve:
East Grove Hills (2010)
Ghosterington Night (2012)
If I Wasn’t Shy (2012)
Nightfall (2008)
Six (2011)
The Z-Machine Matter (2011)
There is also an extension called Native Time Control by Tim Pittman; three games use it:
The Adventures of Super Shlong issue 1 (2012)
Gasparillo (2012)
Peril in Pleasantville part 2 (2012)
Plus there’s Real-Time Delays by Erik Temple; nine games use it:
The Adventures of Super Shlong issue 1 (2012)
Andromeda Apocalypse (2012)
Gasparillo (2012)
Ghosterington Night (2012)
The House at the End of Rosewood Street (2013)
Kerkerkruip (2011)
Narrow Your Eyes (2012)
Peril in Pleasantville part 2 (2012)
Six (2011)
I suspect that this may have historical reasons: D&D emphasizes the success or failure of small local actions, so Zork does, so parser IF does. Since many newer pen & paper RPG / storytelling systems have chosen other granularities, I don’t see why parser IF couldn’t do the same.
It’s true this is a convention, but I actually disagree that that’s the only reason for this. (Having tried a few times to write games that stepped outside that convention, and been unsatisfied with the results. Admittedly maybe I was just Doing It Wrong, but still.)
An example I didn’t see listed is Zarf’s Hoist Sail for the Heliopause and Home. Actions are still small and localized, but there’s a sense of vast tracts of time that pass between each one (and this helps build a sense of “otherness” about the narrator, who we know is, at least, not strictly human).
(I’m not sure where, exactly, this’d fit on David’s list. Possibly under “Message of Time and Change?”)
Right, this was one of the goals of Heliopause.
In a sense actions *aren’t* localized. A “wait” or “fly” command can implicitly mean “wait until we reach the next star system.” Similarly “lower anchor” is shorthand for maneuvering in from the Oort cloud to an inner planet, which may take months. It’s all handled indirectly, of course, but I don’t think the effect would be spoiled if I’d added a calendar to the status line.
Though, I *do* think a calendar would weaken the dreamlike, fairy-tale quality of the piece.
A “wait” in my Persistence of Memory does a “wait until the next thing happens”, but it isn’t at nearly as massive a timescale (also, the clock is right on the status line).
Fair point. I’ve added a link to Heliopause above.
> adapt very early to the rhythm of the inflection points, and every time it shifts there’s a dissonant effect
I don’t think I was very clear here – I mean more the rhythm of events than the passage of time. I think the diegetic passage of time rarely has a primary effect on the rhythm, although it can have strong secondary effect of course – because the kind of events that are interesting over a span of years are not the kind of events that are interesting over a span of minutes.
I only lurk here, but I’d like to say these discussions make for fascinating reading. I understand the people here are likely busy with their own projects, and thus I really appreciate the time and effort they’ve spared for the betterment of IF as a whole, and am very thankful towards everyone contributing here :)
It is a treat seeing so many great figures in IF converse in one place – if Aaron Reed and Porpentine joined the discussion, I would probably have an paroxysm of joy. As it is, I’m already nerding out seeing Zarf and Alexis Kennedy in the same place, and am keen to see more of this happen in the future.
Do any of the standard actions in Inform & TADS have a longer timescale? I don’t think any vary, so games made in the mold of those tools would follow suit.
I was always proud of the extension Permission To Visit for both its timescale thing and for gating space with something other than literal locks and keys. Larger-scale movement like that is a nice hook for creating scene beginnings and endings, and well-formed scenes always read better to me in IF than IF’s usual run of unbroken time.
Well, if we were to try to realistically model them, not all the standard actions in Inform (and I suppose TADS) would take the same time. Searching is going to take longer than examining, going from one place to another depends on the spaces you’re modeling, pushing something could take a long time if it’s heavy, putting on or taking off could take a little while depending on what it is you’re putting on or taking off.
I think a big thing may be the way elapsing time interacts with puzzles. I usually find timed puzzles a pain, which gives rise to the timelessness that David Welbourn talked about in his links; the default is for the world to basically stand still while you try to solve puzzles, which gives you enough time to tinker and experiment to see what to do. And if you had a puzzle that required precise timing (say, the last one before you get in the gate in Christminster), it would be awful to have different actions take different amounts of time. If you have an overall clock that ticks down before you lose, I don’t know that it’d be that much more constructive to have searching take three times more than examine; the goal is more for the player to have a limited amount of time than for the PC to. There are some optimization-style puzzles where it might make sense to have some actions take more than others but I don’t know if this has been done. You’d have to signal pretty carefully which actions took longer.
Besides letting you examine or take inventory or look for free (which is also something that roguelikes do), the main use I can think of for variable time would be interacting with atmospheric effects, or things that were open some of the time but not all of the time, or time limits that were so enormous that players wouldn’t be expected to be in any danger with them (though I’ve often discovered these by running up against the time limit after getting stuck). Realism is likely to be the enemy of rewarding puzzle behavior here, I think.
Agreed. To be clear, my own interest (and I think the interest of the people who originally brought this up) is not in obsessively modeling real-time durations of short-term actions. I don’t care that it takes longer to eat a slice of bread than to cut the slice from the loaf, and I certainly don’t want to inflict such calculations on my player.
The issue as originally raised was rather: novels and choice-based stories both often deal in larger scale actions and in durations of time measured not in minutes but in days, months, or even years. Parser-based IF hardly ever puts the player in a situation to make choices on that scale, and rarely narrates long periods of time, though there are the occasional triggered cut scenes or jump cuts. But often parser games preserve, or come close to preserving, the dramatic unities, taking place in a small contiguous area over a short continuous period of time. That in turn puts a lot of constraints on the type of stories one can tell about developing relationships or character change.
In my opinion, the difficulty of introducing long-scale actions in parser IF is mostly to do with teaching the player about larger, more abstract types of action, and building an underlying mechanic that does something interesting with actions on that scale.
But decisions applying to long scales of time are often also single-use, and single-use verbs are horrible for a parser context: either you can’t get the player to realize he should use the verb at all, or he learns it and then runs around trying to use it in other places where it doesn’t go. Sometimes games get around this by embedding one-off options in dialogue: it might be a long-term action to choose and attend a college, but a short-term action to select a conversation quip to say “Mom, I’ve decided to go to Reed” (and then cut ahead to your first day in your new dorm room). This is effectively a way of inserting a choice mechanic into a parser game.
Heliopause carries the long game off because it does find a context — interstellar travel — in which discrete decisions can be imagined that nonetheless would then have to play out over vast expanses of time and space.
Resource management Ren’Py games may also be good exemplars of mid-length mechanic systems. I know several Hanako Games productions allow the player to decide what she’s doing for the day or the week, gaining different skills and resources depending on which way she decides to go.
The one built-in I7 verb that, if implemented, should advance time on a larger scale is “sleep.”
Anyone else up for a post-chat SpeedIF, wherein we go off and build games demonstrating the crazy ideas that we’ve just come up with?
There is definitely some allure to that idea.