Raph’s Website » Exploiting the space

Exploiting the space

March 27th, 2007 (Visited 3911 times) Tags:

We’re all too ready to do things in familiar ways. You know, when we were working on music systems for Galaxies, we thought very much in terms of loops and whether we could get all the sound to synch up. But maybe this video of the Reactable music prototype shows that instead, a multiplayer music system might be better suited towards soundscapes.

When I look at what they did here, I say to myself, “they exploited the uses of a table.” What are core characteristics of a table? It’s usually large enough to handle multiple people. It is a surface on which objects are placed, and interesting things happen in terms of what objects, where, and how they are positioned relative to one another. Using the traditional metaphor of musical instruments would never get you this instrument. Instead, you’d end up with light-up piano keys on the table surface.

What are the core characteristics of virtual worlds, and how do we exploit the space in a way that simply wouldn’t otherwise?

I think it just goes to show that often, we are so bound up by our preconceptions of how things work that it can be hard to break out of them. So now I am thinking along these lines… the core characteristics of online worlds are

the simulation of space
persistence
multiple users in the space

so we should be thinking of game systems and social systems that are driven above all by these three characteristics. If you’re willing to bear with me a moment — forget all the trappings of combat, elves, spaceships, whatever, and think very abstractly with me.

Space means, among other things,

distance
vectors
relationships
sizes
eyelines

Persistence means, among other things,

duration
constancy
periodicity (e.g., cycles — they only become visible with time)
history
a potential for slow paces

And the quantity of users is, of course a quality of its own.

Are these qualities, then fundamental characteristics of the virtual world that we ought to be exploring more in our designs? These would be things that are to varying degrees hard for other games. Many have commented, for example, that Animal Crossing: Wild World feels like an MMO in some ways. Is this perhaps because it make use of the “slow cyclical behaviors” pattern that can only arise with a certain sort of persistence? And that this pattern is something that is a core element of MMOs?

This suggests that we ought to be leveraging these unique qualities more in MMOs, because they play towards the strengths of MMOs. Among the things that come to mind are:

Systems affected by distance or proximity between players, player emplacements, NPC centers, or even characteristics of the space
Systems affected by spatial relationships between entities
Systems affected by sine wave or other periodic cycles moving in the background of the simulation, perhaps at very slow rates
Systems affected by historical events at given locations
Systems affected by the length of existence of a given characteristic at a given location

Obvious and non-obvious applications towards nifty game systems would include things like

systems that rely on slow cycles, like seasons, annual festivals, and so on. Perhaps magic is affected by the three-month long cycle of the year. Perhaps spawns shift cyclically: still predictable, and therefore not obnoxiously random, but more varied than now. Maybe the pulsating heart of evil is on a one month beating cycle.
systems that rely on long-term persistence and a sense of history, like archaeology of past great deeds. Imagine coming across relics from the first team to ever complete a given raid, a year before or longer. Picture game elements that accumulate only over the course of months of activity by players — and not ones that one player has to manage constantly through the process, like player cities, but things where everyone can do a little here and there, and it still counts. Maybe the more players do a gvien action in a place, the more resistant or susceptible to that action the place becomes.
systems that rely on spatial simulation. MMOs treat space very literally. But we don’t make use of its special qualities. Obviously, there’s things like territory we should be exploring, but what about systems based on proximity, line of sight, and so on? Thinking again in terms of magic systems, mana whirlpools or “gravity wells” that affect topology, or magic based on ley lines, so that tactical positioning mattered. Higher ground actually affecting combat. Or, as in this cool musical table, a group chanting magic system based on people’s proximity and relative positioning. But you could expand this to whatever — formations for combat, performances for music, whatever. Can you imagine a world where every player was actually one of those blocks? As users moved around, they would change what was going on, and they could effectively make dances that actually created the music…
combinations of the above, of course. Picture a ley line node that increases mana, but that remembers what sort of spells were peformed there over time, based on the relative positions of the mages involved…

When I read an article like the recent one on Gamasutra about “fixing MMOs” it strikes me as a curious mix of importation of features from elsewhere, and elements that leverage the real power of MMOs — but I’m unable to put my finger on why. And maybe the Reactable gives me a clue. Always try to leverage the strength of your medium — cut with the grain, not against it.

(the Reactable video came via Todd McKimmey)

You can follow any responses to this entry through the RSS 2.0 feed. Responses are currently closed, but you can trackback from your own site.

shaneneville.com wrote on March 28th, 2007 at 11:27 pm:

Raph’s Website » Exploiting the space

play late wrote on March 30th, 2007 at 3:43 pm:

http://www.raphkoster.com/gaming/etech07.shtml and some relev. data: http://money.cnn.com/magazines/business2/business2_archive/2007/04/01/8403359/index.htm?postversion=2007032305 how game dev’s think: http://www.raphkoster.com/2007/03/27/exploiting-the-space/ New ARG from Jane McGonigal http://www.worldwithoutoil.org/ Twitter…tired Dodgeball…wired (btw, Google bought Dodgeball)

users persist space persist users « Kooneiform wrote on March 28th, 2007 at 12:01 am:

[...] persist space persist users Posted March 28, 2007 Raph Koster threw down a pretty good post today. In the light of developing this RP mush (if you don’t know a RP (roleplay) mush is a [...]

Alathon's BYOND Website: Alathonia wrote on March 28th, 2007 at 1:54 am:

[...] I’m of the latter school, and as such I present the following article he wrote yesterday: http://www.raphkoster.com/2007/03/27/exploiting-the-space/ This presents some really interesting possibilities, and certainly has inspired several additions [...]

COMM394 Daily Notes wrote on March 28th, 2007 at 3:57 am:

[...] Koster’s comments about exploiting space w/ Reactable music prototype (and link to video demo of device) [...]

Dan said on March 27th, 2007 at 9:06 am:

“You know, when we were working on music systems for Galaxies,”

Speaking of which, it’s been years since I was in SWG, but every now and then I find myself humming parts of the in-game music (“waltz” I think). In 6 months of playing it I think I spent 90% of my time inside the cantina, working on my musician/entertainer. I don’t suppose you know anywhere where the musician’s music is available? (Of course, not that it could compare with some of the flourish sequences the guild I was in came up with, but still, it’d be nice to hear some of those tunes again)

chocorisu said on March 27th, 2007 at 10:31 am:

I thought Puzzle Pirates did all the above particularly well. Every archipelago has a slightly different sense of history and community, simply because the journey between each is *slightly* further.

It does an especially good job of being *multiplayer*: not just competitive, but inherently cooperative. You *need* other people. There have been huge rivalries between flags and the larger crews, and the game explicitly supports it, with wars and blockades and so on.

I would love to see more games with the same flavour. I want games that can ONLY be multiplayer right from the start, not tacked onto the end of a piss-poor RPG. Preferably no bloody elfs or goblins, either.

Brask Mumei said on March 27th, 2007 at 10:31 am:

A neat idea for building a sense of history I’ve seen is to automatically build dirt paths according to players actual travel routes. If one couples this with a reinforcing thing, such as giving players a speed boost when on roads, the change of the routes over time will give a sense of history.

Allen Sligar said on March 27th, 2007 at 10:37 am:

Was thinking about a few points, the piece “the more resistant or susceptible to that action the place becomes” where concentration of [N] player activity occurs, [Y] results and contributes to the establishment of [Z]?

This is were is begins to correlate algorithmically to a “live ecology” ala GG&S.

Think farming (requires growth seasons) [N], then think evolution of a village (NPC/PC) [Y] and subsequent spawn of guards [Z] to protect active settlement.

Think deer [N] who like to eat carrots in gardens, of village [Y], and the wolves (or Orcs) [Z] who like to eat the deer. And the young dragon who has a penchant for humans, Deer and Orcs……what does his dynamically accumulated wealth attract?

I think you said something about this, and dragons, and such in development of UO, and issues where the AI capability was not available then, is it available now?

Because a world where player activity effected the meta systems and the subsequent consequences of those activities would be very cool…..

Lots of work nailing down the dependancies though

Great post Raph

chocorisu said on March 27th, 2007 at 10:43 am:

Perhaps worth adding that it eventually lost its charm (for me) for the same reason I only play Animal Crossing in bursts of a month at a time, spread 6 months apart. Eventually the puzzles get old and the goals become meaningless. The world simply isn’t dynamic enough, especially in the case of AC: everyone’s town/island is essentially the same.

I would say that’s a vote in favour of even more dynamic and/or user-generated content, but on the other end of the spectrum is Second Life and I’m not sure that’s quite what I look for in a game either.

Tim Holt said on March 27th, 2007 at 2:41 pm:

Not precisely thinking along the game design route here but…

It would be nifty to take that basic Reactable idea and have it create music from non-direct controlled human items. If you tracked players and NPCs in an MMO live and considered each to be one of those little “sound thingys” could you come up with a way to make them create music that went with “the flow” Or do the same with a live webcam feed of pedestrian traffic?

Bart Stewart said on March 27th, 2007 at 3:08 pm:

Being the systems-geek I am, I also enjoyed this post. It’s great to see someone bring a system-level perspective to MMOs.

In fact, I’d love to see someone take a look at MMOs in general as examples of living things. James Grier Miller’s Living Systems suggested a list of features that all living systems have — I wonder how closely a MMO might fit that classification, and what if anything it might reveal about the results of poking a MMO in various ways to see how it responds.

One addition I’d suggest to the space, persistence, and numbers list of core features of MMOs would be “communication.” We assume it, but maybe we shouldn’t — maybe there’s something special about giving a lot of people in the same space multiple ways to share information. LinkedIn is one example of this in the general Web space — is this something else that’s still very much open for exploitation in MMOs?

Finally, Brask, I actually once suggested that very idea about “emergent roads” in the offical SWG forum. I think it was about 0.3 microseconds before some Black Hat pointed out that players would just use this feature to run around to spell out dirty words.

I still can’t decide whether that was a valid objection or not.

–Bart

Wolfe said on March 27th, 2007 at 11:43 pm:

This is a line of reasoning which appears relatively obvious when listening to what players say they want to play with in an mmorpg. The problem with what players say is usually that they are very direct, or lack the abstract reasoning and end up appearing to ask for impossible features. If you help them by lifting the request into a high level of abstraction and then pull it back down into direct mechanics you will in most cases be able to come up with designs which are very popular ideas and still doable.

Except for the risk aversion problem which prevent the idea from actually being built.

magicback (frank) said on March 28th, 2007 at 6:11 am:

The central question is how to make the world vibrant and live. How do we generate the feeling of going to and revisiting an exotic locale like Thailand or the Grand Canyons?

The list of things Raph noted are the things that antropologists (sp) or archeologist (sp) look for.

The idea of legends, urban myths, and other psychological (non-system based) beliefs that pops up just indicate that players are looking for things that leverage the medium, the “virtual” aspect to make it a bit more tangible, a bit more immersive.

Frank

furrycat said on March 28th, 2007 at 10:10 am:

Quoth Dan,

> I don’t suppose you know anywhere where the musician’s
> music is available?

You can extract it from the game assets if you still have them. You’d need some kind of TRE extractor program to do it. There was one called InsideSWG which you might look for. I don’t know if it deals with the new TOC files that they added later on but it handles the original assets.

Anyway the samples are stored under player_music/sample and there are some in data_sample_00.tre and data_sample_01.tre as well as any later patches.

Each song is split into song intro and outro, loop and 8 flourishes. There’s one WAV file per instrument for each of the above categories.

Once you have those files you can stick them into your favourite audio application and turn the nostalgia knob up to 11.

Andy Havens said on March 28th, 2007 at 12:34 pm:

Raph: I’m gonna be a bit of a twit here and challenge all three of your original assumptions… at least on some pretty crunchy, esoteric level.

Yes, there must be space… or we don’t have a world in our VW. But the way in which we interpret and “virtualize” the geometry of the world really determines if space = space in our RL sense, or *how* space = space in terms of that interpretation. So before we set ourselves the task of developing space-based systems/features for a world, don’t we need to determine some basics about how we are going to understand space to exist in it?

For example, I was struck, last year, upon re-reading, “Snow Crash,” at how… well… dumb it seemed that there wasn’t a simple “think/teleport” system in place. There’s that race from the other side of the world, and it’s a big deal for the plot, and… well… in SL, you’d just click on your bookmark. Yawn. Racing in a world with hyperlinks isn’t possible. In which case, “space” is, essentially, there to serve as visible “scenery” backdrop and, possibly, a part of tactical level issues. As opposed to some MMOs where, with no teleportation, portions of the game become a “running simulator,” which is boring as hell. Well… if you eliminate the “boring running,” but still want to have space/distance be meaningful for other issues — racing, boundaries, borders, etc. — you have some thinkin’ to do. Also… vertical space vs. horizontal space is insanely important in our real world, but less so in some virtual worlds. Can I go underground? Can I fly? Hover? Easily jump over a one-foot tall curb? How much of the 3D-ness is bounded by vertical restrictions. Again… I’m not really disagreeing with you, Raph, but suggesting that there’s some thinking to do about the characteristic, even before we get into desiging with it.

Persistence has the same issues, I think. The real world has levels of persistence. Are we to model them accurately? Or interestingly, but differently? Is the persistence to be predictable and based on classes of behavior and types? Or is it… chaotic?

For example, if I can do “X” and it is persistent, and you can do “Y,” and it is, too, should those things be able to interact in both our absences? IE, can my persistent dog eat your persistent cat. And when we come back… how would we know that that’s what happened? Should we? Unless one of us set up a persistent security camera or a mindful talking parrot?

As to multiple users in the space… I agree that they (we) need to be there. But questions about when and in what numbers and for what reasons go back to the root of what “multiplayer” even means. Is golf, for example, a solo game or multiplayer? I have a friend who says that golf is the loneliest game in the world, and even more so when played with friends.

These cranky musings aside, I’ll try to come back later and provide more helpful thoughts on the topic…

dys said on March 29th, 2007 at 12:15 am:

Well Raph, one of the most annoying thing in SWG actually was music.
I can’t say how much fed up I ended hearing the same old track when entering combat.
It was cool for the first 2 days, then it was just boring and repetitive.

furrycat said on March 29th, 2007 at 3:15 am:

That was the combat music which did indeed get old. But the entertainer music was great. It’s only a shame that they couldn’t implement the system Raph talked about … erm … somewhere else … on this blog … where entertainers could compose their own songs instead of simply chaining together riffs.

I always wished they could do that so to see Raph say that it had been considered was nice.