Hybridization Question

[Wormwood]

I'm currently looking for the theoretical underpinnings of the statement that in a GNS hybrid has a primary and some number of secondary modes. I've viewed this statement as a simple, and extremely unhelpful conclusion of GNS theory, but other people keep using it in a way that argues otherwise. As such, I'm looking for the consensus view of what it actually means.

My current perspective is this is simply the conclusion that two real numbers have zero probability of being equal, and hence one will be larger than the others. This view explicitly does not say that there is some minimal difference between the modes, or that the difference in level is even observable, except in the limiting case.

Since both of these are often argued as corrollaries of the above statement this view would seem disjoint from the one (or ones) currently in use. As such I would like to understand the distinction so that I can better develop the technical play view of GNS.

Thank you for your time,

  -Mendel S.

[Gordon C. Landis]

Mendel -

One the key insights in GNS is (I think) that, ultimately (a vague qualifier, I must admit), you can't truly Prioritize more than one mode.  The fact is, having S as Priority frequently gets in the way of having G or N as Priority.  That doesn't mean there is no "stuff going on" that's consistent with the other modes, but Priority is defined as "the mode that wins when it conflicts with the other modes."  And if you switch which mode that is all the time . . . well, you don't really have a Priority then, do you?  Which is to say, if what you want is G, but S or G win the Priority choice frequently - your G fails to happen.  Hybrid's (as I understand 'em) refer to the fact that, if you're careful, you can let (e.g.) S win some of the Priority-battles and still end up with an overall N-Priority.

Some people claim that what they want is just a bunch of G-stuff, N-stuff and S-stuff going on - they don't care about Priority.  Well, GNS-as-analysis looks at what happens, not what they care about, so I guess a lot of the time a mode will end up being Prioritized anyway.  But as a practical matter, I'm guessing that with the right (or is it wrong? depends on the people, I guess) group play might be able to go on for a long time without a G, N or S Priority surfacing.

NOTE: Especially to Marco, if he happens to read this - I'm thinking that  that GNS-as-analysis thing is the only place GNS has[/i] to be intention-neutral.  GNS-as-overall-theory is entirely willing and able to look at what you intended (with people substituting whatever cognitive replacement for "intended" they feel neccessary) - it's just that what you want does NOT define what happens.  If you "intended" N, but S happened - S happened.  That should not, however, stop us from saying "hey, I wanted N - where did things go wrong?"

Back to no-Priority play - the observation of most folks here (include me in that group) is that such play isn't very satisfying.  Note that I'm not saying that play with mixed aspects of G, N and S is generally unsatisfying, I'm saying play where no true Priority emerges is generally unsatisfying ("generally" because maybe for some folks - those who are really only participating for socialization or the like? - not having a Priority is actually a good thing, because it stops the game Priority from displacing their actual out-game priority.  Hmm - I find that an interesting angle).

So, ruling out the no-Priority possibilities - having G or N or S as Priority is something people want.  They defeat that by having muddled Priorities (I think that's what gets called Incoherent play, but I confess I haven't kept up with that terminology debate).  That's why it's not just about the proportions being unequal - it's about the modes (as Priorities) being mutually  . . . well, if not exclusive, certainly antagonistic.

Does that make ANY sense?

Gordon

[M. J. Young]

Mendel, I'm not sure I'm going to be much help, because I disagree with the statement. I don't think hybrid play is real; I think it's a confusion based on a misunderstanding of the difference between verisimilitude and simulationism.

GNS primarily describes play; all other applications, whether to players or to game designs or to anything else, is derivative. In play, a player generally has one GNS priority, one thing that is his motivation to play the game. It might be the creation of myth (used in the sense of a story which reveals truth by exploring issues); it might be the promotion of the ego (through overcoming challenges); it might be the exploration of imagination (discovering the possible and the dream). One of those things is going to be the driver for why the player plays a particular game at a particular time.

I am not arguing that players can't play for different reasons; goodness knows that I play Reversi for entirely different reasons than Marco Polo, and that my Tristan's Labyrinth is intended to evoke a different kind of play than my Orc Rising.  What I am saying is that at any juncture one of these is the reason for play.

What happens, though, is that in some game circumstances, that thing that motivates me isn't implicated. If I'm gamist, there are going to be moments in which what I do has absolutely no bearing either on my ability to overcome a current challenge or on my preparation for a future challenge. At that moment, I'm making a decision based on a choice between the remaining two GNS concepts. Chatting up the princess could be an option here; it won't help me in the least in my gamist objectives. Do I do it because I think there are themes to explore about a romance between a princess and a commoner, or do I not do it because I think it's entirely unrealistic to think a princess would talk to my character? To some degree you are implicating GNS here; in some sense you have a secondary priority on which to base decisions when your real priority doesn't matter. Yet in another sense, what you're doing here is saying that what matters to you doesn't matter at the moment, so you'll just pick something that feels right on other grounds.

So maybe that's hybrid play; but I think in calling it that, you suggest far more value placed on the second value than it merits.

In game design, hybrid design is an effort to stimulate both a primary mode for the important decisions and a secondary mode for the unimportant ones. It's saying that you don't want the players to fall back to sim when they aren't doing gam--you want them to fill the gaps with nar, so you're going to attempt to promote that as the secondary choice. To do so requires extremely complex considerations of how you keep the primary and secondary from getting in each other's way. It also leaves open the question of why you're bothering to attempt to dictate this lower tier at all, instead of letting players drift according to their preferences.

Anyway, that's my view; it's subject to revision, so I look forward to the comments of others.

--M. J. Young

[Wormwood]

So, if I'm reading this right, the statementis justified by applying the discretization of GNS priority to the creative agenda (i.e. the continuum GNS) and observing the results lie within one of the four regimes (G, N, S, and none-of-the-above). Then the secondary pick occurs by applying the same discretization to the GNS continuum which is not in the image of the primary priority. Essentially, it's double dipping the discretization technique to add a second dimension to lend the prioritization greater analyitic power.

If this is the case, I am uninterested in the discrete theory, since artifacts of this sort appear as a result of the discretization. Rather in discussing the continuum theory, there seems to be singificantly more predictive and analytical power, even if the analysis is more complex. This necessarilly places technical play GNS in the later realm, since discretization would destroy much of it's theoretical advantage, in addition to the usual damage to the continuum theory.

However, I am also interested in determining if there is a dynamical feature which causes the behavior that the discretization generates by accident. I suspect that some people holding this idea have exactly that reasoning, but I could be wrong. Are Gordon and M.J.  in the consensus on this then?

Thank you for your time,

   -Mendel S.

[Gordon C. Landis]

Mendel -

Whoo boy - that last message is just not communicating anything to me.  The only thing I can tease out of it is something about discrete choice from a continuum of GNS.  Best as I can tell, you see GNS as a continuum until that "instance" gets labeled by an observer.  My understanding of the theory is that it rejects that notion; play is nothing-labelable (not a continuum), until it can be identified.  The "discretization" is GNS-analysis - there is nothing until that happens.

But like I said, I'm just not seeing what you're trying to point to in that post, so maybe I should just shut up and see what other folks say.

Gordon

[jdagna]

It was my understanding that the conventional wisdom about hybrids was not theoretical but practical - meaning that you might theoretically be able to balance two or more modes perfectly, but no one had ever seen it succeed in practice.  Coherent/functional games appear to use one primary mode and zero or one secondary modes.

This makes sense to me on many levels - just look at multifunction devices.  Sure you can buy a PDA/cell phone/mp3 player combination, but if it doesn't perform any one of those functions as well as you need, then you're better off with three specialized devices.

Also, I don't think a hybrid game is necessarily played with both of its modes on at any one time.  Instead, I think the idea is that the game facilitates one mode well, and another mode pretty well, and then individual groups pick which mode to use.  For example, Riddle of Steel is primarily N, but with secondary support for S.  However, instances of play are still either N or S, not both.  The game has simply provided the tools with which to play either mode more or less functionally.  Or, another way to look at it: the game has elements that will satisfy the N player in an S group (or vice versa).  This also makes sense in light of my multifunction device example - you can use the PDA/cell/mp3 device for any of those functions, but probably not for two at the same time.

The search for a functional G, N and S hybrid would be something like the quest for a perpetual motion machine.  It's impossible... but not impossible enough for people to stop looking for one.

[AnyaTheBlue]

I think understood what Mendel was saying.

I think he's talking about the difference between GNS being binary, and there being sort of continuous levels of them.  That is, some systems can be more or less G, N, or S than another, even though they'd both perhaps be described as one of the three.  Any two given Sim games may or may not be "as" Sim as one another.  This is opposed to a game being G or not-G, N or Not-N, and S or Not-S.  In this way, any game which prioritizes Sim is in some sense equivalent to all other Sim-prioritized games.

I'm not sure I stated it any clearer, actually.

I personally think it's a thing of degrees.  I also think that when a group is sitting down to play any game, coherent or not, G, N, or S-prioritizing, you will see different players, at different times, exhibiting all of G, N, and S behavior.

I think the particular balance of how much of each, how Intensely the three modes are used, can describe a given set of players in a given game at a given time.  Just because N is prioritized in, say, Sorcerer, doesn't mean there isn't plenty of opportunity for the players to inject G or S.  My understanding of a Coherent game is that it is at worst neutral to a given play mode, although it may favor one.

If you accept that, then a Hybrid would be something that either perfectly balances two (or more, if possible) of G and N priorities simultaneously.  I'm not sure that's possible in a simultaneous way -- one or the other is probably got the upper hand at any given moment -- and that suggests that it's either good for one and pretty good for another, or perhaps some portions of the rules balance in favor of one way while another section balances the other (ie, "Char skills resolution is a Nar system, while combat is a Gam system", or something like that).

I suspect that in the case of a game being good for one mode and pretty good for another, the game would be stable if you used it in a mode "against type".  On the other hand, any game that tries to really balance two or more priorities during actual play I think is 'doomed' to eventually precipitate out to one or the other as a primary mode, with the others subservient to it (although which mode a given group ends up with will be fairly variable).

That may not have made much sense, though.  I blame the medication.  :/

[Mike Holmes]

I believe that "equal" hybridization is possible in a way. First off, as priorities of intent, I'm sure that they're potentially equal. I can certainly want to have play that's, for example, both S and N. What the theory points out is that there will be times when you'll have to make a decision, and at that point the decision will have to be one or the other. And I agree with this completely. For a single atomic decision, you may not be able to satisfy two modes simultaneously.

But then, GNS is about Instances of Play. That is, we don't look at single decisions, we have to look at bunches of decisions over time, perhaps on the session or greater level. So that brings us back to the old atomic decision string theory. So let's look at an example:

S= Simulationist
N= Narrativist
SN= Congruent between Simulatist and Narrativist

SN-SN-SN-S-S-SN-N-N-S-SN-SN-S-SN-SN

This could represent coherent play, IMO. That is, the above could potentially be labled an Instance of Coherent Hybrid Sim/Nar play. Note that the fact that there are slightly more discernable Sim decisions than Nar decisions is irrelevant. Because Sim play in this context is largely irrelevant to the Nar (or gamist player). That is, playing "realistically" or "exploratively" for part of the session doesn't mean that you aren't creating Theme over the long haul. So you satisfy both to an extent. And given the prevalence of indeterminate decisions, the "offenses" of the non-concruent decisions will seem small.

Now, this could also be incoherent for a number of reasons. First and formost would be for the players not to have agreed that the "offenses" are OK. That is, many games will tell you that you're going to have both S and N all the time. Such that when you get a non-congruent moment, that seems to be an offense. Note that, players will each take a side, in this case, I think. That is, deviod of something informing a player that either kind of decision is OK, he'll assume that the non-congruent decisions should be made one way or another. Which is to say that they're essentially playing Sim or Nar likely, or shifting randomly, and the congruence is just by happenstance (which it's agreed happens all the time).

Now, the game can become coherent in this fashion by players forging an unspoken or negotiated coherent creative agenda that includes both. That is, Marco's described play has a creative agenda broght about by tradition, or force of will or something (he refers to it as responsibility), that causes all to understand that the SN hybrid is what's sought.

For a system to be supportive of this, however, requires that the system does not sabotage this in any way. That is, most systems say they want the hybrid, but don't really support it well mechanically. Leaving it to players to either make it work or to fall into incoherent play.

Can a system be made that supports Hybrid play? I'm certain of it. For example, and easy way to do this is to segregate play into sections. TROS, for example allows for extremely Gamist play in combat itself. But tactical decisions are also thematic decisions. So there's a big GN congruence creating factor there. Which means when players come out of combat, they seem to me to then be incentivized to drop the Gamism and move on to Narrativist play until the next combat. Sim is also supported in terms of the "plausibility" dial being very high on all decisions. Lots of congruence, and segregation of modes when congruence is impossible with the system. I'd call this an Asymetrical Hybrid.

Wheras a Symetrical Hybrid would simply push making all decisions congruent between the supported modes. I'm not sure if it's possible to force all decisions to be congruent, but, if not, such a system would somehow make it valid to make single mode decisions. Again, this doesn't seem too difficult. For example, the GM could determie when a decision was single mode, and require the expenditure of a point to validate it. Something of that nature.

In any case, like all games you'll never get to a point where it perfectly promotes coherent play. That's an impossible goal, IMO. What you strive for is a solid game. And, within those parameters, I think Hybrid games are possible.

Mike

[M. J. Young]

A long long time ago in a forum post far far down the list, I suggested that we couldn't discuss hybrid games until we had a definition; this thread has reminded me of that. What are we trying to design when we say we want to design a hybrid game?[list=1][*]A game that is vanilla enough that players can drift it to any one preference they prefer at any time. There are probably several games like this, in which whatever presuppositions the players bring to the table will control, as long as they are in agreement.[*]A game in which players can make decisions each based on his own preference without much interference between them. I'd wager Multiverser lands here, to some degree, in that if the narrativist doesn't like the way the gamist plays, he just moves his character off to do something else--individual play is encouraged, so player interaction doesn't become problematic.[*]A game in which players are guided to make decisions based on different preferences at different times--what Mike is calling the Asymmetrical Hybrid. This seems like a compromise game to me--one that displeases all parties equally. Each is required to play in a way he dislikes at some point in the game, but rewarded with opportunities to play in the way he prefers otherwise. If that were really the way a game was designed, I'd expect it to push the struggle one step further back: now instead of fighting about how to play through all aspects of the game, we're fighting about which part of play we want to emphasize. If TRoS is indeed gamist in combat, then we expect that gamist players will want to have more fights and less of everything else. Such a hybrid is ultimately problematic for this reason.[*]A game in which all GNS motivations lead to the same choice outcomes. In the previous thread, I suggested the Armageddon scenario: the entire world is about to be destroyed, and you are one of the people who have any chance at all of preventing it. Now, whether you're playing nar, sim, or gam, you are probably going to say yes, you'll do it; and if the situation continues to hem you in significantly, nar, sim, and gam decisions are going to continue to be indistinguishable. That means that nar, sim, and gam players can play together without disagreement--as long as they stay within the confines of the situation, which drives all three modes toward the same decisions. This is really the only model I see which makes it possible for players with different priorities to play together without disagreement at some level. (#2 allows them to play simultaneously without disagreement, but does so by eliminating the necessity of interaction between the characters.)[/list:o]
Now, if we're looking for a hybrid, what exactly do we expect to find?

--M. J. Young

[deadpanbob]

M.J.:

Is it possible that there is a fifth type of hybrid?

I'm thinking of one where two different play priorities don't conflict, but can co-exist side by side in different players.  Like your #2 above but including character interaction.

Is it possible that, for instance, the Step On Up priority could be satisfied strictly at the meta-game level and the Story Now priority could be satisfied strictly at the in-game level?

I think I'm on safe ground to say that both G and N paly are more comfortable with significant meta-game level - but that meta-game levels are not required to play either G or N.  So if the Step On Up is limited in terms of the challenge between players at the meta-game level - directed say at driving the addressing of Premise at the in-game level, would that work as a functional hybrid (theorectically?)

I ask because TRoS is often been described to me as either Sim powered Nar or Nar powered Sim (depending on who's doing the describing), which suggests to me that TRoS may not, in fact, be #3 in your list.  Put another way, those I know who have played the game seem to be able to handle either Sim or Nar priorities held by different players without the whole thing falling into that feeling that it must be fair because everyone got equally screwed.  But that could be the result of a healthy and functioning social dynamic that has little to do with the game...

Cheers,


Jason

[Wormwood]

Gordon,

I'd say that the discrete GNS is analogous to a multimeter, exceedingly useful and very important. Many people who regularly use them need not understand or even have heard of Maxwell's laws, but still if you want to start talking underpinnings of why the multimeter works, then you need to get into the guts of those laws. What I worry about with over-accentuating the discrete GNS is that we start making false statements based on generalizations off of the limitations of our measuring device. The purpose of the abstract continuum-style theory is to avoid these fallacies and give the multimeter users a better idea of when they should reconsider their readings.



Justin,

I've heard that metaphor before, and there is a counter one. I like chocolate cake, I also like vanilla icecream. I especially like the icecream on top of the cake. This is, often a better result than either, and it makes inferior cake and icecream much better than they could have been alone. Much of the criticism of hybrids seems analogous to complaining that the cake is under the icecream, rather than the icecream being on top of the cake.

Oh, and a functional three-way hybrid is unlikely to be impossible as you describe, it just seems to require a different approach to design than most people use, but that is really perpendicular to this topic.



Dana,

I generally agree with your points, except the idea of game necessarilly precipitating to one of the three modes. Yes, those are three default attractors for game play. The art of making hybrids suggests that additional attractors for play can be developed. By careful placement the result can be a hybrid that does not suffer from drifting to a pure mode, but rather reliably retains its hybrid qualities.


I hope that helps,

  -Mendel S.

[Wormwood]

Mike,

I concur, congurence is the glue by which hybrids are crafted, and a strong understanding of how to make modes congruent is essential to making a hybrid game. I also would like to stress, perfect promotion of a single mode or a hybrid is equally unachievable, the complexities become insurmountable. But functional games of either type are a important goals for design and play.



M.J.,

I'd define a hybrid as a functional game which supports two or more distinct modes simultaneously (albeit with some bias being possible) and does not require drifting to achieve this.

By this definition 1 and 2 are both drifted and so not true hybrids, while 3 and 4, as well as jason's 5 (which is something of an extension of 4) are hybrids.

What I find interesting about your characterization is that you assume that players of a hybrid are not choosing the hybridization over the single mode, but rather prefer a single mode and play the hybrid grudgingly. In some way it's like assuming that someone gets Neopolitan icecream because they only like chocolate. From the outside view it is stupid, they should just buy chocolate icecream. But what if they like the flavors mixed? Then is makes perfect sense.



Jason,

I'd say your fifth type is mostly just expanding the 4 that M.J. gives to allow constraints on something other than situation. There is also the idea of segementing the modal duties of a game, having players dominate when their mode or modes arrives and then support the play with the other player's modes. In essence there are a variety of ways to integrate modes, and many of them can be successful. The core principle however seems to be interlinking the result of playing one mode with supporting another. Thus, instead of penalizing a player for not having the same mode, they are rewarded (in their mode) for playing accross others in different modes. This is an even more powerful dynamic when players are not fixed in one mode, but instead have hybrid priorities, the result is a far more dynamic situation where hybrid play creates more hybrid play, rather than a conflict or decoherence.


I hope that helps,

  -Mendel S.

[Mike Holmes]

Mendel I think that Jason's #5 is actually MJ's #3. Partitioning play.

Asymmetrical Hybrid. This seems like a compromise game to me--one that displeases all parties equally.

That assumes that games are selected for their GNS supports. I don't think this is true (a fact which may shock some). People don't want a particular support, IMO, they want coherent support for whatever Creative Agenda the game has.

For instance, when introducing people to play of Sorcerer, I think that people see it as substantively different in some ways, but they're content to play it because it does what it does well. They're different but equally enjoyable sorts of activities. Now, some people have strong preferences, certainly. But I truely believe that this only makes a difference for an "entrenched" player who believes that their mode of play is the one true way. That if they gave other styles a chance that they too could easily enjoy them.

Now, this is more or less true. That is, at least I (and I think many others) have come to the conclusion that we're not Simulationists, or Narrativists, or anything like that. We want all three of the modes. But we want them in coherent games. So if I can't have all three in the system that I'm playing, then I'll be satisfied with what the system does promote well.

In any case, Hybrids of any sort simply provide more types of support in some way. Now, I can see it being potentially dangerous or frustrating to some players to say, "Play this way now, then this way later." But I don't think it's really that problematic. The definition of coherent means that it should be satisfying in play. The only question is it really possible to make a coherent Asymetrically Hybrid game. I think so.

Because it's not about players not wanting to see some particular GNS mode. I think that the assumption all along that some players don't like a mode, and therefore are annoyed when it shows is incorrect. Competition, Story Creation, Plausible play, these are all positive things in general. What GNS says is that they become problematic when the system tells different players to do different things at the same time. It's the perception of player A of the Creative Agenda as X that annoys player B who sees Creative Agenda Y. As long as X is substantively different from Y. Other than than difference, it doesn't matter what modes are supported.

That is, a design that promotes coherent play is satisfactory in terms of Creative Agenda to all but the most entrenched player who is imposing their own Creative Agenda despite the system speaking loudly to the first. So, if the game truely can be said to be Coherent, if it truely has a clear and well transmitted Creative Agenda, no matter what manner the modes are supported, the designer has done all he can in this department.

Difficult, yes, impossible, no.

Mike

[jdagna]

Justin,

I've heard that metaphor before, and there is a counter one. I like chocolate cake, I also like vanilla icecream. I especially like the icecream on top of the cake. This is, often a better result than either, and it makes inferior cake and icecream much better than they could have been alone. Much of the criticism of hybrids seems analogous to complaining that the cake is under the icecream, rather than the icecream being on top of the cake.

Oh, and a functional three-way hybrid is unlikely to be impossible as you describe, it just seems to require a different approach to design than most people use, but that is really perpendicular to this topic.

I'm not sure our analogies are entirely in conflict.  They're more of a glass is half full versus half empty comparison - looking at the same thing from two perspectives.  My reaction to your ice cream on chocolate cake is "Why's he ruining perfectly good vanilla ice cream with chocolate cake?"  On the other hand, I would prefer the ice cream and cake better than cake alone...

Basically, I'm a pessimist about how things turn out in real life, whereas you seem to take a more optimistic outlook.  Neither is necessarily wrong until proven otherwise.

I've actually argued in the past that incoherent games may be helped (in terms of market appeal) by their incoherence.  They may please no one perfectly, but they appeal to everyone at some point.  Obviously, a coherent hybrid could the same, but better.  Getting one to work would be quite a breakthrough.

As for three-way hybrids: couldn't we say the same thing about perpetual motion machines? ("a functional perpetual motion machine is unlikely to be impossible, as you describe, it just seems to require a different approach to design than most people use").  In any event, useful discoveries have been made by chasing the impossible.

[Mike Holmes]

No, the perpetual motion machine is fundamentally different. For it to work it would have to violate one of the central laws of science (conservation of mass and energy). There is no central tenet of game design that says anything that would make a G/N/S Hybrid neccessarily impossible. Difficult, yes...

Mike