player-directed exploration of GM-created world

[dindenver]

David,
  Of the games I have actually played, SoY comes the closest to what you are talking about. Although Ensemble might, but in my actual play - even though it is GM-less, it has proven to allow each indivisdual player less control over the over flow of the narrative.
  There probably are some great GM-less games out there that might do what you want.
  As to playing DITV without Dogs, I think it is a bad idea. The ditv mechanics rely on a character type that is highly motivated and relatively inflexible in their views. Look at the escalation path. Why would anyone risk the d10 fallout unless they were willing to die for what they believed in? Don't get me wrong, I think you CAN play ditv rules in another setting, but you have to do it with a class of characters that will die for what they believe in (Star Wars Jedi was a good example I have seen floating around).
  Anyways, good luck Dave and let us know what you find!

[Marshall Burns]

David,

Check out InSpectres.  A player can describe their character, say, looking under a rock to see what's under there.  A roll is made, and based on that roll, either the player defines what's under that rock and what happens with it, or the GM does.  Either way, the outcome still has do deal with "what's under that rock," and whatever is defined becomes FACT, regardless of whether anyone ever planned on it.

My group played InSpectres for the first time last night, and this turns out to be really easy to do (partially because InSpectres is really humorous).  Check it out, and I think you could apply the principles of those rules to what you're talking about pretty easily.

-Marshall

[David Berg]

Dave,
I own SoY, but haven't played it.  I'm kind of at a loss for how to determine "what gets played", but I'm pretty sure that the GM doesn't have sole Content Authority.  Right?  Don't the players get to contribute beyond just saying, "My character pursues X"?

Marshall,
Similarly, I can't see a way to apply InSpectres' mechanism without allowing players to determine world content.

Contracycle,
I hear you on the value of published material!  That would be my ultimate contribution for those GMs with limited time on their hands... but I only have so much time on mine, and writing up (and illustrating) a few more hundred pages of material is gonna have to wait.

I also hear you on the upsides of filling the world with neat stuff and having scenarios lead to each other in satisfying ways.  That was my plan for my Lendrhald game, until I started thinking this new idea might be better.  The problem with even the best-organized web of scenarios is the limit on meaningful player contribution.  In a game where the players have no directorial powers, I feel like giving them total ability to point the camera would be hugely rewarding.  A constrained ability to point the camera is really just freedom of sequence -- i.e. in what order do we play through the scenarios?  This is, IMO, not all that meaningful a contribution.

Does that make sense?

If something could be done to make the prep-on-demand as unburdensome as possible, what would you think?

-David

[Marshall Burns]

Whoops, I didn't realize that you wanted the GM to have sole authority over content.  Now, by "authority" do you mean deciding what's in and what's out, or making all content-based decisions on his own without input from the other players?  If it's the latter, that's, uh, well, honestly it's something I don't understand at all.  I'm not going to assume that it's evil because I don't like it personally, but I'm afraid I can contribute nothing helpful.

But, if it's the former, using an InSpectres-like dynamic the player is allowed to *suggest* things, but the GM gets final say might be a place to start.

-Marshall

[Alfryd]

@ Contracycle

Hence I essentially agree with Alfryd's point about prepared material being non-linear, but this presents other problems that centre on making much more material than will ever be actually called on and thus goes to waste.  This too is a huge burden to place on a GM's time and resources.  But it may become much more do-able, I think, if the scenario/trail-of-crumbs is, again, external to the GM. In this case the GM and players are to some extent co-explorers, both encountering specific interactions that neither of them predicted at the outset, and instead experiencing interactions that were envisioned by an external third party, and seeing, quite literally, "how they play out".

Naturally, a good campaign setting pre-made would be useful, but this doesn't solve the more fine-grained problem of integrating the players into the world, which NPCs they have to deal with, what's the plot-hook, etc.  If the players' actions are meaningful, they have to wind up changing that world on a scale commensurate to their powers, which means you'll inevitably need to fill in the gaps yourself.

I think the problems of content overload can be alleviated partly through using varying levels of detail to prep potential scenes beforehand.  If you find your players have decided to enter scene A, you might write a 40-word description for scenes B, C and D, adjacent to scene A, and an 8-word description for scenes E, F, G, H, I, J, K and L, adjacent to B, C and D.  You then embellish your previous 40-word description for scene A into a 200-word description before the next session, where your players decide to head for scene C.  Again, you embellish your 40-word description for C into a 200-word description, while scenes G, J and K grow to 40 words each, and so on, and so forth.

You might also want to group scenes into acts, acts into plays, etc, and use a similar prep method for these supergroups to keep ahead of the players.  You might have formal rule specifying things like, "the players must have at least 3 formal options available in a given scene.  If they ignore these options, the GM gains an influence point.  If the GM provides fewer than 3 options, or provides options that lead directly to the same scene, the players each gain 1 influence point."

@ David Berg

Yes, but only on a certain scale.  If on their way into a dungeon, the players decide to stop and examine the nearby flowers, those flowers may be boring as hell.  If they decide to search a certain building in a town, that building may be boring as hell.  But if they decide to embark on some sort of multi-step quest that'll take multiple hours of play (i.e., a session)?  Yes, it will be interesting...
...Thus, if the players bail mid-scenario, the session is over.  They stopped caring about the thing that they told the GM would motivate them, so now they need to pick something else so the GM can go prep that.
Hopefully, this can be avoided.  It's the players' job to be honest and consistent in their interests, and the GM's job to create a Scenario that won't bore them while they're exploring it.  Providing tools to facilitate this will also be key.  Again, I think it can be done!

I can see what you're getting at now.
My major concern- aside from the pressure of content generation contracycle mentions- would be that you're not holding the players to any given focal point of interest longer than one session, so it's difficult to ensure a recurrent plot thread, and makes it easy for the players to abnegate responsibilities or commitments they've made to NPCs in prior sessions, or generally elude the consequences of their actions.  You'd need some method by which you can say, "Okay, you can write the plot, and I'll write the content- but I will hold you to what you write."  Otherwise the GM winds up supplying, in essence, window-dressing.

[dindenver]

Hi!
  Dave, There are two modes of play in SoY (as far as I know)
1) All the players work out what scene is about
2) The GM (aka SG) writes a Key scene. The nature of a Key scene is its "about" one of your char's Keys. And you decide what keys are on your char sheet, so you decide what its about.
  So, in the most indirect way, it does what you want. Because the Keys can be very specific. Also, if you change your mind, you can buyoff a key and buy one that matches your new interest.
  Also, I am not sure if this is what you are looking for, but Donjon lets the player decide/create what is in the world. So, when you go to roll to see if there is a secret door, if you succeed, you decide if there is one, not the GM.

[contracycle]

I also hear you on the upsides of filling the world with neat stuff and having scenarios lead to each other in satisfying ways.  That was my plan for my Lendrhald game, until I started thinking this new idea might be better.  The problem with even the best-organized web of scenarios is the limit on meaningful player contribution.  In a game where the players have no directorial powers, I feel like giving them total ability to point the camera would be hugely rewarding.  A constrained ability to point the camera is really just freedom of sequence -- i.e. in what order do we play through the scenarios?  This is, IMO, not all that meaningful a contribution.

Well I agree that merely choosing sequence is not that empowering, but that was not quite what I was thinking of.  I am suggesting something more akin to using an appropriate subset of such scenarios, rather than working through all of therm.  Or perhaps you could run through set A with a given group of characters suited to its problems, and set B with another set of characters suited to those problems.  The main advantage I see is that the players together, including the GM, select what they want to play from a "menu" as it were.  That I can see as a functional version of pointing the camera.  When the players register an interest in investigating orcish tattoos, that is the point at which you crack out the scenarios aimed at the exploration of orcs and cue them up for play.

I'm not entirely sure that I want a more fine-grained camera-pointing than that.  IME, I already have a lot of freedom to investigate different things in a sim game through character action.  I can and have sent my character off to learn things, usually things related to the action of the piece, sure, but nevertheless I am the one putting the questions and obliging the GM to give me answers.  Sometimes I have restrained myself from going "off pitch" on the basis of an educated guess as to where the the limits to the GM's prepared material are likely to be found, which is a limitation, but not one I find overbearingly irksome.

In addition, I did not mean to propose that you, necessarily, should be obliged to write such things.  If a suitable methodology for writing them that is transferable from group to group can be found, then I can write one and you can write one and then we can swap.  Or alternately, for people more interested in actual publishing for money than I am, they could be a serial product in a periodical or similar format.

If something could be done to make the prep-on-demand as unburdensome as possible, what would you think?

Well sure, if it can be done, but I'm not sure it can be done.  Alfryds idea of a flexible network of entries that get fleshed out, perhaps something like the tag clouds that are popular of late, might be viable, but there seem to me to be some essential and inherent difficulties with creating information ex nihilo.  As an analogy, pencil sketch may be a functional, fast way to depict something, but the result is not likely to be as detailed as those obtained by taking the time to set up with an easel and oils.  So sure I'd like to see and discuss any propositions as to how to do that sort of thing, I just can't think of such a mechanism myself at the moment.

[Anders Larsen]

Hi David

I do not know if I really understand what you are going for, because when I read through this thread I ask myself: What is the problem?

For what I can see you want the game to work something like this:

1) The players specify some "elements" they want to explore in the next sessions based on what happened in the previous session.
2) The GM will create a situation based on these elements for the next session.
3) In the next session the players explore this situation.
4) Start over from 1.

I have run a couple of games using this structure and I haven't really had any problem with it. There is a few point which are worth noting, though:

* Two or three elements are more than enough for the GM to construct an interesting situation around (one element could actually do it just fine), so if the players, during a session, want to specify more element than this, it is probably a good idea to make a prioritized list.

* Some of the elements, the GM has build a situation around, may never come into play during the session - the players just never go that direction, because they focus more on some other elements. You should decide how to handle these "abandoned" elements, so the GM does not feel that he has to force them on the players.

* The issue about the preparation time: If you ask the right question and use the right tools (like relationship map), it will not take more than 30 minutes to make an interesting situation with enough material for 4-6 hours of play. (The trick is to make a situation not a story. Not only will it lead to railroading if the GM has a story, but it will also take much longer time to prepare).

The only thing you really need to make a game based on this, is to have some way for the players to write down what they want to explore in the future, and have a guide for how the GM can build a situation around these ideas.

- Anders

[David Berg]

Contracycle, I think we agree on the benefits of having a lot of designer-provided content to choose from, and I think we agree on the downsides of asking a GM to create that much content himself.

As for freedom of investigation, I think that some scenarios will provide a lot, and some (such as dungeon crawls) will provide little.  My assumption is that some dungeon-crawling is fun, and should be supplemented with some camera-pointing, but I, like you, would be content to keep that supplementation at the large "session" scale I've described.  I do worry, though, about creating a game that can get boring in the middle of the session... Situation-creation tools for the GM (or designer) will be key.

Publishing and selling series of scenario-chains actually sounds appealing to me, I'm just not sure how practical it'd be.  In the short-term, swapping creations with other play groups sounds great!

As for making prep-on-demand unburdensome, here's an idea:
1) form general pre-game consensus on what kinds of adventures the players enjoy
2) GM comes up with some situations whenever he has the time, complete with everything except the identity of the reward
3) when the players pick a reward they want, the GM plops that into one of the situations he's already created

I have run a couple of games using this structure and I haven't really had any problem with it.

Cool!  Did your own group just come up with the idea to do it that way, or was there some game text that helped?

The only thing you really need to make a game based on this, is to have some way for the players to write down what they want to explore in the future, and have a guide for how the GM can build a situation around these ideas.

Right.  I think the key part here is the guide.  A good guide will make this game useful; a bad guide will make it no better than plenty of informal practices already being used.

I like your emphasis on "situation, not story", and I like Ryan Stoughton's "make only Problems, Rewards, Assets and Threats" advice.  I would guess that a Relationship Map is not always usable for things like dungeon crawls...  I have little experience with Relationship Maps, though, and may be short-changing their potential.  Have you ever seen a Relationship Map that connected stuff other than people?

-David

[masqueradeball]

Why is prep so important. I've probably run dozens (maybe even hundreds, there were a few years where we games almost every night) of games in my life and I don't remember doing in prep, per se, beyond the bare minimum. It seems like the only pre-requisite for prep free play would be a simple enought system, maybe one that had some way to inspire the GM. I thinking of SAGA (Dragonlance or Marvel (Both of which I feel are heavy on the sim-support) in particular as being good candidates. So, where's the catch. Why is prep neccessary if your not playing a stat/structure heavy game a la GURPS or D&D 3+?

[contracycle]

I suppose you could think of it as texture and reason.  The game world must be textured in a way similar to that of texturing surfaces in computer games, by way of analogy.  It's not just the bare-bone walls and gross structures that are important, but also how they look, and create mood, and integrate with one another.  And secondly the world must be reasonable, it must operate with predictable causality, even , and some might say especially, when those links are not explicit.  So if play is going to occur in a place, then that place must be visualised and 'textured' and brought to life so that it is interesting and rewarding, and some forethought must be invested in causal interactions and how they might affect or prejudice future acts or invalidate established data (otherwise players cannot investigate the world in a reasonable and logical way).

If you stripped a novel down to who said what and went where, it would not be nearly as engaging as one in which its imaginary environment is realised through description, enters the mind almost as as if perceived by the senses, and provides a tangible context to those actions and movements.  The preparation creates all that world-realisation stuff.

[David Berg]

Well said, Contracycle.  That is in fact the type of play I'm going for.  I wasn't explicit about that before, but now it's out there.

Any thoughts on my "plop rewards into scenarios?" idea are welcome.

Experiences with other games that do (not can) accomplish some or all of what I'm going for are welcome most of all.

[Alfryd]

And secondly the world must be reasonable, it must operate with predictable causality, even , and some might say especially, when those links are not explicit.

Well said, Contracycle.  That is in fact the type of play I'm going for.  I wasn't explicit about that before, but now it's out there.

That all sounds well and good, but consider a situation where the players have, for instance, been arrested for slaughtering some guardsmen in the middle of a large city 2 sessions ago.  And then the players indicate that they have no interest in following a 'standing trial or being locked up' scenario during the next session.  It seems to me that you can have a rational, consequential world here, or freedom for the players, but not both.

Alfryds idea of a flexible network of entries that get fleshed out, perhaps something like the tag clouds that are popular of late, might be viable, but there seem to me to be some essential and inherent difficulties with creating information ex nihilo.  As an analogy, pencil sketch may be a functional, fast way to depict something, but the result is not likely to be as detailed as those obtained by taking the time to set up with an easel and oils.  So sure I'd like to see and discuss any propositions as to how to do that sort of thing, I just can't think of such a mechanism myself at the moment.

Well, the basic idea is that you refine a rough sketch by degrees into an oil painting as the players' focus of attention wanders closer to a given portion of the canvas.  Again, if we take your surface texturing analogy, this would be analagous to mimapping of textures at different distances from the viewer- only, instead of taking the highest level of detail and working down, you take the lowest level of detail and work up.

Much of the canvas could even be left entirely blank.  i.e, only the first scene or two needs to be worked out in detail, which a few adjacent scenes, a broad plot outline or act structure roughed out beyond that (and some meaningful, deterministic reasons for why the players would get involved at a given point or level of detail,) like guidelines or perspective markers.  Of course, this has the drawback that you still need to anticipate the players' choices ahead of time, but since you rough out adjacent scenes at a lower level of detail, you can cover your options more comprehensively than usual- particularly if the player's choices meaningfully constrain future behaviour.
An analogy might be the game of Go, which starts off with a seemingly endless field of possibilities, but gradually constrains available choice until play becomes pretty linear.

[contracycle]

I do think that explicit player-chosen rewards may be a good starting point, because they allow the player to set some kind of direction without necessarily revealing the structure of the whole "plot", in whatever form that takes.

I have previously had some thoughts in this direction, like getting rid of XP and handing out rewards directly.  So instead of "adventuring" for experience, you instead choose a mission or whatever that will result in increasing an attribute, or raising some skills.  And then that actual change can be brought about and recorded mid-play in something akin to a training montage.  This is the pattern of movies like say Rocky, in which the main character struggles to achieve this sort of gain and then also puts it to use for some purpose soon after.  Alternately you could build scenarios with character specific goals.  So maybe you have priest-type character and you can over them a mission that leads to them becoming bishop, or another which leads to them acquiring a relic of the saint so-and-so.  Then in choosing which path to pursue they are giving consent to doing "whatever it takes" to achieve that sort of goal.  But integrating these sorts of character-specific constructions with multiple characters seems tricky.

That all sounds well and good, but consider a situation where the players have, for instance, been arrested for slaughtering some guardsmen in the middle of a large city 2 sessions ago.  And then the players indicate that they have no interest in following a 'standing trial or being locked up' scenario during the next session.  It seems to me that you can have a rational, consequential world here, or freedom for the players, but not both.

Well yes and no.  You could choose to see it as, you are totally free to get yourself sent to prison.  Anyway its not an unsolvable problem; if you set it up right you could stage a breakout or rescue before they even arrive at the slammer.  But that brings us back to the problems of planning with a short lead-time, both for the players and the GM.  Obviously I don't see any real way to play a game which consists of "you look through the bars, another day passes", so something would have to be done to avoid that situation.  And yet, whatever is done would have to be Not Silly, not disqualify any established data, etc.  All this said players can often enjoy the sense of impending doom or tight deadlines and work themselves into a fever pitch as a result.  The question is not really whether it can be done, but whether the game can be stand being put on hiatus while you figure out how to do it.

[masqueradeball]

So, prep is necessary for texture and consequences? Well, I think I have to disagree. I'm reminded of Amber: Diceless Roleplay by Eric Wujcik (who, very sadly, is dying of cancer) in the section about how the GM is to handle the plots of the godlike major players (Oberon and Dworkin) and Wujcik advice is of course, don't bother to make their plots. The reason is so that the GM can play it like whatever happens, Oberon knew, he planned it, there you go. This is a technique I've used extensively in running Vampire, and I honestly think Amber is as close to a "good GM's bible" as anything I've ever read.
That whole book is about being good at improve, and in fact, that ability is tied pivotally into the game's "system." Whenever a challenge is faced, the GM actually must know how it works in order to allow the players direct actions to have immediate effect. This is made even more difficult when you consider that this know how it works approach even applies to things like magic and the nature of the universe.
As far as color and texture go, prep can add to these things, but if handle clumsily (over prep) it can actually hinder there delivery. I think any GM who wants to be able to cater to his players wants and needs should spend as much time practicing improvisational techniques (free writing maybe, or playing Once Upon a Time) as they do prepping for play.
Of course, this is from a very biased source. All my experience is with improv, I've never utilized more than bare bones prep work.