D&D 4E The Best Thing from 4E

[AbdulAlhazred]

I'm not sure what you mean by "exploring the nature of magic". When a player declares as an action for his PC of internalising the chaos energy so as to make a magic item, is that "exploring the nature of magic"? If so, then I would expect exploring the nature of magic to be quite common in a game that includes magic-users.

I think he is exactly saying this, probably because that was the way I framed it a couple pages back. I equated exploring the nature of magic with exploring a dungeon. Its an apt analogy I think. [MENTION=6775031]Saelorn[/MENTION]'s theory seems to be that both should be mapped out at a very high level of detail such that any likely question about their nature can be answered by the DM referencing some pre-determined material (or perhaps exercising some already-built random generator perhaps). He's literally saying that the DM's job includes recording in some form a set of 'rules of physics' from which principles any question arising in game can be explored. I assume the alternative being the same thing you'd say if the PCs got to the edge of the dungeon map where you don't know what is behind the door which would be "that door is stuck" and I guess for say ad-libbing some magic it would be equivalently "You're not sure if this is possible, but you lack the understanding to try" or some such.

I don't want to sound condescending to [MENTION=6775031]Saelorn[/MENTION], but IMHO he's at an early point in the trajectory of evolution of gamers mastery of the more abstruse areas of RPG theory and practice. There was a time, probably in the early 80's, after I'd DMed for several years, when detailed process simulation and the achievement of the 'ultimate system' that would somehow present answers to all the possible questions players asked and produce satisfying narrative harmoniously coupled with simple mechanics was the omega point of my conception of RPGs. That day has LOOOONNNNNGGGGG since faded to dust and been replaced by a great appreciation of agenda and cooperative shared world building/exploration. Not that the process sim phase wasn't fun, it was, but it really only worked for a pretty limited realm of play, roughly Gygaxian dungeon exploration.

 

[Tony Vargas]

D&D isn't a particularly broad realm of fantasy. It's a fairly narrow type of fantasy, originating in the works of Tolkien and a few other authors, which has since expanded to codify a number of different-but-similar genres.

No, Saelorn, that's your impression of D&D. The 'further reading' list in the back of the 1e DMG, alone, contradicts that impression.

Deities and Demigods lists gods from myth all over the globe. Monsters and magic items and spells are taken from myth, legend, folklore, and from both well-known examples & obscure corners of the fantasy genre. It's not Tolkien + Vance = all you can have in D&D. It's not even close.

You could do a mythic game, where the Sun doesn't rise because that particular deity has the flu, but it's not something that D&D has tried before. That idea has no particular applicability toward anyone else who might be playing the game at some other table.

I can. I have. I'm not alone in having done it.

There's no reason that any discussion about D&D in general, and settings in particular, should be made to accommodate for your particular wild deviations from precedent.

It's a game, not a law book, there is no 'precedent.'

It's an FRPG, it's trying to model fantasy. Fantasy is a broad genre.

 

[pemerton]

It's not Tolkien + Vance = all you can have in D&D. It's not even close.

Even if we just confine ourselves to Tolkien, the sun and moon are vessels guided across the sky by spirits/demigods, and a hero (Earendil) became a star.

Middle-Earth is not bound by the real-world laws of astronomy!

 

[Tony Vargas]

Even if we just confine ourselves to Tolkien, the sun and moon are vessels guided across the sky by spirits/demigods, and a hero (Earendil) became a star.

Middle-Earth is not bound by the real-world laws of astronomy!

I suppose Saelorn would say that the Silmarilion is 'just' the myths of Middle Earth and not 'real' like LotR.

 

[Manbearcat]

Well, mostly. The GM still has total control over the complications when failures present themselves. And say, for instance, we took your consequence for failing the skill challenge and used it in context with the stated goal of the skill challenge.

The stated goal, in your post: "find a secured shelter, safe from the elements and predators, where she can stow the children".

Your stated consequence for a failed skill challenge when hyenas or mountain lions are preset: "I'm pretty certain that if she would have failed the SC outright with those hyenas or the mountain lion being the relevant threat at the moment, I would have charged her an HS and we would have begun a combat encounter whereby all of the children are one-hit-kill minions (with some kind of attack - like throwing rocks - and utility power defense each) and she has to control/kill the bad guys and keep them off of the kids. Then, down several resources and possibly some kids, she could attempt to start anew (a C1 SC)."

No, this doesn't add up to me. Her goal (to find a secure shelter, safe from elements and predators, to stow the children) has failed. There should be no retries, I would think. That doesn't mean she can't find shelter, or a place to stow the children. It just means that it shouldn't be safe or secure.

Going to give my thinking on this right quick. Couple things:

1) My M.O. for establishing stakes and the adversary/danger interposing itself between a player and their goals stems from GMing Dogs in the Vineyard (I think...it might predate 2004, but that is when it was crystallized). In the above conflict I am playing "the vulnerability of the children" and the "danger that keeps them unsafe and unsecured." I may have just as easily invoked one of the children having a fit of terror (putting pressure on Saerie to have to figure out how to calm/comfort him/her or remove the terror...or deal with the consequences...this would be a "soft move" in Dungeon World).

2) The player in question has a strong maternal instinct. Inevitably, her characters reveal this virtue in play. She will inevitably become emotionally attached to the vulnerable, to the meek, so putting her in such spots resonates emotionally.

3) Given 1 above, shelter is the conduit for guaranteed safety and security. Denying it presently (but not removing it permanently from the table) and putting the children in a spot that highlights their vulnerability and amplifies the danger to them (as the ultimate resolution to the SC) does a few things:

a) It intensifies the moment and escalates the conflict (vulnerability and threat level heightened RIGHT NOW) that the PC is trying to mitigate.

b) It potentially brings some new immediate narrative fallout to deal with (that the PC cares about - eg the death of one or more of the children).

c) It provides the player (through the PC) a legitimate decision-point to attempt to continue to pursue a safe and secure location to secure the children. If it is overtly expressed by me that such prospects of a safe and secure shelter for the children are now completely closed out, the player's decision-points narrow. The action declaration now inevitably becomes "alright, I guess the kids are coming with me while I look for a river, follow it down the mountain, and hopefully find a settlement/village."

Further, as I charge a healing surge (just an open-descriptor resource for heroic mojo/luck, etc) for every failure (and all the PCs a failure on a failed SC), she also has the metagame vector of strategic resource consideration. If things went poorly there, she may have been down 4-5 healing surges (and possibly 1 or more daily) after a failed SC and a combat. Consequently, not only would the children remain vulnerable, but the PC herself would now become vulnerable. That feeds back onto (a) above.
​

Also, even if you did allow her to try again, you are the person coming up with the complications on each failure. You can call for combat, or 1 healing surge lost, or 10 healing surges lost, or magic items or blessing disappearing, or loss of a hand or eye, or someone loyal betraying her, or whatever.

This is empowering. It's just not empowering to the player, in my opinion. It allows for a lot of variation in gameplay, but I can see a GM definitely use the rules to attempt a level of illusionist play. A GM can frame the challenge, targeting bad skills of the PCs, and use failure consequences to keep things on the rails. (Obvious bad GMing is obvious, but we're talking of illuionism and how 4e might combat it at a base level.)

I think what you're saying here is just the inevitability of a game with a GM. Any game where there is a GM who bears primary responsibility for framing situations and responding to player action declarations and evolving the fiction post-resolution, is going to be vulnerable to poor GMing. I mean that is just an inevitability. What is important is how obvious the poor GMing is and/or how transparently manifest the bad faith is. A system with clarity in play procedures (primarily composed of explicit GMing principles/techniques and coherent resolution mechanics) will crank up the obvious and crank up the transparently manifest (hence less vulnerable to illusionism). If the players don't care...well, again, then you're in the realm of "participationism". And if so, then so be it because if we're all in on it then our table agenda is aligned.

Finally, a system that cranks up the obvious and cranks up the transparently manifest (Dungeon World and all of the Power By Apocalypse games do this even better than 4e) will typically yield a lower entry level for sound/good GMing.

I don't know. I mean, I'm definitely against railroading players (for my particular style), but I don't think these principles are enough to fight illusionism off completely. I'm fairly sure that I or my brother could run these types of skill challenges and control the narrative quite effectively while still keeping our players happy, if we really wanted to. Would they see what's going on? They'd probably start to suspect it, if it was a regular occurrence. But as long as they're having a good time with it, they'd probably just think the mechanics are working as intended, and indeed might even blame the mechanics themselves rather than the GM ("I wish skill challenges didn't exist; I'd rather just be able to make skill checks over and over again, rather than be arbitrarily cut off or have something bad happen.")

I'm sure you could. I know I could as well. And I'm certain it would be a great game for people who are "participationists" at heart (they're looking for GM force/illusionism to move the game along and they just want to press an attack button here or say a witty one liner there or rescue a few damsels in distress). However, the real question is (as I framed it above), is how easily could we get away with it from players who are specifically looking for the antithesis of that experience. If I used illusionism/GM force as the primary technique for moving myself (eg me GMing myself and another me - scary - playing a PC), or any other number of folks on this thread yourself included, though a 4e game....well, the jig would be up rather quickly!

I'll do a Dungeon World conflict here when I'm next able (social conflict). I would really be curious about your thoughts on the system...especially in relation to 4e GMing principles (of which I feel are very much inspired by Vincent Baker - who wrote Dogs and Apocalypse World). Perhaps you could take a look at the SRD and just maybe read the section on Gamesmastering. It would be a good primer for when I post the DW conflict.

Again, thanks for the precise and clear reply.

You bet mate!

First, kudos to Manbearcat for the real play SC post. It's good to talk actual example vs. theory.

I agree!

I apologize, but I don't have time to address the entirety of your post. I think that much of what I wrote above addresses the meat of your post however. Perhaps the only addendum (that specifically goes to your post) is that I tend to like to focus the adversarial elements down a bit so I tend to not run long SCs. I would say that perhaps only 1 in 20 of the hundreds of SCs that I've run has been a C4 or C5 (perhaps less). Further, the majority of those other 19 are C2 with C1 and C3 making up the smaller percentage. There are a few reasons for that:

1) Dramatic pacing - you can only get so much mileage out of one thematic conflict and the snowballing thereof toward climax/denouement.

2) More, but shorter bursts of conflict lets me focus in more tightly on the conflict's premise and make more coherent the adversarial elements.

3) Each action declaration and subsequent resolution carries more mechanical heft.

Finally, this was literally only the third conflict of the game. Regarding that particular character, we were still feeling her out. Regarding the setting, it was basically no myth (we had established very little prior to play - you can check out the thread to see the tiny bit that was established) so we were more or less both finding out about this world and its relevant players and history as we went.

I want to say the opening conflicts went like this:

Follow spectral stag through shimmering portal under the Feywild moonlight and wind up on rain-soaked, alpine mountain with bloodcurdling screams just beyond sight. Stumble upon some terrible ritual. Then:

1) Rescue the children/disrupt the ritual
2) Locate a secured shelter for the children
3) Transition scene investigation of lore in the cave and conversation with the children
4) Locate a village on the mountain's river (featured a nested combat encounter with marauders and a nested chase)
5) Combat a vile tentacle monster and save a (soldier) boatman and his (minion) daughter who belong to the island settlement
6) Parley with the settlement's elders

Then things snowballed as a result of things up till that point and the parley's result.

 

[LostSoul]

Yes. That's what I've said, two or three times now. The DM creates the world and all of its laws, and the players explore that world through their characters.

As a player, I absolutely do not want any sort of authorship powers beyond the natural abilities of my character. If 4E includes such a thing, then that is on my list of least-favorite things about 4E. Any game which makes provision for players to influence the game play, after the game has started, and beyond the capabilities of the PC, is a game which I do not wish to play.

I don't want this to sound argumentative; I run a game that's probably closer to yours than [MENTION=42582]pemerton[/MENTION]'s.

If you as DM state certain features of the world's physical or magical laws, and a player proves to you that they are contradictory, how would you respond?

In my case I'd change the setting, trying to limit the "damage" done, but also trying not to pull any punches - if it means that the necromancer is now in a weak position, so be it. This doesn't happen that often because I as DM will defer to players who know more about the subject matter than I (especially relevant in my system where one player trains in the medieval martial arts), though I regret the fact that this causes a conflict of interest for the players.

I'm not saying that this is necessarily an aspect of player authorship, though I guess it could be seen that way; I was just curious to see how you would handle it.

 

[JamesonCourage]

First, kudos to Manbearcat for the real play SC post. It's good to talk actual example vs. theory.

I think the play example is good but the original stated goal is not compelling enough for me. A better choice of success / failure would I think answer JamesonCourage's point as well.

IMO, SC have to have narrative teeth to be worth playing out. This narrative teeth also further serves the purpose of removing the need for illusionism.

So for Manbear's example, I think a much better set up would be something like:

DM: So, you want to keep these children safe? Well, this area is very dangerous -- full of predators and extreme terrain. Success at the SC means you find safe shelter, for every failure you lose 1/3 of the children. (or complete failure means you lose half or whatever -- something with real teeth)

And forget healing surge costs, combat etc. The SC itself has consequences if failed. That's why your bothering to play it out. That's why it's nice and tense. So, sure, the mountain lion could end up eating some kids but that is a result of the failed roll -- an abstraction of a lot of things, including fighting off the mountain lion.

This is good. I still have the same feelings on illusionism and GMing with 4e skill challenges, but I like the way this is framed more (in that, this is more likely how I'd frame it myself for my players).

Another example:

Not worth doing -- Hike through the wilderness skill challenge. Success -- you find your way to Ragamuffin Village. No pressing reason to be in Ragamuffin Village. Failure - everyone loses 2 healing surges.

Worth doing -- Find your way through the uncharted wilderness in three days. No has ever done the trip in less than a week. Success -- you get to Ragmuffin Village in time to defend it from invaders you know are on the way . Failure -- you get there too late and the village is sacked. friends die. (opens up adventure to track down captured inhabitants if desired though!). Or if you want to define micro failures -- each failure is one more day to reach the village giving you a harder time defending it. Complete failure like above.

If you don't define the micro failure consequence ahead of time then during the skill challenge failures (and successes) on individual checks simply influence the narrative, change the framing, and may open up or close down which skills are suitable.

This looks like good stuff to me, too

Yes, the DM still has a lot of latitude on how this unfolds but it doesn't have to work like -- "You can call for combat, or 1 healing surge lost, or 10 healing surges lost, or magic items or blessing disappearing, or loss of a hand or eye, or someone loyal betraying her, or whatever." When the consequences of the SC are decided before hand (whether not to tell the players is another choice but a lesser one) then I think it solves of lot of this issue.

To a degree (and I tend to write up failure consequences for my skill challenges before sessions, so I prefer this method in 4e), but two things pop up. The first is that the GM is the one setting the failure results, even in advance, so he still gets to pick the loss (1 or 10 healing surges lost, magic items disappearing, etc.). The second is that a lot of GMs will run skill challenges on the fly, and stopping to figure everything out (what primary skills can be used, what secondary skills can be used, the complexity, the consequences for failure) can really slow things down.

It also helps you decide if the SC is worth doing in the first place. Often the "success goal" sounds interesting, but if the "failure goal" isn't equally interesting than maybe it's not worth doing. That's why I much prefer narrative consequences to mechanical ones (healing surge lost, combat that doesn't really matter, etc.).

These are the kinds of things I wish they worked out further in a Next Gen Skill Challenge system.

I agree strongly about the "failure goal" needing to be very interesting (at each step of failure along the way, basically), or it's not worth it.

Thanks for the good post

 

[JamesonCourage]

Going to give my thinking on this right quick. Couple things:

1) My M.O. for establishing stakes and the adversary/danger interposing itself between a player and their goals stems from GMing Dogs in the Vineyard (I think...it might predate 2004, but that is when it was crystallized). In the above conflict I am playing "the vulnerability of the children" and the "danger that keeps them unsafe and unsecured." I may have just as easily invoked one of the children having a fit of terror (putting pressure on Saerie to have to figure out how to calm/comfort him/her or remove the terror...or deal with the consequences...this would be a "soft move" in Dungeon World).

2) The player in question has a strong maternal instinct. Inevitably, her characters reveal this virtue in play. She will inevitably become emotionally attached to the vulnerable, to the meek, so putting her in such spots resonates emotionally.

3) Given 1 above, shelter is the conduit for guaranteed safety and security. Denying it presently (but not removing it permanently from the table) and putting the children in a spot that highlights their vulnerability and amplifies the danger to them (as the ultimate resolution to the SC) does a few things:

a) It intensifies the moment and escalates the conflict (vulnerability and threat level heightened RIGHT NOW) that the PC is trying to mitigate.

b) It potentially brings some new immediate narrative fallout to deal with (that the PC cares about - eg the death of one or more of the children).

c) It provides the player (through the PC) a legitimate decision-point to attempt to continue to pursue a safe and secure location to secure the children. If it is overtly expressed by me that such prospects of a safe and secure shelter for the children are now completely closed out, the player's decision-points narrow. The action declaration now inevitably becomes "alright, I guess the kids are coming with me while I look for a river, follow it down the mountain, and hopefully find a settlement/village."

Further, as I charge a healing surge (just an open-descriptor resource for heroic mojo/luck, etc) for every failure (and all the PCs a failure on a failed SC), she also has the metagame vector of strategic resource consideration. If things went poorly there, she may have been down 4-5 healing surges (and possibly 1 or more daily) after a failed SC and a combat. Consequently, not only would the children remain vulnerable, but the PC herself would now become vulnerable. That feeds back onto (a) above.
​

I think if I wanted to do these things, I wouldn't run this skill challenge framed in the same terms you used, then ("find a secured shelter, safe from the elements and predators, where she can stow the children"). I've had the PCs run through various wilderness travel skill challenges before, and I tell them that we're going to see how banged up they get along the way; I'd probably frame it the same way here. "You're going to find shelter, but we're going to see if you lose any of the children along the way." You still have the tension right now (your A) and it puts something the character cares about at risk (your B). It doesn't hit your C (where the players can try again), because they find shelter at the end. But that just means we can move the game forward, so I'd probably go with it knowing that.

I think what you're saying here is just the inevitability of a game with a GM. Any game where there is a GM who bears primary responsibility for framing situations and responding to player action declarations and evolving the fiction post-resolution, is going to be vulnerable to poor GMing. I mean that is just an inevitability. What is important is how obvious the poor GMing is and/or how transparently manifest the bad faith is. A system with clarity in play procedures (primarily composed of explicit GMing principles/techniques and coherent resolution mechanics) will crank up the obvious and crank up the transparently manifest (hence less vulnerable to illusionism). If the players don't care...well, again, then you're in the realm of "participationism". And if so, then so be it because if we're all in on it then our table agenda is aligned.

Finally, a system that cranks up the obvious and cranks up the transparently manifest (Dungeon World and all of the Power By Apocalypse games do this even better than 4e) will typically yield a lower entry level for sound/good GMing.

I agree, which is why the skill challenge system in my game uses objective DCs that the players can look up, and has its formula laid out in the book (how to determine how many successes before 3 failures). The PCs can attempt to go into skill challenges based on certain things they know will likely come up (based on various scenarios that are explicitly resolved using the skill challenge system, complete with lists of skills and uses of those skills to use).

I think this is transparent and player-empowering, whereas I find the 4e skill challenge method mostly transparent, and not at all player-empowering (that is, explicitly giving the players the power to make things happen in the fiction without the GM's approval or permission -at least, according to the rules as written).

I'm sure you could. I know I could as well. And I'm certain it would be a great game for people who are "participationists" at heart (they're looking for GM force/illusionism to move the game along and they just want to press an attack button here or say a witty one liner there or rescue a few damsels in distress). However, the real question is (as I framed it above), is how easily could we get away with it from players who are specifically looking for the antithesis of that experience. If I used illusionism/GM force as the primary technique for moving myself (eg me GMing myself and another me - scary - playing a PC), or any other number of folks on this thread yourself included, though a 4e game....well, the jig would be up rather quickly!

I don't know how easy or hard it'd be, but I'm guessing that there would be too many other conflicts between that player style and my own GMing style for it to easily determined. As a GM, I'm not looking for any player to co-author my world. It's mine, and you get to play in it. It's a creative expression, and I'm going to make it how I envision it. There is no room for any other author. If players ask if they can do something, or make a certain type of character, or whatever, I do try to work with it, but I have zero problem saying "nope, find something else" just to please myself.

I think the kinds of players that are looking for the opposite of "participationist" play would probably already be clashing with the GMing style I just outlined. So I think it'd be really hard to tell if they would have problems with my hypothetical "illusionist" skill challenge style or just my GMing style, generally.

I'll do a Dungeon World conflict here when I'm next able (social conflict). I would really be curious about your thoughts on the system...especially in relation to 4e GMing principles (of which I feel are very much inspired by Vincent Baker - who wrote Dogs and Apocalypse World). Perhaps you could take a look at the SRD and just maybe read the section on Gamesmastering. It would be a good primer for when I post the DW conflict.

I've looked it over, and even incorporated a type of Dungeon World-inspired "moves" system for my 4-page superhero RPG that I use every couple of months for one-shots. I'm not going to look over anything again right now, but if you think anything in particular stands out when you do post that DW conflict, I'll look over that bit then.

 

[LostSoul]

1) My M.O. for establishing stakes and the adversary/danger interposing itself between a player and their goals stems from GMing Dogs in the Vineyard (I think...it might predate 2004, but that is when it was crystallized). In the above conflict I am playing "the vulnerability of the children" and the "danger that keeps them unsafe and unsecured." I may have just as easily invoked one of the children having a fit of terror (putting pressure on Saerie to have to figure out how to calm/comfort him/her or remove the terror...or deal with the consequences...this would be a "soft move" in Dungeon World).

2) The player in question has a strong maternal instinct. Inevitably, her characters reveal this virtue in play. She will inevitably become emotionally attached to the vulnerable, to the meek, so putting her in such spots resonates emotionally.

3) Given 1 above, shelter is the conduit for guaranteed safety and security. Denying it presently (but not removing it permanently from the table) and putting the children in a spot that highlights their vulnerability and amplifies the danger to them (as the ultimate resolution to the SC) does a few things:

a) It intensifies the moment and escalates the conflict (vulnerability and threat level heightened RIGHT NOW) that the PC is trying to mitigate.

b) It potentially brings some new immediate narrative fallout to deal with (that the PC cares about - eg the death of one or more of the children).

c) It provides the player (through the PC) a legitimate decision-point to attempt to continue to pursue a safe and secure location to secure the children. If it is overtly expressed by me that such prospects of a safe and secure shelter for the children are now completely closed out, the player's decision-points narrow. The action declaration now inevitably becomes "alright, I guess the kids are coming with me while I look for a river, follow it down the mountain, and hopefully find a settlement/village."

Further, as I charge a healing surge (just an open-descriptor resource for heroic mojo/luck, etc) for every failure (and all the PCs a failure on a failed SC), she also has the metagame vector of strategic resource consideration. If things went poorly there, she may have been down 4-5 healing surges (and possibly 1 or more daily) after a failed SC and a combat. Consequently, not only would the children remain vulnerable, but the PC herself would now become vulnerable. That feeds back onto (a) above.
​

I'd like to focus on point c. In what way does the player face a decision point? Is the decision point to let the children go (and if so, what benefit is gained by that) or is it simply a matter of choosing the PC's best skill and trying to make that matter? Or something else that I'm missing?

 

[AbdulAlhazred]

I'm trying to follow along. Are you saying that none of the natural laws are known, until a PC checks them? Or just the magical laws, which nobody (including the DM!) would have any pre-conceptions about?

It would be overwhelming to attempt to 'explore' the most simple and mundane basic things which make walking and talking and eating and breathing possible, and from which most of the less common but still mundane things follow (IE gravity works, basic physical mechanics resembles the real world, otherwise you couldn't swing swords or leap off a cliff into a pool of water). Thus in practically every fantasy world the mundane things work in the usual expected way (but note that the reasons might not be the same as in the real world, it could all be because 'spirits' decree it to be so and pull the strings, not because of Quantum Mechanics). Magic is of course not so constrained.

In either case, not only are you handing narrative power over to the players (!), but you're tying that power directly into the skill check of the characters! If the player wants it to be possible to channel firedrake essence into a ring, then that theory works because the character knows a lot about magic. My character knows a lot about magic, therefore he knows that my theory is true.

Sure, nothing wrong with that. Its a bit of an unorthodox view, but perfectly supportable

As opposed to my (much more traditional) method, where that action either is or is not possible, and a successful skill check would allow a character to know which it is (and, if it is possible, how to actually do it), but the narrative control - whether or not it is true - is wielded exclusively by the DM.

You may find it traditional, and it may well be similar in some respects to what -for example- Gygax did, but I think you have taken it to extremes.

I mean, even if you did want players to have some narrative control over the setting, it just seems like a bad idea to tie that into skill checks. Since narrative control is a player resource, you probably want to share that equally, rather than saying that stronger character in-game also has more power across the meta-game.

I think its quite possible to do either or both depending on what you want. I've never done exactly what Balesir is proposing, consider the dice to be a scientific experiment, but I don't see anything wrong with it. In point of fact it wouldn't be materially different from how I play now where we make up what can happen at least partly on the fly anyway and dice for it. 4e's 'say yes' then becomes 'please experiment a lot!'