I'd readily agree that "offscreen failure" is bad thing and to be avoided.
OK - and in iteslf interesting to me! because not everyone agrees. I think it is inherent in Lewis Pulsipher's advice, for example, that there can be failure offscreen, and S'mon endorsed that recently in one or the other of these threads.
Character "on screen" actions during an earlier scene may have consequences which result in certain failures becoming unavoidable. If you only look from the current moment forward then that appears to be "off screen failure". But if you look at everything it is not.
Although this suggests that your conception of "offscreen failure" may be narrower than mine.
In my reply to JamesonCourage just posted on the other thread (and linked above in my reply to The Auld Grump), I talk about the importance of metagame signals in my game. This includes signalling what is at stake in a situation, and who cares about it (so signalling both from GM to players, and from players to GM). If a PC action on screen would cause future failure to be unavoidable,
but at the tme the players had their PCs perform those actions the players didn't know the stakes, then I would categorise that at offscreen failure.
Of course, there is then also room for different conceptions of what counts as the players knowing the stakes. I would tend to set a high bar for knowledge, and hence a low bar for something counting as failure offscreen, and hence want action resolution mechanics which minimise the likelihood that an action in this scene will inadvertantly (as opposed to deliberately) make impossible some resolution to a future scene.
A simple example - the PCs talk to an NPC. One of the PCs mentions something about his/her family. In the GM's notes on the NPC is recorded the fact that the NPC lost all her family in the war, and was devastated by that, and becomes alienated and unresponsive in situations where others talk casually about the pleasures of having a family. So now, how do I (as GM) respond? I would tend to make the alienation and unresponsiveness of the NPC an
active part of the scene, rather than just having the NPC go quite and wander off, leaving it to the players to wonder exactly what is going on and whether or not they have shut down a certain possible avenue for their PCs. I don't mind the players shutting down avenues, but like those to be deliberate choices.
One feature of using skill challenges as an action resolution mechanic is they can create a structure for the GM to play out this sort of "active" alienation/unresponsiveness, because they preclude the GM from "closing" the scene before a certain number of checks - and, therefore, a certain number of responses to complications by the PCs, which underpin those checks - have taken place. They are a type of "discipline" on both GM and players to keep scenes alive at least long enough for their signficance to get a chance to emerge.
Of course, there is always the possibility that no one cares about the situation, and the skill challenge just fizzles. This is the non-combat version of Hussar's bad experience with wraiths. In some ways it resembles the "climactic" final battle that ends in the first shot of the surprise round when a "1" is rolled on the saving throw, but the failure is more intimately bound up with an error of judgement by the GM (and perhaps by the players, in giving the wrong signals to the GM - which can happen, because not everyone can accurately predict how much they will care about something in advance).
I don't know how much other GMs on these boards think about encounter design in these sorts of terms, but for me it is an important and recurring consideration, given how much weight my approach to the game places on the satisfaction gained from participating in an encounter which, simply in virtue of the game's mechanics, is
not going to be resolved in a single die roll.
I think it is also something of a bogeyman that is far and away extremely infrequent in actually happening.
This may be true, but what I'm interested in is the techniques whereby it is avoided. As a good Forge-ite (or FORE*, to use The Shaman's term, I'm especially interest in techniques involving GM force, and the way that force is used (eg situational authority vs plot authority).
Off hand I can't think of any historic exception of true off-screen failure that was not fully intended as a plot device. And even those are vitrually always limited to things like "How the campaign started in the first place..."
Are you talking about your own game here?
* Friend of Ron Edwards
I don't want it to fail, but I think that WotC broke their market with 4e and the way they rolled it out. I don't think that they can mend the divide.
