It is a 'mistake' in classic Gygaxian play to simply let the PCs bypass a 'locked door' (any obstacle). This smacks of going soft on them and letting play progress past some obstacle without testing the players ability against it. In this form of play such a thing is akin to the 'softballing' you describe earlier, and undermines the whole point of that mode of play.
Later, when play progressed into 'story telling' the process had to evolve. Because there were no longer necessarily specific obstacles on the map to be overcome, instead a structure of "the obvious course of the fiction" had to be imagined. So a sort of mythology grew out of the original GM referee role, that the GM could be a 'fair arbiter' of ANYTHING and that there was some definitive set of possibilities that could be discerned by the perspicacious GM that were "the logical possibilities." These became substitutes for the walls and doors and branches of corridor in the original model. Thus the ethos is that Lanefan has concluded that you have 'bypassed an obstacle' which he has determined MUST exist within the fiction, and thus you have committed an error of GMing.
Right.
This is why I call it "puzzle-solving" or
learning what is in the GM's notes (or in his/her
head).
And also why it call it
action resolution by reference to secret backstory - the GM draws upon unilaterally-determined and hitherto-unrevealed elements of the fiction (ie they are not yet part of any
shared fiction) in order to determine whether or not an action declaration has a chance to succeed, and if it has a chance to succeed whether that chance is nevertheless less than certain with the consequence that a check is required.
The matter of
uncertainty appears to be decided by the GM based on a combination of extrapolation from what s/he is imagining about the fiction, and a sense of not letting the players to "get away" with anything.
When play is unfolding in this way, it is hard to see that players are exercising agency. So if this is the principal mode of play, it seems that the game must be one with pretty low player agency.
There is at least one further practical matter around this. One way that this sort of play can become the
principal mode of play is that it requires a lot of time at the table for the players to learn what it is the GM is imagining. To give a concrete example: consider the example from my Traveller game of there being no submersibles available on Zinion. This took a few minutes at the table to establish - the players declared their interest in the possibility and I decided (I think by fiat, having regard to the world profile) that there weren't any. (Maybe I rolled some dice?)
Now the players could have pushed the point by trying to use Streetwise to find an irregular/unauthorised/stolen etc submersible, but didn't. They let my call stand.
But suppose that we had spent many minutes or tens of minutes roleplaying out the PCs' attempt to find an available submersible, in circumstances where I'd already determined that there wasn't one. That would have been tens of minutes of nothing but "characterisation and pantomime". Maybe with an essentially irrelevant brawl or more serious fight inserted as eg the PCs upset someone with overly pointed questions in their hunt for what the GM has already decided can't be found.
That would be low-agency play.
there are no real 'right answers' in terms of what MUST be chosen as obstacles in this mode of play. That choice is made simply on aesthetic grounds, and for the sake of interest in exploring particular possibilities. This is not 'skilled play' which demands each challenge be met, nor DM-directed story telling play which demands that a set of narrative options developed exclusively by the GM for her own reasons be presented and treated as obstacles.
<snip>
Since "avoiding an obstacle" is not some sort of 'softballing' or failure of GMing in this type of game, necessarily, there's no abandonment of anyone's role at the table.
Again, fully agreed.
The relevant question is not
did the GM go easy by letting the players circumvent an obstacle (an obstacle that exists only in the GM's unilateral conception of the fiction). As a GM I can always come up with more stuff for us to think about as we play, some of which will absolutely put the players to the test!
The relevant question is
did I as GM squib the issue and allow everyone at the table to insipidly slide past what really should have been a hard moment. Which connects back to
@Manbearcat and I talking about sentimentality and melodrama, and
@Campbell just upthread articulating what it means for the GM to be a "fan" of the PCs.
I'm happy to plead guilty to being softer as a GM than (I suspect)
@Campbell is! That's probably one thing that makes Prince Valiant appeal to me - melodrama is built into it! And it's why I tend to find BW a bit gut-wrenching, as player but even more as GM because it requires me to be hard!