There's a pretty big difference between:
DM: "You try your best but can't get the lock open. All still seems quiet. What do you do?"
This never occurs in Dungeon World--and, IMO, it should be extremely rare anywhere else.
and
DM: "You try your best but can't get the lock open. Oh, and there's someone behind the door just started screaming her head off, something about 'Intruder!', and lights are coming on within the house. What do you do?"
And now you're doing what all sorts of OSR fans do, following in the path of the execrable "Quick Primer": actively presenting the thing you don't like, don't play, and don't know all that well in the most antagonistic light possible.
This is why people keep telling you you keep
horribly mischaracterizing things, to the point that it becomes willful ignorance when you have been told--and explicitly shown--how your descriptions are wrong, but you
keep doing it.
I respect your intellect. But I'm starting to get very frustrated with your argumentation tactics, where you put overtly, intentionally uncharitable descriptions
that are factually wrong.
Because the thing you described here? That isn't how it works in Dungeon World. You have said something that is
flatly wrong. And I'm pretty sure you know what the rules for a character trained in picking locks look like, because
I posted them and I'm pretty sure you replied to the post where I did so. But just in case I'm wrong about that, here are those rules again.
Tricks of the Trade
When you
pick locks or pockets or disable traps, roll+DEX.
✴ On a 10+, you do it, no problem.
✴ On a 7–9, you still do it, but the GM will offer you two options between suspicion, danger, or cost.
As is the case with most DW moves, the result of failure is not explicitly stated. It is basic system information that (a) if your roll is 6 or less ("6-"), that's a "miss" or a "fail", and (b) the GM must make a "hard" move in response (or, if they feel it more appropriate, they may choose to make a "soft" move). Note, however, that the rules explicitly say that the GM should "never speak the name of your move". This is not because the GM is being secretive, as the move must be in the fiction and thus clearly public in at least some sense. Instead, if you
did speak the name of your move, it would kinda take folks out of the experience, so just
do it, don't put a silly verbal signpost on it.
These are some of the relevant GM moves:
- Reveal an Unwelcome Truth - Lilia fails to pick the lock, and overhears the cook grumbling about how the Master expects fresh food for a guest at this stupidly-late hour, which probably means she's going to sell something. Now Lilia can't wait--she has to get to the ruby as soon as possible, or else it'll be sold and who knows where it'll go then?!
- Use Up Their Resources - The lock is complex and stubborn...and one of Lilia's picks breaks. Great, now she not only isn't in the house, one of her picks is broken, which will make picking any other lock more difficult. Now she's going to have to find a different entrance, and preferably one that doesn't have a lock! (Note: This leans in more a "hard" move direction.)
- Separate the Characters - Lilia isn't going in alone, but the rest of the group expected her to get that door open for the others. Now, she's had to back off, and a guard has come out the door, because the cook got scared hearing noises at the back door. The party is separated, and none of them are inside!
- Put Someone in a Spot - The lock defies Lilia, and now, instead of a guard coming out of the door, one is coming in. There's no time for backup, she's going to have to hide, or fight, or bluff, or something, and it's gotta happen now. (Note: This leans more in a "soft" move direction.)
- Offer an Opportunity, With or Without Cost - The cook opens the door as Lilia is trying to pick it, and the old cook says, "Oh, THERE you are, Marie! Did you lose your key again?! Get up, you slovenly girl, and come help me with the master's midnight snack." Huge opportunity--but now Lilia will be stuck helping the cook, not sneaking around the house, and she'll have to successfully pass herself off as this "Marie" whom she's never met and knows nothing about, or she can run...and the old cook will alert the house, thus making any later entry MUCH more difficult.
There are other moves, but most of them aren't really relevant here, or are too context-specific to be particularly explanatory. As an example of the former, I can't really see how a failure to pick a lock would invite the "Deal Damage" GM move, which is generally the obvious fallback "hard move" in most contexts, it just doesn't really work
here. As an example of the latter, "Use a monster, location or danger move"--this would require us to invent a whole host of other information about Château d'Ys, its environs, its owner, the owner's security measures, etc., etc.--but such a move could be quite relevant in context. Perhaps the Mistress of Château d'Ys is secretly a hedge-wizard, and thus there are magical wards woven into all the defenses of her house. Perhaps, due to wanting to protect the ruby, the Master has hired special mercenaries to protect the house, who have specialized skills for finding and entrapping would-be burglars. Etc. Such things are essential to good Dungeon World play (and likewise good play for any PbtA game), but naturally are really specific and thus might make the example overburdened.
Finally, note that on a partial success, the Thief player gets a choice--but that choice should be just as grounded in the fiction as anything else. Hence my previous examples of "you break a lockpick
but get in unnoticed, OR you get in but the cook finds the door unlocked and gets suspicious" or the like. Both of those outcomes are plausible things that result from having
some success at getting the door open, but not the whole kit and kaboodle. Breaking the lockpick is the player choosing "I get through without causing long-term damage that someone could notice
because I wrench the lockpick, breaking it but not the lock" vs "I damage the lock, but keep my lockpicks, leaving behind evidence that the house has been infiltrated". This is not the player retrocausally manifesting anything; it is the player getting to decide for themselves what partially-negative consequences they'd prefer to deal with. Similar things happen all the time in DW combat (e.g. diving out of the way and thus leaving your friend all alone, vs. eating the attack but keeping your friend safe).
You're quite right that the cook will very likely be mobile and not always in one place; which opens up the possibility that even if the thief fails his attempt to get in, she's not there to notice. Or, she's there but is either asleep or so engaged in what she's doing that she doesn't clue in that someone's trying to break into the kitchen. Or, she's not the only one with keys; maybe other staff members come and go through that door meaning someone unlocking the door is a common thing (and maybe the thief knows this, which is why he's chosen that door as his entry point in the first place). All of these would lead to my first DM narration, above.
All you're doing is inventing reasons why the character
might succeed. Obviously, if the character
did not succeed, those reasons could not apply. This is exactly equivalent to all the reasons that a Fighter's axe might fail to connect, and yet the roll tells us it did, so those reasons
cannot apply in this context.
As with so many of these things, when you attempt to make a rule that excludes this method, it also excludes the methods you clearly accept and use frequently in D&D-alike games. When you carve out an exception meant to only apply to D&D, it turns out that that carve-out also gives an exception to this method. It
really well and truly isn't the horrendous awful bugbear you keep trying to paint it as.
Fail forward is still a failure. It is simply failure which pushes things toward
some kind of terminal point, a conclusion of some kind, from which new things can then spring after. Given it is in fact failure, that essentially always means that "fail forward" pushes things closer to a conclusion likely to be undesirable to one or more players. A single "fail forward" moment does not doom everything, generally speaking. Extreme moments occur but are rare. Usually, that just nudges things toward something worse, or makes a small thing worse right this moment without necessarily making most things crap-awful. But every failure (a) does something bad you'd really rather didn't happen, and (b) pushes things toward a (locally) final conclusion, whatever that conclusion ends up being.
And if he instead succeeds and does get in (i.e. rolls a success on the lock attempt) without attracting anyone's attention, if the cook is there their potentially noticing each other is a separate resolution, followed by resolving any interaction(s) they may have. One step at a time.
This, at least, I freely grant. Just because successfully picking the door lock didn't attract attention, doesn't mean you
cannot ever do so. Indeed, it would be very silly to claim otherwise--the whole point of the adventure is to face the challenges between you and getting the ruby. If those challenges were trivial, then I-as-GM would be failing my duties--I would not be "filling the characters' lives with adventure". The character may or may not succeed in their ultimate goal. Some of the time, even one's best efforts simply aren't enough--but instead of that leaving an empty void, DW pushes both GM and player onward toward something new, even with failure. If you fail grandly--failed roll after failed roll means (say) you got captured and the Mistress of Château d'Ys and now the ruby has been sold to the nefarious Doctor Nekropoleis, or whatever bad thing has occurred, and now you have the local-scale adventure of "escape (or convince the Mistress of her terrible mistake, or what-have-you)" and the longer-term adventure of "stop Doctor Nekropoleis", which (presumably) is part of why you wanted to secure the ruby in the first place. (Obviously, I am here making up further information that would, in an actual game, already be known WELL in advance; but every game is different, so this should be understood as a stand-in for whatever the
real contextual reason was for Lilia & co. to travel to Château d'Ys to steal the Desert Rose.)
