I whole-heartily agree with this, and would be interested in what the other "teachable techniques" are.
Scott Rehm, despite being thoroughly unpleasant to interact with, and before he started to believe his own press, came up with a
nice summary. I'm not sure if he was the first to articulate it, but he did a good job of articulating it. My take on the approach is this:
1. GM presents a situation.
2. Player describe (narratively) an action he wants to take.
3. GM determines whether action is even possible. If not, there has probably been a miscommunication: clear it up and go back to 1.
4. GM determines whether outcome needs to be randomly determined. It only needs to be randomly determined if it does not automatically succeed or automatically fail, AND if there is a consequence of failure. If it would automatically fail you may have a communication problem (eg. you think you said the chasm was 50' wide and the player thought you said 5' and is trying to jump over it); clear it up and go back to 1.
With respect to there being no consequences of failure, I am talking about a situation where there is any chance of success and there is nothing keeping the players from trying over and over again until they succeed. In 1e you often are given only one change to do something, but in other systems you can keep trying. If PC is trying in a dungeon and incurring wandering monster checks and making noise, that is a consequence. If PC is trying something (opening a chest, for example) in the privacy of his castle and nothing keeps him from trying over and over again, don't make him keep rolling. In retrospect it is obvious, but for a lot of people it is only obvious in retrospect.
If it the outcome does not need to be determined randomly, and there is no miscommunication that needs clearing up, skip to 9.
5. GM determines how to randomly determine the outcome. If there is a skill system this means you pick a skill/ability check and assign a target number. If there is no skill system you apply an existing mechanic (secret door rolls, attack rolls, thief ability checks, etc.) or pick a die and the number that would constitute success.
6. GM rolls the die or instructs the player to do so.
7. According to Rehm, the GM determines the outcome of the die roll. Sometimes there are sound reasons for this, but I am in the camp of people who prefer transparency whenever possible. I would roll most die rolls in the open if the VTT framework I use allowed it. That being said, there are times, which I think should be rare, when players should not be aware of whether they succeeded or failed.
8. GM interprets the roll and determines an outcome.
9. GM narrates the results of the attempted action.
However, as the players inevitably continue to probe for the unknowable, I find "How would you know?" a useful mantra. ... When I play, my players do get acess to the PHB, but I try to keep them out of the DMG and MM.
I'm pretty generous with what they would know. Since I use a d20 for pretty much every random roll there is already a lot of variability built in. So I will usually tell my players the target number if they ask for it ("With your skill at climbing you think you can probably get up that cliff" "What do I need?" "You need a 25" "Ok. [rolls] I got a natural 2." "Yeah, that ledge wasn't as strong as you thought it was."). That being said, they rarely ask for more than narrative information.
I am also usually transparent about the power level of NPCs. I figure people who fight for a living are good at sizing up other people's combat ability. Also, if an NPC is wearing plate, a crown instead of a helmet, leather bracers instead of steel vambraces, and dainty slippers, that pretty much screams this dude has a bunch of magic items. In fact, if the players ask, and the NPC is trying to conceal his power level, I secretly roll a bluff check for him.
I also am generous with knowledge regarding common monsters. You can be damned sure that the first fighter to fought a troll and killed it told every damn person he ran into the story of the regenerating monster and how it could only be killed with fire (or acid, whichever he figured out). It then got picked up by bards and become part of the popular culture. Which is why, in the 21st century, we know how Heracles killed the hydra. And if the mentor who trained you to 1st level didn't teach you how to kill a troll, or other common monster (technically "uncommon", but common enough), then he did a crappy job.
Not giving my players access to the DMG or MM wouldn't do much good. Most of us have DM'd a lot. We never used THAC0, but it really wouldn't matter, everyone knows the 1e progression. 4e is a bit different, but like I said, I'm generous with information and will tell them the rumors - although they may be coloured by the cultural prejudices of where the PCs grew up. Looks like the goblins you encountered from the 5th cohort of the last legion of the fallen Dhakaani empire aren't all stupid and cowardly after all.[/QUOTE]