Um ótimo artigo falando sobre o estado atual do playtest de D&D, e fazendo uma comparação bem convincente com programas de computador em fase Alpha:

http://critical-hits.com/2012/06/05/flying-by-the-fill-in-your-gaps/

Acho que vale bem a pena internalizarmos o quanto o sistema ainda está incompleto, e que isso não é necessariamente de propósito; aquilo que achamos que é "deixado vago para o mestre decidir" talvez não seja realmente, e está apenas vago porque são áreas que não foram trabalhadas ainda.

E aqui o novo blog post do Monte Cook, onde ele deixa claro que, se continuasse na WotC, é na direção de "menos regras, tudo nas mãos da 'lógica do Mestre' como era na 1e" que o D&D iria.

http://montecook.livejournal.com/254395.html

Uma ótima resposta nos comentários do artigo do Cook:

Ian Harac wrote:
Jun. 6th, 2012 02:51 pm (UTC)
Sorry, I need to disagree, a lot.
Part I, due to stupid comment length limits.)

I've been DMing for 34 years, in dozens of game systems. The ones I use the most are the ones with well defined rules. I do not find they ever hinder player actions - they're still free to try and do anything. What they hinder is the game of "playing the DM" -- of trying to get better odds of success by asking for DM fiat instead of using rules which might be too harsh.

Your chosen examples are good cases of why "logic", in the absence of mechanics for basic actions, is a bad idea. How hard is to put out a fire by rolling around on the ground? Within whatever action economy the game has, how many actions does it take? Depending on how I "logically" answer those questions, and by "logically", I mean, "based on some stuff I saw in movies, I guess", I change how useful/powerful fire-based effects are. If I establish it takes three rounds to put yourself out and you're helpless during that time, I've made fire more effective than if I say "It takes one round and you're at half defense", or whatever. Given the high degree of abstraction in game rules, there isn't a "right" answer, and "realistic" isn't always right -- but "consistent" and "balanced" are important. If you want fire to be powerful, you make the effects of being on fire harsh -- but that's the game designer's job, not mine. I can edit the rules if I want to change them for my game, but I shouldn't have to design them on the fly, or see my campaign go up in smoke (heh) because I made a bad call in the heat (heh, again) of the moment.

Can you jump the pit? How far can a human jump? If they're wearing plate armor? If they're not a human, but a dwarf? If they get a running start? I don't know! Leaving it up to "what seems logical" creates a powerful, subconscious, tendency to go with "what I want to happen", which is the worst thing for a DM to do.

Or take your itching powder example: Good lord, you're presenting the worst kind of DM pixelbitching as a feature, not a bug? I don't want to play "Guess what the DM will think is the most 'logical' course of action." I want to be told "You can end the effect with a full round action; describe what you're doing." or, at most, "You can end the effect with a full round action, Standard difficulty; If you describe an unusual or especially effective action, the DM may reduce this to Easy difficulty." Hell, I'd even accept something that provided a more concrete base for the DM to work from, starting with a generic "End Condition" action and suggesting "This is a fine powder. Brushing it off is distracting; doing so gives +5 to the roll but the character can take no other action. It can removed by one round in water, or by a Level 2 Wind or Telekinesis effect. Use these as a guideline for similar actions." This doesn't restrict you. It doesn't say, "You can do ONLY THIS, no matter what." If provides a sufficient framework that the DM can make an *informed* judgment about actions not explicitly mentioned, and the listed actions and their consequences set player expectations as to what they can accomplish.(Very important; a lot of players in rules-light games constantly look for the "I Win" button, the single, quick, action that will lead to victory without a die roll or skill use in sight.)

Part II:

You don't need tedious, detailed, rules to accomplish a lot. To use a D&D 4e example, if you're on fire with "save ends", the "save" includes all of the likely quick activities you'd take to end the fire: Patting it out, splashing yourself with water, etc. "Normal" or "quick" actions are subsumed in the basic roll. A DM should grant a bonus to the save if the player sacrifices an action to "do something", going beyond the assumed actions.

Lastly, "logic" isn't as, well, logical, as you'd think. The imagined world exists tenuously, distributed among the DM and all the players. Each is seeing a slightly different world; the DMs description may be interpreted differently by all players, and DMs mental image may be so strong, for him, that he's not aware of the details he isn't explicitly mentioning. We choose a few words to convey a wealth of small details:"A busy factory". "A crowded bar". When the time comes for specifics -- are there conveyer belts I can climb on? Are the tables in the bar bolted to the floor, or loose and easily flipped? -- if the players act on their mental image, they can be stymied if it doesn't match the DMs. Further, things like "What's possible?" depend on different personal experiences, or different tastes in fantasy and SF, or different expectations of what a given character can do, how easy or hard something can be. Pointing out that rules are often incomplete, or so abstract they can lead to comical or implausible situations, is not a good excuse for not having rules. Sure, it's ridiculous that, by pure D&D rules, every foot race between humans results in a perfect tie. Likewise, overly strict rules that try to deny DM adjudication, even when they lead to SOD-breaking results, or which don't even attempt to abstract anything happening in the game world, but are purely floating mechanics, are a problem -- 4e, especially early on, offers a lot of examples of this, too many dead horses to beat now.

My experience with the kind of "DM Logic" gaming you're advocating, and the reasons you give for why it's good, tell me it's a lot like Communism. It sounds great on paper and fails hideously in the real world, but people are so enamored by the idea they keep trying it, ignoring past failures because this time, it will be different. ("But this is how games WERE back in the early days!" Yes, and if you look at early issues of Dragon or A&E, or the first wave of D&D clones and competitors, you'll see that the first focus was changing this -- of adding rules, rules, rules, and more rules, often insanely detailed to the point of total brokenness, but the idea that we all used to live in the Gamer's Paradise where there were no rules until the dark overlords FORCED them on us is utter bull. We (the nascent, late 1970s, gaming community) wrote thousands of pages of bad, incomprehensible, and broken rules -- because they were better than no rules.)