Crazy Jerome
First Post
On the temperature analogy itself, my answer to how the game should be done is both ways.
The problem with the Mearls approach (show the default game as the game, then provide options) is that during design, preference is given so heavily to the default that the options become second-class pieces. Sometimes, they are mere kludges that don't work at all. Good options are built into a design from the beginning (or at least the hooks, and one or two examples to test the hook are).
The problem with the Cook approach (show the options equally, state the default) is, on the other hand, mostly in presentation. S'mon and others are entirely correct that this is a pain to read, especially for beginners. Moreover, it isn't a particularly interesting or fun way to learn a game for anyone. For example, I love the Hero System, but reading it for the first time makes the 4E PHB seem like the 1E DMG by comparison (well, almost
). The more strict Hero got about the neutral framework, the more boring it became in presentation. (In play, it is still great--once you know what you are doing.)
Ideally, from a game design/delivery perspective if not a cost/effort one, the game would be designed the Cook way, but presented the Mearls way. Playtesters would be divided into two groups, some getting one and some the other, with emphasis on the Cook way early and the Mearls way late. Some of the material in the Cook way would become the options/sidebars/appendices in the final Mearls way product.
There is almost never a problem or fuss about sidebars and options based on presentation. When people don't like them, it is almost always because the thing being presented in the sidebars does not work worth a flip.
The problem with the Mearls approach (show the default game as the game, then provide options) is that during design, preference is given so heavily to the default that the options become second-class pieces. Sometimes, they are mere kludges that don't work at all. Good options are built into a design from the beginning (or at least the hooks, and one or two examples to test the hook are).
The problem with the Cook approach (show the options equally, state the default) is, on the other hand, mostly in presentation. S'mon and others are entirely correct that this is a pain to read, especially for beginners. Moreover, it isn't a particularly interesting or fun way to learn a game for anyone. For example, I love the Hero System, but reading it for the first time makes the 4E PHB seem like the 1E DMG by comparison (well, almost

Ideally, from a game design/delivery perspective if not a cost/effort one, the game would be designed the Cook way, but presented the Mearls way. Playtesters would be divided into two groups, some getting one and some the other, with emphasis on the Cook way early and the Mearls way late. Some of the material in the Cook way would become the options/sidebars/appendices in the final Mearls way product.
There is almost never a problem or fuss about sidebars and options based on presentation. When people don't like them, it is almost always because the thing being presented in the sidebars does not work worth a flip.