You were the one discussing "everyone knowing some obscure facts" and "just needing a few well-travelled people so everyone will have heard of all these creatures". We have lots of well travelled people - why do so many people get mauled by Koalas that are "so cute" because they don't realize they have powerful sharp claws?
Because they're cute. If there were a D&D monster with an SoD and a deceptively benign appearance, I bet it would be really dangerous. I don't know of any.
It sets rules for identifying creatures, and dealing with various difficulty levels. The SRD on Knowledge is longer than that on Attacks of Opportunity, so that must be a really light system, huh?
A more apt analogy would be comparing the rules on Knowledge with the rules on attacks, period. That's a more similar scope. By comparison, the former is pretty light.
The flanking character is taking advantage of in-game opportunities to enhance the character's success.
The actual words the character says are in-game.
If the player is a couch potato who has to take two rest beaks to climb a flight of stairs, his character suffers no penalty compared to an iron man competitor when determining success at a feat of endurance. Why should a glib, well spoken player have an advantage over a stuttering wallflower in playing a smooth talking con man or a suave ladies' man spy?
I think there's a pretty clear distinction in D&D that you control the character's mind but not his body.
If we're not clear in what can and cannot be done, then it only comes down to what each of us decides. Why shouldn't a "really good check" get me the whole document, a "pretty good check" get me a page, a good check get a phrase and an OK one get me a word? I'll check one word at a time, by the way. It's easier to roll an OK check!
That would be the sort of thing that sensible DM interpretation of what the DC is and the scope of successful check easily prevents.
I think you set the benchmark too high for mundane tasks performed by trained professionals. There are some skills that require basic training to be viable.
There are, but I am not aware of any D&D skills that meet that description.
We sacrifice some corner case possibilities and accept simplification in the interest of playability.
Getting rid of trained only accomplishes precisely that goal.
Again, the line gets drawn somewhere. I would suggest that the Knowledge skill sets the bar that Arcane Knowledge with a 11+ DC is not common knowledge. It simply is not. Only those trained in this skill are exposed to it. In play, one may pick up some knowledge, but that would be "player memory/notes" knowledge of things the character has already been exposed to, not a knowledge skill.
Wait, so you're okay with a character benefiting from a player's knowledge for substituting the function of knowledge checks, but not his speech construction for diplomacy checks?
(I'm okay with both because you inhabit the character's mind and both are mental)
Emphasis added. It's common knowledge what they can do, how to defend against it and how to deal with them, but the players can't know their hit dice? No double standard here!
No, there isn't. "That's a basilisk, don't look at it" is knowable to characters. Its HD are not readily observable.
Why can't I make a "common knowledge" roll to remember successful and failed attempts? How did the NPC's group know that criminal would do the trick? Why can't the players use the same approach?
I don't understand the first question. As to the second, it's a good question, but they didn't really know for sure. Of course, they have a general idea of relative power levels, and one imagines that high-powered criminals who get the death penalty are a very valuable commodity. But there is a gamble there.
And the last question, the players could use the same approach, assuming they could find and chaperon a suitable sacrifice, pay all the normal costs, and
have no moral compunction about killing someone in cold blood in exchange for a resurrection.
