I have read numerous times about how ideas are a dime a dozen and just about worthless, but is there ever a point when an idea gains value? What if that idea has been refined over the course of months, or even years, to the point where just about every detail of the game can be explained in words precisely, down to each minute aspect? What if the design document is so comprehensive that it can be followed to a T, with little need for interpretation? Say, for instance, this "idea guy" was making an RPG, and in his GDD he has descriptions of every combat mechanic, formulas for every kind of calculation, tables of every item in the game (along with stats, descriptions, etc.), drop tables, blueprints of every map, the storyline progression, detail of each quest, and so on...
Even after all that, would that idea still be just about worthless, or would it have gained some value by then?
There is a point when an idea gains value: When it starts generating money.
As I said before, a thing is worth what someone will pay for it. When we say ideas are worthless, it's not because there is some giant price list somewhere that says "Idea: $0", it's because we know from experience that they're unsaleable. Can you ever imagine buying an idea off someone else?
As you embellish that idea, flesh it out and begin to turn it into an implementation, then the probability of being able to sell it to someone, and the amount they might pay for it - increases. You need to focus on aspects that reduce the buyer's risk and increase their potential reward, in order to make it an attractive investment.
FWIW, a giant monolithic GDD is far riskier and therefore much less attractive than a simple, buggy, but playable and fun prototype with no GDD whatsoever. Getting something implemented is therefore a far better use of your time than refining your GDD to the microscopic levels of detail.