I’ve seen the idea floating around in different places of putting game output in one window and the command prompt in another. For TADS3, would this mean that check() output would go in the game window and verify() in the prompt window? My understanding of the purpose of check() and verify() is that verify() catches input that is parseable but nonsensical in the game world, whereas check() catches input that could make sense but is disallowed for in-game reasons, for instance that it goes against the character’s personality. Is this accurate? Would it be accurate to think of verify() as operating on the conversation between player and parser, whereas check() operates on the conversation between player and character?
I think that’s a reasonable description of the differences between the text output from verify() and check(). Of course, it’s up to the individual author to do it that way. Nothing prevents your doing something like this – let’s say you’re writing a first-person past-tense game:
verify() { if (fishingPole.location != me) illogicalNow ('I thought about going fishing, and realized I didn’t have my pole. '); }
If TADS were set up to output verify messages to an error window (and I don’t know if it can be), the above would look wrong, because the content would belong in the main window.
Check out the article “Action Results” in the Technical Manual. The illogical macros are used inside a verify() block. They don’t call either check or verify.