Fair point.
To build on that, I do think it would be challenging to design a truly arbitrary resolution process, a process that doesn't attempt to simulate something, simply because the usual resolution tools for these kind of games (an encounter table, a weather simulation table, a NPC/faction reaction table, etc.) require their elements to be derived from some source.
Okay, so I am going to be picky about language. While I think I get what you mean, I want to put focus on something here.
Does a shovel
attempt to dig a hole? Of course not. A shovel is a tool a human can use to attempt to dig a hole. The shovel cannot try, or accomplish, anything on its own. ANd a person can use the shovel to kill a snake that surprised them, rather than to dig a hole.
Same for processes. People attempt to simulate things with processes. But, a person can use a process
without attempting to simulate anything, too. Like, I can use an encounter table to simulate a monster ecology, or I can use it because I just don't wanna think much. If the latter, I may, as you note, choose a source that is quick, rather than one that simulates things. I can have Bird Encounter table that I sourced with an internet search, and mixes arctic, tropical, and desert birds, and use it for an encounter in evergreen forest.
I want to avoid the idea that sets of arbitrary systems end up being "simulations". Because then we get...
"Game X is a simulation!"
"Oh? What does it simulate?"
"It simulates that thing that Game X does!"
Which is clearly self-referential (even degenerate, as
all games become 100% simulations) which isn't helpful in our discussions.
Once you put a dragon on your encounter table, that detail alone has specified quite a few facts about your setting. Likewise putting "thunderstorm" on a weather generator.
For me, the table does not establish the facts. You establish facts first, and assemble the table to express those facts in play.