But there is a difference between deciding these things based on what seems like it is plausibly the case (i.e. estimating the time it takes normal humans to hike through wilderness to figure out how long it takes the party to get there, using seasons to decide if the pass is snow-blocked or if the snow has melted, etc) and deciding it because you want something to happen, because you want to control pacing, etc. And of course something like travel time is normally handled by mosts systems anyways.
There may be a difference, yes. But depending on the specific circumstances, there are also similarities. If I'm looking to get to a specific city for some reason, and the way is blocked by the frozen mountain pass, then I'm constrained. It's not an unreasonable constraint... but will it feel different to me if the GM determined the pass was frozen based on his calculation of time and the season or if he just wanted to make it harder for me to get there? The player may not know or care about the difference. He may not even know the process the GM used to determine it.
Now... it's very likely that you have players that are comfortable with this type of game. And I expect that this is where all the mentions about trust come from. And on the one hand, I absolutely understand. If you are playing this game, and want the GM to be making all these decisions, you want to trust that he's doing so based on plausibility and causality and so on. That he's not putting his thumb on the scale, so to speak.
When it comes to this stuff, my area of concern as a GM is that my decision making is influenced to some extent no matter what. When faced with a set of plausible outcomes, if it's up to me to pick one, then I am picking the way things go. That my decision was based on that being the most likely isn't really different from me picking it because it sounded the most exciting. It's still the GM deciding how things will go.
And to be clear... this is not bad. In my Mothership campaign, I made several decisions like this. Many were based on what makes the most sense. A few times, it was definitely about what I thought was the most exciting. None of these are bad reasons for a GM to make a decision. But they both involve the GM directing play... even if they have a different motive for doing so.
Another way to do it, which is not about lack of trust, is to have clear and open mechanics for this stuff. And it's very possible that a living world GM would use such procedures, but from what I've gathered from our discussions, even those procedures may be flexible or ad hoc, depending on circumstances. What I'd advocate for is clear and player facing procedures. To me, this is what maximizes player agency within this game environment. That the information be available and or known to the player, with as little (Ideally nothing) that remains solely within the GM's head.
The reason for this is not lack of trust. It is instead about ensuring that the players have a chance to know and influence the outcome in some way, and that the outcome is then determined not by GM Fiat (however informed that may be by the events of play and the setting) but rather by procedure. That the GM disclaim decision making in these moments.
On time passage, in a sandbox in my experience this is one of the biggest differences than other games. That is something that usually is kind of negotiated. When I am running a sandbox I am very cautious about simply announcing the passage of time, because I don't want to gloss over anything the players might want to do
I expect you're right that this is a big difference. In my game of Spire, and in the game I just started of Blades, we elide time quite often. I don't think this is as potentially problematic as you imply. If you gloss over something... if you jump ahead to Wednesday and a player says "hey, I was hoping to do X on Tuesday..." then you just back it up and let them do what they wanted on Tuesday. We can control the flow of time... speed it up, slow it down, back it up, do it again... it's all malleable.
Now, this may not work at all times... certainly there are times when the passage of time becomes more specific, and we want it to be so. In Blades, this is typically during the Score phase of play. Once that begins, generally then we're playing moment to moment. Of course, the game does allow Flashbacks, but those are exceptions.
Spire lacks distinct play phases like Score and Downtime, though, so when we did it there, it'd always be by consensus. I'd say something like "Okay, so you guys schedule a meeting for tomorrow at noon with Mr. Frost... is there anything anyone wants to do before then, or should we jump to the meeting?" This works quite well, and puts things in the hands of the players.
This is probably one of my big takeaways from games like Blades and Spire... that we don't need to narrate every minute of the characters' lives. That we can jump around a bit, and fill in blanks as needed, and still have functional play and consistency with the events of play. It also allows us to focus on the more interesting bits of play, and not spend so much time waiting for those to emerge.
