Alex, this discussion is very interesting to me. Mind if we continue it for a while? Either way, thanks for taking your time so far. I really appreciate it.
The causes of the known are (1) laws of nature, casuality, history, geography, etc. The need to have knowns is why I reject "quantum ogres" and retroactive continuity adjustments for the sake of "the story"
Oh, me too, that's to me the most interesting thing about sandboxyness and why ACKS is even more appealing than Lab Lord, S&W etc with its more detailed systems for thieves' guilds, domains etc.
That's why this felt so shocking because to me it felt like it went against that. Or did you use some dice system, simplified delving mechanics etc to simulate the I.V.'s progress? I.e. you placed them on the hex map, rolled navigation / movement rolls for them, used a simplified skirmish system for them (Zak suggests 1d4+HD, one-hit-kills, for off-screen battles and that's something I usually use) etc? Or did you just determine by feel which dungeons I.V. had cleared? That can be fine but that feels to me like a bit of a retroactive continuitiy adjustment; or at least it's hard for me with my current perspective / capacity to distinguish it from other forms of quantum ogre. (Am familiar with Courtney's texts on that topic.)
(Side note: Since I believe in the Agency Theory of Fun, I feel that story games which compromise player agency are going down a dead-end design path. They require you to purposefully put obstacles in your character's own path, or purposefully abandon the objectivity of the game world, for the sake of making the character's path more interesting.)
Fascinated by the unabashed objectivism even for this :D
Yeah, we play a mix where the "physics layer" is a hard landscape of rules&consequences (and here, I want the game world to be very objective even though that's so difficult), but the "decision layer" are sometimes made with considerations other than "what would bring the greatest chance of low-risk, high-reward success for the character". "Flashlight dropping", it's refered to here in Europe, when you let your characters be cowardly, incompetent, emotional, messed up etc even though it brings them and their friends more pain than pleasure. For example two of the characters are a divorced couple who still have their hooks in each other, and another two are siblings who are rivals. There definitely is a "play to lose"-mentality. A combination of story gamers' goals in the hard landscape of the sandbox. Sometimes the combination is synergistic, sometimes dysergergistic.
You may be right that it's a dead end, design wise. My game design philosophy is to test various design paths with a detached mindset and see what sort of play it produces and what sort of play it hinders.
Hence, you should be subjective where subjectivity is required, e.g., in making choices about what the NPCs whose free wills you simulate, have chosen.
This trips me up because of all the power I have. If I understand you correctly, I can pretty much decide that all the rulers of Utaqa, Hafayah, Qadim, Muluk, Umara and Liham join forces with either Hawa or Qudra and make a joint strike against the other. That definitely does not sit right with me.
For me, one of several stumbling blocks / mental locks I had to untangle on my way to the sandbox mentality was that I needed separate hats between "prepper/world-builder" and "referee". If we've pressed play and I've got my referee hat on and the players enter a room with one dozen orcs. Then it's my job as referee to push the pedal to the metal wrt those orcs, to not pull any punches. The same Sandra that put those orcs in the room (either on my own volition, or through a stocking algorithm, or through choosing a module) -- that was "prepper Sandra", and now I'm "referee Sandra", the players have chosen to entered that room and now it's all action.
Same for humans with secrets in the cities, the players are interrogating some NPC in order for them to reveal the location of a spy the players want to capture. I can do it, I've got lots of practice at doing those sort of scenes as objectively as I can.
But when we zoom out to a macro level, that's where it breaks down for me. Portraying one dozen orcs -- in a room the players have voluntarily entered -- cruelly and skillfully, going for the casters first etc, doing guerilla tactics, ambushing the party in the dungeon etc -- I don't mess around, the gloves are off once we've put play. (My DM in the Lab Lord game, which was pretty lethal, ran a 5e campaign and noone died. In mine, it's been... pretty extreme. Even though we've run the exact same modules with the exact same monster counts etc. I just don't pull any punches).
But on the whole "world" level... if I decide that 10 000 orcs happen to march on the player's little inn (they own The Inn of the Billowing Sails in Hawa)... at that point I'm just one step away from being a "Rocks fall, everyone dies"-DM, amn't I? If I'm right, then it can't be the referee's free will entirely.
The players are controlling one character each. Even their lackeys, vassals etc (when they get any), they have to roll for on various tables, vagaries etc. The referee is controlling the rest of the world. Yes, army sizes etc are derived from population, realm sizes etc but if I basically have a Mindslaver helmet on all of Zakhara and they do my bidding, it's not going to be a fair fight. Hence the word "referee" rather than "adversary".
I always tell the players: "It's you vs the world, and I'm the referee". My monsters -- or rather, my pretties -- fight dirty AF. They hit downed PCs to kill them, they go for squishies, they hit the party where it hurts, they're cowardly and efficient. But I don't send all the monsters in the world against the party at once.
I've tried explaining my mental lock / stumbling block in a couple of ways here. Maybe it's subtle and hard to get.
(As a similar example of one that I did eventually get past was... the rules for jumping over chasms in some games are pretty complicated. But as GM, I can just put in any chasm width I want. That felt unfair. The metaphors of the two separate hats, and using different rules for myself in prep time vs myself "after we've pushed the 'play' button", have helped me a lot there.)
Hence, you should be subjective where subjectivity is required, e.g., in making choices about what the NPCs whose free wills you simulate, have chosen.
I mean, on page 131 on the core book there's a table for whether or not some NPC:s rebel against the players. Where's the line?
Controlling one or a handful of NPC:s that the PC:s run into in the slums and backwater inns are fine but when those NPC:s are the commanders of thousands?