Here’s the latest version of the display, somewhat shrunken to fit into this page:
The upper speech bubble says “Fate begins a new day”. The lower bubble shows the results of my first decision. It says “Caronycoorck goes to the waterworks.” Underneath that is the backspace glyph and the glyph for executing this command. This all works nicely.
However, when I execute the command, a new problem arises: the subsequent sentence is “Caronycoorck arrives at the waterworks.” It gets displayed in the upper bubble because it’s a trigger sentence: what happens that should trigger some action from Caronycoorck. But how do I show myself doing something? Am I talking to myself? Does the upper panel include my face telling me that I have arrived at the waterworks?
One solution would be to wipe out the upper face in this circumstance and show only the flat statement “Caronycoorck arrives at the waterworks.” Then the player simply reacts to that development.
Or perhaps I should take a more fundamental step: eliminating the sentence announcing the result. All other verbs build consequences into the verb execution; only the verb “depart for” requires a separate verb “arrive at” to conclude its action. This was necessary when there were long delays between departure and arrival. But that condition no longer arises. Instead, there is a new constraint: there can be only two actors on a stage at once. Thus, if you arrive at a stage that is already full, you don’t get to go inside.
Perhaps I should alter this so that there can be any number of actors on a stage. That would mean that the act of movement is an automatic one-step process. The verb becomes “go to” and it simply takes you there. However, in such a case, I’d have to be able to handle multiple actors on a stage. The engine can handle it, but it does raise all sorts of tricky issues with execution. What if two actor agree on a deal to trade information about a third actor, who just happens to walk in just after the first actor has revealed his bit of information? Does the second actor go ahead and fulfill his half of the deal, giving himself away to the third actor, or does he welch on the deal? Neither option seems acceptable.
But then we’d need a way to announce the arrival of the third actor — an event that says, “Third Actor arrives on stage”. That’s the “arrive at” verb! Under what circumstances should the first two actors continue as if everything was normal, and under what circumstances should they abort their conversation?
Clearly, the kinds of conversations intended here are meant to be private; it should not be possible to interrupt them. What to do, what to do?
OK, here’s a wild idea: let’s get rid of stages entirely. All interactions are necessarily pairwise and take place in a space that it custom to the pair. In other words, there are N*N/2 virtual stages, one for each pair. Actors don’t go to places, they go to other actors. With this scheme, we solve both problems.
Of course, the gamers will go nuts over this. “No locations at all?!?!?” they’ll scream. “People just exist in a virtual space?!?!?” Yep, that’ll drive ‘em batty. But it looks to be most dramatically effective solution.
I must sleep on this.