Apps / Agenda Simulator / Support
Everything below applies to Agenda Simulator for iPhone, iPad and Mac. If your question is not answered here, write to me and a human will reply.
hello@no.spam.invalid.handfulapps.com
Reporting a problem? It helps enormously to include: your device and iOS version, what you expected to happen, what happened instead, and if the verdict looked wrong, the bloc positions and weights you had entered. A screenshot of the map and the verdict answers most questions on its own.
Open the app, tap the book icon, choose Load an example, then The 5-voter example (McKelvey chaos). That is the textbook construction: five voters, no stable outcome, and a three-step sequence of votes that reaches a target far outside anything a single vote could carry. Watch where it goes. The status quo starts 105 units from the target, the first vote moves to 138 away, the second to 154 away, and the third lands exactly on it. Every step but the last takes the room further from where the setter is steering, and each one wins a fair majority vote on its own.
The complete user guide covers every control, every reading on the map, all 5 proposal-rights modes, every line of the verdict, and the platform differences. A shorter version of the same material ships inside the app under Learn, including a glossary of every term.
Because the status quo you started from is one every bloc dislikes, so almost anything beats it. From the default position, 3,651 of the 10,201 positions on the grid win in a single vote. Put the status quo somewhere the blocs are content with and chains get longer.
Usually because a core exists: a position nothing can beat. With four equal blocs, "more than half" means three of four, which is a 75 percent supermajority in disguise. Delete one bloc so the same voters need two of three, and the core disappears.
You can, but that is a different question. Unreachable is a statement about the procedure given the preferences you entered. Dragging a bloc changes who the voters are. It is a legitimate sensitivity check. The app also runs the thing you probably mean instead, policy feedback, where each enacted agenda pulls opinion toward it. It works in the open and sponsor modes, and it is worth running because it fails: the reachable set collapses round after round, from 4,070 grid points to 1,480 in 3 rounds, while the closest anyone gets to the target does not move at all. The user guide works through the numbers.
Yes. The Mac version ships alongside the iPhone and iPad one. It is the same app and the same engine, so scenarios and verdicts match exactly.
Everything the app is for stays free: every built-in example, every voting rule, all 5 proposal-rights modes, the whole analysis and the entire Learn section. Whatever is on the map is kept between launches for everyone, so you cannot lose your work by closing the app. The subscription adds keeping several named scenarios side by side, and the Apple Intelligence assistants that suggest bloc placements and axis names.
Tap the book icon, then scroll to Subscription at the bottom. Status, Restore purchases, Manage or cancel, Terms of Use and Privacy Policy are all there, whether or not you subscribe.
They need Apple Intelligence, which needs a recent device and iOS 26 or later. The app itself runs on iOS 17 and up, so on an older device everything else works and the assistants do not. The subscription screen says so before you buy, and the rest of the subscription still applies.
Open the book icon, scroll to Subscription, and tap Restore purchases. If you subscribed with a different Apple Account, sign in with that one first. If it still will not turn on, write to me with the date of the purchase and I will sort it out.
The model assumes voters are sincere, each backing whichever proposal is closer to them, and myopic, judging every vote on its own without looking ahead to where a sequence leads. It assumes the agenda setter knows exactly where everyone stands, that votes cost nothing to hold, and that the policy space has two dimensions.
The computation has its own limit. Reachable verdicts are exact, and every vote in the chain can be checked by hand. Unreachable verdicts, and the claim that a stable position exists, mean nothing was found across a grid of 101 by 101 positions. That is a strong statement rather than a proof, and the app labels it as one.
It models the institutional question, not the mass-communication theory of agenda setting associated with McCombs, Shaw, Rogers and Dearing. Learn explains where the two traditions meet.