Reactive spaces

My recent days have been spent staring at the ceiling, drawing abstract doodles on a whiteboard, or closing my eyes and watching higher-dimensional images fly through my consciousness. No, I haven’t been on drugs. I’m after a very specific piece of mathematics, to solve a specific problem. But I have no idea what that mathematics is.

I’m after the sameness between functions of time and functions that react to events. Here is my best attempt at putting my images into words:

I will call them (generalized) Reactives, to confuse classical FRP people :-). A Reactive is defined over an environment space, which is a set of environments (with some structure). Time is an environment space; it is uniform and boring, so all of its environments are identical or isomorphic. There is also an Event environment space, whose inhabitants are roughly Events from reactive (a set of separated occurrences).

A Reactive takes an environment and outputs values in terms of it somehow. Reactives have a notion of translation. Say you have a reactive over an event space which is “False until the next mouse click and then True”. Translating this switches which mouse click it is talking about, but not when the mouse clicks occur; so the transition point will always be exactly on an external mouse click. However, since time is a uniform space, translation of a reactive over time does correspond to simple translation, since there is no interesting structure to query.

I don’t know yet what exactly an environment is. I am trying to capture the fact that reactives over an event space can only switch on occurrences of events, whereas reactives over time correspond to continuous functions. If an event environment looks like an FRP Event, what does the time environment look like?

About these ads

One thought on “Reactive spaces

  1. Could it be over a branching space? That is, if we consider time as a discrete sequence (if you want it continuous, then just think about some kind of variation on Dedekind cuts), then at each point we branch for each possible input event. Then we define functions on this space, and those functions are pure and total.

    Then environments are a topology on these trees, and Reactives are maps from topology to value.

    Just my initial thoughts, from a problem I’ve been struggling with off and on in stochastic processes for a while.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s