1. 6
  1.  

  2. [Comment removed by author]

    1. 5

      It’s a good metaphor for reactive programming. FRP is much more focused on what Sodium now calls Cells, the continuous behavior changes. For this model, you could imagine each component of a system (even at large) as being continuously modified under input of discrete events.

      I don’t think anyone has experimented with F/RP in the large. That said, I think it might be a good base model. The major deficiency is that F/RP has no notion of message failure whatsoever and that is an important part of the semantics of large message passing systems.

      1. 1

        Conal Elliot is the guy who’s done the most FRP stuff in Haskell, and IIRC, someone wrote a Quake clone in Haskell using FRP. But yes, depending on your definition of ‘large,’ I’m not sure what the largest FRP systems are.

        That said, there have been some single page JavaScript apps written in FRP style. FlowDock uses Bacon.js, if I recall correctly.