On 4/1/07, skaller <[EMAIL PROTECTED]> wrote: > On Sun, 2007-04-01 at 09:56 -0400, Chris King wrote: > > BTW: have you seen JoCaml?
I've heard of it but never checked it out, it looks interesting though. I'll try playing around with it in the next week or so. > Perhaps I misunderstand what > 'reactive programming' actually is? I think "reactive programming" in general has a slightly different connotation than "functional reactive programming"... FRP (as used by the Yale and Brown folks) denotes the system in which programs are constructed in a functional way, but (1) can reference values which change over time and (2) can change their own behavior over time (allowing the system to exhibit state). (1) is accomplished by having some sort of toplevel controller which monitors external stimuli and reevaluates the functional expressions whenever the stimuli changes. The toplevel could be a thread that reads from a channel or it could be triggered by a callback; it doesn't matter much since it is hidden from the programmer. (2) is accomplished by having the toplevel maintain the state of the entire system, allowing the programmer to rely solely on functional forms such as transition functions. Basically the whole question of callbacks vs. blocking threads vs. CPS is avoided with FRP... you don't have the awkward "push" model associated with callbacks but you get to keep the reactivity you "lose" with blocking threads (which are more suited to question/answer models than interactive GUIs), and you get CPS "for free" (since to FRP, CPS amounts to just changing your behavior for the next time cycle). Of course, it's possible to control-invert FRP for things such as dialog boxes if you have either threads or continuations at your disposal. You need only two new functions: one to block until an event occurs, and one which emits an event when some blocking function returns. > Do you mean fthreads? Pthreads are just that: pre-emptive threads, > typically Posix threads. Oh! Yes I did; braino on my part. > > I've > > played around with it a little in Felix... flxcc did a beautiful job > > of wrapping the library :) > > It did? Lol .. hmm.. hard to tell if that's sarcasm or what .. :) Haha no, no sarcasm... after I figured out what it wanted in the config file it did a decent job wrapping everything. Though certainly if I was doing it by hand I'd do it totally differently and make use of Felix's overloading (XCB has a lot of similar functions that differ only in their type). Of course anyone from XCB would tell you to extend the XCB code generator instead, but I'd rather write a 20-line flxcc config file than XSLT stuff any day :) - Chris ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Felix-language mailing list Felix-language@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/felix-language
