Logs for the RingoJS IRC channel, as logged by ringostarr.

2010-04-27

[00:00] <hannesw> i bet wait() will look really bad for lots of long waiting requests

[00:00] <xchl> that'd be super-interesting, actually

[00:00] <earl> well, the intuition would be that threads are far more expensive then the state you keep in then()

[00:01] <earl> xchl: i agree :)

[00:02] <earl> even more interesting would be a comparison to using java's bordmittel for that purpose

[00:03] <hannesw> bord-what?

[00:03] <earl> denglisch! :)

[00:03] <earl> java's highly tuned stuff in java.util.concurrent

[00:05] <hannesw> i don't think there is anything that could help there

[00:05] <hannesw> we're already using a java.uitl.concurrent threadpool for setTimeout()

[00:05] <hannesw> which obviously helps a lot

[00:06] <earl> Future?

[00:06] * xchl is trying to figure out the differences between promises and futures

[00:07] <hannesw> hmmm setTimeout does return futures

[00:07] <earl> xchl: not much to figure out in general, the terms are mostly used interchangably or conflictingly

[00:07] <xchl> so promises add some more functionality over futures, e.g. chaining?

[00:08] <hannesw> setTimeout(..., 0) does basically this: file:///usr/share/doc/openjdk-6-jre/api/java/util/concurrent/ExecutorService.html#submit(java.lang.Runnable)

[00:09] <earl> :)

[00:09] <hannesw> oops, file url :)

[00:09] <hannesw> hehe

[00:09] <hannesw> sorry

[00:09] <earl> http://java.sun.com/j2se/1.5.0/docs/api/java/util/concurrent/ExecutorService.html

[00:10] <hannesw> right

[00:10] <xchl> yeah

[00:10] <hannesw> so with java future, you only have the wait option

[00:10] <xchl> although I have to say that `submit` is clearer than `setTimeout(..., 0)` ;-)

[00:10] <hannesw> yep

[00:11] <hannesw> or maybe execute

[00:11] <earl> submit only makes sense in context of an executorservice

[00:11] <xchl> later ;-)

[00:11] <xchl> earl: yes

[00:11] <hannesw> gotta go too

[00:11] <earl> if setTimeout(,0) is supposed to be a thread-launching idiom, it should probaly be thread.start() or spawn()

[00:12] <hannesw> yes, true

[00:12] <hannesw> we already discussed spawn for processes, i think

[00:12] <earl> of course, those more general names would not really hint at the fact that it's not a real thread that is spawned

[00:12] <earl> but a thread from a pool reused

[00:12] <xchl> isn't spawn in rhino already?

[00:12] <hannesw> might be

[00:13] <hannesw> it is

[00:13] <earl> it's in ringo, in any case :)

[00:13] <earl> and it seems to spawn a fresh thread :)

[00:13] <xchl> it's the thread-launching shorthand in rhino, iirc

[00:13] <hannesw> funny :)

[00:13] <hannesw> ok

[00:14] <hannesw> if we keep that we should use a thread pool for that, and use it

[00:14] <hannesw> instead of setTimeout()

[00:14] <earl> well, maybe we should add another pooled primitive

[00:15] <xchl> splash!

[00:15] <earl> :)

[00:15] <xchl> ok, time to call it quits for today, I think ;-)

[00:15] <hannesw> yap met too

[00:16] <hannesw> good night!

[00:16] <xchl> byebye, hannes!

[00:16] <xchl> see you on wednesday

[08:50] <oberhamsi> ^^ oh nice we have our own promises, i'm getting used to them

[12:22] <hannesw> checked in ringo/promise

[12:22] <hannesw> http://ringojs.org/api/master/ringo/promise

[12:23] <hannesw> also, jsdoc app now uses markdown for @fileoverview and descriptions

[12:37] <robi42> hannesw, great! btw, good time to send out some reminders for tomorrow?

[12:54] * robi42 did so

[13:05] <hannesw> thanks robi42!

[13:05] <robi42> np, now let's also announce on helma ml? :)

[13:23] <hannesw> robi42 why not

[13:23] <robi42> would you do it?

[13:24] <hannesw> nope :) i'm drowning in stuff to do

[13:24] <robi42> k, so i'm doing it :)

[13:24] <hannesw> cool, thanks

[14:42] <waddler> heh some slightly politically incorrect processing.js http://www.uta.fi/~miika.haverinen/nazi/

[14:47] <waddler> hmm pasted on a whim since there was some discussion about processing yesterday without thinking about the nationalities here, no offence intended

[14:54] <oberhamsi> non taken, though better not post such stuff.. probably illegal in FIN too ;)

[14:55] <earl> but certainly not illegal in austria :)

[14:56] <oberhamsi> hu? did you look at it

[14:56] <earl> yes

[14:56] <oberhamsi> i'd avoid it

[14:58] <waddler> oh there was even laws about the symbolism, yep remembering now

[14:59] <earl> in any case, please keep out political stuff wherever possible

[14:59] <waddler> sure, sry

[15:01] <waddler> has anyone btw made XUL applications using rhino or ringo even?

[15:02] <waddler> not entirely sure how it would work just came to mind as something I'd like to try

[15:15] <waddler> there's a few java binding projects running on top of swing, http://jxul.sourceforge.net/ for one

[15:15] <waddler> hmm but that's old

[15:19] <waddler> wazaabi.org supports writing RCP eclipse plugins in XUL, sweet

[15:35] <waddler> actually thinlet http://thinletweb.appspot.com/ probably suits best, doesn't require swing either

[16:36] <waddler> hmm bad idea I guess, XUL applications need the runtime environment anyway

[17:11] <waddler> swt runs nicely though http://www.eclipse.org/swt/snippets/

[21:07] <earl> i just pushed some improvements to the recent changes rendering in ringowiki

[21:08] <earl> it's also live on ringojs.org: http://ringojs.org/wiki/recent

[21:08] <earl> enjoy

[21:10] <robi42> yay, we got a bliki :)

[21:11] <earl> nope, just a slightly more useful recent changes display :)

[21:11] <robi42> right, it's slick the same :)

[23:14] <hannesw> earl: wonderful!

[23:22] <earl> thanks :)