[00:02] JimBastard_: me too [00:02] JimBastard_: im gonna try to make it to scurvyconf [00:05] deanlandolt: JimBastard_: where are you located? [00:05] JimBastard_: nyc [00:05] deanlandolt: you're ny right? that's not such a bad drive [00:05] deanlandolt: i hear the buses have wifi now :D [00:05] JimBastard_: yeah yeah, i just didnt get a ticket for jsconf [00:05] alexiskander has joined the channel [00:05] deanlandolt: yeah, just saying -- a trip down for friday night wouldn't be so bad [00:05] JimBastard_: yaya im gonna do it [00:06] deanlandolt: hell, you could sign up for the free yahoo classes and make it worth your while [00:06] JimBastard_: maybe see if i can do a lightning talk on hook.io [00:06] deanlandolt: or hang around all weekend and do the sat night cruise dealie and i think the closing ceromonies are open to everyone too [00:06] deanlandolt: deadline for talks is april 1 so definitely get something submitted [00:06] JimBastard_: for sure, im gonna try to figure something out [00:07] deanlandolt: hook.io's pretty damn slick -- play your cards right and you might get that last golden ticket :) [00:07] JimBastard_: haa haa [00:07] JimBastard_: ;-) [00:07] JimBastard_: thanks man [00:09] deanlandolt: in any event, get yourself down here -- as a native washingtonian i swear we'll show you a good time, even if it's just a night [00:10] JimBastard_: haha sounds like a plan, i might come down with charlie robbins [00:12] creationix: javajunky: that's some crazy nested if statements in your github ticket [00:16] gianni has joined the channel [00:26] dnolen has joined the channel [00:37] tmpvar has joined the channel [00:38] tmpvar: yo [00:43] JimBastard_: sup [00:44] magcius: _ry: you have any idea? [00:44] tmpvar: whats going on? [00:46] JimBastard_: grumble grumble hook.io [00:46] tmpvar: ^_^ [00:47] JimBastard_: had epic battle with ajax iframe uploads today [00:47] JimBastard_: i think i lost [00:47] tmpvar: lol [00:47] tmpvar: i know that battl [00:47] tmpvar: e [00:47] JimBastard_: everything works until i get the response back from the iframe, it fucks the formatting everytime [00:47] JimBastard_: the net response looks great [00:47] JimBastard_: but the client smashes it [00:47] tmpvar: weird [00:48] tmpvar: wrong content type? [00:48] JimBastard_: thats the hack [00:48] JimBastard_: since application/json will tell the iframe to load a mime type it doesnt know what to do with [00:48] JimBastard_: it has to be something else.... [00:48] JimBastard_: been trying a few tricks [00:48] tmpvar: text/plain :P [00:48] JimBastard_: yeah [00:48] tmpvar: haha, yeah.. pain [00:48] JimBastard_: so then my HUGE json string that contains html fragments see's eval() and is like FUUUUUUUUUUU [00:49] JimBastard_: except eval on the net response is perfect [00:49] JimBastard_: JSON.parse() is like no fucking way guy, no line number for you [00:49] JimBastard_: but its turning " into " [00:49] JimBastard_: on the client [00:49] JimBastard_: thats the main problem [00:49] tmpvar: weird, get a better browser ;) [00:50] JimBastard_: thats not the right answer [00:50] JimBastard_: its use SWF for file uploads [00:50] deanlandolt: JimBastard_: turning into "? something ain't right [00:50] JimBastard_: its the ajaxSubmit() form plugin thingy [00:50] deanlandolt: perhaps try something like dojo's window.name protocol [00:50] JimBastard_: http://jquery.malsup.com/form/ [00:51] tisba_ has joined the channel [00:51] JimBastard_: no clue deanlandolt [00:52] JimBastard_: dojo isnt an option [00:52] deanlandolt: i've been in that rodeo before -- bottom line is if you're doing an iframe file submit you should probably assume you're not getting a response [00:52] deanlandolt: you control the server? [00:52] JimBastard_: yeah [00:52] deanlandolt: hrm...yeah, window.name could help -- you don't need to actually roll in dojo [00:52] JimBastard_: ohhh [00:52] JimBastard_: stuff the data in window.name ? [00:52] deanlandolt: precisely [00:53] JimBastard_: yeah i started to dig through the bad code that is that plugin [00:53] tmpvar: heh [00:53] tmpvar: i rolled my own, which was less than horrible with jquery [00:53] tmpvar: barely less. [00:53] JimBastard_: you u snd codez plz? [00:53] JimBastard_: i need 4 class [00:54] deanlandolt: it's a disaster...hurray html5 file api :D [00:54] tmpvar: i dun have it tonight, i could check at work tomorrow [00:54] JimBastard_: ill see whats hood tomorrow i might just give up and do a 3 step form [01:00] drostie: JimBastard_: JSON.org has a parser which is rigid and non-eval. You could convince it to give you a line number. [01:00] tmpvar: heh [01:00] ditesh|cassini has joined the channel [01:01] deanlandolt: if you want a line number you could just log the actual json somewhere and throw it into a linter [01:02] deanlandolt: i think the problem is it "s unnecessarily... [01:02] deanlandolt: if you don't behave douglas crockford will escape your json #crockfordfact [01:02] crockbot: Douglas Crockford files his taxes in JSON, and the IRS thanks him. ara_p [01:02] deanlandolt: thank you crockbot :D [01:02] crockbot: Everytime you create a global variable, Douglas Crockford roadhouse kicks an intern erudianart [01:04] drostie: I was at some point trying to have a hybrid really-lax really-strict library of my own. The idea was to use HTML with and to indicate the parts that needed to be stripped away for valid JSON to occur, and then style it really bright. [01:06] drostie: The first draft basically just turned into a regex-driven JSON parser and I got somewhat frustrated with the slowness of progress. :< [01:08] deanlandolt: drostie: you're better off writing a json line scanner -- no need for regex, really... [01:08] deanlandolt: i did a streaming json parser in .net in an hour and i suck at .net [01:08] deanlandolt: k.i.s.s. [01:09] drostie: My regex parsers amount to line scanners. :D [01:09] bpot has joined the channel [01:11] drostie: I just find it nice to say "seek to the next control character, which in non-strict mode can be either :, \n, an unescaped ", or the end of the file" all at once rather than testing for them in a switch statement. [01:11] drostie: Blame python. I now hate any switch where the options can't ultimately be written as python lambdas. :x [01:12] drostie: Node.js might change that someday though :3 [01:12] tmpvar: you could do that now im guessing [01:13] tmpvar: it wouldnt be a "switch" per se, but its possible ^_^ [01:13] deanlandolt: python lamdas are just the redheaded stepchild of javascript [01:14] deanlandolt: a foolish insistence on avoiding block delimiters [01:14] deanlandolt: s/lamdas/lambdas [01:14] drostie: deanlandolt: the insistence is on avoiding python constructs which don't evaluate. :x [01:16] deanlandolt: that's your insistence...but python refuses to accept the value of multiline anon functions, and for that i started cheating on guido with my javascript mistress [01:16] deanlandolt: you can eval a multiline statement in python (using python's defintion of eval) [01:17] deanlandolt: you can exec a function -- and /that's/ where the action is [01:18] deanlandolt: the stubborn refusal to accomodate inline anon fns holds python back IMVHO [01:18] JimBastard_: drostie: i figured out whats breaking it, jQuery and or the plugin I'm using thinks its a good idea to replace " with " [01:19] blakemizerany has joined the channel [01:19] deanlandolt: JimBastard_: hack the plugin to set dataType: 'text' [01:20] JimBastard_: lol [01:20] tmpvar: JimBastard_, is it by chance doing a .html() ? [01:20] JimBastard_: text, json, xml, script [01:21] JimBastard_: http://github.com/malsup/form/raw/master/jquery.form.js?v2.43 [01:21] JimBastard_: ughh soo many innerHtml references [01:21] drostie: deanlandolt: I don't think forced naming really holds python back much. :/ [01:22] tmpvar: my god JimBastard_ this is horrible [01:22] JimBastard_: i know [01:22] tmpvar: my god, its full of fsck [01:22] deanlandolt: drostie: moving the /action/ away from the call to action can be very confusing [01:23] deanlandolt: but fair enough [01:23] drostie: deanlandolt: I mean, if we compare to say PHP, which prohibits /subfunctions/, we can really see "held back" in action. :D [01:24] deanlandolt: drostie: fair enough, but compare python's linevomit to an elegantly defined javascript object, functions and all...there's no comparison [01:25] deanlandolt: (and bear in mind, i've considered myself a pythonista for years) [01:28] drostie: I just find something aesthetically pleasing about the one-liner lambdas. [01:29] deanlandolt: drostie: same, and you can do that in js or ruby too...and it's more explicit (queue the zen of python: explicit is better than implicit) ... but the ability to spread that one liner to something more legible shouldn't be discounted [01:30] tmpvar: JimBastard_, have you seen this? http://valums.com/ajax-upload/ [01:30] tmpvar: worked for me :) [01:30] JimBastard_: yeah ive seen a lot of them we have a replacement good to go but there is this whole issue with the backend. ill just fuck with it tommorow [01:31] drostie: I should really be getting to bed though. :/ [01:31] drostie: friggin' time. :/ [01:33] tmpvar: yep [01:33] tmpvar: im racing the clock myself heh [01:33] tmpvar: gotta hussle, kid! [01:33] dnolen_ has joined the channel [01:38] JimBastard_: deanlandolt you make any progress on your ORM thing [01:42] deanlandolt: you mean pintura? that's not really mine as much as kriszyp's...i'm just a very heavily invested user [01:42] rudebwoy has joined the channel [01:42] deanlandolt: as far as progress on node, no -- not until after a conference next week -- and then i'm going all in on node+mongo [01:43] technoweenie has joined the channel [01:43] deanlandolt: node+sql's gonna have to wait -- nobody seems to care for sql in the node community anyway :) [01:45] devinus has left the channel [01:50] gf3 has joined the channel [01:51] r11t has joined the channel [01:56] r11t has joined the channel [02:00] softdrink has joined the channel [02:00] r11t has joined the channel [02:00] sh1mmer has joined the channel [02:11] binary42 has joined the channel [02:29] BryanWB has joined the channel [02:30] sh1mmer has joined the channel [02:35] blakemizerany has joined the channel [02:47] Ramone has joined the channel [02:48] Ramone: hey all... I get a seg fault streaming a 50K file over http. known issue? or should I whittle it down to a reasonable test case? [02:49] tmpvar: test case pl [02:49] tmpvar: plz* [02:50] Ramone: k... might take me a while to whittle this down, but I'll get on it now [02:50] tmpvar: well.. [02:50] JimBastard_: can you setup a minimal test case that reproduces problem? [02:50] tmpvar: i guess you could configure node --with-debug and gdb node script.js [02:50] Ramone: JimB: yeah that's what I'm going to do [02:50] JimBastard_: gist that shit [02:50] tmpvar: see where and why its blowing up [02:50] JimBastard_: someone will help [02:51] Ramone: sure... gimme 15 or so [02:51] Ramone: thanks guys [02:51] Ramone: there's a bunch of code to cut away ;) [02:51] tmpvar: ACTION low fives JimBastard_ [02:51] tmpvar: hah [02:51] JimBastard_: ACTION blocks [02:52] JimBastard_: i cant wait till i have hook.io hooked into the node irc bot [02:52] JimBastard_: @createHook{} [02:53] JimBastard_: added prettyPrint to the test cases : http://hook.io/test.html [02:54] tmpvar: lol [02:55] tmpvar: @createHook("twitter","tmpvar","x4fj9000", function(data) { sys.out(data)}); [02:58] tmpvar: JimBastard_, nice at the test cases.. sorry I was too involved with making up a fake pass heh [02:59] JimBastard_: thats not a valid hook config [02:59] JimBastard_: im polishing up the front some more for the v0.1 release [03:00] JimBastard_: i really need help getting the docs for creating protocols and definitions going, Tim_Smart's code is pretty straight forward but i'm still a bit confused on best practice stuff [03:02] JimBastard_: like the http protocol [03:02] JimBastard_: http://github.com/Marak/hook.io/blob/master/hookio/protocols/http.js [03:02] JimBastard_: or the timer protocol [03:02] JimBastard_: http://github.com/Marak/hook.io/blob/master/hookio/protocols/timer.js [03:03] JimBastard_: maybe even a protocol / definition explorer [03:05] Ramone: here's the segfault-causing code: http://gist.github.com/340807 ... it serves files from a dir named "public"... the file I was trying to serve is just jquery's js file. smaller files don't cause a segfault... [03:06] pedrobelo has joined the channel [03:07] tmpvar: Ramone, what version of node? [03:07] Ramone: latest from github [03:07] tmpvar: node -v [03:07] JimBastard_: question : line 114, is the request still going to close if there was an error? [03:07] tmpvar: k [03:08] Ramone: well I grabbed it yesterday [03:08] Ramone: v0.1.33-131-gbfdc421 [03:10] Ramone: JB: response, you mean? [03:11] JimBastard_: yeah [03:11] JimBastard_: i mean its probably not your problem but you should probably close those [03:11] JimBastard_: res.close(); [03:11] JimBastard_: amirite? [03:12] Ramone: hmmm maybe... still segfaults though [03:13] tmpvar: yikes, i cant compile the latest [03:15] dnolen has joined the channel [03:17] tmpvar: just needed a cleaning/love [03:18] mjr_ has joined the channel [03:21] tmpvar: Ramone, looks here [03:21] tmpvar: wow. [03:21] tmpvar: works here [03:21] Ramone: I'm recompiling too [03:21] tmpvar: pulling jquery.js development down [03:22] tmpvar: if you cant figure it out [03:22] tmpvar: run it with gdb node_g [03:22] Ramone: I figured it was filesize related... I can't see how it'd be content-related [03:22] tmpvar: (means you'll have to ./configure --debug) [03:22] tmpvar: im out for now, need to play with my apc. peace! [03:22] tmpvar: peace JimBastard [03:23] Ramone: thanks man [03:23] tmpvar: hit me up if you make plans :) [03:27] isaacs has joined the channel [03:33] mikeal has joined the channel [03:43] RayMorgan has joined the channel [03:51] gwoo has joined the channel [03:55] Ramone: can anyone throw me a clue on how to use gdb to figure out where I'm segfaulting? [03:58] JimBastard_: someone should be able to [04:03] rtomayko has joined the channel [04:08] Ramone: alright... got it segfaulting in gdb: [04:08] Ramone: Program received signal SIGSEGV, Segmentation fault. [04:08] Ramone: [Switching to Thread 0xb7d026d0 (LWP 15602)] [04:08] Ramone: 0x081ad934 in v8::HandleScope::RawClose (this=0xbfcca504, value=0x0) at /usr/local/node/deps/v8/src/api.cc:540 [04:08] Ramone: that's the absolute latest from github [04:19] BryanWB has joined the channel [04:20] drostie has joined the channel [04:23] isaacs has joined the channel [04:23] PyroPete1 has joined the channel [04:24] dgathright has joined the channel [04:27] RayMorgan has joined the channel [04:37] bpot has joined the channel [04:37] brainproxy: hmm getting: [04:37] brainproxy: node.js:815:9 [04:38] sveilin has joined the channel [04:48] dnolen has joined the channel [04:51] konobi: brainproxy: looks like you're throwing a string error rather than a `new Error("...");` [04:55] blakemizerany has joined the channel [04:56] sh1mmer has joined the channel [04:59] blakemizerany has joined the channel [05:02] christkv has joined the channel [05:03] brainproxy: seems to be related to setTimeout [05:03] brainproxy: like the timeout value is being ignored [05:04] brainproxy: and it just loops as fast as possible [05:04] brainproxy: until it dies [05:25] brainproxy: setInterval doesn't result in a loop either [05:25] brainproxy: it just dies [05:25] brainproxy: maybe I'm not using these correctly, but i think I am :( [05:28] isaacs: brainproxy: gist the code? [05:28] brainproxy: tomorrow, too tired, going to bed, going to go back to stable too [05:28] Tim_Smart: Anyone here used the socketpairs and friends? [05:28] isaacs: brainproxy: this worksforme: setInterval(function () { puts("hi") }, 500) [05:29] brainproxy: stable [05:29] brainproxy: or master [05:31] brainproxy: i tried: setInterval(puts(text), 5000); [05:31] brainproxy: and it prints once to the console [05:31] brainproxy: then exits [05:35] bronson has joined the channel [05:39] mjr_: brainproxy: are you finding that setInterval doesn't work with head? [05:42] comster has joined the channel [05:42] keeto has joined the channel [05:43] charlesjolley has joined the channel [05:47] brainproxy: it seems not to [05:47] brainproxy: maybe user error [05:48] mjr_: I can run it just fine on Darwin: http://gist.github.com/340872 [05:49] mjr_: Just checked Linux, and it works there too. [05:50] brainproxy: strange strange [05:50] brainproxy: I'm probably just too tired to be messing with this, will come back to it tomorrow; thanks guys, for taking a look [06:07] dgathright has joined the channel [06:08] aguynamedben has joined the channel [06:19] comster: anyone have trouble with the latest ajaxim? After the first load, the ajax aborts. [06:24] admc has joined the channel [06:25] admc: anyone come up with a solution for the {task: libv8.a SConstruct -> libv8.a} issue? [06:27] charlesjolley has joined the channel [06:35] micheil has joined the channel [06:44] r11t has joined the channel [06:51] Tim_Smart has joined the channel [06:51] BryanWB has joined the channel [07:01] brainproxy: btw, i never went to sleep, and was able to solve my issue thanks to ur guyz help; felt real good to push from a desktop to mobile phones online on 3g network, with resoultion down to 0.1s; node.js doing all the hard work :-D [07:01] brainproxy: node.js off on a server in the Cloud [07:01] brainproxy: friggin sweet [07:03] BryanWB has joined the channel [07:09] Tim_Smart: brainproxy: Now go to sleep :p [07:12] micheil has joined the channel [07:13] dgathright_ has joined the channel [07:14] nsm has joined the channel [07:16] nsm has left the channel [07:32] javajunky has joined the channel [07:39] jansc has joined the channel [07:40] kennethkalmer has joined the channel [07:52] mjr_: well, I just stayed up too late and rewrote repl.js [07:59] piranha has joined the channel [08:02] hassox has joined the channel [08:02] blazzy has joined the channel [08:04] markwubben has joined the channel [08:16] cpojer has joined the channel [08:18] nsm has joined the channel [08:33] QuietKnight has joined the channel [08:37] pdelgallego has joined the channel [08:39] javajunky has joined the channel [08:51] qFox has joined the channel [08:55] jansc has left the channel [09:06] teemow has joined the channel [09:09] Gruni has joined the channel [09:22] QuietKnight: I know some of you guys were talking about writing bindings to libxml the other day. I just stumbled over this: http://libxmljs.squishtech.com/ [09:24] QuietKnight: doh. I see it in the modules list. Nevermind. [09:26] Tim_Smart: QuietKnight: Unfortunately it doesn't have much regard for the DOM spec [09:29] nsm has joined the channel [09:29] hellp has joined the channel [09:30] hellp has joined the channel [09:30] rnewson has joined the channel [09:46] rolfb has joined the channel [09:55] cpojer has joined the channel [09:57] kjeldahl has joined the channel [10:08] maushu has joined the channel [10:10] maushu: Some projects should be moved to the modules page. [10:15] Tim_Smart has left the channel [10:16] rockstar has joined the channel [10:16] rockstar has joined the channel [10:24] nsm has joined the channel [10:37] cpojer has joined the channel [10:56] bronson has joined the channel [11:13] cedricv has joined the channel [11:22] hober has joined the channel [11:31] micheil has joined the channel [12:04] gwoo has joined the channel [12:12] stepheneb has joined the channel [12:13] gf3 has joined the channel [12:22] dnolen has joined the channel [12:23] binary42 has joined the channel [12:51] jfd has joined the channel [12:55] pmuellr has joined the channel [12:55] shreeve: anyone know if express has an irc channel? [12:58] maritz has joined the channel [13:07] tisba_ has joined the channel [13:13] davidsklar has joined the channel [13:16] javajunky: shreeve: I don't think so, but I'm about [13:18] shreeve: javajunky: btw, tried to install node/express/etc. today and everything works... still have a few problems with permissions, but at least things *seem* to be working now [13:20] javajunky: cool [13:21] jherdman has joined the channel [13:31] kennethkalmer has joined the channel [13:35] sh1mmer has joined the channel [13:40] QuietKnight has joined the channel [13:55] KungFuHamster: I'm having a hard time finding a lot about OO javascript.. I looked through 8 or 10 books at Micro center and the amount of space devoted to OO programming was minimal [13:57] maushu: Are you looking for OO in general or OO specific to javascript? [13:57] KungFuHamster: specific to js.. since it's a little different from other languages [13:57] maushu: http://www.phpied.com/3-ways-to-define-a-javascript-class/ [13:57] maushu: That helps. [13:58] KungFuHamster: thanks, I'll check it out [13:58] maushu: The good parts javascript book is also good. [13:58] maushu: Right crockbot? [13:58] crockbot: Douglas Crockford doesn't wait for onDomReady, the DOM waits for him.... http://bit.ly/91mQxH jmathai [13:58] blackdog` has joined the channel [14:01] ashb: warning: dont take everything crockford says as gospel. he's as wrong as he is right [14:01] crockbot: JSLint *WON'T* hurt Doug Crockford's feelings. thisisroot [14:01] webben: KungFuHamster: Crockford's videos at YUI Theatre are also ace. [14:01] crockbot: Douglas Crockford doesn't waste time with "just in time" JavaScript compilers, he just writes compiled JavaScript. ara_p [14:02] ashb: maushu: that example is crap. it doesn't list prototype [14:02] KungFuHamster: ok, thanks.. making a note [14:02] webben: KungFuHamster: as ashb says don't take him as gospel (or anyone else for that matter...) [14:02] stepheneb has joined the channel [14:07] maushu: ashb: Prototype? People still use that? [14:08] ashb: not that prototype [14:08] ashb: MyClass.prototype.methodName = function() {} [14:08] ashb: that prototype [14:08] maushu: Right. People still use that? [14:08] maushu: :p [14:09] ashb: why wouldn't they? [14:09] javajunky: maushu: um yes? [14:10] KungFuHamster: it's used extensively in one of the sample ircd server someone posted [14:11] KungFuHamster: well, I only know of one sample ircd in node, so I guess "one of" is superfluous [14:13] dnolen has joined the channel [14:16] alex-desktop has joined the channel [14:21] wereHamster has joined the channel [14:21] wereHamster: does anyone know if node-mongodb-native is threadsafe? [14:23] ashb: is any of node? [14:23] ashb: i thought v8 itself wasn't even thread safe [14:23] jherdman: ashb: iirc Node is process-oriented and not thread-oriented [14:23] ashb: yeah exactly. the only time it uses threads is for posix APIs that can be async-ified any other way is my understanding [14:25] wereHamster: so, do I need to open a new connection to mongod for each client that connects to node? [14:26] ashb: thats got nothing to do with threads [14:28] wereHamster: I guess my question is, can I open a single connection to the db at the top of my script and then use it in the callback passed to createServer or do I need to open the connection inside that callback? [14:29] jherdman: i think you can just open the one connection. [14:29] jherdman: iirc, if you look at the node-redis code they take that approach in their sample app [14:31] ashb: mongodb protocol certainl has request/response numbers so you can match things up [14:31] javajunky: wereHamster: its all good [14:32] javajunky: just do a db open and re-use, my express mongodb session manager has a reasonable example usage in it, bear with [14:32] aryounce has joined the channel [14:33] javajunky: wereHamster: http://github.com/ciaranj/express/blob/mongodb_session_provider/lib/express/plugins/session-mongodb.js [14:33] christkv: wereHamster: I'm the responsible for that driver. I'm finishing off some other projects but once that's done I'll start writing some articles about the driver. [14:33] javajunky: there is no connection pooling though so you can overflow the queue I think [14:33] javajunky: christkv is normally about [14:33] ashb: javajunky: look up :) [14:34] christkv: yeah if you need pooling just create a group of db's [14:34] christkv: and round-robin over them [14:34] christkv: for execution [14:34] KungFuHamster: hmm more hamsters [14:34] ashb: you can issue more than one command over a conneciton at once can't you? [14:34] javajunky: ashb: I Swear that wasn't there when I started typing ;) [14:34] christkv: ashb yes [14:34] ashb: corse that probably messes up the getLastError [14:35] ashb: (horrible idea, that) [14:35] christkv: probably yes [14:35] christkv: but only if you get multiple other errors in a row [14:35] christkv: also getLastError is only really needed for insert/update/delete [14:35] christkv: as they are executed async [14:35] christkv: by mongodb [14:42] ashb: christkv: 10gen people just recommended issuesing a getlasterror immediately after the up/in/del op (before returning control to the user/even loop) [14:43] christkv: hmm that sounds like a major pain in the ass on nodejs. I would have to queue the messages in the connection [14:44] christkv: right now the order is irrelevant it just goes through [14:44] ashb: wouldn't have to queue, surely [14:44] Yuffster has joined the channel [14:44] ashb: just dont fire the callback form the op, fire it from the getlasterror respons instead [14:44] christkv: ashb: jupp :D [14:44] christkv: ashb: contribs are welcome ;) [14:44] ashb: my interest isn't that high :D [14:45] christkv: ashb: it might take me a little time to get around to it. I'll look at it once net2 is released as a 0.1.3.X something or other [14:46] christkv: anybody using slicehost ? have you gotten node processes randomly killed ? [14:46] ashb: not node specifically, but slicehost does that for presistent servers [14:47] quirkey has joined the channel [14:47] ashb: i've heard people complain about that in other projects [14:47] ashb: i just dont use slicehost - i resell my own vms instead ;) [14:51] steadicat has joined the channel [14:52] ditesh|cassini has joined the channel [14:54] alexiskander has joined the channel [14:55] christkv: gaarg lol [14:55] christkv: who's as cheap and allows me to set up nodeblogs.com [14:56] javajunky: I'm using linode seems fairly cheap and hosts my node stuff fine [14:56] christkv: javajunky: and no random killing of your running code ? [14:56] javajunky: not so far that I've noticed .. btu I am running monit, and have a persistent session store :) [14:57] javajunky: plus you can pay monthly/pro rata which means if it doesn't work out you can exit nice an early (some hosts require 6 month minimums I believe) [14:57] tisba has joined the channel [15:01] christkv: yeah sounds good. I might give them a try and get rid of slicehost [15:03] cpojer has joined the channel [15:09] ashb: christkv: how much are you playing for slicehost? [15:10] softdrink has joined the channel [15:11] mjr_: I just started playing with a couple of slicehost instances last week. I'm pretty pleased with the price/performance ratio. [15:12] ashb: wow thats... expensive [15:12] softdrink: have you played with linode? [15:12] mjr_: $20/mo is expensive for your own private instance? [15:12] ashb: yeah [15:13] ashb: oh not quite as expensvei - did currency conversion wrong [15:13] ashb: but still more than i'd be willing to pay, and more than i'm charging [15:13] ashb: but they probably have an SLA ;) [15:13] mjr_: I'm not sure that they do. [15:13] softdrink: my current provider: http://www.linode.com/ (i love 'em) [15:14] ashb: also i have to olo at the price of the highest slice [15:14] ashb: thats 10x the price i'm playing for an entire server [15:15] mjr_: Sounds like a business opportunity for you. [15:15] ashb: effort and marketing ;) [15:15] ashb: things i dont do [15:15] CIA-77: node: 03Ryan Dahl 07master * r54d4efd 10/ (8 files in 2 dirs): [15:15] CIA-77: node: Upgrade http-parser [15:15] CIA-77: node: Now at version 6f72c780f0a237a775150a9963bcdf5299685cde - http://bit.ly/9jvDTu [15:15] CIA-77: node: 03arlolra 07master * r6f84063 10/ (3 files): Migrates benchmarks to the new api. - http://bit.ly/czObQm [15:16] mjr_: Yeah, I guess that's how they end up charging US$20/mo for something that appears to cost them $5 or whatever. [15:16] ashb: oh it probably costs them even less [15:16] ashb: in terms of raw hardware [15:16] mjr_: How many instances can they squeeze onto one computer? [15:17] ashb: how long is a piece of string [15:17] ashb: depends how much over-booking you do and what size the instance are [15:17] ashb: and how powerful the server [15:17] mjr_: They can't over subscribe the RAM though. [15:17] ashb: yeah they can :) [15:17] ashb: and often do [15:17] mjr_: Well, I guess they could, and then grind the whole thing down into swap land. [15:18] ashb: swap on an SSD wouldn't be *too* painful [15:18] ashb: but i dont think Xen will let you subscribe the RAM mind [15:18] ashb: never tried mine [15:19] ashb: *d [15:19] mjr_: I'm glad someone else is figuring this out. $20/mo to be free from the hassle of hardware management is a huge win for me. [15:19] ashb: sure. [15:19] KungFuHamster: as long as the service is good enough [15:19] ashb: its $5 more than what i've been charging people [15:20] ashb: (tho i work in GBP, not USD) [15:21] mjr_: Once things get down into this price range, Is the monthly fee really what decides it for people? $20 or $15 or $25 all sound about the same to me, given how much different the performance is likely to be. [15:21] ashb: yeah - thats why i just sell to friends/colleges whatever. its too much effort to try to market at that price point [15:22] KungFuHamster: lots of competition at that level [15:22] ashb: yup [15:23] ashb: er *colleagues [15:23] ashb: not colleges [15:23] mjr_: Sounds like you should raise your price up to the US$20 range and then treat yourself to a nice meal. [15:23] mjr_: Maybe have a cocktail. [15:23] ashb: i mainly just do it to cover the cost of my own server uses [15:25] KungFuHamster: there are hundreds of stock templates for web providers... someone can set up a professional-looking reseller site in 10 minutes and you wouldn't be able to tell it's a teenager from Delhi on a 128k DSL line.. you can't even trust review sites because those are just people making referral dollars through an automated script with false reviews that push you towards the one with the biggest signing bonus [15:26] KungFuHamster: pardon my rant [15:26] ashb: heh ;) [15:29] mjr_: Oh, also AOL is going to kill the Internet. [15:29] mjr_: Once they allow all those dialup users access. [15:37] bpot has joined the channel [15:41] mjr_: KungFuHamster: do you ever worry that by working on things like node that we are hastening the crapification of the Intenret? Because we sort of are. [15:41] jherdman: why do you say that mjr_? [15:42] mjr_: Powerful tools can be used for good or evil. [15:42] jherdman: go on [15:43] QuietKnight: node installed on a botnet for DDOS attacks. Fun. [15:43] KungFuHamster: no.. adsense and other easily-available advertising is what is killing the internet [15:43] mjr_: But with things like $20/mo VPS hosting, nginx, and node, any random asshole can make things that would have taken entire racks of gear and tends of thousands of dollars just a few years ago. [15:44] QuietKnight: just don't ever get around to porting to windows :P [15:44] KungFuHamster: meh, good IDS software can track that stuff down... any single server/botnet that raises its head above the crowd will get it shot off [15:44] mjr_: Right, and then there's the abuse potential of being able to handle so many concurrent connections. [15:45] jherdman: by that logic we should go back to rubbing sticks together because matches have the potential for making more fires irresponsibly [15:50] confounds has joined the channel [15:52] mjr_: Exactly. When you advance technology, everybody gets to use it, even jerks. [15:53] RayMorgan has joined the channel [15:56] inimino: christkv: I've had a slicehost VPS up and running for two years, and never had any downtime or any random processes killed [15:56] inimino: also they seem to not over-commit their hardware at all as far as I can tell [15:56] ashb: they probably do CPU [15:57] ashb: but thats not really over-commiting [15:57] inimino: that's just good sense :) [15:57] ashb: yeah [15:58] KungFuHamster: I'm spoiled, I've had access to dedis for like 8 years now [15:58] KungFuHamster: *free [16:05] kennethkalmer has joined the channel [16:06] JimBastard: zzzz [16:06] ashb: q [16:09] halorgium: hmmm, is there a buffered tokenizer written for node? [16:09] mikeal has joined the channel [16:09] halorgium: i guess it is pretty straight-forward [16:10] jlilly has joined the channel [16:11] indiefan has joined the channel [16:17] kriszyp has joined the channel [16:19] charlesjolley has joined the channel [16:28] technoweenie has joined the channel [16:32] technowe_ has joined the channel [16:35] silentrob has joined the channel [16:35] gwoo: halorgium: does jslint do anything like that? [16:38] _ry: x/window 11 [16:42] aguynamedben has joined the channel [17:01] JimBastard: fuck you XHR iframe file uploads, i defeat you [17:01] JimBastard: i am master of all client side hacks [17:02] BRMatt has joined the channel [17:02] erichocean has joined the channel [17:10] Gruni has joined the channel [17:12] stephenlb has joined the channel [17:17] arlolra has joined the channel [17:19] technoweenie has joined the channel [17:20] TheEnd2012 has joined the channel [17:20] dnolen has joined the channel [17:25] admc has joined the channel [17:27] dandean has joined the channel [17:28] siculars has joined the channel [17:29] dgathright has joined the channel [17:29] rnewson has joined the channel [17:30] juvenn has joined the channel [17:30] juvenn has left the channel [17:31] dgathright_ has joined the channel [17:34] mikeal has joined the channel [17:34] maritz has joined the channel [17:38] cpojer has joined the channel [17:45] MattJ has joined the channel [17:55] kjeldahl_ has joined the channel [18:01] isaacs has joined the channel [18:01] aryounce has joined the channel [18:07] charlesjolley has joined the channel [18:08] charlesjolley: hi yall [18:08] charlesjolley: was wondering if anyone has looked at getting node.js to run with JavaScriptCode instead of V8. [18:10] jherdman: what's that, charlesjolley? [18:10] charlesjolley: oops - sorry I mean JavaScriptCore - i.e. Nitro - the JS engine from WebKit [18:11] jherdman: ohh [18:11] mikeal: node makes heavy use of the C++ bindings in v8 [18:12] mikeal: it's not trivial to port it to another javascript vm [18:12] charlesjolley: yes I know. My team would rather we use Nitro for some reasons I can't discuss. I'm trying to gauge if there are any other options that would satisfy this requirement. [18:12] mikeal: and v8's optimizations line up well with the programming model [18:12] charlesjolley: orly? [18:12] charlesjolley: like what? [18:12] mikeal: function inlining [18:12] joshbuddy has joined the channel [18:12] joshbuddy has joined the channel [18:13] mikeal: make functions super cheap and highly optimized, you end up creating a lot of functions in node that you run once and then get garbage collected [18:13] mikeal: so they need to be as cheap and optimized as possible [18:13] jherdman: charlesjolley: can you hint at why your time might prefer Nitro? [18:14] mikeal: they work in Cupertino? :P [18:14] jherdman: lol [18:14] charlesjolley: no [18:14] charlesjolley: sorry :) [18:15] charlesjolley: doesn't function inlining only really help when you call the same function over and over? [18:15] mikeal: i smell the fog of NDAs [18:15] charlesjolley: it doesn't get inlined the first time through right? [18:15] mikeal: I bet Apple is around here somewhere [18:15] mikeal: charlesjolley: yes it does [18:15] mikeal: it's not JITd [18:15] mikeal: er JIT'd [18:15] charlesjolley: hm ok. [18:16] mikeal: that's why it's faster for this case than Spidermonkey's tracer [18:16] charlesjolley: I see [18:18] charlesjolley: mikeal do you mean inline caching? [18:19] mikeal: yes] [18:19] mikeal: http://portal.acm.org/citation.cfm?id=800017.800542 [18:19] charlesjolley: yes ok [18:19] mikeal: everything always comes back to smalltalk doesn't it :) [18:19] charlesjolley: by function inlining I thought you meant as in C/C++ where the function code is copied into the caller function [18:20] charlesjolley: yep [18:20] charlesjolley: hm I thought nitro did that also [18:20] charlesjolley: pretty common trick. Ruby 1.9 has it too I think [18:20] charlesjolley: yes it does always come back to smalltalk. The most important language [almost] no one has ever used. :/ [18:20] charlesjolley: too bad too - it is a fun lang to write in [18:20] jherdman: woof. i clearly need to do my homework... [18:20] mikeal: Ruby can't do it efficiently with closures [18:21] charlesjolley: why can V8? I thought their closure model was pretty much the same. [18:21] mikeal: even v8 starts to show how the optimization is more overhead that it's worth if you have a loop that references closure data [18:21] bpot has joined the channel [18:22] mikeal: nah, Ruby just has all of it's var lookups in every scope look through the scopes that would be accesible in the closure, just like Python [18:22] dgathright has joined the channel [18:23] mikeal: in all modern js implementations the function is inspected and only referenced variables will have pointers included for the closure [18:23] mikeal: it's a lot faster [18:23] jherdman: that's pretty spiffy [18:24] charlesjolley: oh so when the closure is generated it come up with a specific list of available vars [18:24] charlesjolley: that explains why with() is so slow [18:24] charlesjolley: it breaks that [18:24] mikeal: right, and the garbage collector can release all the other crap in the parent scopes if it's not referenced anywhere else [18:24] charlesjolley: you never know what vars should be visible [18:24] mikeal: if you include an eval() in your closure it fucks it all up tho :) [18:24] charlesjolley: well I'm told you revert to the Ruby/python model if you use a with() [18:25] charlesjolley: ha [18:25] charlesjolley: so nitro does not do inline caching? [18:25] mikeal: i don't know about with, but that's basically what happens when you use eval() [18:25] mikeal: i'm not nearly as familiar with nitro [18:25] charlesjolley: oh no - it does [18:25] charlesjolley: http://webkit.org/blog/214/introducing-squirrelfish-extreme/ - polymorphic inline caching [18:25] charlesjolley: exact same trick [18:25] mikeal: i used to be at Mozilla so i'm pretty familiar with spidermonkey, and i use it daily on couchdb stuff [18:26] charlesjolley: yes I wish couchdb were using Nitro or V8 [18:26] charlesjolley: for some of these [18:26] charlesjolley: though I guess Mozilla's next thing is getting a lot of this too right? [18:26] mikeal: it shouldn't be hard to make it work [18:26] charlesjolley: r u working on couchdb? [18:26] mikeal: charlesjolley: i don't know if it's public but that was the plan before i left Mozilla [18:26] mikeal: the places where the tracer is inadequate are the places that inlining works really well [18:27] mikeal: yeah, i work at couch.io [18:27] charlesjolley: interesting [18:27] charlesjolley: oh cool [18:27] charlesjolley: I know jchris [18:27] charlesjolley: I work on the sproutcore project [18:27] mikeal: he's like 10 feet away from me :) [18:27] mikeal: oh cool, i had to deal with that a bit when integrating some Bespin stuff [18:27] charlesjolley: nice. yes I'm very excited about your CommonJS work [18:28] charlesjolley: I'm in the process of converting SproutCore over to use CommonJS as well. [18:28] mikeal: the thing is, about tracing, is that it's perfect for CouchDB [18:28] charlesjolley: Looking forward to having a CouchDB data source [18:28] mikeal: it's like the optimization was MADE for the view server [18:28] charlesjolley: that makes sense [18:28] mikeal: but my hope is to make the view server engine agnostic [18:28] charlesjolley: though really does it matter? [18:28] mikeal: we mostly just pipe stuff to the js bin [18:29] mikeal: it's going to start to matter on mobile [18:29] charlesjolley: Tracemonkey is C++? [18:29] charlesjolley: my biggest reason for not using Spidermonkey in the past was the java startup time [18:29] jherdman: i thought spidermonkey was C++ and Rhino was Java? [18:29] inimino: correct [18:30] charlesjolley: oh that's right [18:30] charlesjolley: hm. somehow our tools always end up with Rhino though. :) [18:30] jan____: ouch :) [18:31] mikeal: spidermonkey isn't java [18:31] mikeal: at all [18:34] stepheneb has joined the channel [18:45] blakemizerany has joined the channel [18:50] charlesjolley has joined the channel [18:55] al3x__ has joined the channel [18:58] mikeal: this is so weird [18:58] mikeal: how come, on this ubuntu machine, when i clone a repository with git the branches don't come down with it [18:59] mikeal: on my Mac when i clone I get all the branches [18:59] mikeal: they are only available by the remote references [18:59] jherdman: same version of git? [18:59] mikeal: Mac is slightly newer [18:59] mikeal: 1.6.6 vs 1.6.3.3 [18:59] mikeal: that shouldn't make this large of a difference [18:59] jherdman: what about your config files? are they the same? [19:00] mikeal: both are default [19:00] mikeal: but maybe mac git sets something [19:00] mikeal: whoops [19:00] mikeal: someone else should really host crockbot [19:01] mikeal: JimBastard: you up for hosting? [19:01] JimBastard: yeah man i told you [19:01] JimBastard: im all ready, i tried to launch him but the triggers don't allow multiple bots in the same room [19:01] JimBastard: once one of them triggers the other its loop tim [19:01] mikeal: ok, well, mine is down [19:01] JimBastard: e [19:01] mikeal: make sure you pull the latest [19:02] mikeal: cause i think i did make some changes [19:02] JimBastard: we had it working last thursday [19:02] JimBastard: i just gotta go home and link it up [19:03] KungFuHamster: is the source for crackpot posted somewhere/ [19:05] JimBastard: yeah its a Jerk script [19:05] JimBastard: no mikeal i can turn him back on tonight when i get home aight? [19:05] gsf has joined the channel [19:05] JimBastard: really we should add a conditional in there though [19:05] KungFuHamster: I remember seeing Jerk listed, I wasn't sure if it was that or a different one [19:05] JimBastard: to prevent triggering loops [19:05] JimBastard: or we could just not run more then one bot [19:05] mikeal: oh yeah, having two of them in one channel would flood [19:05] KungFuHamster: throttling would be nice, heh.. I was in here when they went berserk [19:06] mikeal: i can't work on it today [19:06] mikeal: but it's easy code to write [19:06] mikeal: github.com/mikeal/crockbot [19:06] mikeal: JimBastard: you have write access to my repo so just push to it if you fix something [19:07] JimBastard: cool thanks [19:07] JimBastard: tonight for sure, im at work now [19:08] JimBastard: i see nodejs_v8 is back online too [19:08] JimBastard: im gonna get a hookio bot as well so you can interact with the hookio api from IRC [19:08] JimBastard: and via IRC triggers [19:10] codeswin_ has joined the channel [19:15] piranha has joined the channel [19:18] mpoz2 has joined the channel [19:22] teemow has joined the channel [19:28] pedrobelo has joined the channel [19:32] mikeal: when is that patch gonna get merged that removes all the http request prints? [19:43] arlolra has joined the channel [19:45] drostie has joined the channel [19:45] mpoz2 has joined the channel [19:46] kriskowal has joined the channel [19:47] codeswin_ has joined the channel [19:51] paul_ has joined the channel [19:55] JimBastard: Douglas Crockford does not respond to email. You can only reach him through a JSONP callback. [19:55] mattly: hah [20:04] dgathright_ has joined the channel [20:05] JimBastard: When Douglas Crockford says "window.open" all the windows in his house do just that [20:06] JimBastard: (im the crockbot until i get home) [20:06] mrjjwright has joined the channel [20:06] jherdman: "Not even Douglas Crockford could find your good parts" [20:08] Tim_Smart has joined the channel [20:17] pdelgallego has joined the channel [20:17] javajunky has joined the channel [20:22] mrjjwright has joined the channel [20:25] QuietKnight has joined the channel [20:32] dnolen has joined the channel [20:34] kriszyp has joined the channel [20:42] JimBastard: morning Tim_Smart [20:43] Tim_Smart: Hello JimBastard [20:46] gwoo: JimBastard: is hook.io like pubsubhubbub? [20:48] charlesjolley has joined the channel [20:49] admc has joined the channel [20:50] mikeal: is anyone else having issues with node master http ? [20:51] gwoo: segfault? [20:51] tlrobinson has joined the channel [20:52] mikeal: HTTPParser.onIncoming (http:606:20) [20:52] mikeal: HTTPParser.onHeadersComplete (http:82:14) [20:52] mikeal: Client.ondata (http:566:12) [20:52] mikeal: IOWatcher.callback (net:320:31) [20:52] mikeal: node.js:815:9 [20:52] gwoo: oh [20:52] mikeal: i'm gonna roll back to before ry did the parser upgrade [20:52] mikeal: and see if it's still a problem [20:57] JimBastard: gwoo: similiar yeah, but not just for Atom and RSS. i plan on making hookio a a pubsubhubbub node as well [20:58] gwoo: nice [20:58] JimBastard: gonna add the pupsubhubh spec for sure [20:59] JimBastard: maybe v3 or v4 [20:59] JimBastard: 0.3 and 0.4 that is [20:59] JimBastard: would like to link up with someone more familar with pubsub [20:59] JimBastard: looking for more contributors in general [21:00] maushu has joined the channel [21:02] mikeal: hrm.... [21:02] mikeal: it's not the http parser checkin that broke it [21:03] gwoo: oh interesting [21:04] jherdman has joined the channel [21:05] gwoo: JimBastard: i hear ya [21:05] mikeal: is the http client in node master expected to be working right now? [21:06] mikeal: or is most everything in flux at the moment [21:06] rtomayko has joined the channel [21:06] rictic has joined the channel [21:06] Tim_Smart: mikeal: Yeah, except everything is a stream now [21:06] mikeal: does that effect the top level API much? [21:06] rtomayko has joined the channel [21:07] JimBastard: gwoo: hopefully we can gain more traction with a stable version being released [21:07] JimBastard: then ill turn hook.io loose on twitter, that should be fun [21:07] gwoo: JimBastard: ha [21:07] Tim_Smart: mikeal: It should, it means all IO has a unified interface [21:08] Tim_Smart: unless http is abstracted more than the others [21:08] mikeal: it is a bit [21:09] charlesjolley has joined the channel [21:09] mikeal: the docs haven't been modified in the master yet [21:10] JimBastard: yeah the docs are out of date for sure [21:10] javajunky: the changes that actually surfaced out of the apis were pretty minimal in the end though which was surprisingly cool ;) [21:11] mikeal: i'm pretty sure that npm works with the latest and it's http code looks unchanged [21:11] gf3 has joined the channel [21:14] JimBastard: Is anyone here involved with ScurvyConf? I'd like register to speak, and try to win that last ticket [21:14] JimBastard: i think deanlandolt mentioned something [21:15] QuietKnight1 has joined the channel [21:16] mattly: JimBastard: talk to voodootikigod [21:16] voodootikigod: what [21:16] voodootikigod: email scurvyconf@jsconf.us [21:16] JimBastard: will do thanks [21:17] voodootikigod: do you have the ticket already? [21:18] stepheneb has joined the channel [21:21] JimBastard: whats up voodootikigod ? i just sent an email [21:21] JimBastard: i dont have a JSConf ticket [21:21] voodootikigod: hmm [21:21] JimBastard: it would be nice to win won, but really i just want to meet some people and get the word out about hook.io [21:21] JimBastard: win one* [21:21] voodootikigod: i think we are requiring jsconf ticket [21:21] JimBastard: waaah? [21:21] voodootikigod: well you can come to the parties [21:22] voodootikigod: http://jsconf.posterous.com/jsconf-scallywag-party-tickets [21:22] voodootikigod: pick up the scallywag ticket [21:22] voodootikigod: get it for friday [21:22] voodootikigod: present [21:22] voodootikigod: if you win [21:22] JimBastard: For ScurvyConf, we be inviting significant others free of charge, but stowaways will have to pay $30.00 to board and they get unlimited food and drink as well. Stowaway tickets will go on sale the week before the event (April 9, 2010). [21:22] QuietKnight has joined the channel [21:22] voodootikigod: we can figure it out from there [21:23] voodootikigod: where do you see that? [21:23] JimBastard: aight ill defintely by the 30 dollar ticket, it was @ http://jsconf.us/2010/scurvy.html [21:23] JimBastard: i was confused, didnt think tickets were on sale for friday night party [21:23] voodootikigod: ahh sorry [21:23] voodootikigod: i will have it updated [21:24] JimBastard: cool cool thanks [21:25] voodootikigod: updated [21:25] hukash has joined the channel [21:26] rnewson has joined the channel [21:28] mikeal: where are the node javascript files installed now? [21:29] mikeal: cause they aren't in /usr/local/lib anymore [21:31] javajunky: they're part of the shipped 'binaries' now aren't they ? [21:32] javajunky: and the only active lib directory is ~/node-libraries iirc [21:32] mikeal: that's annoying [21:32] KungFuHamster: if one of you wonderful programmers could look at a simple script I wrote that expands on the simple tcp server on the node front page... http://pastebin.ca/1850792 ...when using broadcast to send to all connected clients, I drop messages sometimes, even when it's only one client... when just using write() back to the originating client, I don't. [21:33] mikeal: is there a way to run my dev directory [21:34] javajunky: repeated make installs is what I"ve been doing ;) .. or alias node to that folder perhaps ? [21:37] QuietKnight has joined the channel [21:43] charlesjolley has joined the channel [21:48] rnewson_ has joined the channel [21:50] inimino: mikeal: set NODE_PATH [21:51] QuietKnight1 has joined the channel [21:52] JimBastard: hey whats up inimino , did you ever make a github account [21:55] inimino: JimBastard: I have a github account, I just don't have any github projects [21:55] JimBastard: got ya [21:55] JimBastard: we've been hacking away at http://hook.io a bunch, getting close to release [21:55] inimino: cool [21:55] JimBastard: http://github.com/Marak/hook.io [21:57] inimino: looks like it's coming along [21:58] JimBastard: for sure, one day at a time [21:59] inimino: ^.^ [21:59] JimBastard: i'm getting really strange people following the project [21:59] JimBastard: people i dont know, im thinking they are coming in from google [21:59] JimBastard: not anyone related to my github circle [21:59] JimBastard: or this room [22:00] JimBastard: i think pubhubsub support is in the near fiture [22:00] JimBastard: future [22:00] inimino: I vaguely remember that [22:03] silentrob has joined the channel [22:04] sh1mmer has joined the channel [22:06] dgathright has joined the channel [22:06] dnolen has joined the channel [22:06] mikeal: _ry: you around? [22:11] micheil has joined the channel [22:14] dgathright_ has joined the channel [22:16] al3x____ has joined the channel [22:20] Zizoodiesel has joined the channel [22:20] Zizoodiesel: Hi everybody [22:21] Zizoodiesel: 12anyone can help me? [22:21] JimBastard: sop [22:21] charlesjolley has joined the channel [22:22] JimBastard: Zizoodiesel: [22:22] Zizoodiesel: 12i want to ask if node.js can make peer to peer connection for file transfer [22:22] Tim_Smart: Zizoodiesel: Sure, no reason why you can't [22:23] mikeal: so [22:23] Tim_Smart: You will have to implement the protocol yourself, though [22:23] Zizoodiesel: 12over websocket? [22:23] mikeal: is anyone else using concurrent http clients that are reused for keepalive? [22:23] Tim_Smart: websockets for file transfer? That would be interesting [22:23] Zizoodiesel: 12i have windows and dont know how to install it [22:24] technoweenie: the blue font means he's serious [22:24] JimBastard: Zizoodiesel: no windows support atm [22:24] JimBastard: yeah i mean really, whats up with that [22:24] Tim_Smart: haha, that blue font is giving me a headache [22:24] mikeal: it just shows as bold on adium [22:24] JimBastard: i wonder if i still have my rainbow script lol [22:24] Tim_Smart: it's blue and bold [22:24] JimBastard: this is making me remember my mIRC hacking days [22:24] Tim_Smart: VERY SERIOUS [22:25] mattly: man now i want to use a red font [22:25] Zizoodiesel: sorry guys [22:25] JimBastard: lolz [22:25] JimBastard: Zizoodiesel: you can do p2p transfers, it would be epic if someone ported torrent protocol to node.js [22:25] JimBastard: i think that would be amazing [22:26] mikeal: voodootikigod: who is on the commonjs panel? [22:26] pedrobelo has joined the channel [22:26] voodootikigod: ryan, james duncan, kevin dangoor, tom robinson [22:26] mikeal: cool [22:26] Zizoodiesel: i want to make file transfer over the browser in peer to peer but as i saw in node.js site there is a server that i would install [22:26] mikeal: want a couchdb representative :) [22:27] voodootikigod: haha [22:27] voodootikigod: wouldn't jan sort of fill that? [22:27] maushu has joined the channel [22:27] mikeal: i wrote the module support tho [22:27] konobi: james is on the panel? cool [22:27] voodootikigod: haha [22:27] mikeal: and i'll be working the next rev of the modules and packages spec [22:27] voodootikigod: agree [22:27] voodootikigod: but we can't have everyone up there [22:27] mikeal: i know :) [22:27] voodootikigod: so let me go with a maybe [22:28] mikeal: couchdb is an interesting case just because it's way more sandboxed than anyone else [22:28] JimBastard: Zizoodiesel: perhaps you should do more research as to what node.js is and how it works [22:28] JimBastard: either that or work on the engrish [22:28] mikeal: modules can't even do IO to load other modules, the design doc is just one big sandbox [22:29] Zizoodiesel: ok guys you are cool :) [22:29] Zizoodiesel: thanx [22:29] voodootikigod: no Zizoodiesel [22:29] javajunky: its like watching a puppy being kicked, harsh ;) [22:29] voodootikigod: you are cool [22:29] voodootikigod: you are supper cool [22:29] mikeal: haha [22:29] Zizoodiesel: sorry for my bad english i am french [22:30] Zizoodiesel: bonsoir [22:30] Tim_Smart: Guns don't kill people, I kill people, with guns. [22:30] voodootikigod: fuck me [22:31] voodootikigod: i have been using my bare hands [22:31] voodootikigod: i never thought about using the guns [22:31] voodootikigod: its so obvious [22:31] JimBastard: Douglas Crockford can kill people with XmlHttpRequests [22:31] voodootikigod: I heard he did it once with his mustache [22:31] mikeal: oh man this is broken [22:32] mikeal: all my http clients are broken in node master [22:32] Tim_Smart: lol ownt by the node [22:32] voodootikigod: all ur httpclient r belong to us [22:32] mikeal: the rewrite is broken for concurrent keep alive clients [22:32] mikeal: grrrr [22:32] JimBastard: Douglas Crockford doesn't need to long poll, all servers polls him constantly for updates. #crockfordfacts [22:33] JimBastard: poll him* fuck [22:33] mikeal: ok, i get the hint [22:33] JimBastard: keeping up with the releases is hard [22:33] voodootikigod: JimBastard: AHAHHAHAHAHAHA [22:34] voodootikigod: thats what she... [22:34] JimBastard: hee hee [22:34] voodootikigod: outtie [22:34] JimBastard: me too [22:34] crockbot has joined the channel [22:34] JimBastard: peace [22:34] mikeal: welcome back crockbot [22:34] mikeal: haha [22:34] mikeal: breakage on latest node :) [22:42] mjijackson has joined the channel [22:44] unomi has joined the channel [22:46] sh1mmer has joined the channel [22:50] stepheneb has joined the channel [22:51] rictic has joined the channel [22:54] rnewson has left the channel [23:02] charlesjolley has joined the channel [23:08] JimBastard has joined the channel [23:13] JimBastard: VOLCANO TIME [23:14] _ry: build bots look in bad shape [23:17] blakemizerany has joined the channel [23:24] mikeal has joined the channel [23:29] isaacs: Tim_Smart: i pushed some changes to help your weary eyes. [23:29] Tim_Smart: haha cheers [23:31] isaacs: also, I stilly haven't heard from herby at all. I sent him a github message, hopefully he's still interested. [23:31] isaacs: he did a lot of interesting explorations in his node fork, so i really want to get his feedback [23:52] mikeal: isaacs: you around? [23:52] isaacs: yo [23:53] mikeal: i'd like to get your comments on my latest email to commonjs [23:53] mikeal: especially since i'm using npm as a use case :) [23:53] mikeal: it's on the packages 1.0 comments thread [23:54] _ry: mikeal: hey you pinged me? [23:55] mikeal: i have this strange problem on master [23:55] mikeal: i've got a bunch of http clients, that i'm reusing for keepalive [23:56] mikeal: and currentRequest.emit("response", resp) [23:56] mikeal: is throwing undefined errors [23:56] CIA-77: node: 03isaacs 07master * r6e603cc 10/ lib/http.js : Fix debug statement that was using puts instead of debug() - http://bit.ly/dlt7lB [23:56] CIA-77: node: 03Ryan Dahl 07master * rb35d72d 10/ (96 files in 13 dirs): Upgrade V8 to 2.1.7 - http://bit.ly/bPK3Jm [23:57] mikeal: i did some testing and it seems like my client is being reused outside of my own reuse, if that makes sense [23:58] softdrink: node + opengl + bullet ... please? :) [23:59] RayMorgan_ has joined the channel