[7:53]<ondras> kris kowal here? [17:41]<deanlandolt> Wes: hey, how goes it? [17:41]<Wes> deanlandolt: Awesome! [17:41]<Wes> deanlandolt: Two things I wanted to ask you.. One I've forgotten [17:41]<deanlandolt> your wrist still works? [17:41]<Wes> The other is "do you have a file full of ES5 equivs?" [17:42]<deanlandolt> yes...it's from narwhal [17:42]<deanlandolt> ..coming up... [17:42]<Wes> deanlandolt: Yeah, about 95% today. Still working on it, physio guy has been hurting me but it's working [17:42]<deanlandolt> https://github.com/kriskowal/narwhal-lib/blob/master/lib/narwhal/global-es5.js [17:42]<Wes> deanlandolt: Awesome [17:42]<deanlandolt> nice...good to hear [17:42]<Wes> deanlandolt: BTW, did oyu know that labelled dependencies make a nice "mappings"? [17:43]<deanlandolt> hadn't thought of that...i think of mappings more in a package context [17:43]<deanlandolt> but that's an interesting point [17:43]<Wes> module.declare([ { a: "/www.page.ca/~wes/big/long/path/to/module/a" } ], f(rem) { module.deps('a').hello(): ) [17:43]<deanlandolt> one thing that link from narwhal doesn't have is the global JSON, which should obviously also be available in an es5 context [17:44]<Wes> deanlandolt: Additionally somebody (stefaan?) proposed an interesting alternative to module.deps today -- require [17:44]<Wes> so [17:44]<Wes> module.declare([ { a: "/www.page.ca/~wes/big/long/path/to/module/a" } ], f(rem) { require('a').hello(): ) [17:44]<Wes> module.declare([ { a: "/www.page.ca/~wes/big/long/path/to/module/a" } ], f(rem) { require('a').hello(); } [17:44]<deanlandolt> i'm, i'll have to catch up...i've been heads down for a few days...i've got a big ML backlog [17:44]<Wes> deanlandolt: Entertainingly, the spec doesn't mention that in the ES5 section either :D [17:46]<Wes> you program in ML?! ;) [17:46]<deanlandolt> heh [17:48]<Wes> deanlandolt: FWIW - I've got a working script-tag, module.declare loader that handles 404s properly [17:48]<Wes> And passes the modules/1.0 test suite [17:48]<deanlandolt> already? that's awesome [17:48]<Wes> yeah. working on labelled deps today [17:55]<Wes> deanlandolt: module.declare([ {a: "a", b: "b"}, "c" ] [17:55]<Wes> deanlandolt: What is module.dependencies[0] + module.dependencies[1] ? [17:56]<deanlandolt> i remember asking that when we first talked about it...i still don't ahve an answer :) [17:56]<Wes> "ab", "ac", "cundefined" [17:56]<deanlandolt> module.deps("1") -> "c" [17:56]<Wes> deanlandolt: What do you think of the latter case? Make module.deps depends not show in module.dependencies at all? [17:56]<deanlandolt> module.deps("a") -> "a" [17:57]<deanlandolt> why can't we have a place where the full dep array exists? i thoguht that was what module.dependencies was... [17:58]<Wes> deanlandolt: Well, yeah, we could just copy the full array including the labels -- I suppose that's not too annoying to use [17:58]<Wes> because the guy uysing it will be th eguy writing it [17:58]<deanlandolt> if you want the mapped path use require, right? [17:58]<deanlandolt> then there's no need for module.deps [17:58]<Wes> deanlandolt: Exactly - I think that's a marvelous idea [17:58]<deanlandolt> was that what you were saying earlier? [17:59]<Wes> yep [17:59]<deanlandolt> ah, well in that case, i get what you mean now :) [17:59]<deanlandolt> and i agree [17:59]<Wes> it actually solves some of the same problem space as require.paths [17:59]<Wes> which is good, beceause r.paths does not work in the browser [17:59]<Wes> (for reasonable values of work) [18:04]<deanlandolt> Wes: do you know if e4x is slated to be removed from SM? [18:04]<Wes> deanlandolt: there is no schedule AFAIK, but lots of hate from the SM devs [18:04]<deanlandolt> thanks [18:05]<Wes> stefaan's invention has another bonus [18:06]<Wes> I don't need a 'module.require' in my implementation [18:06]<Wes> although I need a way to tie the module object to the require instance [18:06]<Wes> that's not so bad [18:06]<Wes> actually, I don't [18:06]<Wes> I only need to pass the deps array into the require factory [18:06]<Wes> OMG! [18:06]* Wes dances [18:07]<deanlandolt> hmm, so less crap...nice