00:00:00  * ircretaryquit (Remote host closed the connection)
00:00:07  * ircretaryjoined
00:06:23  * ITproquit (Ping timeout: 245 seconds)
00:07:05  * ITprojoined
00:11:04  * jez0990_joined
00:11:04  * jdiezjoined
00:35:47  * tomshredsjoined
00:43:22  * mikealjoined
00:44:18  * mikealquit (Client Quit)
00:44:46  <Raynos>adding print statements to node core and running the tests ruins your computer
00:51:06  * ralphtheninjaquit (Ping timeout: 276 seconds)
00:52:20  <Raynos>Ok I managed to fix most of the regressions
00:57:06  * mikealjoined
00:58:28  * ITproquit (Ping timeout: 245 seconds)
00:59:03  * ITprojoined
01:03:22  <isaacs>Raynos: dest.on('finish', unpipe) would be silly
01:03:29  <isaacs>Raynos: 'finish' can only come after end() is called
01:03:40  * mikealquit (Quit: Leaving.)
01:03:49  <Raynos>isaacs: but you have examples in your test where you call end manually outside the pipe cycle
01:04:11  <isaacs>Raynos: yeah, but that's a rare stupid edge case
01:04:22  <Raynos>then that's fine.
01:04:51  <Raynos>there are some other questions
01:05:24  * tomshredsquit (Quit: Linkinus - http://linkinus.com)
01:05:26  <Raynos>this guy ( https://gist.github.com/180ed4fa7ba12a521ec4 )
01:06:19  <Raynos>also I don't like pipe anymore it's a horrible state machine with lots of undefined behaviour on edge cases :P
01:10:06  * yorickquit (Remote host closed the connection)
01:10:50  <chrisdickinson>mbalho: so, i'm working on a collision api for 2d tilemaps, but it should be portable to voxeljs: https://gist.github.com/8d591cbd28c93511783e
01:12:32  <chrisdickinson>player.box is an aabb (a la https://github.com/chrisdickinson/aabb-2d), it's up to the client to change the vec in-flight in the callback and return whether or not further tiles should be checked
01:12:47  <chrisdickinson>(basically "return true" halts the check along the axis.)
01:27:26  * jibayquit (Remote host closed the connection)
01:32:14  * ralphtheninjajoined
01:35:09  <Raynos>objects for streams2 -> https://github.com/joyent/node/pull/4581
01:35:16  <Raynos>Let me know if there are any other edge cases I should test
01:42:39  * mikealjoined
01:46:50  * mikealquit (Client Quit)
01:53:13  <mbalho>chrisdickinson: thats awesome
01:53:13  <Raynos>isaacs: streams2 semantics question
01:53:32  <Raynos>with a high water mark. it only means "only call _read if less then hwm"
01:53:46  <Raynos>it doesn't mean "call _read until we are at least at the hwm"
01:54:04  <Raynos>so with the pull stream model (_read) we don't aggressively pull on a source to fill up our entire watermarked buffer
01:54:47  <Raynos>Oh wait that's the lwm
01:54:55  <Raynos>the lwm causes aggressive pulling
02:04:24  * mikealjoined
02:17:34  * defunctzombie_zzchanged nick to defunctzombie
02:35:31  * thatguydanjoined
02:41:42  <gozala>Raynos: could you please post your feedback on https://github.com/Gozala/reducers/issues/35
02:44:33  <mbalho>substack: just got to urban blend, gonna be here till 10 when my friends arrive on an amtrak train
02:48:29  * h0kejoined
02:49:29  * h0kepart
02:51:47  <substack>mbalho: ok I might stop by once I warm up
02:54:18  <Raynos>gozala: vague feedback posted
02:56:44  <gozala>Raynos: txn
02:56:45  <gozala>tnx
02:58:08  <gozala>Raynos: I don't think you got the whole point of it
02:58:21  <Raynos>i have no need for finalize that returns results
02:58:31  <Raynos>if finalize returns an error then your fucked :p
02:58:37  <gozala>In fact it's something you have being asking for several times
02:59:04  <Raynos>ive been asking for "notify me on end / finish" not for allow me to do stuff on "end / finish" and then push values / errors back into the thing that just ended
02:59:06  <gozala>Raynos: it just returns end or error
02:59:10  <gozala>no data
02:59:51  <gozala>Well it wan't be same thing
02:59:57  <Raynos>well there are two use cases
03:00:02  <gozala>function will return new value
03:00:04  <Raynos>I have a reducible, run this function when it's closed.
03:00:11  <gozala>that is original + do my thing
03:00:28  <Raynos>I am creating a reducible, ensure this function runs before it's closed
03:01:12  <gozala>Yes if thing you do is sync that's fine
03:01:40  <gozala>but if it's async then you need to direct error somewhere if there is an error
03:02:05  <Raynos>that depends
03:02:07  <gozala>in a way attempt to close input may cause error
03:02:16  <Raynos>on whether your finalizing something as a folder or transformer
03:02:38  <Raynos>for the former use-case, as a folder notify me when this reducible is closed I can handle error myself
03:02:43  <Raynos>and that's what I want
03:02:45  <Raynos>for the latter I dont know
03:04:03  <gozala>Raynos: do you know what happens if you interrupt ReadStream from fs ?
03:04:19  <Raynos>how do you interrupt it
03:04:23  <Raynos>close it early?
03:04:23  <gozala>I suspect it closes fd and if it fails to do so it emits error on readStream
03:04:30  <gozala>yes
03:04:32  <Raynos>yes
03:04:38  <Raynos>but the contract there is
03:04:44  <Raynos>close() -> emit("error") OR emit("close")
03:05:02  <gozala>so this is identical
03:05:11  <Raynos>so maybe you want EOF, CLOSED, ERRORED states
03:05:20  <gozala>finalize can only give end or error
03:05:22  <gozala>nothing else
03:05:34  <Raynos>so that means error and end is allowed after consumed
03:05:50  <gozala>yes
03:06:05  <gozala>well that's a topic of discussion I mean
03:06:43  <gozala>Raynos: actually not even that
03:07:05  <gozala>return reduced(value) is close
03:07:24  <gozala>so close ends an input with `end` or `error`
03:07:28  <Raynos>what happens if you try to close and ended thing
03:07:43  <gozala>depending weather closing it down was successful
03:07:44  <Raynos>input -> end, consumer <- reduced(value)
03:08:03  <Raynos>i guess you can just ignore that because end happens after close
03:08:03  <gozala>ended input won't call your reducer so you won't be able to do that
03:08:23  <gozala>Raynos: return value on end and error does not matters
03:08:48  <gozala>it's currently is ignored
03:09:23  <gozala>so question really is should close require confirmation from input
03:09:25  <gozala>or not
03:10:05  <gozala>by confirmation I mean `end` or error
03:10:25  <gozala>and error is actually end with indication for problem
03:10:46  * dguttmanquit (Quit: dguttman)
03:24:11  * mikealquit (Quit: Leaving.)
03:42:13  <jjjjohnnny>lol https://npmjs.org/package/softmotions-jazz
03:48:30  * st_lukejoined
03:50:04  <Raynos>gozala: we need to send end / error after reduced
03:54:29  * mikealjoined
04:04:09  * mikealquit (Ping timeout: 265 seconds)
04:09:28  * defunctzombiechanged nick to defunctzombie_zz
04:30:39  * defunctzombie_zzchanged nick to defunctzombie
04:30:45  * mikealjoined
06:51:08  * st_lukequit (Remote host closed the connection)
07:01:53  * st_lukejoined
07:11:01  * mikealquit (Read error: Connection reset by peer)
07:15:56  * mikealjoined
07:23:41  * AvianFluquit
07:33:14  * thatguydanquit (Remote host closed the connection)
07:33:28  * thatguydanjoined
08:00:35  * defunctzombiechanged nick to defunctzombie_zz
08:19:52  * captain_morganjoined
08:41:24  <fotoverite>wow everyone is so quiet. Nothing happening in node at all!
08:43:11  * substackis whipping up a hairy ui module
08:43:21  <substack>factoring out some pricing code so I can reuse it for the testling-ci profesh
08:43:29  <substack>and for team plans
08:43:38  <fotoverite>very nice
08:43:50  <fotoverite>I'm trying to adapt what I learned about callbacks and flow control to threads.
08:44:02  <fotoverite>Bring the wisdom of node at a least a little bit to ruby.
08:44:26  <substack>oh right I can show what I have even http://browserling.com:9005/
08:44:45  <substack>push state on an embedded widget woo
08:45:34  <fotoverite>I like the look
08:45:44  <fotoverite>I think the whole N thing might go over too many peoples head
08:45:50  <fotoverite>Yes I know… :(
08:46:00  <substack>our customers are highly technical folk
08:46:10  <substack>not too worried about that
08:46:13  <fotoverite>Not the ones buying in bulk not always.
08:46:30  <fotoverite>And better messaging is never a bad thing
08:46:49  <substack>I'll put a (number of developers) parenthetical to it
08:48:26  <substack>done
08:53:01  <Raynos>substack: that n + 1 thing should be like a draggable scale
08:53:11  <fotoverite>O good idea.
08:53:27  <Raynos>you can build a nice slider widget
08:53:31  <Raynos>with <input> and <output>
08:53:37  <Raynos>and make it REAL TIME AND STREAMING
08:53:53  <fotoverite>Ummmmm why real time or streaming
08:53:57  <fotoverite>it's just client side
08:54:06  <fotoverite>Are the prices changing every n seconds?
08:54:07  <Raynos>because buzzwords
08:54:12  <fotoverite>BUZZWORDS!
08:54:35  <Raynos>well you want data to stream from <input> {{ n }} to <output> {{ price }}
09:14:34  <Raynos>> https://npmjs.org/~grimen
09:14:52  <Raynos>That grimen guy just decided "this week I will build an ORM for EVERY DATABASE EVER"
09:16:13  <rvagg>Raynos or anyone else: do you have a go-to module for an Object#forEach style iterator? something that sends key/value pairs to a callback would be nice
09:16:41  <Raynos>rvagg: https://github.com/Raynos/for-each
09:16:46  <Raynos>But fuck that shit
09:16:58  <rvagg>Object.keys().forEach() gets old pretty quick
09:17:00  <Raynos>Just do Object.keys(obj).forEach(function (key) { var value = obj[key]; ... })
09:17:13  <substack>yes
09:17:15  <substack>just do that
09:17:16  <rvagg>Raynos: yeah, my code is littered with that, it's just too verbose
09:17:26  <substack>it's not actually especially verbose
09:17:35  <substack>rvagg: maybe factor some of those out into maps, filters, and folds?
09:18:00  <Raynos>or ragequit forever
09:18:02  <Raynos>that works too
09:18:03  <rvagg>yeah, but you still need to start with Object.keys() cause we're missing iterators on non-arrays
09:18:19  <rvagg>this is the main reason lodash & underscore are so popular
09:18:23  <substack>meh
09:18:24  <Raynos>no
09:18:28  <Raynos>they are popular because ES3
09:18:29  <rvagg>but they're too bloated
09:18:34  <Raynos>but if you want it
09:18:36  <Raynos>Raynos/for-each
09:18:43  <Raynos>single function export, exactly what you want
09:18:51  <rvagg>they're popular in Node, !ES3
09:18:59  <rvagg>yeah, thanks Raynos, that's what I want
09:19:27  <Raynos>rvagg: they are popular in Node because "lol how i npm"
09:19:56  <rvagg>perhaps, also perhaps because npm discovery still sucks
09:20:17  <Raynos>I also have [reduce](https://github.com/Raynos/reduce) and [map](https://github.com/Raynos/map-sync)
09:20:28  <substack>rvagg: this project makes me sad https://github.com/thomaslanciaux/Slider
09:20:28  <Raynos>Nobody needs filter.
09:20:50  <substack>dependency without putting it in the package.json dependencies
09:21:42  <rvagg>substack: yeah well, that's not how you're spozed to do it for Ender so he's at fault
09:21:46  <rvagg>open a PR if you're so sad
09:22:16  <Raynos>better just rewrite the entire thing
09:22:23  <substack>yeah probs :(
09:22:24  <Raynos>rvagg: Also https://github.com/Gozala/extract
09:22:58  <Raynos>transform hash into array of key values and extract to destruct tuples into multiple arguments
09:29:00  <rvagg>hmm.. not sure I see a whole of utility in that beyond a standard reduce()
09:29:29  <rvagg>the second example is interesting but only because of the keyValues() function
10:06:07  <Raynos>rvagg: https://github.com/Gozala/oops/blob/master/key-values.js
10:07:13  <Raynos>rvagg: combined with the notion that "every data structure can be represented by a reduce function" ( https://github.com/Gozala/reducers#reducers )
10:18:16  * dominictarrjoined
10:49:30  * jibayjoined
11:15:30  * antix_quit (Quit: Lost terminal)
11:29:14  * st_lukequit (Remote host closed the connection)
12:03:39  * captain_morganquit (Remote host closed the connection)
12:14:15  * thatguydan_joined
12:18:09  * thatguydanquit (Ping timeout: 276 seconds)
12:18:10  * thatguydan_changed nick to thatguydan
12:19:55  * thatguydanquit (Remote host closed the connection)
12:20:07  * thatguydanjoined
12:51:03  * No9joined
12:56:26  * thatguydanquit (Quit: thatguydan)
12:57:26  <substack>Raynos: ok done http://browserling.com:9006/
12:57:35  <substack>09:22:17 < Raynos> better just rewrite the entire thing
12:57:50  <substack>except this is all from scratch because whatevs
12:59:36  <dominictarr>It's a slider widget?
13:00:16  <substack>yep
13:01:34  <dominictarr>good. we need more widgets
13:02:58  <dominictarr>is the code up yet?
13:10:06  <substack>writing the docs right now
13:16:16  <substack>https://github.com/substack/slideways
13:21:53  <ralphtheninja>substack: nice one
13:22:24  <ralphtheninja>easy to customize?
13:25:44  <substack>should be
13:25:51  <substack>just do your own css for it
13:26:27  <substack>it checks computed styles for the widths so it should be robust against different sizes
13:32:54  <dominictarr>substack: can you document the .element property?
13:34:08  <substack>why?
13:36:50  <substack>fotoverite: now using a slideways slider http://browserling.com:9005/pricing/team
13:40:43  <dominictarr>substack: I just find that passing elements, keeping them in scope, etc more natural than using selectors, etc
13:41:18  <substack>appendTo() takes an element or a selector
13:42:03  <dominictarr>sure, but I might want to stick it in the middle, or the start
13:42:26  <substack>ah true
13:42:34  <dominictarr>substack: you should make the value slider go down to 1, but start on two
13:42:51  <dominictarr>it feels like a bug when it only goes to 2
13:43:07  <dominictarr>although it makes sense, you have to think about it.
13:43:31  <dominictarr>looks great though!
13:49:33  * jez0990-quit (Quit: EliteBNC - http://www.elitebnc.net/)
13:52:54  * mirkokjoined
13:55:02  * jez0990-joined
13:57:32  * mirkokquit (Client Quit)
13:57:41  * mirkokjoined
13:59:55  * mirkokquit (Client Quit)
14:00:52  * yorickjoined
14:00:52  * yorickquit (Changing host)
14:00:52  * yorickjoined
14:44:17  * mikolajoined
14:45:35  * mikolapart
14:56:03  * mirkokjoined
14:56:56  * jez0990-quit (Quit: EliteBNC - http://www.elitebnc.net/)
15:01:34  * jez0990-joined
15:11:43  <substack>dominictarr: agreed, I folded that into the developer plan http://browserling.com:9005/pricing/developer
15:11:51  <substack>now it goes from 1-9
15:18:21  * defunctzombie_zzchanged nick to defunctzombie
15:27:44  <defunctzombie>substack: we need a better way to package css and html with the components
15:28:22  <defunctzombie>substack: I have been thinking about some possible approaches in that space that still make use of package.json and can be integrated easier
15:29:09  * devaholicjoined
15:35:08  <substack>defunctzombie: I'm not so sure anymore.
15:35:18  <substack>throwing css and html into js files is pretty easy
15:35:30  <defunctzombie>no way, all that useless quoting and commas
15:35:37  <defunctzombie>that just isn't gonna happen
15:35:44  <substack>you can write a simple script for it if you care about such things
15:35:46  <defunctzombie>I mean you did it that way cause you could just ship the js
15:35:48  <defunctzombie>I get that
15:35:51  <substack>it makes reuse so much easier
15:35:54  <defunctzombie>but it isn't very practical
15:36:05  <defunctzombie>I know we can do better without much fuss
15:36:17  <substack>it's practical in that module authors build this stuff once and then can forget about it
15:36:22  <defunctzombie>I don't want html.js .. I want html.html and style.css
15:36:23  <substack>it's not for app development
15:36:37  <defunctzombie>you can't forget about it until it works
15:36:45  <defunctzombie>and while you develop it, it is a pita
15:36:50  <dominictarr>substack: defunctzombie so you just have a precomple step
15:37:02  <substack>use something like exportify to compile those files to .js
15:37:07  <substack>yes exactly
15:37:12  <defunctzombie>dominictarr: meh, I am usually not fond of those but yes, you can do that
15:37:21  <defunctzombie>I think that is fine
15:37:36  <defunctzombie>just need to add style: and template: or something to package.json maybe
15:37:48  <substack>that sounds horrible
15:37:51  <defunctzombie>why?
15:37:53  <dominictarr>if you are talking about doing this for browserify at the end of the day that is what is gonna happen anyway
15:38:09  <defunctzombie>the tool doesn't have to be browserify
15:38:10  <dominictarr>what about a prebundle script?
15:38:16  <defunctzombie>but why is package.json horrible?
15:38:27  <defunctzombie>you are making a widget
15:38:35  <defunctzombie>describe the various components about it as metadata
15:38:39  <defunctzombie>there is nothing horrible about that
15:38:56  <dominictarr>that could be used to run the exportifiers at bundle time
15:39:00  <defunctzombie>dominictarr: that is one approach, but apparently scripts are going away eventually in npm?
15:39:12  <defunctzombie>dominictarr: I personally prefer more seemless solutions
15:39:28  <dominictarr>because they want you to use gyp
15:39:33  <defunctzombie>all my js currently autobundles
15:39:46  <dominictarr>this is a different case
15:39:50  <defunctzombie>I don't like manual work
15:39:53  <defunctzombie>:)
15:40:01  <dominictarr>in js there is compile to js langs, etc
15:40:19  <defunctzombie>?
15:40:48  <defunctzombie>I don't mind building the html or css into js files, but they should not be written as such
15:41:00  <dominictarr>defunctzombie: of course
15:41:03  <defunctzombie>they are not js files and it makes for terrible editing
15:41:09  <dominictarr>I don't think substack is advocating that...
15:41:18  <defunctzombie>I think he was for a moment ;)
15:41:40  <dominictarr>well, for very small html files it's probably okay
15:42:00  <defunctzombie>the way component did it is a bit better imho by having metadata in component.json, I just think package.json is perfectly fine for that
15:42:13  <substack>sure I am
15:42:40  <dominictarr>the only html I write is <!DOCTYPE html><html><body></body><script src=/bundle.js></script></html>
15:42:58  <defunctzombie>dominictarr: I write a lot more html for templating
15:43:04  <defunctzombie>dominictarr: and page layouts, etc
15:43:26  <defunctzombie>and when you do that, writing it as regular html is much more natural
15:43:39  <dominictarr>I'm really not sure about this whole "templating" thing at all.
15:43:55  <substack>dominictarr: same
15:44:03  <dominictarr>there is nothing _natural_ about html
15:44:08  <defunctzombie>dominictarr: don't think about templating using {{ }} stuff
15:44:13  <defunctzombie>just write html as semantic markup
15:44:21  <defunctzombie>about what you want to render
15:44:24  <defunctzombie>and style it with css
15:44:28  <defunctzombie>I am sure about the html stuff
15:44:39  <defunctzombie>it is very readible
15:44:40  <substack>I just use querySelectors to update html elements
15:44:43  <defunctzombie>and very straightforward
15:45:02  <substack>and then this module makes bulk querySelector updates somewhat less verbose https://github.com/substack/hyperglue
15:45:03  <defunctzombie>I hate writing boilerplate html as javascript strings in my js
15:45:10  <defunctzombie>it is far less maintainable
15:45:16  <dominictarr>I'm just keeping references to elements i create, at the moment...
15:45:22  <defunctzombie>again
15:45:27  <defunctzombie>your html there is trivial
15:45:39  <substack>only write trivial html
15:45:41  <defunctzombie>and not real world larger project html imho
15:45:44  <defunctzombie>:/
15:46:00  <defunctzombie>anyhow, I write html to describe a page
15:46:09  <substack>"argument from big apps" should be a formal fallacy
15:46:12  <defunctzombie>I have yet to find a better way
15:46:14  <substack>your argument notwithstanding
15:46:19  <defunctzombie>substack: it has nothing to do with big apps
15:46:20  <substack>big apps === failing at modularity
15:46:37  <defunctzombie>it has to do with divs and other elements just to get page basics going
15:46:42  <defunctzombie>and I will not write those in js
15:46:47  <defunctzombie>that *is* just stupid
15:47:03  <defunctzombie>and html works great as basic markup
15:47:12  <defunctzombie>I know it, I can read it, and it is easy to write
15:48:03  <defunctzombie>if I want tabs I can easily do <tabs><tab></tab></tabs> and this works as html too
15:48:55  <defunctzombie>but more importantly, with your js css and html file I have thrown out all syntax highlighting as well
15:49:30  <dominictarr>one thing that I don't like about templating is having to put the view, and code that is tightly coupled to it in separate files.
15:49:53  <dominictarr>that is the one thing I like about coffee script, multiline strings.
15:50:00  <defunctzombie>dominictarr: this happens with anything that has UI vs control code
15:50:19  <defunctzombie>I will not write html in js, it is just nonsensical to do so
15:50:32  <substack>I just want to put all this shit in black boxes so I don't need to think about any of it ever again.
15:50:43  <substack>I don't even care how these work so long as the interface is nice.
15:50:43  <defunctzombie>look at Qt .ui files as an example
15:51:01  <defunctzombie>substack: sure, except when they break and to write them in the first place
15:51:14  <dominictarr>defunctzombie: sure, that doesn't mean it's a good pattern though.
15:51:32  <dominictarr>currently, i'm using a dsl thing to write html in js
15:51:46  <defunctzombie>dominictarr: why?
15:51:50  <defunctzombie>there is no advantage
15:51:55  <substack>defunctzombie: when the black boxes are tiny enough they're usually pretty easy to fix
15:52:05  <substack>no matter how they're written really
15:52:06  <dominictarr>h('a', {href: '#'}, 'linktext')
15:52:13  <defunctzombie>not true really
15:52:20  <dominictarr>returns a HTMLElement
15:52:36  <substack>at a certain degree of narrowness in scope everything pretty much is fine
15:53:01  <defunctzombie>dominictarr: I don't see the point of this as you really want to be writing html but have now invented some random syntax to make it even more involved
15:53:02  <dominictarr>defunctzombie: there is less indirection, because I can see the html and the code that uses it at the same time.
15:53:20  <dominictarr>defunctzombie: it moves it closer to js.
15:53:22  <defunctzombie>split your windows?
15:53:34  <defunctzombie>if you scroll away from the html you no longer see it anyway haha
15:53:42  <dominictarr>also, I can easily grab references to things
15:53:53  <dominictarr>so I don't need to give things ids
15:54:04  <defunctzombie>dominictarr: you can easily grab references after the fact too
15:54:08  <defunctzombie>many times without ids
15:54:33  <dominictarr>nothing is easier than var thing = createThing()
15:55:17  <dominictarr>anyway, I don't need to persuade you to do it my way.
15:55:25  <defunctzombie>right
15:55:35  <defunctzombie>you won't ;) I looked at dsl html
15:55:41  <defunctzombie>like jsonml and stuff
15:55:43  <defunctzombie>it is all crap
15:56:41  <dominictarr>well, somepeople like to park their bikes in blue bike-sheds, some red...
15:57:31  <substack>this is why we have npm
15:57:46  <substack>so you can just require() modules and not at all care how they're written
15:57:50  <defunctzombie>dominictarr: it is really annoying whenever people just end arguments with that crap honestly
15:57:55  <substack>without even looking at the code in many cases
15:58:59  <dominictarr>defunctzombie: if this was something more important I wouldn't have
16:00:02  <dominictarr>It's not anything personal, or anything againts html, if you wanna write html, it's your app.
16:00:43  <dominictarr>but it's so hard to come to agreements, best to reserve your energy for when it's actually important.
16:01:02  <defunctzombie>dominictarr: I am not trying to tell you to write html or not use your dsl
16:01:09  <dominictarr>okay
16:01:12  <defunctzombie>I am just trying to tell you that when other people see it
16:01:15  <defunctzombie>it will not be familiar
16:01:32  <dominictarr>true
16:01:37  <defunctzombie>you should always do what works for you
16:02:04  <defunctzombie>but from a UI standpoint my *personal* opinion on it is that it is not a viable approach (same with the .js css and html stuff)
16:02:04  <dominictarr>prehaps I should use jquery instead :)
16:02:33  <defunctzombie>development/syntax and bikeshedding is all UI/UX for a coder
16:02:53  <defunctzombie>we care about it in the real world for physical products
16:03:05  <defunctzombie>why wouldn't we in the products we consume (code and libraries)
16:03:49  <dominictarr>right, but there are different users here
16:04:24  <mirkok>defunctzombie: having a html dsl in your programming language can go a long way
16:04:29  <dominictarr>and this thing of doing modular html *widgets* is not well solved by any approach so har
16:04:34  <mirkok>look at what seaside is able to achieve with this: http://seaside.st/documentation/callbacks
16:04:37  <dominictarr>mirkok: hey!
16:04:44  <mirkok>hey dominictarr
16:05:17  <dominictarr>agree, seaside was the first framework I saw that I felt actually made sense from a high level standpoint
16:05:36  <mirkok>yes I haven't seen another framework that solved the modularity problem in such a clean way
16:05:57  <defunctzombie>mirkok: this is just another form of X template language
16:06:12  <mirkok>no its not at all
16:06:18  <mirkok>it goes way beyond that
16:06:24  <dominictarr>defunctzombie: it's way more than that
16:06:25  <mirkok>it actually abstracts the client-server barrier
16:06:40  <defunctzombie>like opa or whatever?
16:06:42  <dominictarr>it's not a RESTful framework
16:06:56  <dominictarr>it's the anti-rails
16:07:04  <mirkok>yes its not RESTful at all so its certainly not suitable for any kind of app
16:07:15  <mirkok>but its super productive :)
16:08:02  <mirkok>you really describe your app in a way as if you were coding a desktop app
16:11:38  <mirkok>I actually believe that its possible to use seaside's concepts to generate RESTful apps out of their kind of declarative code
16:15:04  * yorickquit (Remote host closed the connection)
16:22:58  * gozalaquit (Remote host closed the connection)
16:26:55  * AvianFlujoined
16:31:08  * dguttmanjoined
16:32:16  * gozalajoined
16:46:03  * dominictarrquit (Quit: dominictarr)
16:58:26  * ITproquit (Ping timeout: 244 seconds)
17:03:30  * dguttmanquit (Quit: dguttman)
17:04:48  * st_lukejoined
17:11:32  * dguttmanjoined
17:16:19  * dguttmanquit (Client Quit)
17:18:05  * dguttmanjoined
17:26:40  * ITprojoined
17:28:12  * dguttmanquit (Quit: dguttman)
17:31:15  * ITproquit (Ping timeout: 260 seconds)
17:33:19  * mikealquit (Quit: Leaving.)
17:55:27  * mikealjoined
18:01:10  * mikealquit (Quit: Leaving.)
18:29:16  * mikealjoined
18:33:23  * mirkok_joined
18:34:08  * mirkokquit (Ping timeout: 252 seconds)
18:34:09  * mirkok_changed nick to mirkok
18:35:49  * dguttmanjoined
18:43:58  * st_lukequit (Remote host closed the connection)
19:08:23  * chadskidmore_joined
19:09:01  * tilgovijoined
19:09:28  * chadskidmorequit (*.net *.split)
19:09:30  * chadskidmore_changed nick to chadskidmore
19:10:42  <Raynos>dominictarr, defunctzombie: modular html widgets are trivially solved with multiline strings, just wait a few years
19:12:21  * mikealquit (Quit: Leaving.)
19:35:50  * defunctzombiechanged nick to defunctzombie_zz
19:42:36  * mikealjoined
19:54:45  * tilgoviquit (Ping timeout: 276 seconds)
20:22:27  * nk109quit (Quit: Computer has gone to sleep.)
20:23:50  * captain_morganjoined
20:27:20  * ITprojoined
20:32:35  * nk109joined
20:35:35  * nk109quit (Client Quit)
21:04:41  <Raynos>http://www.hackfind.com/#13.00/37.7294/-122.3908
21:04:57  <Raynos>does anyone get that
21:06:44  * thatguydanjoined
21:22:00  * mirkokquit (Quit: mirkok)
21:25:16  * dominictarrjoined
21:31:13  <fotoverite>Why do I get the feeling I saw an argument that would have inspired this. http://christianheilmann.com/2013/01/05/a-total-web-recall/
21:31:58  <fotoverite>Also seriously dominictarr I feel that the web needs more then just a body to function correctly most of the time. DSL are great but you need a templating language also.
21:33:35  * nk109joined
21:34:57  <Raynos>fotoverite: https://gist.github.com/25cadf7725af6105e77a
21:44:27  <dominictarr>Raynos: looks interesting, but I can't understand how it works
21:45:04  <dominictarr>oh, the content gets inserted into {{text}}
21:45:31  <dominictarr>Raynos: have you looked at shadow-dom?
21:46:46  <dominictarr>Raynos: also don't make require add to the scope invisibly. we hate that.
21:47:12  <fotoverite>shadowdom only works in chrome
21:54:52  <Raynos>dominictarr: But then you would have to export all those macros
21:55:06  <Raynos>dominictarr: shadow dom is verbose and not a compile time macro
21:55:58  <Raynos>dominictarr: I should rename require to include / import
21:56:07  <dominictarr>sure
21:56:25  <Raynos>dominictarr: btw apparently <script src="bundle.js"></script> is a valid index.html
21:57:15  <dominictarr>but does it need $(ready)
21:57:33  <dominictarr>you persuaded me to just put the js at the bottom
21:57:47  <Raynos>i dont know
21:57:55  <Raynos>whether that puts the script in the head or body
21:58:16  <dominictarr>head probably
21:59:18  <Raynos>yeah its puts in the head
21:59:25  <Raynos>blargh
21:59:40  <Raynos>then just do <body><script ... ></script></body> :D
22:12:21  * tilgovijoined
22:17:33  <dominictarr>you also need the doctype so some querks work right
22:21:00  * st_lukejoined
22:47:21  * dominictarrquit (Quit: dominictarr)
23:31:09  <mbalho>substack: AHOY STACKMAN I AM DOWNTOWN WITH MY ODDLY NAMED FRIENDS FROM PORTLAND NIM AND FRITZ
23:35:06  * gwenbelljoined
23:44:04  * xerverjoined
23:46:09  * rolnaabaquit (Ping timeout: 248 seconds)
23:46:09  * xerverchanged nick to rolnaaba