00:00:12  * jxsonquit (Ping timeout: 240 seconds)
00:00:12  * ircretaryquit (Remote host closed the connection)
00:00:13  * ircretaryjoined
00:00:29  * thealphanerdquit (Ping timeout: 250 seconds)
00:03:54  * thealphanerdjoined
00:11:17  * captainplanetjoined
00:11:29  * thealphanerdquit (Ping timeout: 265 seconds)
00:11:29  * captainplanetchanged nick to thealphanerd
00:12:39  * thlorenzjoined
00:13:03  * i_m_caquit (Ping timeout: 250 seconds)
00:14:07  * wolfeidaujoined
00:14:53  * wolfeidauquit (Remote host closed the connection)
00:15:22  * wolfeidaujoined
00:18:10  * ednapiranhaquit (Quit: Leaving...)
00:21:14  * i_m_cajoined
00:26:51  <rowbit1>Hourly usage stats: [developer: 1, free: 14]
00:26:57  * thlorenzquit (Ping timeout: 276 seconds)
00:32:27  * i_m_caquit (Ping timeout: 240 seconds)
00:38:13  * kumavisquit (Ping timeout: 245 seconds)
00:40:02  * phated_joined
00:40:23  <Raynos>jesusabdullah, ogd: thanks :)
00:43:13  * phatedquit (Ping timeout: 276 seconds)
00:43:43  * phated_quit (Read error: Connection reset by peer)
00:43:54  * phatedjoined
00:44:45  * thlorenzjoined
00:46:53  * thlorenzquit (Remote host closed the connection)
00:50:03  * mikolalysenkojoined
01:02:01  * ralphtheninjaquit (Ping timeout: 250 seconds)
01:10:43  * addisonjquit (Quit: Connection closed for inactivity)
01:11:42  <jesusabdullah>Raynos: I try XD
01:12:58  <rowbit1>substack, pkrumins: These encoders are STILL down: 184.106.99.27(dev-ie7-3)
01:26:03  <ogd>hey, anyone know how to spawn a node program on windows with child_process?
01:26:27  <ogd>e.g. if i have a file called foo.js, on OSX i can just do child_process.spawn('foo.js') and if foo.js has usr/bin/env node at the top it will work
01:26:27  * ferossjoined
01:26:32  <ogd>but that doesn't seem to work on windows
01:26:41  <substack>ogd: spawn(process.execPath, [ file, args... ])
01:26:51  <rowbit1>Hourly usage stats: [developer: 0, free: 9]
01:27:02  <ogd>substack: ahhh!! thanks
01:27:15  * phatedquit (Remote host closed the connection)
01:27:52  * phatedjoined
01:29:19  * dybskiyquit (Remote host closed the connection)
01:32:35  <ogd>substack: ahhhh for it to work on windows you have to manually pad it with double quotes
01:32:37  * phatedquit (Ping timeout: 276 seconds)
01:34:17  <substack>weird
01:35:41  <ogd>substack: its cause the path is like: C:\Documents and Settings\blah blah
01:35:48  <ogd>and it thinks the first arg is "C:\Documents"
01:36:41  <mikolalysenko>does anyone know why typed arrays might be faster in chrome than node?
01:37:05  <mikolalysenko>I have some weird results from a benchmark
01:37:13  <mikolalysenko>which is waaaay faster in the latest chrome
01:37:29  <ogd>mikolalysenko: whoa weird
01:37:34  <mikolalysenko>https://gist.github.com/mikolalysenko/10758605
01:37:42  <mikolalysenko>it is from the kdtree thing
01:37:45  <mikolalysenko>which is almost ready
01:37:53  <mikolalysenko>the one weird case is the nearest neighbor query
01:38:14  <mikolalysenko>look at the kNN, k=1 case
01:38:36  <jjjohnny>mikolalysenko: isn't node v8 many version behind chrome v8?
01:39:09  * ednapiranhajoined
01:39:37  <mikolalysenko>not sure
01:39:58  <mikolalysenko>but I do know that on the higher knn queries static-kdtree is faster than ubilabs' kdtree now
01:40:03  <mikolalysenko>in the latest version anyway
01:40:08  <mikolalysenko>currently running the experiment
01:40:41  <mikolalysenko>but knn,k=1 on node is slow for some reason
01:41:32  * kriskowalquit (Quit: kriskowal)
01:42:31  <mikolalysenko>in chrome though static-kdtree totally smokes ubilabs kdtree on all benchmarks
01:42:41  <mikolalysenko>and gives node-kdtree a run for its money
01:45:58  * i_m_cajoined
01:53:18  * ednapiranhaquit (Remote host closed the connection)
01:53:44  * ednapiranhajoined
02:03:04  * addisonjjoined
02:03:07  <ogd>mikolalysenko: you should try on [email protected]
02:07:01  * ednapiranhaquit (Ping timeout: 258 seconds)
02:14:31  <mikolalysenko>ogd: maybe later, not a huge priority right now
02:14:40  <mikolalysenko>just finished writing up the results
02:14:40  * thealphanerdquit (Quit: thealphanerd)
02:14:41  <mikolalysenko>https://github.com/mikolalysenko/static-kdtree
02:15:09  * Maciek416quit (Remote host closed the connection)
02:26:51  <rowbit1>Hourly usage stats: [free: 3]
02:29:48  <mikolalysenko>also in chrome 36, static-kdtree is faster than node-kdtree
02:30:03  <mikolalysenko>where the latter uses libkdtree which is written in c++!
02:30:35  * naneaujoined
02:34:22  * i_m_caquit (Ping timeout: 276 seconds)
02:37:40  * i_m_cajoined
02:42:23  * i_m_caquit (Ping timeout: 245 seconds)
02:43:12  * thlorenzjoined
02:43:36  * ednapiranhajoined
02:43:55  <jjjohnny>mikolalysenko: are kdtrees all right for geodata?
02:44:51  <mikolalysenko>jjjohnny: depends on the data and the queries, but they can be
02:45:04  <mikolalysenko>the main advantage of kdtrees are low space usage and fast construction
02:45:18  <mikolalysenko>and they can answer closest point queries relatively fast
02:45:29  <mikolalysenko>for range queries though range trees are better
02:45:35  <jjjohnny>so you might take a bucket, say a rectangle of coordinates, construct a kdtree out of them, and do some queires
02:45:46  <mikolalysenko>that can work
02:46:01  <mikolalysenko>though the one thing that you can do in 2D (but only in 2D!) is build a voronoi diagram
02:46:07  <mikolalysenko>that is the fastest way to do closest point queries
02:46:32  <mikolalysenko>basically take points -> construct voronoi diagram -> construct point location data structure -> logarithmic closest point queries
02:46:48  <mikolalysenko>in 3D though kdtrees are better
02:47:07  <jjjohnny>neat
02:47:32  <mikolalysenko>the problem with that technique in 3D is that voronoi diagrams in 3D are just too big
02:47:37  <mikolalysenko>so preprocessing not efficient
02:47:44  <mikolalysenko>and point location in 3D is waaaay harder than 2d
02:47:59  <mikolalysenko>I still don't know the optimal way to do it
02:48:21  <jjjohnny>we need a good databasable model for geodata
02:48:41  <mikolalysenko>simplicial complexes!
02:48:53  <mikolalysenko>that's all you really need in 2D/3D
02:49:30  <mikolalysenko>or generally cell complexes, but that can be heavier weight and you probably don't need all the extra topology baggage that comes with that
02:49:51  <jjjohnny>is that a good indexing model for points?
02:50:09  <mikolalysenko>you build the index from the data generally
02:50:23  <mikolalysenko>but simplicial complex can help indexing points
02:50:28  <jjjohnny>yeah, or you build the index around the model of how you store the data
02:50:45  <mikolalysenko>ie if you run delaunay triangulation on the points it gives you a cell complex you can use for indexing
02:51:21  <mikolalysenko>I think of indexing as sort of a post process you run on the data to speed up specific types of queries
02:52:01  <mikolalysenko>though in js if you want to figure out where a point is in a collection of regions you can use this: https://github.com/mikolalysenko/point-in-region
02:52:19  <defunctzombie>Domenic_: ping
02:53:24  <jjjohnny>mikolalysenko: nice one
02:54:09  <jjjohnny>its not clear to me if that indexing is efficient
02:54:57  <mikolalysenko>it is asymptotically efficient
02:55:06  <mikolalysenko>for small point sets you can just scan though
02:55:20  <jjjohnny>this is my best guess at solving basic geo data store, so that you can query a point, and get neighbors in some range:
02:55:37  <mikolalysenko>it takes O(log(n)) to locate a point, which is the best you can do short of invoking dynamic optimality/cache arguments which require more sophisticated models
02:55:42  <jjjohnny>put everything in small buckets, or zones
02:56:04  <mikolalysenko>that takes exponential space in the worst case though
02:56:10  <jjjohnny>query a point, grab that bucket, and range over nearby buckets
02:56:43  <mikolalysenko>that approach is alright for some data, but has problems in general
02:56:59  <mikolalysenko>if you have features of different sizes for example, it is hard to pick the right bucket size
02:57:15  <mikolalysenko>and if some are close together you might have some buckets with lots of stuff and others with very little
02:57:30  <mikolalysenko>also if you pick a bucket far away from any points you might have to look very far to find the closest point
02:57:46  * ednapiranhaquit (Ping timeout: 276 seconds)
02:57:47  <mikolalysenko>for some situations though it can work
02:58:05  <mikolalysenko>specifically the distance between any two points is at most size of a bucket or so
02:58:10  <mikolalysenko>and same for all edge lengths
02:58:20  <mikolalysenko>and the points have to be uniformly distributed
02:58:49  <jjjohnny>the last isn't going to happen, but if you make the buckets small enough
02:58:56  <mikolalysenko>then you have too many buckets
02:59:00  <mikolalysenko>and searching will take too long
02:59:16  <mikolalysenko>in fact, it could be exponentially many
02:59:31  <mikolalysenko>that is exponential in the number of bits used to encode the coordinates of a point
02:59:44  <jjjohnny>number of buckets is fixed and flat
03:00:00  <mikolalysenko>to make the buckets small enough to prevent doubling up it could be exponential
03:00:33  <mikolalysenko>suppose the points are in 1d and you have [0000000000000, 00000000000001, 111111111111111]
03:00:44  <mikolalysenko>then you need a bunch of buckets to separate the first two points
03:02:00  <jjjohnny>bascially zip codes
03:02:06  <jjjohnny>except smaller
03:02:55  <jjjohnny>perhaps an index that changes the shape and size of "zip codes" according to point density
03:03:49  <jjjohnny>mikolalysenko: hwo would you do this with leveldb/
03:03:49  <jjjohnny>?
03:03:59  <mikolalysenko>for point location?
03:04:05  <jjjohnny>you have a low level store, and then various indexing methods
03:04:14  <mikolalysenko>ah a bst right?
03:04:17  <mikolalysenko>it is tricky
03:04:58  <mikolalysenko>the method I used in this would be hard to adapt I think, but it might be doable: https://github.com/mikolalysenko/point-in-region
03:05:25  <mikolalysenko>but probably some index on a voronoi diagram is what I would do
03:05:46  <mikolalysenko>it would not be dynamic though, that problem is much harder to solve
03:06:04  <mikolalysenko>efficient methods are known, but they are very hard to implement
03:06:55  <jjjohnny>thats why I say start with small, equal sized buckets, don't worry about density. then you can easily find the bucket of any given point
03:07:41  <jjjohnny>and range over all the points nearby
03:07:54  <mikolalysenko>but that doesn't really work
03:08:03  <mikolalysenko>at least it won't make things faster
03:08:10  <mikolalysenko>and could use a lot of space
03:08:21  <jjjohnny>the index would be tiny
03:08:52  <mikolalysenko>linear scan might be just as good
03:09:05  <mikolalysenko>or you could use a kdtree
03:09:07  <jjjohnny>see im thinking the approach is to find a good low level start, and the build the right index on top, per usage
03:09:22  <mikolalysenko>kdtree is a simple, generic low level way to get closest points
03:09:27  <mikolalysenko>it would be faster than that grid
03:09:41  <mikolalysenko>but not as good in 2D as voronoi diagrams
03:10:04  <mikolalysenko>though this might not matter too much, depending on the data and number of queries
03:10:36  <jjjohnny>it matters. for instance, rtrees seem to be completely useless
03:10:44  <jjjohnny>cuz the whole index is the thing
03:11:02  <jjjohnny>i mean, the index is the whole thing
03:11:24  <jjjohnny>you dont want to start there, and then need something that can handle more data or queries
03:11:55  <mikolalysenko>you only add an index once you need to answer a query
03:12:09  <mikolalysenko>it doesn't make sense to start going crazy with indices before you know what you want to do
03:12:55  <mikolalysenko>rtrees though are a different story from all of this, they are often used for frustum tests, though I don't know if they are much better than kdtrees
03:13:17  <mikolalysenko>range trees are different though, and are good for box queries in low dimensions
03:15:16  <mikolalysenko>in the end, if you don't need to repeatedly test some query or another you should not build an index over your data
03:15:42  <mikolalysenko>it will just slow down updates for no good reason, and increase space/processing requirements
03:22:16  <jjjohnny>yeah and indexes are complex
03:22:23  <jjjohnny>i want to model and script
03:23:19  * ednapiranhajoined
03:23:23  <mikolalysenko>work backwards to the index. start from a problem
03:26:51  <rowbit1>Hourly usage stats: [free: 8]
03:27:54  * ednapiranhaquit (Ping timeout: 258 seconds)
03:30:08  * ednapiranhajoined
03:34:23  * thlorenzquit (Remote host closed the connection)
03:34:56  * thlorenzjoined
03:39:24  * thlorenzquit (Ping timeout: 258 seconds)
03:42:28  * ednapiranhaquit (Ping timeout: 258 seconds)
03:52:22  * i_m_cajoined
03:57:46  * mbrevoortquit (Remote host closed the connection)
03:58:12  * mbrevoortjoined
03:58:48  * mbrevoortquit (Remote host closed the connection)
04:00:29  * tilgoviquit (Read error: Connection reset by peer)
04:09:34  * mbrevoortjoined
04:16:59  * i_m_caquit (Ping timeout: 252 seconds)
04:18:03  * defunctzombiechanged nick to defunctzombie_zz
04:21:35  * i_m_cajoined
04:26:51  <rowbit1>Hourly usage stats: [free: 10]
04:27:56  * ednapiranhajoined
04:32:27  * ednapiranhaquit (Ping timeout: 240 seconds)
04:40:43  * addisonjquit (Quit: Connection closed for inactivity)
04:41:26  * naneauquit (Quit: Quit.)
04:48:11  * calvinmetcalfquit (Quit: Connection closed for inactivity)
04:57:01  * mbrevoortquit (Remote host closed the connection)
05:04:39  * i_m_caquit (Ping timeout: 252 seconds)
05:06:17  * i_m_cajoined
05:08:03  * shamaquit (Remote host closed the connection)
05:16:03  * brianloveswordsjoined
05:18:07  * ednapiranhajoined
05:18:58  * wolfeidauquit (Remote host closed the connection)
05:19:18  * Maciek416joined
05:19:57  * ferossquit (Quit: feross)
05:21:52  * ferossjoined
05:26:51  <rowbit1>Hourly usage stats: [free: 15]
05:34:21  * shamajoined
05:42:27  * i_m_caquit (Ping timeout: 258 seconds)
05:42:29  <rowbit1>substack, pkrumins: These encoders are STILL down: 50.57.171.229(dev-ie6-2)
05:46:49  * shamaquit
05:58:49  * brianloveswordsquit (Quit: Computer has gone to sleep.)
06:07:43  * mikolalysenkoquit (Ping timeout: 252 seconds)
06:16:56  * ednapira_joined
06:17:10  * wolfeidaujoined
06:17:10  * ednapiranhaquit (Read error: Connection reset by peer)
06:17:17  * ednapira_changed nick to ednapiranha
06:19:03  * ins0mniajoined
06:19:10  * thealphanerdjoined
06:22:10  <ogd>substack: i get "Heroku | No such app" when I go to http://ci.testling.com/
06:22:12  <ogd>pkrumins: o/
06:23:11  * ednapiranhaquit (Quit: Leaving...)
06:24:05  <terinjokes>ogd: works here
06:24:11  * thealphanerdquit (Quit: thealphanerd)
06:25:46  * thealphanerdjoined
06:26:28  * thealphanerdquit (Client Quit)
06:26:51  <rowbit1>Daily usage stats: [free: 20]
06:26:51  <rowbit1>Hourly usage stats: [free: 23]
06:47:16  * Maciek416quit (Remote host closed the connection)
07:12:59  <rowbit1>substack, pkrumins: These encoders are STILL down: 184.106.99.27(dev-ie7-3)
07:13:09  * hayesquit (Ping timeout: 258 seconds)
07:13:09  * heathquit (Quit: No Ping reply in 180 seconds.)
07:13:10  * heathjoined
07:13:10  * heathquit (Changing host)
07:13:10  * heathjoined
07:18:25  * hayesjoined
07:26:20  * kirbysayshijoined
07:26:51  <rowbit1>Hourly usage stats: [free: 11]
07:48:55  * kirbysayshiquit (Quit: Leaving...)
08:15:25  <ogd>made a pr to make shields.io badges less ugly https://github.com/badges/shields/pull/159
08:26:51  <rowbit1>Hourly usage stats: [free: 25]
08:27:48  * fotoveritequit (Quit: fotoverite)
08:35:55  * Maciek416joined
08:40:55  * Maciek416quit (Ping timeout: 265 seconds)
08:48:20  * thealphanerdjoined
08:52:40  <pkrumins>ogd: we don't use Heroku
08:53:24  * ralphtheninjajoined
09:02:53  * thealphanerdquit (Quit: thealphanerd)
09:21:24  <ins0mnia>ogd: sending a new pr to filereader-stream
09:21:36  * thealphanerdjoined
09:26:47  * ralphtheninjapart
09:26:51  <rowbit1>Hourly usage stats: [free: 18]
10:24:45  * Maciek416joined
10:26:52  <rowbit1>Hourly usage stats: [free: 20]
10:29:34  * Maciek416quit (Ping timeout: 258 seconds)
10:30:45  * thealphanerdquit (Quit: thealphanerd)
10:31:43  * yorickjoined
10:41:48  * no9joined
10:59:33  * no9quit (Quit: This computer has gone to sleep)
11:26:51  <rowbit1>Hourly usage stats: [free: 26]
11:41:51  * calvinmetcalfjoined
11:42:30  <rowbit1>substack, pkrumins: These encoders are STILL down: 50.57.171.229(dev-ie6-2)
11:54:34  * dybskiyjoined
11:59:06  * dybskiyquit (Ping timeout: 240 seconds)
12:07:07  * mafintosh_joined
12:10:33  * ehd_joined
12:11:10  * dsfadfjoined
12:11:10  * rannmannquit (Disconnected by services)
12:11:35  * ins0mnia_joined
12:13:39  * Maciek416joined
12:13:42  * no9joined
12:14:51  * ins0mniaquit (*.net *.split)
12:14:52  * terinjokesquit (*.net *.split)
12:14:52  * mafintoshquit (*.net *.split)
12:14:53  * ogdquit (*.net *.split)
12:14:53  * _hendrikquit (*.net *.split)
12:14:53  * ehdquit (*.net *.split)
12:15:18  * terinjokesjoined
12:15:21  * mafintosh_changed nick to mafintosh
12:15:32  * ehd_changed nick to ehd
12:18:27  * Maciek416quit (Ping timeout: 250 seconds)
12:20:21  * ogdjoined
12:21:26  * _hendrikjoined
12:25:29  * no9quit (Quit: This computer has gone to sleep)
12:26:52  <rowbit1>Hourly usage stats: [free: 17]
12:54:41  * dybskiyjoined
12:59:27  * dybskiyquit (Ping timeout: 258 seconds)
13:08:41  * othiym23quit (Ping timeout: 255 seconds)
13:13:00  <rowbit1>substack, pkrumins: These encoders are STILL down: 184.106.99.27(dev-ie7-3)
13:13:53  * othiym23joined
13:26:52  <rowbit1>Hourly usage stats: [free: 55]
13:42:46  * defunctzombie_zzchanged nick to defunctzombie
13:43:52  * thlorenzjoined
13:44:16  * pfrazejoined
13:45:34  * thlorenzquit (Remote host closed the connection)
13:48:11  * calvinmetcalfquit (Quit: Connection closed for inactivity)
13:54:44  * dybskiyjoined
13:59:15  * dybskiyquit (Ping timeout: 258 seconds)
14:02:31  * Maciek416joined
14:03:18  * thlorenzjoined
14:07:18  * Maciek416quit (Ping timeout: 258 seconds)
14:14:44  * mikolalysenkojoined
14:16:45  * thlorenzquit (Ping timeout: 250 seconds)
14:26:52  <rowbit1>Hourly usage stats: [free: 28]
14:46:03  * naneaujoined
14:51:43  * calvinmetcalfjoined
14:54:40  * dybskiyjoined
14:55:21  * no9joined
14:59:06  * dybskiyquit (Ping timeout: 240 seconds)
15:16:40  * thlorenzjoined
15:19:29  * dsfadfchanged nick to rannmann
15:22:04  * rannmannquit (Changing host)
15:22:04  * rannmannjoined
15:23:25  * brianloveswordsjoined
15:26:52  <rowbit1>Hourly usage stats: [developer: 1, free: 35]
15:29:17  * no9quit (Quit: This computer has gone to sleep)
15:33:39  * naneauquit (Quit: Part.)
15:40:43  * AvianFlujoined
15:46:30  * no9joined
15:51:25  * Maciek416joined
15:54:43  * dybskiyjoined
15:56:10  * Maciek416quit (Ping timeout: 258 seconds)
15:58:51  * dybskiyquit (Ping timeout: 240 seconds)
16:25:50  * fotoveritejoined
16:26:51  <rowbit1>Hourly usage stats: [developer: 0, free: 21]
16:30:24  * Maciek416joined
16:32:30  * contrahaxjoined
16:43:13  * ednapiranhajoined
16:49:00  * no9quit (Quit: Leaving)
16:54:46  * dybskiyjoined
16:55:28  * thealphanerdjoined
16:59:07  * dybskiyquit (Ping timeout: 245 seconds)
17:05:07  <mikolalysenko>does anyone know of any pure javascript bitmap parsing libraries?
17:13:25  * thlorenzquit (Remote host closed the connection)
17:22:30  * mbrevoortjoined
17:26:15  <mikolalysenko>also, what is the current best browserify xhr stream wrapper?
17:26:52  <rowbit1>Hourly usage stats: [developer: 1, free: 20]
17:27:27  * mbrevoortquit (Ping timeout: 276 seconds)
17:32:00  <ogd>mikolalysenko: require('xhr')
17:32:24  <ogd>mikolalysenko: xhr doesnt actually stream
17:32:40  <ogd>mikolalysenko: so i've found it isn't worth it to have the stream API on top
17:33:42  <mikolalysenko>ogd: well, there is a mozilla extension for it
17:34:02  <mikolalysenko>ogd: and it would be nice to have some kind of optional upgrade if available policy
17:34:27  <mikolalysenko>also sometimes I want to turn an xhr into a stream to interface it with existing code, not worrying about efficiency
17:35:26  <ogd>i see where you're coming from, I just have tried it in practice and it felt very square-peg-round-hole to me
17:35:54  <ogd>mikolalysenko: require('http') w/ browserify implements the stream API
17:36:03  <ogd>for reading at least
17:37:13  <mikolalysenko>that might be enough
17:38:19  <ogd>https://github.com/substack/http-browserify/blob/master/lib/response.js#L105-L116
17:42:30  <rowbit1>substack, pkrumins: These encoders are STILL down: 50.57.171.229(dev-ie6-2)
17:53:30  * kirbysayshijoined
17:53:54  * kirbysayshiquit (Client Quit)
17:56:55  * peutetrejoined
18:07:11  * ednapiranhaquit (Quit: Leaving...)
18:07:56  * thlorenzjoined
18:19:59  * h0kejoined
18:24:40  * Maciek416quit (Remote host closed the connection)
18:26:52  <rowbit1>Hourly usage stats: [developer: 0, free: 15]
18:31:27  * i_m_cajoined
18:35:58  * peutetrequit (Quit: peutetre)
18:50:09  * peutetrejoined
18:50:52  * ogdchanged nick to mbalho
18:54:31  * tanepiperquit (Ping timeout: 258 seconds)
18:54:31  * svnltoquit (Ping timeout: 258 seconds)
18:54:32  * svnltojoined
18:54:32  * svnltoquit (Changing host)
18:54:32  * svnltojoined
18:54:48  * cpupquit (Ping timeout: 258 seconds)
18:54:48  * mrb_bkquit (Ping timeout: 258 seconds)
18:54:50  * dybskiyjoined
18:55:05  * mbalhochanged nick to ogd
18:55:28  * cpupjoined
18:56:09  * mrb_bkjoined
18:56:11  * tanepiperjoined
18:59:31  * dybskiy_joined
18:59:36  * dybskiy_quit (Remote host closed the connection)
19:01:04  <owen1>what's my best shot at finding why my CPU spikes to %100? node-tick?
19:02:35  * ferossquit (Quit: feross)
19:03:00  * dybskiyquit (Ping timeout: 276 seconds)
19:07:38  * peutetrequit (Quit: peutetre)
19:10:31  * i_m_caquit (Ping timeout: 258 seconds)
19:13:01  <rowbit1>substack, pkrumins: These encoders are STILL down: 184.106.99.27(dev-ie7-3)
19:25:13  * h0kequit (Remote host closed the connection)
19:26:52  <rowbit1>Hourly usage stats: [developer: 0, free: 44]
19:33:12  <mikolalysenko>ugh. buffers don't work in browserify well do they :(
19:33:38  <jjjohnny>mikolalysenko: wym?
19:34:10  <jjjohnny>have you updates to the latest browserify, which uses the new browser buffer implementation by feross?
19:35:46  <mikolalysenko>I'll give it a shot
19:36:01  <jjjohnny>also check this out, prepublish version if a dangerous new module: https://github.com/NHQ/bibimbap
19:36:18  <jjjohnny>which converts almost any JS to ArrayBuffer, including nndarrays
19:36:37  <jjjohnny>all tests passing
19:36:49  <mikolalysenko>jjjohnny: it works!
19:36:55  <mikolalysenko>was using an old browserify
19:36:58  <mikolalysenko>happy again
19:38:41  <mikolalysenko>check it out! more file formats: https://github.com/mikolalysenko/get-pixels
19:38:47  <mikolalysenko>also 100% js, no native dependencies
19:39:01  <mikolalysenko>would be nice to have more formats eventually
19:39:54  <jjjohnny>yay!
19:42:16  <mikolalysenko>another nice feature in the latest version is that animated gifs are supported to
19:42:35  <mikolalysenko>you get a higher dimensional ndarray of all the slices of the gif
19:44:48  <jjjohnny>should support local usage, like if I create pngs from canvas, load DOM Images, or files
19:45:07  <jjjohnny>files in the browser
19:46:11  * i_m_cajoined
19:51:55  <substack>multidimensional gifs O_O
19:52:36  <substack>https://hatonnspeaksthetruth.files.wordpress.com/2009/12/block_4210988.gif?w=474
19:52:54  <jjjohnny>whoa
19:53:21  <jjjohnny>that is the shape of a proper 2d game world
19:54:15  <substack>mikolalysenko: it might be better if get-pixels could take a buffer of the data instead of doing the readFile itself
20:05:55  * ins0mnia_quit (Ping timeout: 252 seconds)
20:11:03  * scottjabquit (Remote host closed the connection)
20:12:52  * i_m_caquit (Ping timeout: 245 seconds)
20:13:38  * Maciek416joined
20:16:19  * i_m_cajoined
20:16:30  * h0kejoined
20:17:39  * Maciek416quit (Ping timeout: 240 seconds)
20:18:11  * calvinmetcalfquit (Quit: Connection closed for inactivity)
20:18:31  <mikolalysenko>substack: I've thought about that
20:18:41  <mikolalysenko>substack: but you have to identify the file type somehow
20:18:50  <mikolalysenko>I suppose you could pass it another parameter
20:20:03  <mikolalysenko>substack: ideally it would take a stream I think, but not all image parsers use streams or operate in a consistent way
20:20:09  <mikolalysenko>also there are issues with the dom version
20:20:21  <mikolalysenko>since it uses the Image() object to load the data, and that takes a URL
20:20:35  <mikolalysenko>so, the situation is not great but it works ok for now
20:21:26  * AvianFluquit (Remote host closed the connection)
20:23:36  * No9joined
20:25:40  * ferossjoined
20:26:51  <rowbit1>Hourly usage stats: [developer: 0, free: 23]
20:33:24  * dguttmanjoined
20:33:54  * i_m_caquit (Ping timeout: 240 seconds)
20:36:18  * phatedjoined
20:37:32  * phatedquit (Remote host closed the connection)
20:38:00  * i_m_cajoined
20:45:53  * kirbysayshijoined
20:47:48  * ferossquit (Quit: feross)
20:49:06  * scottjabjoined
20:52:29  * gozalaquit (Quit: Connection closed for inactivity)
20:54:21  <Domenic_>substack & everyone else. Ember wants to standardize on npm for packages. They really want an indication in package.json that they can bug everyone to add for browser-compatible packages. Should we do `browser: true` or a `browser` tag in `keywords`?
20:55:47  * h0kequit (Remote host closed the connection)
20:56:16  <ogd>Domenic_: they should have a browser CI badge in all the readmes
20:56:38  <ogd>Domenic_: IMO thats the single most important thing
20:58:49  <substack>Domenic_: browser is already used for things
20:58:57  <substack>https://gist.github.com/defunctzombie/4339901
20:59:16  <substack>also use keywords
20:59:18  <Domenic_>ogd: buying into an unreliable CI service is not a realistic ask for everyone.
20:59:30  <ogd>Domenic_: i just said 'browser ci'
20:59:33  <substack>http://npmjs.org/browse/keyword/browser
20:59:50  <Domenic_>substack: sure I was wondering about a simple "browser: true" version that would not do any overrides but just be an indicator
21:00:08  <Domenic_>substack: that link is pretty compelling, browser keyword seems like a winner
21:00:50  * addisonjjoined
21:01:41  <mikolalysenko>the real issue is if it uses any native dependencies, not whether it works for browser
21:01:57  <mikolalysenko>so maybe the other way to say it is "native": false
21:02:30  <ogd>mikolalysenko: i view them both as equally opt-in, as in the majority of people today who publish stuff to npm don't consider other platforms than the one they are currently on
21:02:49  <ogd>mikolalysenko: and the platforms are: windows, mac, linux, browsers (and all flavors of each)
21:02:56  <mikolalysenko>I think that the fs stuff though can probably be wrapped eventually
21:03:14  <mikolalysenko>same with udp/tcp which hardly anyone touches directly
21:03:21  <ogd>mikolalysenko: have you seen https://github.com/mafintosh/level-filesystem ? sponsored w/ my grant money :)
21:03:32  <mikolalysenko>ogd: haha, that's great!
21:04:03  <mikolalysenko>the real problem I think is not using core node.js modules, but using native dependencies in node
21:04:07  <ogd>mikolalysenko: i still think for many cases authors will have to make a conscious effort to support all runtimes
21:04:22  <mikolalysenko>hmm
21:04:32  * i_m_caquit (Ping timeout: 245 seconds)
21:04:38  <mikolalysenko>ogd: for some things, maybe
21:05:07  <mikolalysenko>ogd: one big issue I think is canvas/webgl, where there aren't really good node analogs
21:05:38  <ogd>agreed. I also think there are usability issues
21:05:48  <ogd>or maybe 'marketing' is the right term
21:05:54  <mikolalysenko>depends on what the module does mostly I think
21:06:11  <ogd>like... theres no 'brand' to represent modules that work everywhere
21:06:17  <ogd>if that makes sense
21:06:31  <mikolalysenko>hmm... I sort of think of this as the default situation, but I agree that this perception is not widely held
21:06:54  <mikolalysenko>(hence things like component...)
21:06:59  <ogd>yea we are living in moduletopia already :
21:08:03  <mikolalysenko>so, from the perspective that compatible is the default, it makes more sense to brand things as either "browser specific" or "native specific"
21:08:14  <mikolalysenko>and if unspecified, assume that it works anywhere
21:09:15  <jjjohnny>johnny's first file type
21:09:42  <substack>fixed the testling issue, the ci process crashed for some reason but it was still running tests
21:11:44  * thlorenzquit (Remote host closed the connection)
21:12:59  <jjjohnny>substack: is there latest browsers only for testling
21:13:07  <jjjohnny>like chrome/latest
21:13:39  <jjjohnny>all i want is a html5 check
21:13:48  <jjjohnny>testling : html5
21:16:03  <substack>yes
21:16:08  <substack>chrome/latest
21:16:10  <substack>that works
21:16:33  * i_m_cajoined
21:18:09  <jjjohnny>substack: doing ipad/7.1 yet?
21:25:04  <jjjohnny>service timeout on the github hook :\
21:26:41  * defunctzombiechanged nick to defunctzombie_zz
21:26:52  <rowbit1>Hourly usage stats: [developer: 0, free: 24]
21:39:07  * brianloveswordsquit (Quit: Computer has gone to sleep.)
21:42:19  <ogd>substack: im not seeing new pushes showing up for https://ci.testling.com/maxogden/dat, are there any errors on your end?
21:42:29  * phatedjoined
21:44:52  * ferossjoined
21:47:51  <jjjohnny>ogd: check yr github hook status, mine shows a timeout error pinging the service
21:48:33  <ogd>jjjohnny: mine is green, plus i was using `testlingify test` which sends it directly
21:50:44  * taterbasequit (Quit: Connection closed for inactivity)
21:51:35  * phatedquit (Remote host closed the connection)
21:54:07  * phatedjoined
22:02:20  * Maciek416joined
22:05:19  * AvianFlujoined
22:06:42  * Maciek416quit (Ping timeout: 240 seconds)
22:08:54  * mikolalysenkoquit (Ping timeout: 276 seconds)
22:13:31  * i_m_caquit (Ping timeout: 252 seconds)
22:18:08  * i_m_cajoined
22:23:02  * dguttmanquit (Quit: dguttman)
22:26:18  * i_m_caquit (Ping timeout: 240 seconds)
22:26:51  <rowbit1>Hourly usage stats: [developer: 0, free: 7]
23:08:29  * defunctzombie_zzchanged nick to defunctzombie
23:21:31  * AvianFluquit (Remote host closed the connection)
23:25:37  * Maciek416joined
23:26:51  <rowbit1>Hourly usage stats: [developer: 0, free: 11]
23:38:45  * yorickquit (Remote host closed the connection)
23:42:31  <rowbit1>substack, pkrumins: These encoders are STILL down: 50.57.171.229(dev-ie6-2)
23:46:08  * ednapiranhajoined
23:46:10  * ednapiranhaquit (Remote host closed the connection)
23:46:17  * ednapira_joined
23:46:39  * ednapira_changed nick to ednapiranha