00:02:12  * jaz303_quit (Remote host closed the connection)
00:22:44  * ralphtheninjajoined
01:05:52  * ralphtheninjaquit (Ping timeout: 264 seconds)
01:06:40  <substack>mikolalysenko: all set up with the server space
01:06:58  <substack>now reading about slippy tiles and deciding how much OSM baggage I am comfortable taking on
01:07:54  <substack>I don't like how all the tile techniques are based around mercator ideas and capped at ±81/82 deg
01:08:44  <substack>I would rather use something not so coupled to a particular shitty projectionp
01:12:49  <substack>the earth is a goddamned approximate oblate spheroid
01:13:14  * ralphtheninjajoined
01:13:51  <substack>I mean the real question here is how to uv texture an oblate sphereoid
01:15:47  <substack>if you have an approximately correct representation of reality you can always convert it to some worse model later
01:18:57  * ralphtheninjaquit (Ping timeout: 276 seconds)
01:22:27  * phatedjoined
01:26:59  * phatedquit (Ping timeout: 250 seconds)
01:29:21  * ralphtheninjajoined
01:37:22  <substack>but! I can cut out trapezoids with osmconvert's border polygon tool (-B)
01:47:41  <substack>idea for naming: "planetary tiles"
02:03:22  <substack>mikolalysenko: have a favorite spherical polyhedron?
02:03:42  <substack>I think it will be easy to take any of them and squash by the polar flattening
02:28:48  <mikolalysenko>cube maps are simple, but icosahedron is probably the best option
02:28:55  <mikolalysenko>I kind of lean toward cube maps
02:29:52  <substack>oh right I remember you were telling me about these
02:30:31  <substack>yes, those make sense and are simple and not too distorted
02:31:11  <substack>well, they are very distorted, but it seems easy enough to project them onto a much more accurate wgs84 sphereoid
02:33:29  <mikolalysenko>it is actually an area of ongoing research
02:33:29  <mikolalysenko>there are a bunch of formats that have been proposed for parameterizing spheres
02:41:46  * contrahaxquit (Quit: Sleeping)
02:59:38  * pfrazequit (Quit: ZNC - 1.6.0 - http://znc.in)
03:10:37  * phatedjoined
03:14:57  * phatedquit (Ping timeout: 244 seconds)
03:23:40  <substack>https://en.wikipedia.org/wiki/HEALPix is an equal-area cube map
03:30:00  * contrahaxjoined
03:31:49  <ogd>without clicking link i will try to remember what HEALPix stands for: hierarchical longitudinal equal area projection!
03:32:04  <ogd>AH DANG its isoLatitude Pixelisation
03:32:07  <ogd>i always forget that part
03:32:34  <substack>favored projection of cosmologists!
03:32:38  <substack>that's how you know it's good
03:34:00  <substack>nothing on npm except some d3 things, so that's what I'm going to be doing tonight
03:34:10  <ogd>i worked with it a few years ago but from python
03:41:04  <substack>cool, it's built into proj
03:44:38  * phatedjoined
03:45:25  <mikolalysenko>I think healpix is the current best practice
03:46:01  <mikolalysenko>BUT I remember from a conversation with gregory chirkijian that they had worked out some alternative scheme based on subdivisions of icosahedrons that had much better properties
03:46:09  <mikolalysenko>though I cannot recall the details
03:49:04  <mikolalysenko>also it is almost 5am here and my mind is not working right
03:49:16  * phatedquit (Ping timeout: 252 seconds)
03:51:23  <substack>well, for this problem equal area is more a nice-to-have rather than something that is absolutely required
03:59:24  * pfrazejoined
04:04:08  * pfrazequit (Remote host closed the connection)
04:23:21  * contrahaxquit (Quit: Sleeping)
06:27:17  * warbrettquit (Quit: Connection closed for inactivity)
06:37:25  <substack>ok I have a ang2xy healpix implementation
06:37:29  <substack>now to invert it
06:38:06  <substack>what I don't get is my implementation reading directly from the paper is 16 lines, but http://healpix.sourceforge.net/ is thousands of lines in each lang
06:38:25  <substack>including bizarrely OOP python and c++?
06:39:31  <substack>also this js implementation seems way too big for the algorithm as described in the paper https://github.com/kapadia/healpix.js/blob/master/healpix.js
06:40:40  <substack>oh maybe because those libs try to implement the FIT file format as well
06:40:46  <substack>which is silly, that should be a separate lib
06:40:55  <substack>even though the paper talks about FITS
06:46:25  <substack>some severe cargo culting must be going on
08:11:00  * trevnorrisquit (Ping timeout: 244 seconds)
08:11:16  * trevnorrisjoined
08:11:31  * dlmanningquit (Ping timeout: 244 seconds)
08:12:02  * perlbotquit (Ping timeout: 244 seconds)
08:12:03  * ec\quit (Ping timeout: 244 seconds)
08:12:21  * hyperirc-9d25a90quit (Read error: Connection reset by peer)
08:12:28  * hyperirc-9d25a90joined
08:12:33  * mmaleckiquit (Ping timeout: 244 seconds)
08:13:01  * mmaleckijoined
08:13:20  * ec\joined
08:14:10  * dlmanningjoined
08:14:58  * perlbotjoined
08:15:36  * beaugundersonquit (Ping timeout: 276 seconds)
08:16:29  * beaugundersonjoined
08:31:50  <substack>https://github.com/substack/healpix
09:04:22  <substack>this grid will be really good for aerial and satellite photography
09:07:03  * fotoveritequit (Quit: fotoverite)
09:34:15  * jiangplusquit (Ping timeout: 276 seconds)
09:41:49  * jiangplusjoined
10:27:13  * thealphanerdquit (Quit: farewell for now)
10:27:43  * thealphanerdjoined
12:47:41  <mikolalysenko>nice
13:20:57  <robertkowalski>where do the cool remote javascripters work these days?
13:21:25  <robertkowalski>location-wise and company
13:21:49  <robertkowalski>i'm going to live in seville soon, for two months
14:06:25  * pfraze_joined
15:02:30  * pfraze_quit (Remote host closed the connection)
15:26:36  * pfrazejoined
15:26:36  * pfrazequit (Read error: Connection reset by peer)
15:27:57  * pfrazejoined
15:29:40  * pfallenopquit (Ping timeout: 272 seconds)
15:30:48  * hyperirc-9d25a90quit (Remote host closed the connection)
15:30:54  * hyperirc-9d25a90joined
15:31:19  * pfallenopjoined
15:31:44  * ekristen_joined
15:33:17  * ekristenquit (Ping timeout: 272 seconds)
15:33:17  * ekristen_changed nick to ekristen
15:36:33  * pfrazequit (Quit: pfraze has joined)
15:36:55  * pfrazejoined
15:39:46  * beaugundersonquit (*.net *.split)
15:39:46  * ec\quit (*.net *.split)
15:39:46  * mk30quit (*.net *.split)
15:39:46  * toddselfquit (*.net *.split)
15:39:47  * ferossquit (*.net *.split)
15:39:48  * creationixquit (*.net *.split)
15:47:10  * toddselfjoined
15:47:11  * mk30joined
15:47:22  * ec\joined
15:47:38  * creationixjoined
15:50:12  * xipliasquit (Ping timeout: 260 seconds)
15:51:07  * ferossjoined
15:51:47  * beaugundersonjoined
15:53:43  * xipliasjoined
15:59:29  <jfhbrook>you could become a barber robertkowalski
16:39:07  <mikolalysenko>robertkowalski: never been to seville, though it sounds nice
17:08:01  * fotoveritejoined
17:52:42  * contrahaxjoined
18:06:21  * warbrettjoined
18:43:05  * pfrazequit (Remote host closed the connection)
19:07:22  * pfrazejoined
19:21:13  <robertkowalski>jfhbrook: what is barber?
19:55:50  * contrahaxquit (Quit: Sleeping)
21:21:28  * pfrazequit (Remote host closed the connection)
21:35:48  <jfhbrook>robertkowalski: the barber of seville? FIGARO FIGARO FIGARO
21:36:00  <jfhbrook>robertkowalski: but y'know, a barber is a hair-cutter
22:13:24  * phatedjoined
22:34:40  * ralphtheninjaquit (Ping timeout: 264 seconds)
22:39:56  * ralphtheninjajoined
22:45:42  * contrahaxjoined
22:48:28  * ralphtheninjaquit (Ping timeout: 264 seconds)
23:22:56  * saijanai_joined
23:29:24  * ralphtheninjajoined
23:37:29  <jfhbrook>substack: can I interest you in a fun hack project? https://github.com/jfhbrook/microservice-fizzbuzz
23:42:11  <substack>jfhbrook: it should use kafka
23:42:25  <jfhbrook>agree, not sure how though
23:42:52  <jfhbrook>I'm thinking about what would actually be sane for a joining interface where you can input some range and get some array as output
23:43:00  <jfhbrook>I feel like a queue would be useful here
23:43:15  <jfhbrook>and kafka is time series data, right? so maybe similar fit, maybe something tricky there
23:43:28  <substack>it should use kafka AND rabbitmq
23:44:50  <jfhbrook>use rabbit for 2-way rpc