17:38:20  * isaacstopic: Manta: Big Data Unix | http://apidocs.joyent.com/manta/ | http://logs.libuv.org/manta/latest
17:39:08  * daviddi__quit (Remote host closed the connection)
17:39:46  * daviddiasjoined
17:39:56  <nahamu>(and then a browser plugin that can take a bunch of rows of json and turn it into an index with hyperlinks ;)
17:40:43  <mcavage>well, eventually :\ -- I think _everybody_ really wants that.
17:41:11  * daviddiasquit (Remote host closed the connection)
17:41:22  <mcavage>and the http signature stuff got picked up by W3C, so it's not even far-fetched.
17:41:53  <bahamas10>mcavage: that's awesome news... i'd love to see that in the browser
17:42:39  * daviddiasjoined
17:43:16  * notmattquit (Remote host closed the connection)
17:43:20  <mcavage>yeah, the w3c payments group were the ones that got it into an ietf draft, actually: http://tools.ietf.org/html/draft-cavage-http-signatures-00
17:43:41  <mcavage>it sounds like the ietf has their own favorite horse for http 2.0, so i doubt they'll actually make this an RFC, but W3C is betting on it, at least.
17:43:52  <mcavage>they're baking it in everywhere with the web payments spec.
17:45:55  * ryancnelsonquit (Quit: Leaving.)
17:47:05  * ryancnelsonjoined
17:48:09  * mcavagequit (Remote host closed the connection)
18:05:33  * marsellquit (Ping timeout: 245 seconds)
18:09:23  * marselljoined
18:12:58  * ryancnelsonquit (Quit: Leaving.)
18:13:32  * notmattjoined
18:42:07  * mcavagejoined
18:42:23  * mcavagequit (Remote host closed the connection)
18:42:36  * mcavagejoined
18:42:52  * mcavagequit (Remote host closed the connection)
18:43:14  * mcavagejoined
18:51:58  * abraxasjoined
18:56:29  * abraxasquit (Ping timeout: 252 seconds)
19:12:04  * notmattquit (Remote host closed the connection)
19:28:34  * mamashjoined
19:43:41  * notmattjoined
20:04:39  <isaacs>bahamas10: yo
20:04:42  <isaacs>this might be faster?
20:04:43  <isaacs>:)
20:11:02  <bahamas10>isaacs: hey yeah definitely
20:11:25  <bahamas10>i like the idea for the sync module, feel free to rip out as much stuff as you can from `manta-sync`
20:12:20  <bahamas10>but yeah, definitely an eventemitter that can emit a 'completed', or 'finished' event (idk a good name for it), that gets called with an optional Error object
20:15:22  <isaacs>bahamas10: i was thinking of emitting 'file' for each file completed
20:15:39  <isaacs>bahamas10: also, do you want to do all-or-nothing?
20:16:01  <isaacs>bahamas10: like, if there's 10 files being sent, should it try to kill any that are in progress if one fails? i guess that's probably stupid.
20:16:06  <isaacs>since you can just run again anyway
20:16:55  <isaacs>bahamas10: and then a final "complete" event for "all files are done", which will pass the first error, and then a {filename:<Error or 'ok'>} hash
20:17:22  <isaacs>sync(...).on('complete', function(er, data) { ... }) <-- data is the {filename:<Error or 'ok'>,...} list
20:18:27  <bahamas10>isaacs: perfect, i like that
20:18:31  <isaacs>kk
20:18:35  <bahamas10>so 'file' and 'end' events?
20:18:37  <bahamas10>er, complete
20:26:29  <isaacs>yah
20:26:40  <isaacs>and if you pass a cb to the ctor, it gets registered as teh 'complete' event handler
20:26:52  <isaacs>so you can still use it as a "lemme know the first failure" type of approach
20:34:23  <bahamas10>isaacs: also, will this module handle parallelism/concurrency? being able to set an option for that, like {concurrency:30} would be great
20:35:13  <isaacs>bahamas10: oh, nice, i hadn't thought of that
20:35:32  <isaacs>bahamas10: my usual approach is to effectively do "concurrency: ALL OF IT" and then deal with any problems that ensue
20:35:56  <isaacs>bahamas10: ie, it'll just make all the requests, and do all the things, and deprecate the counter, and tell you when it's done
20:36:44  <isaacs>(for those following along at home, this is re https://github.com/bahamas10/node-manta-sync/issues/4 )
20:36:47  <isaacs>mcavage: ^
20:40:40  * bahamas10quit (Ping timeout: 264 seconds)
20:43:51  * bahamas10joined
20:51:43  * chorrellquit (Quit: My iMac has gone to sleep. ZZZzzz…)
20:52:50  * abraxasjoined
20:57:19  * abraxasquit (Ping timeout: 246 seconds)
21:10:12  * mamashpart
21:43:48  * tim_smartjoined
21:44:05  * chorrelljoined
21:44:09  <isaacs>yunong: tim_smart here is looking for mill.
21:46:09  <yunong>tim_smart: happy to chat about mill!
21:46:40  <isaacs>21:42 < tim_smart> My current stack is haproxy, thalassa, redis, mongo, node
21:46:41  <isaacs>21:42 < tim_smart> Just need to add in some sort of monitoring service and log aggregation
21:46:50  <isaacs>so, we don't have parsers for all those.
21:46:55  <isaacs>and a lot of teh code is hackathon-quality
21:47:02  <yunong>no we don't.
21:47:09  * mamashjoined
21:47:18  <isaacs>but i'm wondering how far you'd get with just the rotater, and mill grep?
21:47:28  <yunong>the code is open source — and we will definitely by making it better in time.
21:47:30  <isaacs>oh, i guess it's not monitoring, either
21:48:42  <yunong>he could get started with our mill hackathon code as an example.
21:49:01  <yunong>https://github.com/joyent/mill
21:49:39  * AvianFlujoined
21:52:30  <isaacs>yunong: tim_smart is here, you can address him directly :)
21:52:43  * tim_smartis looking at the source
21:53:15  <isaacs>tim_smart: also, there's a few pretty decent for-pay services (technically mill is for-pay, since you use your manta account)
21:53:20  <isaacs>tim_smart: but they tend to be really expensive.
21:53:42  <isaacs>tim_smart: the one that people seem to really like, but it is pricy, is Splunk
21:53:50  <tim_smart>We are currently hosted with digital ocean
21:54:07  <yunong>tim_smart: and you're not going to get the near real time latency you'll get with mill. And that's because we're using Manta, which gives you in-situ compute.
21:54:11  <isaacs>mcavage: Honest question: "size" or "length"?
21:54:21  <isaacs>mcavage: Does it matter which one I pass into mput?
21:54:30  <mcavage>you mean programatically?
21:54:33  <isaacs>mcavage: it looks like the examples all use "length", but the info i get OUT of manta is "size"
21:54:36  <isaacs>mcavage: yeah
21:54:37  <mcavage>(mput is the CLI)
21:54:40  <isaacs>oh, riht
21:54:47  <isaacs>var mput = myMantaClient.put
21:54:53  <isaacs>.bind(myMantaClient)
21:54:54  <isaacs>:)
21:55:03  <mcavage>size
21:55:06  <isaacs>ok
21:55:11  <isaacs>so, "size" is canonical
21:55:12  <isaacs>yes?
21:55:20  <mcavage>either size, or just set {headers: {content-length: 123}} directly
21:55:26  <mcavage>yes, it's just sugar over content-length
21:56:16  <isaacs>kewl.
21:56:58  <isaacs>thanks
21:59:13  * mamashpart
22:33:18  * iraquit (Quit: Computer has gone to sleep.)
22:33:53  * irajoined
22:39:42  * elijah-mbpjoined
22:43:48  * iraquit (Quit: Connection terminated.)
22:57:04  * paulfryzelquit (Read error: Connection reset by peer)
22:57:04  * papajuansquit (Read error: Connection reset by peer)
22:57:34  * papajuansjoined
22:57:42  * paulfryzeljoined
23:24:35  * papajuansquit (Read error: Connection reset by peer)
23:25:00  * papajuansjoined
23:27:39  * AvianFluquit (Ping timeout: 272 seconds)