14:56:25 RRSAgent has joined #dap 14:56:25 logging to http://www.w3.org/2017/02/23-dap-irc 14:56:27 RRSAgent, make logs world 14:56:27 Zakim has joined #dap 14:56:29 Zakim, this will be DAP 14:56:29 ok, trackbot 14:56:30 Meeting: Device and Sensors Working Group Teleconference 14:56:30 Date: 23 February 2017 14:56:35 Agenda: https://lists.w3.org/Archives/Public/public-device-apis/2017Feb/0006.html 14:56:40 fjh has changed the topic to: das https://lists.w3.org/Archives/Public/public-device-apis/2017Feb/0006.html 14:57:00 Chair: Frederick_Hirsch 14:57:49 Present+ Frederick_Hirsch 14:58:00 Regrets+ Dominique_Hazael-Massieux 14:58:18 Regrets+ Mikhail_Pozdnyakov 14:58:33 Topic: Welcome, scribe selection, agenda review, announcements 14:58:37 wanming has joined #dap 14:59:17 Regrets- Mikhail_Pozdnyakov 14:59:27 Present+ Mikhail_Pozdnyakov 15:01:34 anssik_ has joined #dap 15:02:59 sure 15:03:11 ScribeNick: anssik_ 15:04:25 topic: Minutes approval 15:04:26 Topic: Minutes approval 15:04:42 Approve minutes from 9 February 2017 15:04:42 https://lists.w3.org/Archives/Public/public-device-apis/2017Feb/att-0003/minutes-2017-02-09.html 15:04:43 proposed RESOLUTION: Minutes from 9 February 2017 are approved 15:04:53 RESOLUTION: Minutes from 9 February 2017 are approved 15:05:10 Topic: Wake Lock API 15:05:34 s/sure// 15:05:36 Present+ Tobie_Langel 15:05:40 Andrey not on call, but noted on email “I have integrated the comments into the spec, including the Privacy & Security section. My plan is to move on with the implementation in Chromium." 15:05:45 Present+ Anssi_Kostiainen 15:05:59 topic: Generic Sensor API 15:06:04 Present+ Wanming_lin 15:06:30 fjh: I don't understand the frequency thing, privacy I haven't read yet 15:06:52 haven’t read privacy papers shared on list re side channel attacks 15:07:05 tobie: frequency, implementers and security teams at implementers tell us the higher the freq, the more privacy and security risks 15:07:25 ... consider if you have high freq, can figure out what someone is typing on their (onscreen) keyboard 15:07:48 ... want to strike a balance between features and security and privacy risks 15:08:39 ... extra point to make, contrary to what opening access to mic or camera, the understanding of the user is not yet there what the risks are for the motion sensor 15:08:52 ... more responsibility on the user agent vendors 15:09:08 ... if they expose too high a frequency, may expose users to risks they're not aware of 15:09:54 ... question, can we find a frequency that strikes the balance, after which things would pivot 15:10:21 ... answer seems to be, not much research done on that specific area 15:10:53 ... no single paper focusing on how the increasing frequency impact threats 15:11:01 ... already pretty bad risks at around 20 Hz 15:11:32 ... hope we're able to allow higher frequency than 60 Hz, because it opens up important use cases 15:11:59 papers https://arxiv.org/abs/1602.04115 and https://arxiv.org/abs/1611.03748 15:12:15 ... spec makes a distinction between polling frequency and reporting frequency 15:12:23 ... reporting freq more problematic 15:12:59 ... one solution to say, you can increase polling frequency, but you still won't provide more than 60 samples / second 15:13:11 ... caters for the low latency use case, VR and AR 15:13:38 ... Navisense folks we're talking to doing indoor position with sensors, they actually need more than 60 samples / second 15:14:24 ... examples use case 60 samples per second on very low latency, poll 120 times a second but report only 60 samples when reaching rAF 15:14:34 ... reduce latency without exposing more data 15:14:36 Present+ Tobie_Langel 15:14:42 ... that's what VR typically does 15:15:03 ... or that was the state of the art (in VR) in 2016 15:15:40 ... it seems from what I'm looking at, VR is moving to more complex approaches, for neck position, for example 15:16:07 ... steps 1) identify use cases and requirements 15:16:27 ... 2) take them to browser security folks 15:16:49 GitHub digest: https://lists.w3.org/Archives/Public/public-device-apis/2017Feb/0004.html 15:18:51 ScribeNick: 15:19:01 ScribeNick: fjh 15:19:32 tobie: could also use permissions API to set levels, but that might be very hard to explain to end users 15:20:00 mikhail: event performance matters so too high a frequence would break things 15:20:48 tobie: need to discuss use cases, e.g. indoor navigation, VR, all ok with tying reporting of data to animation frames, so never get higher frequency of polling 15:21:04 … get 2-4 samples with every animation frame 15:21:20 mikhail: animation rate can also be flexible 15:21:24 tobie: another issue 15:22:02 … buffer question - what if buffer is ready to accept 2 readings, but get 5 readings during that time - more to figure out 15:22:21 ScribeNick: anssik_ 15:22:40 tobie: all use cases point it is OK to funner all readings through rAF 15:22:49 s/funner/funnel/ 15:23:36 ... polling freq you provide to the underlying layers 15:23:52 ... reporting freq, is freq at what point you bring this information to the JS layer 15:24:20 ... given these things happen on separate threads and you don't want to block 15:24:25 ... must poll super fast 15:25:59 ScribeNick : fjh 15:26:13 tobie: webVR has similar issue, original occulus paper 15:26:23 … using off the shelf sensors 15:27:20 … key problem was tatency, since people get sick from latency, traditionally combo of gyroscope and others, different values 15:27:56 … use 3D vector + acceleration, only data points used in v1 15:28:16 … calculate where head would be positioned in next frame 15:29:08 … issue is gyroscope drift over time 15:29:28 … correct with magneometer which is slow and imprecise 15:30:03 … poll gyroscope at 1kHz and poll magnetometer at 200Hz and combine to fix drift when fast head movement 15:30:35 … now can also track movement in room, using external sensors in room 15:31:20 … polling at 1000 per sec, using 60 per sec, to get low latency 15:32:17 ScribeNick: anssik_ 15:33:13 tobie: want to create dedicated MotionSensor space, collect use cases for it 15:33:23 s/space/spec/ 15:33:36 ... derive requirements, and design an API that works for the security people 15:33:55 ... key use cases are WebVR and AR, indoor navigation using motion sensors 15:34:25 ... input from people doing pedometers would be useful, also audio 15:35:09 ... report at rAF max frequency, and complement it with buffering system 15:36:21 mikhail: experimenting with Orientation sensor that provides alternative representations quaternion and matrix 15:36:36 tobie: agree that's a good direction 15:36:37 mikhail: to simplify use, avoid mathematics in Javascript 15:37:02 s/matrix/matrix, to simplify use, avoid mathematics in Javascript/ 15:37:18 s;mikhail: to simplify.*;; 15:37:51 topic: HTML Media Capture 15:37:56 https://lists.w3.org/Archives/Public/public-device-apis/2017Feb/0009.html 15:38:14 ScribeNick: fjh 15:38:34 anssik_: people wanted to be able to use front or rear camera, added, reviewed by implementers 15:38:45 … in process up updating test suite and implementations 15:39:17 … implementation reviews from google and apple implementers 15:40:11 topic: Any Other Biz 15:40:20 Topic: Other Business 15:40:22 (none) 15:40:28 s/topic: Any Other Biz// 15:41:15 rrsagent, generate minutes 15:41:15 I have made the request to generate http://www.w3.org/2017/02/23-dap-minutes.html fjh 17:15:35 Zakim has left #dap