Fwd: RfC: Wide review of Generic Sensor and Ambient Light Sensor API

Dear all, especially SSN folks,

This looks of direct relevance to our work here...

Phil


-------- Forwarded Message --------
Subject: RfC: Wide review of Generic Sensor and Ambient Light Sensor API
Resent-Date: Tue, 30 Aug 2016 10:23:40 +0000
Resent-From: chairs@w3.org
Date: Tue, 30 Aug 2016 12:23:22 +0200
From: Dominique Hazael-Massieux <dom@w3.org>
To: janina@rednote.net, mandyam@qti.qualcomm.com, runnegar@isoc.org, 
tjwhalen@google.com, Brad Hill <hillbrad@fb.com>, dveditz@mozilla.com, 
rstreif@jaguarlandrover.com, peter.winzell@melcogot.com, 
adrianba@microsoft.com, chaals@yandex-team.ru, tink@tink.uk, 
Joerg.Heuer@siemens.com, ratan@microsoft.com, stearns@adobe.com, 
appelquist@gmail.com, peter@linss.com
CC: chairs@w3.org <chairs@w3.org>

Dear APA WG, Geo WG, Privacy IG, WebAppSec WG, Web Platform WG, Auto WG,
Wot IG, CSS WG, TAG chairs,

The Device and Sensors Working Group just published two updated Working
Drafts of:
* the Generic Sensor API
   https://www.w3.org/TR/2016/WD-generic-sensor-20160830/

* the Ambient Light Sensor
   https://www.w3.org/TR/2016/WD-ambient-light-20160830/
on which the group is seeking wide review, especially from your groups.
Preferably, we would like to receive your reviews before TPAC (September
19) so that we can use our meeting there to make progress on the
received feedback. We will not request transition to Candidate
Recommendation before mid-October, so any feedback received by then will
be taken into account as part of the "wide review" process.

For context, the Generic Sensor API is a an abstract API that can serve
as a basis for any kind of sensors; the Ambient Light API is the most
advanced of our APIs based on that framework, and demonstrates how a
concrete API is built on top of the sensor API.

Other APIs to be built on top of the Generic Sensor API include
accelerometer, magnetometer, gyroscope, proximity [1], but we're not
specifically calling for wide reviews on these yet.

For sake of clarity, we're calling the following groups for reviews with
the following expectation:
* for APA WG, to determine if the specific ambient light API under
review brings accessibility concerns, and more generally, if the generic
sensor API needs to highlight generically applicable accessibility
concerns with exposing sensor data

* for WebAppSec, we are seeking likewise review on the general impact on
security of exposing sensors data, the specific risks associated with
Ambient Light; we are also seeking input on the generic security
considerations in the Generic Sensor API, in particular the relationship
with secured contexts, permissions and permission delegation

* for the Privacy IG, we're seeking input on the privacy risks that are
generally applicable to sensor data (generic sensor API), and the ones
that are specific to ambient light, and whether the current privacy
considerations satisfactorily cover them

* for the Geolocation WG, we're seeking input based on the group's
experience with the first widely-deployed sensor APIs (geolocation and
deviceorientation), to see in particular if the generic sensor API
appropriately covers some of the issues that emerged in this prior work

* for the Web Platform WG and the TAG, we're seeking overall review of
the shape of the API, its integration with the rest of the platform and
its usage of WebIDL (noting that the TAG has already provided input to
the generic sensor API)

* for the Auto WG, while we understand that the group is moving away
from a WebIDL-based approach to gathering sensor data, a review of the
generic sensor API as a potential wrapper around the new approach
explored by the group might usefully expose limitations of our specification

* for the WoT Interest Group, we're seeking feedback on whether the
Generic Sensor API is applicable as is (and if not, what would need to
change) in the contexts the group is exploring for communications with
sensors

* for the CSS Working Group, we're specifically seeking feedback on the
Ambient Light API, since it exposes a similar functionality as the one
that was at some point considered for CSS Media Queries (as "luminosity"
or "light level")

If you determine your group does not need to provide input, we would
also appreciate if you could let us know.

Thanks,

Dominique Hazael-Massieux, Staff Contact for Device & Sensors Working Group
for Frederick Hirsch, Chair

1. https://www.w3.org/TR/2016/WD-ambient-light-20160830/

Received on Tuesday, 30 August 2016 10:38:34 UTC