Privacy report on sensors, for generic sensors API.

Dear all!

I am working on a sensors privacy (impact, risk, ...) assessment for a
while now. And I think now it has little sense to withhold it for any
longer, as most of the work I did some time ago, anyway.

It is primarily intended for Devis APIs WG (DAP), with whom I have the
pleasure to work on the privacy aspects of sensors API.

I invite you to take a look on the document [1]. I hope it will be useful,
and I primarily hope this can be an appropriate starting input in privacy
considerations of sensors.
Often, as indicated in the PDF report, even perhaps far-fetched scenarios
are considered. Same for cross-device risks, where plausible scenario could
be pointed to.

As advised in private correspondence with (and by), Tobie Langel (DAP), it
would be good if specific pull(s) request(s) follow. I'll look into that
next.

Also of note. It is not included in the PDF (should it?), but I believe it
is worthy to require a secure (i.e. TLS) connection for having access to
sensors ('secure contexts') - all of them, generically and just like that.
I can't imagine a scenario where this could cause any issues, apart from
the need to set up a TLS, that is.

I also highlight my view and want to ask a question. Can W3C give
guidance/recommendation/note regarding the transparency UIs (sometimes
called "privacy user interface")? A method for a straight-forward
user-verification of: what/how was being used, how frequent, etc.

Please, enjoy ;-)


Best regards
Lukasz Olejnik

[1] http://lukaszolejnik.com/SensorsPrivacyReport.pdf

Received on Tuesday, 29 March 2016 09:50:29 UTC