Discussed at London F2F; serious doubts about viability based upon testing at TAG F2F.

😠

@torgo to chase up current status of WebRTC, ORTC with Adrian.

Discussed in NYC; Dan still coordinating.

Discussed at April 2015 f2f - mostly considering rechartering - Dan to report back at next telco

Discussed in SF; still didn't work talking to Fluffy.

Hmm.

Had general discussion of re-charter. @torgo to report back.

This issue is a bit vague -- what about WebRTC? Charter? Technology? Privacy?

Related WebRTC issue. https://github.com/w3c/webrtc-pc/issues/179

Current topics of discussion at Berlin F2F are: WebRTC rechartering and privacy issues.

Proposed charter: http://www.w3.org/2015/06/webrtc-charter.html

As discussed at the Berlin f2f: if parties (e.g. ad networks) are using WebRTC technology to aid fingerprinting and not in service of actual services delivered over WebRTC then maybe this is an indication that an additional permission step should be required.

One threat model discussed: in the use of e.g. Tor browser or VPNs by activists who may want to conceal their source IP, WebRTC may enable attackers to determine those IP addresses.

relevant: https://tools.ietf.org/html/draft-ietf-rtcweb-security-arch-11#section-5.4

Tor disables webrtc at FF compilation time currently but are thinking of re-enabling it with IP address hiding: https://trac.torproject.org/projects/tor/ticket/16221 (really interesting idea to use hidden services for ICE!)

Agreed today: Mark will report back on the privacy issue. We have no substantive feedback on the rechartering.

Picked up at Boston f2f: https://pad.w3ctag.org/p/09-16-2015-minutes

https://wiki.mozilla.org/Media/WebRTC/Privacy

http://www.w3.org/2015/09/09-webrtc-minutes.html#item06

https://www.w3.org/2011/04/webrtc/wiki/images/d/da/WebRTC_IP_Address_Privacy.pdf

@mnot to provide a summary on the next call.

Discussed 2015-09-23 telcon follow up at TPAC

Demo of local IP address discovery: https://diafygi.github.io/webrtc-ips/

Potential positive use cases include stuff like https://www.sharedrop.io/

Discussed at Melbourne f2f: https://github.com/w3ctag/meetings/blob/gh-pages/2016/01-mel/agenda.md

Will continue to revisit this issue and get updates from WebRTC to assess potential privacy impact.

Some data: https://freedom-to-tinker.com/blog/englehardt/the-web-privacy-problem-is-a-transparency-problem-introducing-the-openwpm-measurement-tool/

A preliminary analysis [5] of a 50,000 site pilot measurement from October 2015 suggests that WebRTC local IP discovery is used on the homepages of over 100 sites, from over 20 distinct scripts.

Chrome mitigation pref: https://chromium.googlesource.com/chromium/src/+/7c98bab02128ae9fdc4fcc9a9df38588af86290e%5E!/chrome/common/pref_names.cc

Firefox discussion: https://groups.google.com/forum/#!topic/mozilla.dev.media/L6Rx9ubSjMc

Ref: https://github.com/w3c/webrtc-pc/issues/179 it looks like they have closed this issue. @dontcallmedom can you elaborate here and explain how the issue was resolved so that we can close this issue off please?

The RTCWeb IP Handling draft describe the strategies that browsers are invited to adopt to avoid revealing sensitive IP addresses.

Essentially:

• by default, only "safe" IP addresses get exposed; this includes so called "private" IP addresses (e.g. 192.168.0.1) since in practice they don't reveal much, but not hidden-by-VPN IP addresses
• once camera/mic access has been granted, more IP addresses get exposed to find more network efficient paths
• browsers should provide a mechanism to never reveal any other IP address than the public one

Discussed at IETF 95; minutes here (search for "IP-address handling").

Do we need to follow up at all here?

We can get someone (EKR or MT) to come in and update us in Stockholm on Thursday, if wanted. I'll need a heads up to ask them soon, though.

EKR presented at the Stockholm f2f, including new IETF Draft intended to address this issue.