[saag] Ubiquitous encryption draft feedback - mobile case

["Smith, Kevin, (R&D) Vodafone Group" <[e-mail]>]

Hi Kathleen & Al,

Thanks for publishing this draft, which makes a lot of sense. I support approaches that allow network management to persist without breaching encryption or introducing any security/privacy weakness, and this paper provides a sound reference for such work.

I'd like to offer an additional (sub) section  on the particular case of traffic management for mobile networks, along the lines of:

"Bandwidth in cellular radio networks tends to be more volatile than in fixed networks. This is a result of variance in radio signal strength as a user moves around a cell, the rapid ingress and egress of connections as users handoff between adjacent cells, and sudden congestion at certain cells at certain times. Mobile networks account for this by queuing traffic according to its required bandwidth and acceptable latency, and hence spread the available bandwidth sensibly across users: for example, a user is unlikely to notice a 20ms delay when receiving a Web page, email or instant message response, but will likely notice video buffering or VoIP call jitter. The network manages the queue so that each user has an acceptable experience as conditions vary. Application and transport layer encryption makes the traffic type detection less accurate, impacting queue management."

Also section 4.1 highlights many similarities between Enterprise and a government-regulated mobile network.

A couple of minor typos in the Introduction:

"These efforts are necessary to improve end users expectation of privacy,"
s/improve end users expectation/improve an end user's expectation

"Many attackers and those that pose a greater threat are already using strong encryption and tools like TOR [TOR] to prevent active attacks from on their data streams."
s/from/

All best,
Kevin

Kevin Smith, Vodafone R&D