Why A Hardware Root Of Trust Matters For Mobile
Even with mobile device management, enterprises still lack control over devices
September 24, 2013
As the IT industry grapples with the security implications of mobile devices, some experts believe one of the most important first steps it can take is to stop getting caught up in irrelevancies.
"We are lost in a conversation of mobile versus PC or phones versus tablets or whatever else, but that's not what's important," says Steven Sprague, CEO of Wave Systems, explaining that the really important piece is, "How are we going to manage multiple tenant trusted devices, and what are the basic foundation principles for that? Then you've got to stick to your guns. I don't care if they have the slickest marketing program under the sun -- we've got to continue putting on our glasses and calling out when the emperor has no clothes."
And one of the most important duds that mobile is missing, according to Sprague, is a standards-based hardware root of trust. Together with Dave Challener, security architect for Johns Hopkins University Applied Physics Laboratory, and Dan Griffin, president of JW Secure, Sprague discussed the deficiencies of mobile device technology in a panel earlier this month at the first annual Trusted Computing Conference in Orlando, Fla. The running theme in their discussion was the enterprise relinquishment of on-device control.
"Mobile is a step backward from a couple of perspectives," says Griffin, explaining that, first and foremost, the major mobile device vendors have not baked enough security features into their operating systems or provided the kind of development platforms that encourage developers to build security into their applications.
"Finally, the carriers and the implementers of these operating systems are super-nervous about providing system-level access to the device, but you can't do antivirus or other security without system-level access," he says. "So we're just in this weird state right now where, OK, we have all this fun stuff we could do to make a PC really locked down -- you just can't do that on a mobile device."
Challener views the state of things even more dimly.
"You look at mobile devices, and you see that you don't control the network, you don't control the hardware, you can't select hardware subsystems that are in it, you don't get to control when firmware is updated, you don't get to select the OS, and the app selection in an app store is uncontrolled," he says. "Boy, if I were an IT guy, I would be panicking."
However, mobile has done one very good thing for IT security: bringing the discussion squarely back around to the importance of device security.
"A device-centric view of the network is really useful," Sprague says. "The enterprise has been trying to ignore the device because devices are complicated and messy. And so we have control in the network, and hope and prayer in the device."
But control is the key word in device security; as things stand, there's no real control on the mobile device, whether it is owned by the employee or the enterprise. Take MDM, for example.
"You don't buy mobile device control software -- you buy mobile device management software," he says.
One of the biggest impediments today is the fact that at the hardware level, the device is either controlled by the carrier or the vendor itself. This is most visibly seen in the transition from iPhone to iPad as Apple got out of its single-carrier relationship with AT&T.
"As a carrier, AT&T controlled the iPhone with absolute power. They could shut it off at will, terminate service, and change the OS," Sprague says. "The brilliant maneuver by Apple was to take control of the initial hardware root of identity of the subscriber."
Nowadays, the only way to get full use out of the iPad is through that connection with iTunes, with Apple having ultimate control over the device and the ability to shut down its functionality remotely.
"The reason why a standards-based, independent hardware root of trust is important is that it allows someone else to take control of the device before the carrier," he says. "If you look at almost every use case and application out there, this is the fundamental capability that's being requested, even if it is being requested in a language that is not as clear as that."
Unfortunately, the real difficulty is convincing carriers or vendors to loosen their grasp of control. It is an issue of leverage and one that Sprague believes only one entity exists capable of wresting control away for the betterment of the industry.
"The only way we can wrestle control back from Verizon is through a requirement placed on the environment by a player strong enough to do that," he says. "The only player -- emphasis on the word 'only' -- is the U.S. federal space."
Have a comment on this story? Please click "Add Your Comment" below. If you'd like to contact Dark Reading's editors directly, send us a message.
About the Author
You May Also Like