Samsung Nixes Knox: The Android Security Saga Continues

After 18 months of going it alone and spending untold sums of money on development and marketing, Samsung is throwing in the towel on Knox.  Google is stepping up to take the lead on Android security.

The move by Samsung is not surprising, considering Samsung’s 24% drop in operating profit and 10% drop in sales for the three months ended June 30th. And despite a lot of hype, Knox market take rate is a miserable at <2%. Ironically, this comes just a few weeks after the U.S. Department of Defense approved the use of Knox

So why did Sundar Pichai, Google’s head of Android development, talk up Knox during his remarks at Google I/O. as if there is some new found co-development love between Samsung and Google –and that this collaboration is centered on Knox? Probably because Samsung is Google’s largest Android partner, and Google wants to help Samsung save face. Plus, it makes sense that Google will use whatever Marcom benefit Knox may have. Frankly, if I were in Pichai shoes, I’d do the same; I’d want to put the best spin on what clearly is a precarious situation.

What’s Google up to? As I understand it, Google has built its own organic container. It’s Samsung Knox-like. But it’s not Knox. It’s new. It’s not yet field-tested. Further, the initial specs of Google Android L specifications appear very anemic. In the release I’m expecting this fall, there appears to be no on demand VPN capability –and no per app VPN. To me, that means the “secure” container will hold apps and data safe and separate at rest, but not while in transit where consumer and biz apps will flow across the same secure connection into the enterprise. It also remains unclear how Google will integrate features and or product from its recent acquisition of Divide.

The idea of an organic security solution built in the Android OS is a good one. And I think Pichai is sharp and gets the enterprise space. (Many Google partners tell me he is very engaged and accessible.) Even so, it appears the road ahead will be long for Google. Why? Because the scene seems eerily similar to many of the false promises, the Android ecosystem has made to enterprise interests.

We await answers in at least two important areas:

Backwards compatibility is a known, unknown. Google says that Android L will support handsets with the Ice Cream release of Android and newer. The problem, of course, is that Android was built by a search company, for advertisers, and then distributed by telcos. All eyes are on Pichai to see how he navigates the various enterprise interests – internal and external – to fulfill his commitments to business users.

Distribution is complex and political. Don’t forget: OS versions of Android like Ice Cream and Kit Kat may be created equal, but they are not alike. Let’s image Google releases a generic and (to its credit) clean-build to the handset makers. Then, the handset makers have to butcher the generic release to meet each requirement by each telco, and by each handset model. As an example, I think Samsung maintains about 150 current images of each version of Android. That mess is what causes division and delays in the user base. Fragmentation and badly or staged synchronization of OS release is an unacceptable artifact in the enterprise market.

Granted, Google has been updating handset issues at a quicker pace – particularly when it comes to security patches, via Play Services –and so far, the telcos have not played spoilers. But remember: Google has not initiated a move to push an entirely new OS directly to users except to those who own Google’s telco independent Nexus brand devices. Keep in mind that there’s a big difference between updating a feature or security patch and producing an entirely new OS. OS updates typically up the Kernel and the radios. It will be interesting (and historical) if the telcos continue to stay out of the way.

That’s why I’ll be watching Pichai, Androids’ captain, to see how well he navigates these choppy waters. Whatever else, satisfying enterprise users is going to take time, and in the end it appears that Google still will not be able to offer an end-to-end security solution required by so many enterprise sectors and probably IoT. Perhaps I need to pinch myself, but is a natural synergy between BlackBerry and Google appearing on the horizon?”

This fall Google looks to be hosting a major event around the release of Android L that appears to include partners like Airwatch, Citrix, Good Technology, MobileIron and SAP .