As analysts, we have a nasty habit of looking to past failures and assuming they are indicative of future events. However, we are often proved wrong by one company that manages to do it right. Consumer smartphones were all stupid until the iPhone. Tablets were a joke until the iPad, which was basically a Netbook that replaced the keyboard with touch technology. The Netbook was even considered a failed product in mature markets (it did OK in emerging markets for years though). Electric cars were a joke for a century until Tesla showed it could make a good one. And until the Kindle, anything with a screen smaller than 10” was a door stop.
In the world of failed ideas, though, none are more infamous than the dual boot system. Yes, folks still use it, mostly on Macs running Windows, but the graveyard for products that have shipped with this feature (mostly Windows notebooks running a Linux variant) is much larger than the market has ever seen. At least so far.
Well apparently, a number of vendors are going to try this again soon and my old friend Tim Bajarin and I think one could finally get dual booting right. Let’s talk about what that product might look like and why the attempt should concern you.
Why Analysts May Not See It
As analysts, we are trained to look back at our experiences and analysis and then project that knowledge on the future. This makes it very difficult for us to predict a success in an area like dual booting, which has mostly been about failures. The way you anticipate a success is to assume success and then think about what was necessary to create it. That’s not particularly hard to do once you have a successful product like the iPhone because you can go back and see what actually happened. But for a product that doesn’t exist yet, it requires a great deal of creative thinking and analysts tend to be more like engineers--we tend to be creatively challenged. If we weren’t, we’d likely be in more creative lines of work. So you’ll mostly see analysts, and the reporters we talk to, talking about why these products will fail. But in the area of future dual booting products, I think we may be wrong in that assessment--either this time or the next.
A dual boot system is a PC (and tablets are really PCs) that can run two OSes. On paper, this has always looked far better than in reality because the extra OS reduced limited storage capacity, generally could only run if the primary OS wasn’t running, and had a tendency to do ugly things like corrupt open files because the two OSes didn’t talk to each other. The benefit was that you could gain access to applications that ran on two different platforms from one system, and there have always been advantages unique to one platform over another.
Now ARM can run a version of Windows and Intel’s Atom and Haswell processors showcase battery life that is close to ARM. This means you no longer have to run dual processors. In addition, we’ve been doing virtualization for some time on servers running Windows and Linux side by side without issue, and while there is a slight performance hit, processor performance has continued its rapid advancement so that people coming from older systems likely wouldn’t notice. Finally, the Android platform has nearly as many apps as the iOS platform, making it a stronger alternative to Windows on the desktop than Linux has ever been. So the benefits to dual booting between Windows and Android on a tablet are far stronger.
The Key Is the User Experience
And marketing, actually. If the user experience is such that the user doesn’t actually know they are running Android and Windows (it just feels like they have access to Android apps while in Windows, or Windows apps while in Android), this will fly. Granted, the advantage will have to be well marketed and few companies outside of Samsung, Apple and recently Microsoft, market at the needed level, but they do showcase OEMs that can fund such programs. But given the right user experience and no major price penalty, a Windows tablet system that could seamlessly run Android apps, or an Android tablet that could seamlessly run Windows apps, should be preferred over peers. And, something to think about, an Apple system that could do both, if attractively priced, could be the most attractive of all—but right now, only Apple could do this and I won’t hold my breath.
Wrapping Up: Security Concerns
While such a system could be amazing, several security concerns come to mind. Both OSes would need equivalent security technology and that may not exist. More important, malware could be running on the guest OS and probably wouldn’t be identified by any antimalware product, and a network access product might see only the Windows OS and not even notice Android, which could be compromised. In other words, a system running both OSes and being used to do hostile things might bypass several of your security protections, which would otherwise prevent them from getting on your network. So you may want to consider blocking these systems as a class (actively scanning for this capability) before you can mitigate the risk. This might be a wise initial move, because some of you will likely be surprised by what these systems can do, and not in a good way. The good news is, though, that generally just because someone can do it right doesn’t mean they will and it is more likely firms won’t think through the user experience and necessary marketing before launching, so I think you have some time.
In the end, I think we are very close to having PCs in a variety of form factors that can access apps from several platforms simultaneously and transparently to the user. Unfortunately, I also think we should block them until our security technology can catch up to the related risks.