SILOS AND PLATFORMS

Dear Linux Pro Reader,

Sometimes I'm amazed at how this column just slips out into the world and no one says much about it. One thing I do hear occasionally is that "It's refreshing that you aren't just summarizing the contents of the issue, which is what most welcome columns do." I do plead guilty to having a great many opinions, some of which I manage to share from time to time in this space. However, in this case, my opinion does point to an article you'll see in this month's issue.

One of the things the high-tech sector has witnessed in recent years is the return of restrictive silos. Today, especially in the FOSS community, we tend to remember Microsoft as a vast, intimidating monopoly, and they certainly were, but at least, they never tried to own the whole ecosystem of hardware and software together. The most extreme silos came before – like IBM and less successful competitors such as Honeywell and DEC. Microsoft actually came into its dominant position by artfully resisting IBM's efforts to impose the kind of control over the PC industry that it already had over the mainframe industry.

Apple, on the other hand, did try to control a complete silo of hardware and software, and they almost went bankrupt back in the 1990s. Ten years ago, which is around the time I started writing this column, we in the open source community believed that the battle against monopolistic hardware practices had already been won, and all we had to do was win the battle against monopolistic software practices.

However, the return of Apple and the rise of mobile technology has put proprietary control of the hardware/software environment back in the foreground. Others have tried to copy Apple's success, assuming that behind every victory is a formula, but the results have been quite mixed.

Counter-balancing Apple's success story is the Nokia/Microsoft failure story. The once-great phone vendor and the once-dominant software vendor formed an alliance for a hardware/software proprietary platform that was supposed to rocket them into competition with Apple, but the experiment was a total washout. According to some reports, Microsoft wasn't willing to adapt enough, and Nokia couldn't adapt quickly enough. Radical, mid-course corrections are difficult in any industry, but this story underscores the fact that a complete production chain, from hardware to app store, has a lot of moving parts, and it is pretty hard to get all those parts to line up perfectly. And, as the FOSS world has been saying all along, if you lock down control, you lock out innovation.

Android, on the other hand, really is challenging Apple for leadership of the smartphone market. The open nature of the Linux-based Android system leads to great flexibility, but one problem looms: The freedom to develop for Android makes it a popular medium for malware. The good news is, the flexibility and openness of Android lets the system adapt, and we are watching that process unfold right now.

As you will learn in this issue, the upcoming Android L release integrates several new features designed to crack down on security problems. These features include the Knox security environment (created by Samsung) and SELinux, originally created by the US National Security Agency and adapted extensively by Red Hat and other vendors. Do other companies contribute code to fix the problems with Windows Phone OS? Not really, unless Microsoft buys the company, as they eventually had to do with Nokia.

Maybe that's why Android is still standing and Windows Phone is still falling off the burning platform [1].