I'm at Embedded World in Nuremberg, Germany, where the EETimes/Embedded 2013 survey results we're going to present Wednesday indicate that Android is ascendant. Spoiler alert: It's first among operating systems embedded developers are considering using in the next 12 months.
That's significant, and it gives me an excuse to dive into a discussion of the highest-profile Android application extant. That would be as the OS embedded in smartphones and tablets, particularly those made by Samsung.
My objective — or, my excuse — in talking about the market position of Samsung's devices is to show that embedded technology is not an island divorced from real-world forces. Actually, it's more like an iceberg, at least insofar as users are concerned. That's because that can't “see” it, like they think they can with Windows on a PC. And, if they could see it, they wouldn't care, because they're not developers.
One might assume that Android has caught fire because it's free. There's clearly some truth in this, because another embedded OS that's highly popular is FreeRTOS. However, if lack of licensing costs were the sole determinant of market success, most laptops would run Linux.
They don't. Nor is Samsung the only Android smartphone vendor. (HTC, Motorola, LG, and Sony are among the others.) But Samsung is on everyone's lips, and it's that only company that can seemingly catch Apple.
Snow snarls Nuremberg, Germany the day before Embedded World opens.
It's only recently thus. Remember when Samsung, LG, and Hyundai were the s**t? No, I don't mean as per the phrase's current colloquial meaning. I'm talking early 1980s — at the time, LG went by the Goldstar brand name — when all three were synonymous with cheap in price, and low in quality, too.
Fast-forward 30 years and we've got a collection of chaebol capable of competing with the best. (I was gonna write, alliteratively, “trio of top-notch keiretsu,” but the “k” word refers to Japanese conglomerates, not South Korean.) I'd add that the 180-degree reversal in brand perception, which all three have accomplished, is highly unusual. Typically, tainted names fade away.
Instead, we see Samsung beginning to take a bite out of mighty Apple. Samsung's Android-based Galaxy line of smartphones has caught fire. True, its tablets are less potent competition to the iPad than Samsung's screen-stretched handsets are to the iPhone. And it's also true that Samsung unsportingly reinstantiated the iPhone's user interface. But Apple's anger about that doesn't have a PARC mouse to stand on. And, hey, what's a little patent litigation amongst competitors?
It's not precisely fair to say that Samsung and Apple have swapped trajectories, and that the latter has lost its case of cool. However, I am indeed arguing that the Android part of the Samsung product equation is a key element behind Galaxy's popularity. There's a lesson therein for the embedded arena. (I'll get to what that lesson is later.)
Thus, all Google Maps routes lead to the question: Is Samsung's strength sustainable? On the flip side, will Apple become Cooked without Jobs. (And will that be good or bad?)
Quick sidebar: There's also Nokia, poor Nokia. For years, it produced what were inarguably the hippest devices of their day. For example, the E95 from 2006 had stellar (actually, silver) industrial design, a sleek form-factor, and forward-looking features, notably the ability to connect to a projector. It was perhaps the first phone which a corporate road warrior could take on a trip and leave the laptop at home, as long as it was a short trip. (If I did write so myself in InformationWeek cover story in October, 2008. See “Is the Smartphone Your Next Computer?“)
Unfortunately, Nokia's high-end products never seemed to get much push in the United States, where the vendor was known mainly for its commodity feature phones. We all know what happened in the intervening years. Nokia sagged, bagging the Symbian OS, and in 2011 got into bed with Microsoft.
Few thought that Nokia and Windows Phone could turn the two companies' fortunes — both separately and combined — around. And maybe they haven't, sales-wise. But one can't deny that, after a decade of not-quite-successful steps — through CE, Pocket PC, and Mobile — Microsoft finally got its smartphone OS right. Indeed, the tile-based Windows Phone user-interface is the first “think different” UI in ages. It's great, and Nokia has paired it with solid hardware.
Sadly, it's the disruptive sibling at grandpa's dinner table — those elders being Apple and Samsung — and no one will pass it the salt. (Blackberry has been banished to the children's table.)
Isn't it ironic that, when Microsoft finally does what everyone's been prodding it to do for years — i.e., something original — it falls on deaf virtual ears? Oddly, a similar fate seems to have befallen Microsoft's Azure cloud offering, which makes one wonder whether timing and marketing way outweigh technology as the determinants of success. (Paging Sony Betamax.)
Yet in the case of Windows Mobile, sales belie influence. We're seeing the OS's tile metaphor — which originated in the under-adopted Windows 8 PC OS — spreading like wildfire in Web site designs. That's a good sign, though a good sign of what I can't tell you.
OK, back to Samsung and Apple. It's highly probably that the “cool” factor attendant to Galaxy stems first from its large screen and second from the fact that Android Apps carry the imprimatur of Google. Also, there are tens of thousands of them. So Android doesn't suffer from the apps scarcity which has plagued Blackberry. Nor are potential users off-put by a perceived association with the Borg (http://www.urbandictionary.com/define.php?term=borg )
In conclusion, am I saying that Android's advance is an accident? Not at all. Actually, it's a case study in how to build an ecosystem. If you — or Google — stoke it properly, eventually it reaches a tipping point after which it becomes ubiquitous. That's about where we seem to be now.
But the path beyond ubiquity leads, dangerously, to commodity. So I'd worry more about Samsung than Android. The analogy I'd make is to ARM and its ecosystem of licensees. It's not a perfect comparison, because ARM hung around (some would say, hung on) for over 20 years before it really caught fire. In contrast, Google seems more of a mind to give any idea six months, and if it hasn't captured 100-million users, it's back to the gym to think of something else. (I'm kidding, but not completely.)
Anyway, so, to use the more recent ARM Cortex cores as an example, the licensees which brought these to market originally all had something original. Now, all they've got are commodity parts. This means that the licensees are scurrying to add differentiation around their cores — with features like flexible I/O, SoCs with multiple cores, highly configurable FPGAs with Cortex computation engines, etc. And while the licensees struggle to break from the pack, ARM sits pretty collecting licensing fees.
I can't say whether any other Android vendors will be able to dethrone Samsung. (HTC, for one, hasn't.) Nor can I predict whether Apple will thrive post-Jobs. What is clear, however, is that the embedded OS one selects is important.
More important still is that one should not infer from the smartphone arena that any of the above has any bearing on which embedded OS you choose for your application in a different vertical. This means that Micrium, VxWorks, Integrity, MQX, TI-RTOS, QNX, etc. all have their place. (As does the perennial “roll your own,” even though it's not 1992 anymore.)
More in future posts, after we roll out the EETimes/Embedded 2013 embedded market study.