One of my regulars pointed me at Is Android Evil?, an article by one Andreas Constantinou which purports to be a brave and hardhitting contrarian take on Android.
I read this, and I’m asking myself “Wow. How many different ways can one guy be wrong in the same article?” Particularly entertaining, and the main reason I’m bothering to rebut this nonsense, is the part where Mr. Brave Contrarian Guy claims that the success of Android has nothing to do with open source and then lists three “key factors” of its success in every one of which open source is critically involved.
“With the unprecedented success of the iPhone and the take-it-or-leave-it terms dictated by Apple to network operators, the carriers have been eagerly looking for cheaper alternatives…” Damn straight they have been. And two of the key advantages of an open-source cellphone stack are: (1) avoiding per-unit licensing costs, and (2) you get to leverage the fact that somebody else spends most of the NRE. One wonders how Brave Contrarian guy thinks these could ever be duplicated by a closed-source OS.
“Android provides the allure of a unified software platform supporting operator differentiation at a low cost (3 months instead of 12+…)” Yes, it does. It’s not like there’s any secret about open source cutting time-to-market; embedded-systems vendors for things that aren’t cellphones have been relying on this as a key part of their business strategies for years now. One wonders to what else Mr. Brave Contrarian Guy wants us to attribute this time-to-market advantage in the cellphone case. Are we supposed to think it was left under the telecomms operators’ pillows by the Tooth Fairy?
His third point is mostly repetition. “In other words, in an Android handset, most of the OEM budget goes into differentiation; compare that to Symbian where most of the OEM budget goes into baseporting.” Well, duh. This is somebody-else-paid-most-of-the-NRE again. Mr. Utterly Oblivious Contrarian somehow fails to notice the central reason that investment could be spread across multiple stakeholders in the first place. You don’t get competitors covering each others’ engineering costs unless everybody rationally expects to get more out of the pool than they put in, and that’s exactly the promise open-source development both makes and delivers on.
Mr. Brave Contrarian Guy then proceeds to list eight control points that he claims make Android “closed” even though the SDK is open. All of these miss the central constraint on Google, which is that if participation in Android doesn’t return more value to its development partners than they’re investing they can fork the codebase. His failure to grapple with the implications of this is even funnier since he notices that China Mobile is actually doing it.
You have to think game theory about the second-order, third-order, and nth-order effects of irrevocable strategic commitment to really get what’s going on here, something Mr. Brave Contrarian Guy seems unwilling or unable to do. Everybody knows that one of the constraints of the open-source game is that overcontrol leads to forking; because it’s so, Google’s business partners can form justified expectations about its behavior that enable them to make billion-dollar bets with much more confidence in the stability of Google’s behavior than they could have otherwise. These expectations, in turn, create future value for Google in ways I’ve previously described…and so on, out through several more layers of strategic minimaxing by both Google and its partners. The bottom line is that these selfish agents can form a stable cooperative equilibrium that wouldn’t be stable without the open-source commitment.
Against this background: Wow. So Google has process, partnership-agreement, and trademark constraints that push against any attempt to fragment the platform. How shocking! How unexpected! How courageous Andreas Constantinou is to write about them! I’d say the real question here is how anybody this dim manages to operate a keyboard, except I don’t actually think Constantinou is as stupid as he appears. What he’s done here is adopt the rhetorical posture of Mr. Brave Contrarian as a way of sexing up a business-case analysis that would otherwise rather boring and obvious. Well, except for the part he doesn’t get: those control points create value for the Android OEMs, too by stabilizing the cooperative game that all parties are playing.
His failure to get that would be OK, because the factual material about specific Google control points isn’t completely useless even without the insight that they’re game stabilizers, if Constantinou hadn’t felt he needed to set up his rhetorical ploy with a quite idiotic series of claims about open source being irrelevant. But Google knows better and so do its partners — and if they somehow managed to forget that, the China Mobile fork would be there to remind them.