"Closed can beat open," writes Tim Wu in a paradigm-unpacking story for the New Yorker, "but you have to be genius."
We're told (and we've published) stories about how open systems are superior to their closed counterparts. But not always, Wu says. First, we need to understand how a company can be closed or open. Wu counts the ways.
- How permissive a company is connecting with other brands: Anyone can make a Linux-running machine, while the Apple store will never carry a Kindle
- How impartially it treats itself and competitors: Firefox treats (almost all) sites equally, Apple always treats its products better
- How transparent products and processes are: Open source makes every line of code available, while Google won't share its search code.
Back in the day, everyone wanted to be closed, vertical, integrated. Andrew Carnegie integrated the aspects of steel production, the old Hollywood studios did the same for filmmaking, and the AT&T monopoly closed telephones. But as tech accelerated, openness started to hold sway.
Wu writes that Windows beat Apple's and IBM's operating systems because it ran on any hardware setup, and most of the '80s to aughts winners like Dell, Palm, and Netscape were open--including the Internet itself, which eclipsed AOL.
Open systems are more resilient--or even antifragile--because they distribute human error. Open systems excel because no one person can wreck the product, plus they can crowdsource, or, in New Yorker-ese "take advantage of collective, voluntary contributions of the masses."
Wu argues that Apple beat the "open beats closed" rule because they had Steve Jobs: a genius-dictator, "a corporate version of Plato’s ideal: the philosopher-king more effective than any democracy." The concentration of power works, in other words, if you're right (most of the time).
And then Jobs left us, leaving us with the question of how open or closed to be.
Wu has a few points to guide you to the answer:
- There's a tradeoff between open and closed, so don't be extreme in either direction
- Open systems, even Wikipedia (or Github), need some point of control
- The better your vision and design, the more closed you can be
- If you can be as prescient as Jobs, try going closed, but if you're merely mortal, the "economics of error suggest an open system is safe."
- Or "rely on this test," Wu suggests, "look in the mirror, and ask yourself, Am I Steve Jobs?"
[Image: Flickr user Ben Stanfield]