Monday, January 12th, 2015

Recently there has been a lot of discussion about a noticeable decline in the robustness and reliability of software and services at Apple over the past two years (well, poor services has been a long-running thing at Apple). Marco Arment kicked off the latest round of angst with “Apple has lost the functional high ground”, a piece that he has some come to regret but which largely speaks the truth about the state of software at Apple (for a great discussion, listen to the latest episode of ATP). His diagnosis of the issue was that the marketing-driven schedule of annual releases of OS X and iOS is at fault. This blog post sparked off a number of replies from other bloggers.

• Federico Viticci called for more balance between adding new features and providing stability. Here’s his diagnosis:
  

  The reality is that it’s hard to come up with a single reason for Apple’s modern software woes. And the most likely explanation is that they’re the result of a series of trade-offs, deadlines, and wrong decisions that contributed to the overall product.

• Daniel Jalkut reviews the long history of angst over software quality at Apple. In other words, there’s nothing new here.
  

  A company like Apple, moving at a breakneck speed, will undoubtedly continue to give us plenty to obsess about, both positively and negatively.

• Joe Cieplinski strikes similar notes in Compromises Have Consequences. His diagnosis:
  

  You can’t stay way ahead of the curve and not introduce some bugs along the way. This decline in software quality is a side-effect of the current strategy. It’s a compromise Apple has made in order to reassure the general public that the company isn’t “doomed” without Steve Jobs.

• Guy English agrees with Arment that Apple is doing too much, too fast.
• John Gruber sees these issues as belonging to an era of flux primarily caused by the convergence of Apple’s different platforms:
  

  My hope is that the reliability issues we are seeing in iOS and Mac OS X in recent releases are largely the inevitable result of Apple going through numerous transitions simultaneously. Extensions, XPC, iCloud Drive, Continuity — these things require coordination between all three of Apple’s platforms (mobile, desktop, cloud). That what we’ve been seeing the last few years is this decade’s equivalent of the first few years of Mac OS X — rapid development and flux that precedes an era of relative stability and a slower pace of change.

• Glenn Fleishman doesn’t provide a diagnosis, but he does enumerate the many issues users are encountering with Apple’s software and services.
• Ashley Nelson-Hornstein, Dr. Drang, and Craig Hockenberry also chime in.

There are a couple names that I haven’t seen pop up in these discussions, but that might prove relevant for arriving at an accurate diagnosis of what’s causing software issues: Jony Ive and Scott Forstall. Since Forstall’s departure two years ago, Ive has consolidated even more power within the company. Though Craig Federighi is now in charge of both OS X and iOS, Ive’s design portfolio also overlaps with software engineering since he now leads the Human Interface teams. If you read Leander Kahney’s bio of Ive, you’ll know that Ive has more power than anyone in the company and that a series of executives have basically lost battles with him and left the company as a result (he seems like a great guy, but I wouldn’t want to be on the opposite of him in an executive face-off). I’m sure he and Federighi “collaborate” very well, but we all can be fairly sure who will have the edge in any potential conflict. Forstall and Jobs could provide a counterweight to Ive, but it’s unclear if anyone at the company can play that role now.

This is not to say that conflicts between executives are causing software quality issues at Apple. Or that Tim Cook’s era of collaboration, after years of Jobs’ practice of pitting executives against each other, is smothering dissensus and not allowing software issues to come to the fore (some speculated this might happen when Forstall was first ousted).

Rather it is a reminder that an industrial designer is largely leading the product decisions of a company that purports to provide solutions that equally integrate hardware, software, and services. You can see the consequences of this organizational structure in the issues that come up in the first generation of a new form factor:

• the iPhone 4’s antenna,
• the iPhone 5’s slate color chipping and its faulty sleep/wake button,
• the iPhone 6’s increased “bendiness” (say what you will about the validity of this one, the iPhone 6 is certainly bendier than the previous generation),
• the Retina MacBook Pro’s screen retention problems,
• the first generation MacBook Air’s failure on multiple levels as a viable product.

All these instances are evidence of a company where the hardware division is pushed to the limit by the dictates of the industrial design group. Software and services are similarly always playing catchup, even if it’s not as visible as the hardware issue. For example, in the iPhone 4S and the iPhone 5s, the camera software is much better able to cope with the diminished physical space given to the camera unit in each new form factor, compared to the photos that could be taken with the iPhone 4 and iPhone 5. Some of this is just the march of progress in technology – but imagine the quality of photos we would have in an iPhone today given the space available in the iPhone 4 form factor? This and battery life are but some of the compromises made by virtue of a form factor-first design approach. As for services, is it possible to imagine how little attention they get from Jony Ive or how much of a corporate priority they are? (And remember, design is “how it works,” so, yeah, this falls within Ive’s purview too).

Most likely, as Nelson-Hornstein and Gruber predict, things will settle down and a more normal standard of stability will return to Apple with the next release of OS X and iOS. iOS 7 was a huge visual change that entailed a wide-ranging cultural change at Apple, with divisions beginning to work together and secrecy between groups diminishing. iOS 8’s tremendous functional expansion begins to show the benefits of increased collaboration between divisions at Apple.

But how a company is organized largely determines what it is capable of doing. Org charts still matter. I have a feeling that so long as Ive has the most clout, software and services will always lag somewhat. Perhaps this will change as Federighi comes more into his own.

Till then, one suggestion: wait six months to update your OS after a major release. Do that every year and you’ll have a better chance at maintaining a year-long stretch of a stable OS. Oh, and don’t buy first generation Apple products. Wait for the second generation, or the “s” model for iPhones. I’m looking at you, Apple Watch…

---