Unpacking the Effects of Apple’s Privacy Stance
In my previous piece, I detailed how I reached my current position regarding the need for Apple to relax their hardline privacy stance by deconstructing their faulty reasoning. However, I feel the need to go further and detail its effects on three key areas:
- Talent Attraction and Retention
- Product Progress
- Competitive Position
1. Talent Attraction and Retention
Apple’s poor ability to attract and retain artificial intelligence and services talent is the most serious effect. Apple rarely acquires AI or services companies; however, the odd occasion when they do, key staff depart soon after— often to work on competitors’ products:
- Siri Inc: Almost everyone. Many departed to create Viv, which Samsung acquired. Another key employee left to develop what became… Google Assistant. Finally, two engineers left to work on a similar cloud platform for GE.
- Chomp: The co-founder left, as well as others. She is currently a Director of Engineering at Google.
- Perceptio: The co-founder left to work on Google TensorFlow.
- Beats: Beats Music CEO left less than a year after acquisition.
- Cue: The co-founder, a former Director at Apple, has moved to Y Combinator. His company’s technology was the basis of iOS’ Proactive features.
And even when they somehow manage to attract major talent, they seem to depart alarmingly quickly:
3 December 2016
Before joining Apple [in May 2016], Yoky Matsuoka led technology efforts as a vice president at Alphabet Inc.’s Nest Labs. She is credited with developing the technology that helps Nest thermostats automatically adapt to things like environmental conditions and past usage. Matsuoka was a pioneer in robotics, having helped invent a lifelike robotic hand. She is also a co-founder of Google X, the lab that produced Google Glass.
Or refuse to work on Apple’s own projects:
We would like to welcome Chris Lattner, who will join Tesla as our Vice President of Autopilot Software. Chris’s reputation for engineering excellence is well known. He comes to Tesla after 11 years at Apple where he was primarily responsible for creating Swift, the programming language for building apps on Apple platforms and one of the fastest growing languages for doing so on Linux. Prior to Apple, Chris was lead author of the LLVM Compiler Infrastructure, an open source umbrella project that is widely used in commercial products and academic research today.
Besides repeatedly losing a key benefit of their acquisitions, Apple is rapidly narrowing the pool of possible candidates, particularly since the pipeline is already tiny and their competitors are hoarding whomever it produces:
Former employees have squarely blamed Apple’s stance on privacy on their departures:
As for current employees, their job is made unnecessarily difficult by requiring “the three feared privacy czars” to approve everything.
They are not even permitted to collaborate:
Apple appears to be aware of this issue, mentioning how their “ML is produced by many people who weren’t trained in the field before they joined the company.” This seems like a poor last-ditch attempt at bolstering their ever-leaking team.
2. Product Progress
I have exclusively used Apple’s services for years — from iWork.com to Siri to iCloud Mail. I was unsure why they had progressed so little all this time; however, after learning more about their methods and staunch anti-data practices, the cause became clear.
I think the most pertinent example is Siri. Tim Cook introduced Siri on the eve of Steve Jobs’ death over five years ago. I will grant that Siri has improved — but only on the surface, and just enough every year for it to occupy a slide on Apple’s keynotes. Indeed, five years later and people are still complaining about the same issues: poor natural language processing (even though Apple repeatedly touts this), persistent amnesia, hardcoded queries, limited functionality, and constant unreliability. Rather than solve any of these issues, Apple has fragmented the Siri experience on watchOS and tvOS, adding to its growing list of problems.
3. Competitive Position
Right around this point, someone usually sends me a link showing an arbirtrary comparison between Siri and Alexa/Google Assistant/Cortana, where Siri somehow manages to always just win the face-off. Notwithstanding the numerous problems with such an approach, people seem to forget that Apple has had Siri on the market for over five years — longer than any of these other products. If Google Assistant is “only” as good as Siri already in less than a year, that should be greatly concerning to Apple. Furthermore, it is fallacious to assume the competitive landscape will remain the same forever.
This illustrates that since Apple is the only company using the abovementioned counterproductive development methods, it puts them at a major competitive disadvantage. The other companies’ late market entry and rapid progress means Apple will likely be overtaken in the near future; they will then find it increasingly difficult to catch up since R&D, data collection, and data processing cannot happen overnight. And that is the best case scenario — soberingly, this is more likely:
Apple cannot afford what amounts to continual experimentation with their services division, with Differential Privacy being the latest. It has resulted in a talent crisis, which has directly benefited their competitors; their product development has remained hampered; and they have purposefully put themselves at a competitive disadvantage.
In his final keynote, Steve Jobs never distinguished between software and services — because they are interrelated and equally important — and he called them the soul of Apple’s products. Apple must take this to heart and realise that world-class services is equally important as delivering world-class operating systems and beautiful hardware.