Aaron Tilley and Kevin McLaughlin of the Information (this article is behind a paywall):
To determine how Apple squandered its own head start over rivals Amazon and Google in the digital assistant realm, The Information interviewed a dozen former employees who worked on various teams responsible for creating Siri or integrating it into Apple’s ecosystem. Most of them agreed to speak only on the condition that they not be named, citing non-disclosure agreements they had signed or concerns about retaliation from Apple executives.
Many of the former employees acknowledged for the first time that Apple rushed Siri into the iPhone 4s before the technology was fully baked, setting up an internal debate that has raged since Siri’s inception over whether to continue patching up a flawed build or to rip it up and start from scratch. And that debate was just one of many, as Siri’s various teams morphed into an unwieldy apparatus that engaged in petty turf battles and heated arguments over what an ideal version of Siri should be — a quick and accurate information fetcher or a conversant and intuitive assistant capable of complex tasks.
Even if you view this as a half-true gossip piece — and I don’t think it is, for what it’s worth — it’s still a fascinating look into the struggles Apple has faced with improving Siri’s capabilities.
For example, Tilley and McLaughlin report that separate teams worked on Siri and Spotlight’s suggested answers, which explains why the same query would sometimes return different results in each. On iOS, Apple rebranded some Spotlight features as Siri features: Siri App Suggestions, and Siri Search Suggestions, for example.
And then there’s Apple’s acquisition of VocalIQ two and a half years ago:
The VocalIQ team viewed Siri as a “manually-crafted system” and felt their technology could help improve it, said a former VocalIQ employee. VocalIQ’s technology is designed to continually finetune its accuracy by ingesting and analyzing data from voice interactions, he said. Apple has successfully integrated the VocalIQ technology into Siri’s calendar capabilities, sources familiar with the project said.
It’s interesting that Siri’s capabilities are set up in such a way that something like VocalIQ can be applied to just one feature. I don’t know how much this says, if anything, about why Siri often feels like its capabilities are so fragmented, but it struck me as odd.
Siri has been the responsibility of Craig Federighi since last year, transferred from Eddy Cue’s online services oversight. This year’s WWDC seems too soon to see that particular branch of discussion bear fruit; but, then again, the inconsistencies and general untrustworthiness of Siri make it feel like it cannot be soon enough for real changes to be made.
Update: The mysterious ATP Tipster:
The only thing you need to know about Siri is that the people who used to build it feel the need to absolve themselves of personal responsibility for the state that it is in. That they are doing so in the press is almost an implementation detail.