4 ms·

That stil doesn't explain why the keynote was so boring. For the first time, I had had enough and closed it and walked away. Usually, Apple's iPhone announcemen

by neya 7y ago

That stil doesn't explain why the keynote was so boring. For the first time, I had had enough and closed it and walked away. Usually, Apple's iPhone announcements are pretty exciting. This time around, even the presentation itself felt like a giant corporate presentation sprinkled with keywords like "innovation" to make me believe they were innovating. There was nothing innovative about the keynote. The iPhone is a laggard at this point and there are far superior offerings at this point. A few years ago, I wouldn't have said the same and would have even ridiculed if someone claimed Android phones were better, but after playing with an S10+, iPhones really don't impress me anymore. "But it offers more privacy?" Nope. Didn't you read about the part where all your personal siri recordings were played out by contractors? Including sex acts and arm deals? What happens on your iPhone does NOT stay on your iPhone because it's proprietary software. Duh. I'm done with Apple. But I'm also looking for something beyond Android at this point. Wish there was a truly 100% open source phone I could trust.

saagarjha 7y ago
> there are far superior offerings at this point Such as? > all your personal siri recordings were played out by contractors Nit: not all, just a fraction.
simonh 7y ago
Pretty much all the voice assistant companies have been caught having people listen in at this point. Amazon, Samsung and Google for sure at least. It is pretty disappointing to find Apple in on the act as well though.
- snowwrestler 7y ago
  There is literally no other way to apply machine learning to such a large problem. The system must have training data to learn from; as you expand the scope of your ML system, you must expand the scope of your training data. Every large commercial ML system works this way. Google for years has paid people to manually score and categorize search results pages, to provide training data for ML web search systems. It would have been better if companies were up-front about this when talking about "artificial intelligence" in their voice products, so it doesn't look like a nasty surprise. And in the long run, everyone will benefit from some consistent regulations on ML data scoring, like HIPAA for health care (another industry where for-profit companies handle very sensitive customer data).
- neya 7y ago
  Yet only Apple advertises "What happens on your iPhone, stays on your iPhone"

by neya 7y ago

saagarjha 7y ago

> there are far superior offerings at this point Such as? > all your personal siri recordings were played out by contractors Nit: not all, just a fraction.

simonh 7y ago

Pretty much all the voice assistant companies have been caught having people listen in at this point. Amazon, Samsung and Google for sure at least. It is pretty disappointing to find Apple in on the act as well though.

snowwrestler 7y ago
There is literally no other way to apply machine learning to such a large problem. The system must have training data to learn from; as you expand the scope of your ML system, you must expand the scope of your training data. Every large commercial ML system works this way. Google for years has paid people to manually score and categorize search results pages, to provide training data for ML web search systems. It would have been better if companies were up-front about this when talking about "artificial intelligence" in their voice products, so it doesn't look like a nasty surprise. And in the long run, everyone will benefit from some consistent regulations on ML data scoring, like HIPAA for health care (another industry where for-profit companies handle very sensitive customer data).
neya 7y ago
Yet only Apple advertises "What happens on your iPhone, stays on your iPhone"