Soli Radar-Based Perception and Interaction in Pixel 4
ai.googleblog.com
ai.googleblog.com
From the current leaks it sounds like Soli isn't going to make it into the Pixel 4a; I'm curious whether it makes it into the 5 (and TBH hope it doesn't, unless they can figure out how to not make it take up a huge amount of bezel space).
Soli makes a lot of sense for the Google/Nest Home line of products, where you rarely are close enough to touch them. But for your phone? I just don't get it, and it seems like few people find it useful. I'm pretty surprised it shipped and would be very curious to learn how that happened, since it seems to have negatively impacted the product pretty heavily.
Responding to critical feedback is a good and sensible thing to do. But ideally you would be able to get feedback on large issues internally, without needing the market to reject your product (Pixel 4 sales have been reported to be quite low): Soli seems so clearly useless, and caused such obvious product design tradeoffs, that I'm curious what led to it shipping in the first place. Was it an exec who was excited about Soli pushing it through? Was there a design/engineering communication breakdown? Would be very interesting to be a fly on the wall in the retro.
Aside from that, the Pixel 3a doesn't have wireless charging, so if that also applies to the 4a it's a total deal-breaker for me. And I really enjoy the performance of the Snapdragon 855 - especially with 90Hz forced at all times and animations tweaked to run at 2x speed, it's extremely noticeable how fast this phone is.
Not saying the Pixel 4a won't be a good phone, but the 4 is definitely better for premium features imo.
Overall, it feels like this is a tech demo lacking a killer app.
The user-facing features are just there to encourage slightly more usage and justify including it.
disclaimer: This is pure speculation.
The only truly useful thing is waking up the phone on detecting motion nearby, which could probably be done with a much simpler sensor.
It would be a cool thing to play around with, and I wish they would make it available. The problem is the raw data isn't likely to be very useful without using ML to filter out false positives for any gestures you want to implement.
I’d like to use an ATM without touching anything.
The coronavirus scare should open up markets.
Google is always afraid that someone else would overtake them as the #1 global aggregator of user behavior. They always acted swiftly to be on the forefront of new user tracking technologies, with more or less success. From Analytics to Doubleclick to Android to Maps To Plus. Sometimes they even tell themselves they are the good guys. I just wish they'd sometime realize the proper good guy behavior is to prevent the genie being released out of the bottle, not rush to be the ones releasing / acquiring the first / largest one.
I don’t want it to follow me into the bank. I only want to register my fingers when I’m a few inches away.
I've never tried one, but everyone I know that did said it was so bad they stopped using it after maybe 10 minutes of frustration.
It occurs to me that I like the model of carrying my own universal physical interface around with me more than I like sharing a domain- and location-specific one with other people.
Adding it to a phone seems like an unintelligent move -- maybe they should release a smart watch (if they know how)?
Guess I will have to wait at least a few years more.
Touchscreens are such crap to use.