The Unseen Cost of Google's New "Free" Photo Service

Is it just me, or does something not seem quite right about Google's new Photos service?

There was quite a bit of talk at Google IO this week about how Google Photos is using Machine Learning to help you to be able to search through your images.  You search for "Cronuts" and voila! Every photo that you have taken of those delectable cronuts that you stood on line hours for will show up!   From Business Insider:

“Google uses machine learning to power the scary-accurate image recognition within the Photos app, and based on the demo I’ve seen, it’s incredibly fast and accurate. Other services let you search photos based on when they were taken, where they were taken, and how you’ve tagged them.

But with Photos, you can simply type in the word “pizza,” and any photos you’ve taken of pizza will pop up whether you’ve labeled the pictures or not. If you went rock climbing last summer, you can type in the phrase “rock climbing” to find those images. Google also automatically sorts photos based on these types of phrases in the Categories section. So, all of the photos of any given person will appear in one organized album, all photos of food would be in another, etc.

Google already uses machine learning — a term that refers to a type of algorithm that learns on its own without human intervention — for many of its products, including Google Now, Google Maps, and Search. But now we’re seeing it being applied to photo storage.”

— http://www.businessinsider.com/how-google-created-the-worlds-smartest-photo-app-2015-6#ixzz3c3zY2qP5

Super Cool right?   On the surface, YES!   The amount of computation and engineering that has gone into a product like this is staggering.

But then I had a thought.  (Dangerous, I know.)

Why is it free?

No, seriously.   Why is it free?   Why is Google going to give away terabytes of VERY expensive cloud storage away for free?   Why is Google going to give away use of a photo recognition engine that required that massive engineering effort for free?

Answer?  They aren't.  Of course, YOU won't have to pay a thing for the service.  Google will happily index every photo you take from your smartphone,  apply it's machine learning to it,  and give you an infinite amount of storage for all of those beautiful memories.   What does Google get out of it?

I have a feeling Google is using its "machine learning" to index quite a bit more than most people think.  They say it can tell the difference between palm trees and pizza, and mountains, and people.   If it is really that intelligent will it also be able to tell, based on your photos, what kind of beer you like to drink at a summer barbecue with your friends?   How about if you prefer hot dogs or hamburgers?  Will it be able to tell based on your photos that you like to visit Disney World and go on Carnival Cruises for vacations?  I bet people buying advertising on Google would LOVE to have information like that.

Google has spent years developing applications and services to write up a profile of your online activities and persona.  Google Chrome / Google search track your web browsing habits to deliver you advertising.  Gmail algorithmically scans your email for advertising purposes.  Google Maps uses your direction requests and GPS data to deliver you advertising.  and now our last shred of real life private moments are probably being used to deliver us relevant ads.

It has gotten to the point where Google will now know more about the average Android user than his/her partner does.   The profile they have built on us contains our conversations, our web searches, our travel habits, who our friends are, and now everything we have taken a photo or video of.   Google knows what things we want to keep a memory of.   I don't know about you, but to me, that's creepy.

Tracking my online use was one thing, but somehow knowing that Google is scanning my photos and videos to identify everything in them is a step too far.   It seems today more than ever, that the services Google deliver to us are not really the products they sell.  

We are.

Steve Jobs Movie Trailer

Amazing how so many incredibly talented people can come together to make a film about one of the most influential people of the last 30 years and make it look utterly uninteresting.

Crazy WebGL Water Demo!

I must have missed something.  I wasn't aware WebGL had advanced to quite this point.   This is a really cool proof of concept demo from MadebyEvan.com showing that WebGL can accurately simulate water,  as well as the behavior of lighting in water.   The fact that this runs as well as it does even on my lowly 2012 Macbook Air is super impressive.    I'm embedding a youtube video of the effect below in case your browser or computer doesn't support the effect,  but I highly suggest you check it out at the site instead.  Really, REALLY, cool.

A realtime pool of water rendered using WebGL with reflection, refraction, caustics, and ambient occlusion. The pool is simulated with a heightfield and contains a sphere that can interact with the water's surface. http://madebyevan.com/webgl-water/

50 years of Space Exploration in one beautiful infographic

View fullsize

Click to Embiggen

Amazing what we have accomplished as a species in just half a century.  Just imagine what we can accomplish in another 50.

Sony hocking $160 "Premium Sound" SD cards→

You know what they always say.  "A fool and his money are soon parted."

Seriously though, Sony has decided to take the Monster Cables scheme to another level by selling absurdly expensive SD cards that supposedly "produce less electrical noise" than say... any other SD card.  How any company can have the balls to actually release something so scammy blows my mind.  From an article at PCWorld:

“Need the perfect expandable storage solution for your $1,200 Walkman? Sony might have a bridge to sell you.

The company is now selling a 64 GB Micro SDXC card “for Premium Sound” in Japan. At $160, it’s four or five times more expensive than a typical 64 GB card, but as the Wall Street Journal reports, it’s supposed to produce “less electrical noise.”

Now, the idea of audiophiles obsessing over barely-perceptible details is nothing new. The speaker market is filled with products whose frequency response exceeds the 20 Hz to 20 kHz range of the human ear, and you can spend thousands of dollars on audio cables in pursuit of eliminating noise.

But while those expenses at least have some technical justifications behind them, the case for audiophile storage is flimsy at best. As writers at The Register and PC Perspective have pointed out in the past, a storage device couldn’t affect digital sound quality without actually changing the actual data being transferred. And if that happened, it be a much bigger problem for all kinds of applications—not just music.

Still, that hasn’t stopped some people from believing the storage device makes a difference, perceiving “more organic” tones in some products and “edgy grain” in others. So when Sony tells the Journal that “we thought some among people who are committed to great sound quality would want it,” it might be onto something.”

— PC World

Thoughts on Apple's Alleged Car Project

The week of Valentine's Day 2015.  The week when speculation on Apple's next big thing began to reach fever pitch.  Apple is building a self-driving car.  Or at least that's what "people who know people" are saying.   A lot of the speculation really began a week earlier.  On November 4th, Business Insider (lol) ran an article about a "mysterious Apple van" that has been seen roaming around San Francisco.  Of course, going past the obvious conclusion (as they do), they immediately assumed that there is no way that this vehicle was just Apple's attempt at Street View, but in fact is Apple secretly (out in the open in San Francisco!?) testing it's driverless car technology.  From the article:

“A camera-mounted van spotted driving around the Bay Area is apparently registered to Apple, and has sparked speculation that the Cupertino company is developing its own Google Street View competitor, Apple Insider reports — or even self-driving car technology.
First spotted by Claycord, the vehicle is a Dodge Caravan with a four-camera rig mounted on the roof. KPIX reached out to the Department of Motor Vehicles and received confirmation that the vehicle is leased to Apple.

When Claycord asked the driver what the vehicle was for, he “would never give an answer.”

There are two possibilities as to what it could be: A Google Street View competitor under development, or a secretive self-driving car project. Analyst Rob Enderle told KPIX that he suspects the latter, saying that it has “too many cameras” to be a Street View-type technology. Apple doesn’t have a license to test self-driving cars itself, but according to Enderle, this doesn’t rule it out — they could be working in partnership with a company that does.”

Apple's alleged "self driving car" (read: Apple's probably Street View Car) - Claycord

The chance that this is part of an Apple self driving car project is nearly nil.   This is very obviously (at least to me) Apple's attempt to catch up with and possibly surpass Google Street View in its Maps application.  Business Insider's assertion that there are "too many cameras" is ridiculous because they neglect to consider what those cameras might be FOR.  

Capturing a 360 degree panorama bubble of an area every few feet like Street View is one thing.  Capturing the actual 3D data of the area is another.   Those extra cameras could very well be kinect-style depth mapping cameras that are building up a high resolution three dimensional mesh of everything that is photographed.   If Apple used the high res photo data in conjunction with the 3d mapped data,  they could create an almost photorealistic 3d world that could be smoothly driven through.   Very much unlike Google Street View which requires that you jump from one 360 degree bubble to the next.

That being said, even if this project is that.  It will still be a long time before Apple is able to capture enough roads to make a useful addition to it's app.  Of course, even with the smart money chiming in that this was bogus,  the hype train didn't stop.   And of course Business Insider was there once again to fan the flames; reporting that an un-named Apple employee said they were working on something that "would give Tesla a run for it's money".  From the article:

“Last week we reported on a mysterious Apple van making its way around San Francisco.

After writing about how the van could be used for a self-driving car, we got an unsolicited email from an employee at Apple about “vehicle development” at the company.

This person said Tesla employees were “jumping ship” to work at Apple.

”Apple’s latest project is too exciting to pass up,” the person said. “I think it will change the landscape and give Tesla a run for its money.”

Apple has about 50 employees who previously worked at Tesla, according to LinkedIn. Many of those hires were engineers who interned at Tesla. Most of the engineers Apple has hired from Tesla specialize in mechanics, manufacturing, and robotics.”

— Business Insider

Now, it IS interesting that Apple has been poaching Tesla employees.  And of course, where Apple is concerned, where there is smoke, there is generally at least some fire.   In fact, the idea that they would be looking into building a car is not all that ridiculous.   Though there is almost no chance it would be a self-driving one.   At least not yet.

Think about it.   Building a new car from scratch for a major car maker is a project that can take 5-6 years.  That's 5-6 years for people who are highly versed and experienced in automotive design.   Apple lacks almost entirely the skill set to develop any car, let alone a self driving one.  A few engineers from Tesla isn't going to be anywhere near enough to get them up to speed.

That being said, it might just be enough to get them started on the road.

If I had to guess, this is part of a VERY long game for Apple.  One that may or may not hinge on the acquisition of an upcoming name in the industry.  Yep.  That's right...

Tesla.

To me, Apple has always been about one thing.   Taking something that sucks (PCs, cell phones, MP3 Players, Tablets) and taking them back to the drawing board to do them RIGHT.  One can take a look at the HORRENDOUS UI of the average smart tv or cable box and see why TV has been a growing interest of theirs as well.   So what else sucks?   Car Tech.

Cars mechanically speaking have been refined to near perfection over the past 100 years.   But to an environmentally-concious, design driven company like Apple, cars have two big problems.  

First, gasoline.  Oil and natural gas have become the baseline source of most of our conflcts of the past 20 years.   That shows no sign of stopping, and to Apple, it's time to get off gas.  That means electric, and so far there is only one company with any real skill in getting high range electric cars on the market.   Tesla.

Second, like many of the objects we use today, software quality in cars is becoming a more important part of the experience than it ever has before.  Unfortunately for us, car-makers tend to be horrendous at software/computer hardware.   Terrible slow resistive touchscreens,  slow CPUs, low memory, cluttered UI design, and graphics that look like something from Prodigy in 1992 just add up to something that makes me wish that most cars just went back to AM/FM radios.   Don't even get me started on the pain of pairing a bluetooth device with half of these vehicles.  Disasterous.  

Apple is trying to fix this problem with CarPlay in the same way that Google is trying to fix it with Android Auto.   Unfortunately, that means having to support these automakers terrible hardware.   More than likely this will end up as just another MotoROKR situation.  Apple likes to own the whole stack.   So either they could partner with automakers to design their center consoles (a disaster waiting to happen for a company that likes things it's way),  or they could just build a car.

The big problem with building a car is that it's a project that will take Apple close to a decade to come to fruition, and even then there is no guarantee that it will be good.  So what to do?

Simple.  Buy Tesla, or at least try to.

Apple has $180+ BILLION dollars in cash on hand for a reason.  Big purchases to keep themselves in the fray if need be.  They know that like PCs and the iPod,  the iPhone is a business that won't last forever.  Something will eventually supplant it, and Apple always wants to be ahead of that.

If Apple is interested in getting into the car business, why not simply use some of that war-chest to buy the most Apple-like of the auto manufacturers?    I mean let's be honest here.   The Tesla Model S already looks like a car that Apple would have designed.   Steve Jobs was an unabashed fan of the Mercedes SL550, and lead designer Jony Ive is a longtime fan of Aston Martin.   Put those in a blender with a bit of Apple DNA and you have something closely resembling the Tesla Model S.

Not to mention the side benefits that an Apple acquisition of Tesla would bring.

Apple and Tesla are two of the biggest companies researching battery technology today.  Teaming the two up would be beneficial for both, and battery technology would only improve because of it.

The other major side benefit is Elon Musk.

Perhaps Apple could bring Musk in as a "visionary-in-residence".  If there is one person out there who has that kind of Steve Jobs-ian clout and charisma with the public, it's Musk.  I'm honestly not sure he would be right for Apple's leadership, but having him on campus dreaming up crazy things like SpaceX, and the Hyperloop might be that spark of creativity that some people think Apple is missing.   Maybe he could resurrect Apple's "pirate" division that Jobs' helmed to develop the Macintosh.   An elite team of engineers that develop "moon shot" style products in the way Google Labs does.   Things that may not directly come to market, but inform Apple's internal thinking.

I'm beginning to think that is Apple's basic plan.   Get the ball rolling on WHAT EXACTLY an Apple Car could be.  Do the research, make the designs, get everything in place for doing a Tesla acquisition that would be as important for Tesla as it would be for Apple.  While they are doing that, they will do more work on developing and improving CarPlay and getting it out to as many cars as they can. 

Considering that Apple could get Tesla for around $75 billion, and still have over $100 billion in the bank,  the only question is why not?  I would be on a list for an Apple Model S.

Is Apple Watch's 3d Touch Tech coming to the iPhone?

This makes nothing but sense to me.   If this technology works as well as I've been hearing, its a no brainer to bring it to iPhone.   Something tells me this will eventually tie into what Apple is doing with that stylus for the "iPad Pro".   Pressure sensitivity via touch or stylus throughout the OS would be a game-changer for artists who like to paint or sketch on an iPad or iPhone.   From Cult of Mac:

“Apple Watch will borrow a lot of tech from the iPhone when it ships in April, but according to a new rumor from supply chain sources in China, Apple is planning to bring one of its wearable’s coolest features to the next iPhone.
The Economic Daily News has reported that Apple is considering adding ‘3D touch’ technology to the iPhone 6s, similar to Apple Watch’s Force Touch. According to the sources, Apple’s is planning to tap US-based Avago Tech as the main supplier for the iPhone 6S 3D touch technology.
The report doesn’t fully explain how Apple will implement the technology on the iPhone 6s, but if it’s the same as Apple Watch, the new displays will use tiny electrodes to distinguish between a light tap and a deep press to trigger contextually specific controls.”

— Cult of Mac