On the Timing of Apple’s Map Switch

Wednesday, 26 September 2012

Let’s assume the timeframes being reported about Apple and Google’s maps license are accurate. The various reports coming out yesterday and today are in general agreement in this regard, and my own sources (who in this case are, as they say, directly familiar with the matter) back this up:

• Google found out earlier this year, before WWDC, that Apple would be switching from Google Maps to Apple’s own mapping service for iOS 6.

• The existing deal between Apple and Google still had a year remaining at that time — that is to say, at the time Google found out. Sticking with that deal through its expiration date would have left Maps in iOS 6 exactly where they stood in iOS 5: no turn-by-turn directions or vector map tiles for Apple, no additional Google branding or Latitude/Google Plus integration (and, thus, user location data collection) for Google.

• WWDC took place in June this year (as usual). That suggests the old deal ran through, at the latest, somewhere around June or July 2013. (It would make sense if the deal expired June 29, the anniversary of the original iPhone’s ship date.) During a keynote interview at the AllThingsD conference on 31 May 2011, Google chairman Eric Schmidt said that Google had recently renewed its maps partnership with Apple. So it would appear that was a two-year deal.

If you think about it, it makes strategic sense that, if Apple were going to break out on its own for mapping data, they would do so while there was significant time remaining on the maps license with Google. iOS is on a more or less annual development schedule. iOS 6 just arrived last week. iOS 7 is probably not coming until a year from now, and even if it’s on a more aggressive schedule, Apple would surely seek the luxury of having the option to wait until a year from now to ship it.

An all-new maps back-end is the sort of feature that Apple would only want to ship in a major new OS release. Technically, they could roll such a thing out in a 6.1 or 6.2 update, but major changes — and I think everybody can agree this has been a major change, for users and app developers alike — should be delivered only in major new OS updates.

But if the old agreement between Apple and Google expired in the first half of 2013 (which, again, my own sources familiar with the matter agree to be the case), that means the deal was set to expire halfway through the expected year-long life cycle for iOS 6. If Apple had stuck with Google Maps for another year they would have been forced to renegotiate with Google in a situation where both sides at the table would know that Apple either (a) had to agree to whatever terms Google demanded to extend the deal; or (b) would be forced to swap the mapping back-end of iOS 6 midway through its development cycle.¹ However tumultuous a change this has been in iOS 6.0, it would have proven more tumultuous and controversial if Apple had been forced by failed contract negotiations to squeeze it into a 6.1 or 6.2 update come May. And, that would have forced Apple to devote significant engineering resources for an iOS 6 update that could otherwise have been applied to iOS 7. Big changes come in the major release; bug fixes, security updates, and minor improvements come in post-major-release .x updates.

Whatever chance there was for Apple and Google to agree to a longer-term deal for iOS to continue using Google Maps, the effective deadline for Apple to make that decision was earlier this year, not next year when the existing deal expired. Apple wasn’t going to wait to negotiate until their backs were to the wall with the currently-shipping version of iOS reliant on Google Maps when the old deal expired.

---

Update: Timeline-wise, and regarding Google’s purported surprise that Apple made this switch, it’s worth pointing out that Apple began making mapping-based acquisitions in July 2009, when they acquired Placebase. Apple then acquired Poly9 a year later, and then, a year ago, acquired C3 Technologies for $267 million. What exactly did Google think Apple was acquiring these companies for if not to replace Google Maps with their own offering? ★

Get the Fainting Chair

Wednesday, 26 September 2012

Chris Ziegler, The Verge:

Apple’s decision to ship its own mapping system in the iPhone 5 and iOS 6 was made over a year before the company’s agreement to use Google Maps expired, according to two independent sources familiar with the matter.

So far, so good. I’ve heard the same thing.

The decision, made sometime before Apple’s WWDC event in June, sent Google scrambling to develop an iOS Google Maps app — an app which both sources say is still incomplete and currently not scheduled to ship for several months.

We’re supposed to believe that Google was surprised that Apple left the deal a year early? That I don’t buy. It implies that Google’s executives are either stupid, lying, or blindingly arrogant. I’ve never heard anyone suggest that Google executives are stupid. Ziegler continues:

For its part, Apple apparently felt that the older Google Maps-powered Maps in iOS were falling behind Android — particularly since they didn’t have access to turn-by-turn navigation, which Google has shipped on Android phones for several years.

I’d say where by “apparently felt” Ziegler meant “were well aware”.

The Wall Street Journal reported in June that Google also wanted more prominent branding and the ability to add features like Latitude, and executives at the search giant were unhappy with Apple’s renewal terms. But the existing deal between the two companies was still valid and didn’t have any additional requirements, according to our sources — Apple decided to simply end it and ship the new maps with turn-by-turn.

So Apple’s options were:

1. Continue for one more year with iOS Maps as it was — no turn-by-turn directions, no vector-based map tiles — and thus go for another year at significant feature disadvantage to all other major phones on the market, and then face the same decision one year from now.

2. Work out a new deal with Google, granting Google more prominent branding and more (and more personally identifying — that’s what Latitude is all about) location data from iOS users in exchange for modern features like turn-by-turn and vector map tiles. More Google branding, less user privacy.

3. Cut the cord with Google now and go ahead with Apple’s own mapping services — which have been in development for years, with several prominent acquisitions of mapping technology companies — so that they can add turn-by-turn directions, greatly improve the aesthetics of the map graphics, use vector map tiles, add the visually impressive Flyover mode, and, most importantly, take control of their own destiny.

We’re supposed to believe Google was surprised that Apple chose #3? Ask yourself this: Were you surprised Apple chose #3? This option came with a now-well-known price: the accuracy and comprehensiveness of the underlying cartographic data. But all three of those options came with a heavy price. #1 meant iOS Maps would spend an entire year falling further behind all of Apple’s competitors. #2 would grant Google — now Apple’s archrival — more branding, more control, and more access to user data on Apple’s platform.

But surprised is exactly what Google apparently wants us to believe they were. Nick Wingfield and Claire Cain Miller of The New York Times report the same story, citing what must be sources within Google:

Google is developing a maps application for iPhone and iPad that it is seeking to finish by the end of the year, according to people involved with the effort who declined to be named because of the nature of their work. There has been widespread speculation about whether and when Google would release a maps application for Apple devices since Apple released a new version of its iOS operating system with an Apple-made maps service.

One reason that it will take Google some time to build the iPhone app: it expected the app with Google’s maps to remain on the iPhone for some time, based on the contract between the two companies, and was caught off guard when Apple decided to build a new application to replace the old one.

Everyone else in the tech world expected Apple to release its own mapping service for three years, but Google, the one company that stood to lose the most, did not? OK, sure. If you believe that, I have a melted bridge to sell you.¹ ★

The iPhone 5

Tuesday, 18 September 2012

The iPhone 5 is really nice.

It feels great, looks great, has the best display I’ve seen at any size, runs noticeably faster, networks noticeably faster, is way thinner and lighter than any of its predecessors, takes better photos, and, in my six days of testing, gets totally decent iPhone-4S-level battery life.

But you don’t even have to turn it on to see how nice it is. Just hold it. You really have to. Apple boasted during last week’s event that they now measure the precision of the iPhone 5 assembly in microns. A micron is one-millionth of a meter. On the web page promoting the iPhone 5’s design, Apple states:

iPhone 5 is made with a level of precision you’d expect to find in a finely crafted watch, not a smartphone.

Never before has this degree of fit and finish been applied to a phone. Take the glass inlays on the back of iPhone 5, for instance. During manufacturing, each iPhone 5 aluminum housing is photographed by two high-powered 29MP cameras. A machine then examines the images and compares them against 725 unique inlays to find the most precise match for every single iPhone.

iPhone 5 in my hand, this talk of micron-precision, fine watch craftsmanship, and the computerized selection of best-match inlays sounds not the least bit bullshitty or blustery. It simply sounds like an explanation of the level of obsession that it takes to create a mass-produced device that feels this, well, nice. It even feels as though they’ve put some serious work into the iPhone’s one historical weak spot: the home button.¹

The iPhone remains the flagship of Apple’s entire product line. It exhibits not merely the highest degree of fit and finish of any smartphone, but the highest degree of fit and finish for anything Apple has ever made. When first you hold it — where by you I mean “you, who, like me, is intimately familiar with the feel and heft of an iPhone 4 or 4S” — you will be struck by how light it feels, yet in a premium, not chintzy way. Within a week, it will feel normal, and your old iPhone 4/4S will feel like a brick.

Both aspects — the weight and premium feel — are related to materials. The aluminum unibody harkens back to the original 2007 iPhone, which, until now, was my all-time favorite in terms of how it felt in hand. The plastic 3G/3GS body seems like an anomaly in hindsight — it surely offered engineering and cost benefits, but looked and felt (and sounded, for that matter, when, say, you tapped your fingers on it or set it down on a hard table) pedestrian. The glass back of the 4/4S looks and feels very nice, but glass is heavier than aluminum. Volume and weight tend to correlate pretty closely with gadgetry. The iPhone 5, despite being more than a quarter-inch taller than the 4/4S to accommodate the larger display, is 12 percent smaller volumetrically. It’s thus fair to say the new iPhone is “smaller” — and in all but one way (see below) it really does feel smaller. But it’s 20 percent lighter, far greater than the reduction in volume. I’m sure there are dozens of engineering feats that contribute to this reduction in weight, but the biggest is the switch from glass to an aluminum unibody.

The pattern after 2007 has been for a tick-tock design schedule: new hardware design (iPhones 3G and 4), followed the next year by faster, more refined versions of the same design (iPhones 3GS and 4S). No one should be surprised by a same-size-and-shape iPhone 5S next year, but, it’s foolish to treat Apple as a creature of habit. But whatever Apple does with its iPhone designs over the next few years, I’d be very surprised if they move away from aluminum (or at least some sort of textured metal) as the primary housing material. It just feels right, and offers several practical advantages. As stated before, it’s lighter. And compared to the 4/4S’s glass back, it should prove far more durable. (Would you not love to know the number of cracked 4/4S glass backs Apple has replaced?) I believe this is a big reason Apple has not updated its bumpers for the iPhone 5 — this is a phone that was meant to be used without a case. There’s also a marketing advantage: few of the competing phones on the market are made of metal. (No competitor followed Apple’s move to glass; it will be interesting to see if any follow their move back to metal.) A typical consumer could easily identify an iPhone by touch alone, much less by sight.

Is it worth devoting the first 750 or so words of this piece to the iPhone 5’s surface appeal? I don’t know how else to convey the niceness of this thing. This iPhone 5 review unit is the single nicest object in my possession. I own things that cost and remain worth more (e.g. my car). But I own nothing this nice. It sounds hyperbolic to put it that way, but I offer this observation with no exaggeration.

And so thus the meta story surrounding the iPhone 5 is the same as that of the iPhone 4S a year ago: a gaping chasm between consumers so excited to buy it that they stay up until (or wake up in) the middle of the night to pre-order it, and on the other side, a collective yawn from the gadget and tech press. That story a year ago was lost amid the tributes to Steve Jobs, who died the day after the 4S was unveiled.² If anything, that chasm is growing. The collective yawn from the tech press was louder this year; the enthusiasm from consumers is stronger.

Niceness is my explanation. The bored-by-the-iPhone tech press/industry experts surely value niceness, but they do not hold it in the same top-tier regard that Apple does. They are not equipped to devote an amount of attention to niceness commensurate with the amount of effort Apple puts into it. Apple can speak of micron-level precision and the computer-aided selection of the best-fitting of 725 identical-to-the-naked-eye components, but there is no benchmark, no tech spec, to measure nice. But you can feel it.

And that is what resonates with millions of people around the world.

The Display: Size

The new size takes some getting used to. For one thing, the new dimensions look weird at first. My first few days with the 5, it continually struck me each time I took it from my pocket that it looked too tall — like if my son went away to summer camp and came home several inches taller.

The letterbox mode for not-yet-updated-for-the-new-display apps kind of sucks. It’s not so much that it looks bad (my review unit is white; I’d wager money that the letterboxing is almost hard to notice visually on the black ones), but that it really throws me off while typing. My muscle memory knows where the keys are supposed to be relative to the bottom of the phone; letterboxing moves them all a row higher. This will surely be a non-issue within a few weeks as updates roll into the App Store, unless you’re a devoted user of any apps that are no longer maintained by their developers.

Video playback is better for the utterly obvious reason that it’s nice not to have to choose between filling the display and maintaining the video’s true aspect ratio. Reading is better — more words per page, more tweets or messages per screen.

The bigger display is a total win while using the iPhone 5 two-handed. But navigating the full screen while holding the iPhone in one hand is worse, for exactly the one reason why, even one year ago, I did not expect Apple ever to increase the size of the iPhone display: my thumb no longer easily reaches from corner to corner. (My hands are at least somewhat larger than average. Perhaps, counter-intuitively, this issue will not be noticed by the smaller-handed, whose thumbs don’t easily stretch from corner to corner even on a 3.5-inch iPhone display.)

Consider the windshield wipers on a car, and how, because they swing in a radial arc, they can’t reach the passenger-side top corner. Using the iPhone 5 is like that. There are two specific touch targets where this gives me trouble, both of which I invoke frequently. First, back buttons in the top left corner. I keep mis-tapping underneath them with my fully-outstretched thumb and then need to subtly re-grip the phone so that my thumb can reach. Second, tapping the status bar to scroll to the top of the current view. The top-left back-button issue is only a problem when holding the iPhone 5 one-handed in my right hand, but, I’m right-handed and so that’s the hand I tend to use it with.

There’s a reason Apple emphasizes typing in its justification for why the iPhone 5 display is larger but not too large:

Anyone can make a larger smartphone display. But if you go large for large’s sake, you end up with a phone that feels oversize, awkward, and hard to use. iPhone 5 features a 4-inch display designed the right way: it’s bigger, but it’s the same width as iPhone 4S. So everything you’ve always done with one hand — typing on the keyboard, for instance — you can still do with one hand.

Typing on the iPhone 5 does feel exactly the same. And in my experience testing big-screen phones (mostly with the 4.65-inch Galaxy Nexus), it really is far more difficult to do anything on them one-handed, including typing. And I can reach top-left-corner back buttons and the status bar one-handed with the 5, it just isn’t as easy, and requires an ever-so-slightly different choked up grip on the device than I’ve used for the past five years.

There is no argument that some people really do like these big closer-to-5-than-4-inch Android and Windows phones. I was in a Verizon retail store yesterday (long story; don’t ask) and overheard a relatively small woman buying a Galaxy S III. A companion asked if she wasn’t worried that it was too big, and she said no, big was exactly what she wanted, because she doesn’t have a tablet and wanted to do a lot of reading on whatever phone she got. She even said she was thinking about the 5-inch Galaxy Note (which Verizon doesn’t carry). It was like a conversation out of a Samsung commercial. Such people surely think the iPhone 5’s display remains too small. But, trust me, there are going to be many long-time iPhone users complaining that it’s too big after they upgrade.

In an ideal world, perhaps Apple would offer two iPhone sizes — like they do with products such as MacBook Pros, MacBook Airs, and iMacs. A smaller one with the classic 3.5-inch display, and a larger (say, 4.5-inch?) one for people who want that. On the logistics side, this doesn’t align with Apple’s interests — economies of scale and the marketing simplicity of just one new iPhone per year.

But there’s another factor. I believe many people would choose poorly. Bigger looks better. It’s like the old chestnut about TV sets in big box stores — side-by-side, standing in the store, people tend to choose TVs that are oversaturated, the ones with the boldest colors, rather than the ones with the better, more accurate colors. I can’t help but think that many people would choose the big-ass iPhone in my hypothetical two-sizes scenario, and later regret it with tired thumbs sore from stretching. (My thumbs feel sore just by looking at photos like this one of the LG Optimus G.) Design is making decisions, and Apple has always decided what the best size is for an iPhone display.

So the question is, if a 4-inch 16:9 display is better than a 3.5-inch 3:2 display, why hasn’t the iPhone been using 4-inch 16:9 displays from the start? Cost must have been a factor. Bigger screens are more expensive, and the 2007 iPhone display was like nothing else on the market. Bigger displays also consume more power. But I think it’s really mostly about a subtle change in priorities, a reordering of the tradeoffs — and, let’s face it, a response to marketing pressure from the aforementioned bigger-seems-better retail showroom factor.

The new size is not a radical change. Both the display and the phone itself feel exactly like what they are: the same width but taller. The iPhone 5 remains one of the smaller-screened smartphones on the market. What Apple has arrived at is a reasonable compromise.

But if Apple offered me an otherwise identical iPhone 5 with a 3.5-inch 3:2 display, which one would I choose? Last week, in the first few days of use, I’d have chosen the 3.5-inch one. Now, though, one week in, I’m not so sure. My trusty old iPhone 4S feels better to use for tapping those back buttons and the status bar, but, it really is starting to look squat to my eyes. Give me another week and I suspect I won’t look back.³

Display: Quality

As I wrote in my first impressions after last week’s event, the integration of the touch sensor into the display moves the pixels significantly closer to the surface of the glass. This is very good. The ideal is pixels right on the surface, like ink on paper, and Apple is getting closer and closer every two years.

Color quality is amazing. (Schiller claimed on stage that the iPhone 5 display covers the full sRGB color gamut.) It makes the iPhone 4S display look dull and dim. Brights are brighter, colors are more saturated (but not grossly so, a la AMOLED displays), and blacks are incredibly black. It is almost impossible for me to discern black pixels along the display edge (like, say, the black status bar) from the thin black area surrounding the display.

No single display size can please everyone. But in terms of quality, I honestly can’t imagine how anyone could deny that this is the best phone display in the world. And I still think the iPhone 4/4S display ranks second. Apple has a lead here, which is interesting, because they buy these displays from companies like LG.

Camera

The camera improvements over the 4S camera are subtle, but real. Images look sharper and colors are more vivid. Low light performance is significantly improved, not just in terms of exposure, but also for autofocus. The lens is also ever-so-slightly wider angle (4.1 mm vs. 4.3 mm). I’ve uploaded a small set of comparison photos between the iPhone 5 and 4S to Flickr. This low-light shot of an action figure in my office is remarkably better than the corresponding shot from the iPhone 4S. In a nut, iPhone 5 photos look slightly better in daylight; they look dramatically better in low light.

Without selling a single dedicated “camera” since the groundbreaking but discontinued-in-1997 QuickTake, Apple has become one of the leading camera companies in the world.

LTE and Carrier Choice

Apple asked which carrier I preferred for this review unit, and I asked for and received a Verizon model. I’ve been an AT&T customer since the pre-iPhone Cingular days. My pre-ordered personal iPhone 5 is on Verizon.

Verizon LTE was fast and ubiquitous in San Francisco, and the same has been true here at home in Philadelphia. Your mileage may vary, but I’m switching for simple, practical reasons. Verizon has stronger signals in the places I visit most, better family plans with shared data, broader LTE coverage, and AT&T’s network quality seems to be getting worse, not better. Also, my iPad (3) is on Verizon, and it always has as good or better a network connection than my AT&T iPhone 4S. I couldn’t be happier with Verizon’s LTE on the iPad, and, after six days of use, the same is true with the iPhone 5.

The only two factors in AT&T’s favor are: (a) the simultaneous voice and data thing; and (b) the laziness factor — it would’ve been easier to order a new phone on my existing account than to do this switching thing (cf. the aforementioned anecdote about visiting a Verizon retail store yesterday).⁴

Using the iPhone 5 on LTE is nearly indistinguishable from using it on Wi-Fi. Web pages load in a snap, Siri parses input and responds promptly. It’s as big a difference from 3G (and whatever bullshit AT&T calls “4G”) as 3G was from EDGE.

Battery Life

I’ll leave the battery-specific testing to others. I just used the thing as usual. On Sunday I watched the entire Yankees game over 3G (I was at a family outing in an area sans LTE coverage), browsed the web, read email and Twitter, and by midnight the phone was in the red but hadn’t yet hit the 10 percent charge remaining warning. If anything, I’ve been giving the phone fewer sips of power throughout the day than was typical with my iPhone 4S, simply because I have only one Lightning plug. (I didn’t order any of the 30-pin-to-Lightning adapters, but I did order an extra Lightning-to-USB cable.) Battery life seems good, exactly on par with my 4S.

Performance

After my item the other day pointing out that the iPhone 5’s Geekbench score (1,600-ish, which I can confirm) is far higher than that of any PowerPC laptop Apple ever built, a few readers pointed out that Geekbench’s baseline of 1,000 is the “2003 entry-level Power Mac G5”. So, as of this week, we have computing performance in our pants pockets that nine years ago required a professional desktop workstation.

I wrote about this back in 2008, while making the case that RIM was screwed because the future of the phone market was really about the future of portable computing, and that RIM was a phone/messaging company, not a computing company:

Along the lines of can’t-really-be-answered-but-gosh-they’re-fun-to-ponder questions like, say, “Who’d win in a fight, Batman or Spider-Man?” or “Star Destroyer vs. U.S.S. Enterprise?”, here’s one regarding the iPhone: What historical Mac is a current iPhone most analogous to, spec-wise? I.e, complete this sentence: “An iPhone is like having a tiny __ in your pocket?”

Think about this: eight or nine years from now, we should have phones that are computationally equivalent to today’s Mac Pro. (Maybe even sooner, given the sorry state of the Mac Pro at the moment.)

Bottom Line

The question everyone who hasn’t yet pre-ordered wants answered: Should you upgrade? My answer is simple. If you can afford it, yes.

There’s a reason why, just as with all five of its predecessors, it just says “iPhone” on the back. The iPhone 5 is all new technically, but it’s the exact same thing as an idea. Apple is simply improving upon that idea year after year in infinitely finer detail, like a fractal. It’s nice. ★