Safari is the new Internet Explorer? Or maybe not.

Today ArsTechnica posted a weak opinion piece by a Squarespace employee named Nolan Lawson. In it, Mr. Lawson uses some weak arguments wrapped in a cliched, flawed narrative about how Apple hates the web to try and rally web developers against Apple’s Safari browser.

What follows is the text of a comment I left in response:

Rapid change is great when everything sucks (for example IE 1-5), and when something that was formerly adequate has fallen years behind (ie IE 6 for of the the 2000s). Other times, steady incremental change is better for for users and developers. Well, most of them. It’s obviously not great for the ones that have an agenda they are pushing. On the web, in 2015, I suspect steady incremental change is best.This OpEd doesn’t really give me much reason to think otherwise.

The author makes a weak case for his assertion. It’s clear from some of the other comments here that some of the specific technologies the author mentions aren’t exactly fully baked or widely supported yet. I know from my own experience that, as a user, Safari is a hell of a lot better for my useage than Chrome on OSX from the point of view of responsiveness, memory use, and battery life (very important to me on my laptop). Apple’s focus on this actually makes web apps more viable for my use — I can have significantly more tabs open in Safari than I can in Chrome without bringing my computer to a grinding halt.

His piece seems driven by the cliched narrative that Apple hates the web because it threatens its platform dominance and the OS X app store. This narrative, while oft repeated, doesn’t really hold up to close scrutiny. For one thing, Apple’s web-friendlyness has a long and continuing history. They helped advance WebKit as a cross-platform alternative to IE and Firefox before Google Chrome was even a Windows-only beta. The original iOS app strategy was Web-app only. They were right in the thick of things trading off with Chrome in javascript performance. The developer tools that people loved about Chrome were originally Apple’s doing, and even though Apple took them in a direction that wasn’t exactly loved, they continued to refine and improve them. Despite the success of the App Store, Apple has continued to advance Safari on iOS as well. Things may have changed recently, but it wasn’t too long ago that Mobile Safari was better for developers and users than the original android browser, or Chrome for Android.

The pace has surely varied, but over time, Apple continues to push steady forward, and has continued to do so even as the App store has prospered, even as people like the author have flogged the “Apple hates the web” narrative.

Its hard not to think that the author might be predisposed to embrace the flawed “Apple Hates the Web” narrative because he identifies on his website as an “Android Fan,” but that’s just speculation on my part. All I know is that he makes a weak case, built on a weak foundation.

I have an alternate explanation that he may want to consider: Apple behaves as it does out of self-interest, yes, but self-interest that is largely and consciously aligned with the interests of end-users. That framing is consistent with my observation that, as an end user, Safari development in recent years has clearly focused on things that are better for me, like responsiveness, memory footprint, CPU utilization and battery life (all obviously interconnected). Putting it another way, Apple is striking a balance between its support for mainstream web standards and nascent web standards in a way that favors the majority of its end-users.

It also gives a hint for another reason why Apple may not be as all-in on the web as some others would like: The web is great, but it isn’t everything, at it isn’t even unambiguously the best at everything it does. Even a bleeding edge web app taking advantage of all the features of whatever browser the author thinks is best is likely to lag what is possible with a iOS or MacOS app using well worn frameworks and APIs. Apple, quite reasonably, thinks it is best for the company and its customers to balance its support of mainstream and nascent web standards with its support for developers of native iOS and OS X apps.

The author is, of course, free to have his own priorities, and to try to persuade Apple to adjust theirs, but in both cases, he’d do well to broaden his perspective, acknowledge his own biases, and think more critically and less dogmatically about the issues.

The Economist Doesn’t Get It: Apple Watch Edition

I don’t pretend to know how the Apple Watch will do, but I’m sure that many of the people predicting its failure don’t know what they are talking about.

Add to the list, The Economist.

Yet in spite of Mr Cook’s bouncing optimism, Apple seems unlikely to turn its watch into the next big must-have gadget. Certainly, the watch will not match the success of previous products, such as the iPod or iPhone. This is true for two main reasons. First, Apple’s newest creation replicates many of the functions that the smartphone already makes so seamless, such as checking e-mail, receiving calendar alerts and communicating with friends. People are unlikely to want to shell out a sum between $350 (for the most basic model) and $17,000 (for the fanciest version) for something with so few extra functions. Second, the Apple Watch is dependent on a nearby smartphone, which means that users will just be adding another device to their growing menageries instead of replacing one. This is not unlike selling someone a wristwatch that requires a pocket watch to work.

from Launching the Apple Watch: The time machine | The Economist.

Their first point could have been made about the iPhone, and certainly was about the iPod. Neither didn’t anything that something else didn’t do before them.

Their second point though is particularly foolish.

Once upon a time, computers were expensive. People bought one, and used it for everything they could. Sometime between the iPod and the iPhone, computing devices became cheap enough that people bought more than one, and used them for specialized use cases.

And yet, some people, don’t get it. Those people seem to be overrepresented among technology reporters and pundits. They expect new devices must replace earlier devices, like the personal computer replaced the typewriter, etc. Certainly new devices may replace older devices — the iPod replaced walkman, and the iPhone replaced the iPod, but must? That’s pure superstition.


On Apple Watch Upgrades

While waiting for the release of the first Apple Watch, Apple watchers are killing time by speculating about future versions. Today John Gruber quoted a piece by Serenity Caldwell at iMore about the upgrade scenarios for the $10,000+ Apple Edition watches. She hopes for an upgrade program so that those who can afford to spend $10,000+ for a watch, the poor dears, don’t suffer when Apple releases inevitable improvements. (Meanwhile, she thinks its fine for those who pay $350 for a watch, many of whom don’t have the option of spending $10K+ for a watch, to buy a new Apple Watch every few years.

Gruber added his own take:

I hope Apple Watch — at least the Edition models — is upgradeable. I would bet that it’s not. The single most frequent question I’ve received this week is how can Apple justify $10,000+ prices for a watch that will be technically outdated in a few years. The simplest answer is that it’s for people who don’t care.

I say I’d bet against upgradeability simply because it’d be so unlike Apple. But, the whole idea of a solid gold $10,000 watch is also unlike Apple. We’re in new territory here. And I do wonder why Apple called out the modular design of the S1 on their technology page. Why does this image exist? An “upgrade” would probably require new sensors and antennas and battery too — more or less replacing everything inside the watch case.

This prompted a short exchange between Gruber and I on Twitter:

This has come up before. Back in December, Jean-Louis Gassée explored the question of Apple Watch upgrades through the lens of the arrival of the dSLR in the camera market, and I turned my comment on his piece into an earlier blogpost on the subject.

I thought I’d take this opportunity to restate and update my thinking on the subject:

To make the Apple Watch a reality, Apple has had to make some key compromises in the service of battery life. Despite significant efforts to miniaturize the electronics, the Apple watch seems plump, while every other Apple device is slim. Despite the satisfaction of having a personalized watch face visible to passers by, the Apple Watch shows a blank black face until the wearer raises it to his or her face. Despite the desirability of having all the functions of the watch available all the time, many of them, including third party apps, rely on a Bluetooth connection to the owners iPhone. These compromises are all, ultimately, in the service of a full day’s battery life under typical use.

Given this, I think battery life will be a primary consideration in Apple Watch updates. Batteries are incredibly important for portable electronics like the Apple Watch, smartphones and notebook computers, but despite their importance, the pace of improvement is about 10% a year. Battery life of portable devices has improved at a faster pace though. This has been accomplished by making the devices more efficient in their use of power, and much of this improvement is due to Moore’s Law.

Moore’s law can produce a halving of power consumption every 18-24 months. People worrying about the Apple Watch updates seem to be assuming that updates will come at a similarly rapid pace. I think this is unlikely, because Apple has already made significant architectural and user experience accommodations to achieve reasonable battery life. Because of this, I suspect that the core electronics of the Apple Watch account for less than half of the daily power budget of the device, while the display probably makes up the majority. Display power consumption is driven primarily by the power consumption of the light emitting diodes, which also only improve at a slow pace.

My guess is that the existing Apple Watch battery life is good enough for now, but needs further improvement. Therefore, I would expect that any Moore’s law improvements will go to battery life rather than performance or functionality. Battery capacity improvements will be put to similar use.

Once battery life is truly sufficient, Apple will next spend further improvements in thinness, by reducing the size of the battery.

The arrival of a significant improvement in the computing abilities of the Apple Watch may not arrive until complimentary communication capabilities can fit within the form factor and power budget.

I wouldn’t be surprised if it takes 5-10 years before truly stand-alone Apple Watches are available. On the other hand, I think that identity and fashion are increasingly important drivers of consumer tech product cycles, and I think the Apple Watch will bring that into much sharper focus. I will probably have more to say about that soon.

Daring Fireball: Dazzling Results

It’s amazing how some people can just not get it when it comes to Apple in particular, and the technology industry in general.

John Gruber tends to get it, which he demonstrates in this post dismantling someone clearly doesn’t get it. I’m going to be lazy and rather than carefully selecting some quotes to comment on, I’m going to post pretty much excerpt pretty much the whole thing and interject my own commentary here and there.

Juan Pablo Vazquez Sampere, writing for Harvard Business Review, “We Shouldn’t Be Dazzled by Apple’s Earnings Report”:

But one thing has changed. Apple used to revolutionize industries, announcing record sales numbers because it had introduced a new technology, feature, or product that we had never imagined but that, when we saw it, we all instantly wanted. That Apple seems no longer present. In this instance, all Apple has done is copy a feature for its own best customers. While that’s very effective for today, it does not solve the problem of tomorrow for a company that competes on serial innovation.

That one feature he’s talking about is the larger display sizes for the iPhone 6. I’ll reiterate that Apple has never been a company that serially produced revolutionary product after revolutionary product. Their revolutions have been very few and far between: Apple II, Macintosh, iPod, iPhone/iPad. Everything else is constant iteration and refinement.

Exactly right. Revolutionary products come when the industry as a whole gets lazy and slow.

The Macintosh built on innovations created at Xerox PARC (along with plenty of Apple’s own) that Xerox had failed to capitalize on.

I’d argue that the iPod took advantage of a series of incremental changes to the technology and media landscape that the incumbent players had neglected to take advantage of. The music labels shunned online distribution, I think, in part, because they couldn’t make it work on their own, and didn’t trust Microsoft enough to work with them. On the other side, the Wintel duopoly didn’t have the chops to make a true consumer oriented product, and the PC OEMs they held in their orbit couldn’t pull it off either.

Apple wasn’t ready for it either, not at first. They needed to get their own house in order before they could pull off the iPod, but once they did, their ability to launch the iTunes music store was primed by the desperation of the music industry for a partner that could give them a path into the future.

I’d suggest that Apple’s opportunity to do something revolutionary with the iPhone came because of the conservatism of Blackberry, Microsoft’s habit as a fast-follower, rather than an innovator, and Palm’s decline.

If there is a steady pace of innovation, often leavened by some competition, the sorts of holes that you can launch a Mac, an iPod, an iPhone, or even an iPad into just don’t exist. Constant steady improvement carry’s the day.

So I’d argue Sampere is provably1 wrong on Apple’s history. And it seems doubly weird to publish this two months before Apple Watch is set to hit. Potentially, Apple Watch is clearly another “we had never imagined but that, when we saw it, we all instantly wanted” product.

I would also argue that Apple’s record-shattering results last quarter are remarkable. Not because the iPhone 6 and 6 Plus are revolutionary, because they’re not. But because it shows that design can matter in the mass market. For decades the industry’s conventional wisdom held that design wasn’t important. The industry’s leaders created shitty software and shitty hardware. Apple’s success has upended the industry’s value system. Almost all of Apple’s competitors value design more today than they did a decade ago: Microsoft, Google, Samsung, HP — all of them.

Indeed, and the industry leaders created shitty software and shitty hardware because they weren’t selling to individual end-users, they were selling to IT managers, and CIOs.

There’s no reason to buy an iPhone 6 or 6 Plus other than because you’re willing to pay a premium for superior hardware and software quality. And last quarter 74 million people around the world did just that.

As its products evolve, Apple pours ever more effort into incremental improvements in the details. The bigger displays are the most noticeable differences in the iPhones 6, but everything else was improved too: the camera is better, both in terms of speed and image quality; the CPU is faster; the GPU is faster; battery life is better; the display quality is better; Touch ID is better. And then there’s Apple Pay.

Most of these are things that Clayton Christensen and his acolytes would call sustaining disruptions, and then there is Apple Pay. I’m not sure what to think of Apple Pay. To be clear, I think it will be huge, but part of the reason it will be huge is because it does such a good job of accommodating all the various players in the payment chain. Such accomodations usually require either a great deal of duress (like the dire straits of the music industry when Apple arranged iTMS), or a lot of kowtowing that ends up leaving an opportunity for something truly disruptive. Which isn’t to say there isn’t a middle road, but it remains to be seen how Apple Pay plays out.

Again, none of those improvements are revolutionary. But it’s a solid list of year-over-year improvements, and the results show that consumers agree. The most telling — dare I say dazzling — number Apple revealed last week wasn’t the number of iPhones they sold during the quarter, but the price people paid for them. Average selling price went up year-over-year, in an industry where average selling prices are going down.

A fundamentaist Christensen Disruptionist, will know that an excessive focus on average selling price (ASP) is what ends up leading innovators astray, leaving them vulnerable to disruption from below. Of course, Apple’s success in the computer market is a reminder that a company with a rising ASP in a market where ASP is otherwise trending downward isn’t necessarily doomed.

The problem isn’t that Apple has changed. The problem is that Apple has not changed, and their continuing success is proving that conventional disruption theory does not apply to consumer-driven markets in which outstanding design and integration (as opposed to modularity) can drive demand.

I’d suggest that there are two important factors that fundamentalist disrupptionists ignore at their peril. The first, is, as Gruber notes, that consumer markets are different from B2B markets. The other is that a lot of disruption theory focused not just on B2B markets for high-tech, they did so at a time when those markets were still on the long climb to reaching global scale, a time when computing was relatively precious, when Microsoft’s vision was a computer on every desk.  Now, many have one on their desk, another in their bag, one in their pocket, and one in their living room. Other parts of the world aren’t so well served, but mobile phones are widespread, and are being quickly followed by smart phones, and inexpensive PCs are spreading as well.

Apple is disrupting the conventional tenets of business even more than they are any particular product category in consumer electronics. There is something fascinating — in several ways unprecedented — going on with Apple right now. Rather than study it, understand it, describe it, and teach it, Sampere2 has chosen to deny that it’s happening.3

via Daring Fireball: Dazzling Results.

When you have a hammer, everything looks like a nail. I think disruption theory was a very useful tool for exploring business dynamics, and it remains so, but it is merely an imperfect attempt to model the world. I think that the mistakes Clayton Christensen and his acolytes make in understanding Apple, and the failed predictions they’ve produced to date, are, unfortunately, being repeated as the try to mis-apply the theory to other spheres, like education.

On Moore’s Law’s and the Apple Watch

Today, Jean Louis-Gassée shared some thoughts on forecasting demand for the Apple Watch.  He is thoughtful about the impact Moore’s Law will have on the higher-end models, writing:

But the biggest question is, of course, Moore’s Law. Smartphone users have no problem upgrading every two years to new models that offer enticing improvements, but part of that ease is afforded by carrier subsidies (and the carriers play the subsidy game well, despite their disingenuous whining).

There’s no carrier subsidy for the AppleWatch. That could be a problem when Moore’s Law makes the $5K high-end model obsolete. (Expert Apple observer John Gruber has wondered if Apple could just update the watch processor or offer a trade-in — that would be novel.)

Gasse’s comments were picked up elsewhere, including The Loop. So far though, I haven’t seen anyone consider the impact of the WatchKit software architecture on this question. So I commented with my own thoughts, which I’m revising and publishing here.

The current version of WatchKit only gives developers access to the the watch as a smart terminal interfacing communicating with code running on an iOS device over Bluetooth Low Energy. In this model, the processing demands on the watch are pretty flat across different apps and over hardware generations, since they are bounded primarily by display resolution, which is itself bounded by the optical characteristics of our eyeballs, and UI update rates. As such, Moore’s law improvements would probably accrue to battery life and component cost. I doubt that will have a significant impact on the end-user experience — I suspect that the screen and wireless make up a large portion of the power budget, and neither are going to be tightly linked to Moore’s law improvements. As for expense, I doubt that annual SoC price reductions will have that much impact on anything but the lowest end models (if it impacts them at all).

We’ll see what happens “later next year” when Apple allows native apps to run on the watch. My guess is that their execution will be tightly managed, as they were for web apps in iOS 1.0, and native apps when they were enabled by iOS 2.0. As a result, Apple will have a lot of ability to manage the way the platform and apps takes advantage of Moore’s law.

I’d guess that there will still be major generational discontinuities, but they will come every 5-10 years, rather than every year, as they do with iOS devices. That still creates issues for a device that some may expect to last lifetimes, but perhaps that assumption must itself be revisited.

For all the talk about the timelessness of high-end timepieces among analog watch aficionados, it isn’t all that relevant in the larger sense. The fundamental issues, when deciding whether to spend thousands of dollars on a watch, is: do I have thousands of dollars to spend on a watch when a $50 watch would tell me the time just as well; if so, does spending thousands of dollars on a watch feel good to me (which is sometimes a question of whether it sends the “right” message to others).

I suspect that for all the people who, today, spend thousands of dollars on a luxury watch because of their quality and timelessness, a significant portion could find a different reason to spend thousands of dollars on a watch. Along side of them, a significant number of other people who could spend thousands on a luxury watch, but are unpersuaded by whatever appeal drives traditional high-end watch buyers. Some of these people could find other reasons to buy an Apple watch.

I am at this point unlikely to spend thousands, or even hundreds on any watch, particularly since I finally gave up on wrist watches all together when I started carrying a “pocket watch” (read: cell phone). I will not be surprised though if I am wearing a lower-end Apple Watch a year from now. As a kid in Utah, when I used to go skiing, my friend and I would be freezing our asses off riding the ski lift, and would find ourselves checking our watches for the temperature. Our watches didn’t tell the temperature, and, as I recall, there was not yet any (affordable) watch on the market that told temperature. Still, it was natural to us that a wrist device should provide useful information like that. So, I’m willing to give the (lower end) Apple Watch a try.



The future of nations

This morning, in my car on the way to an appointment, I was listening to NPR’s The Takeaway. I felt like I caught a glimpse of the future of the nation-state in a common thread connecting their story about Indian Prime Minister Narendra Modi’s recent US visit, and another about ISIS.

In the piece on Modi, they talked with Arvind Rajagopal, a sociologist and media theorist at New York University. Rajagopal talked about how Modi made a strong nationalistic appeal to the Indian diaspora. What was notable, was that the appeal wasn’t for them to come back to India to support its development, but rather to support India’s interest wherever they were.

In the piece on ISIS, they interviewed Louise Shelly, the executive director of the Terrorism, Transnational Crime and Corruption Center at George Mason University. Shelly talked about how ISIS daws on age old smuggling routes in order to finance and supply its efforts.

What struck me is that both these stories hint at an interesting disconnect between geography, identity, power, influence and the economy, things that, in my naive view, are key ingredients in the classical nation state. And yet, here they are, present in forms that don’t quite fit that mold:

  • India, a nation with borders, redefined as a global nation with a geographic center, and a other aspects superimposed on other nation-states.
  • ISIS, a broad-based enterprise with aspirations of geographic borders whose assets are also superimposed on other nation-states.

And somewhere, in the same primordial goo, mixed with cryptocurrency, and social networks, and ragefaces, what else might emerge?