Month: January 2018

Tim Stevens of CNet has been digging more into BMW’s annual CarPlay fee:

And then I read that Matt Bubbers, an automotive journalist for the Globe & Mail, was given a curious statement by a representative from BMW Canada. He was told that Apple will be changing its fee structure such that manufacturers would need to pay on a per-car, per-annum basis to keep CarPlay running. That statement has since been retracted and a BMW Canada representative told me that Mr. Bubbers was given “inaccurate information.” However, in the confused hours in between initial statement and subsequent retraction, I was left wondering: just what does Apple charge for CarPlay, and indeed what does Google charge for Android Auto?

The answer, as I’d find out, is basically nothing — though it is a little more complicated than that.

In speaking with multiple sources at various manufacturers who offer cars with Apple CarPlay and/or Android Auto, I was quickly able to confirm that such fees, at least right now, do not exist. CarPlay and Android Auto, which are free for we consumers to use, are also provided for free for manufacturers to embed into their cars.

Here’s the crazy thing: even if Apple were not currently waiving the MFi per-device licensing fee, my understanding is that automakers would still only have to pay for the cost of a single Lightning connector component — which includes Apple’s royalty — per car, if they supplied their own connector cables. If they mount a USB port in the car instead and require drivers to supply their own Lightning-to-USB cable, I don’t believe that Apple would charge a royalty anyway. There are additional development costs for automakers to integrate and test CarPlay with their own system, and Stevens says that Apple charges a fee to participate in the MFi/Made for Apple accessory program.

Even with those costs, though, I can’t see a reasonable justification for BMW to charge owners $80 per year to use CarPlay — other than, of course, because they can.

Update: There is an authentication co-processor component as well.

Filip Struhárik, editor and social media manager at Slovakia’s Denník N newspaper:

Our traffic decreased by three percent in November and by nearly six percent in December 2017 (real users, year-on-year). Traffic to some other (mostly smaller) sites fell by tens of percentage points after the Explore Feed test started.

For a long time, Facebook was a main source of traffic for Denník N — around 40 percent of our readers came from Facebook. But this has changed. In December, less than 30 percent of our traffic came from Facebook. In November and December 2017, we had more visitors from Google than from Facebook for the first time (and it’s happening everywhere).

Although our reach, engagement, interactions and consumption have fallen dramatically, something interesting is happening. When we look at our “Reach Engagement Rate”, we can see that it‘s growing, especially after the Explore Feed test started.

What this suggests is that Facebook is concentrating visitors into audiences. This may reduce traffic and minimize the spread of biased and misleading news links amongst casually-interested users, but Struhárik’s post indicates that it could reinforce more active users’ news bubbles too.

Alanis King, Jalopnik:

While a lot of manufacturers stick Apple CarPlay into their vehicles as standard equipment these days, The Verge reports that it’s been a one-time $300 charge for BMW buyers since BMW started offering it on cars with built-in navigation in 2017. But BMW North America’s technology product manager Don Smith told The Verge that’ll change next year, and CarPlay will cost owners $80 a year.

To be clear, there’s nothing remotely subscription-based in CarPlay from a consumer’s perspective. The phone connects to the car’s screen, displays its own UI, and routes its audio through the car’s speakers — that’s pretty much it. There’s no justification for this other than nickel-and-diming iPhone users.

Apple:

Combining new investments and Apple’s current pace of spending with domestic suppliers and manufacturers — an estimated $55 billion for 2018 — Apple’s direct contribution to the US economy will be more than $350 billion over the next five years, not including Apple’s ongoing tax payments, the tax revenues generated from employees’ wages and the sale of Apple products.

Planned capital expenditures in the US, investments in American manufacturing over five years and a record tax payment upon repatriation of overseas profits will account for approximately $75 billion of Apple’s direct contribution.

As of right now, I am no longer paying my taxes like everyone else. I am contributing to the Canadian economy, and will be issuing a self-congratulatory press release every April.

By the way, Apple estimates that they will pay $38 billion to repatriate $245 billion in income stored internationally, so the actual increase in expenditures and investments in American manufacturing will be $37 billion over five years, or about $7.2 billion per year.

Apple expects to invest over $30 billion in capital expenditures in the US over the next five years and create over 20,000 new jobs through hiring at existing campuses and opening a new one. Apple already employs 84,000 people in all 50 states.

The company plans to establish an Apple campus in a new location, which will initially house technical support for customers. The location of this new facility will be announced later in the year.

This is pretty big news. I’ve seen a handful of reports stating that this will be Apple’s “second” campus. It is not. Apple already has two well-known campuses — Infinite Loop and Apple Park — plus at least one more, in Austin, Texas.

There’s a lot to love about this press release; but, like many of the corporate gestures following last month’s U.S. tax cuts, I don’t see anything here that couldn’t have been done at the previous tax rate if companies like Apple were unable to withhold income internationally. I’m going to get emails for writing that, right?

Update: Tim Bradshaw of the Financial Times breaks down how Apple is calculating their $350 billion economic contribution:

Wednesday’s headline $350bn figure, though, does not include that kind of thing. What it does include is its annual spending with US-based suppliers and manufacturers over five years, capital expenditure plans for its new campus and data centres and a record tax payment related to its repatriation of overseas profits.

Spending with US suppliers was $50bn last year and will be $55bn this year, Apple says. Cynics might argue that this is money Apple would have spent anyway.

Bradshaw on Twitter:

Breaking down Apple’s $350bn “direct contribution” to US:
$275bn+ of spending with US suppliers at $55bn+/year
+$38bn tax bill (estimated) for repatriation of overseas profits
+$30bn capex on new campus, data centres etc
+$5bn adv manuf fund
=$350bn(ish)

A 10% year-over-year increase in supplier spending certainly doesn’t have the impression of that eye-popping $350 billion figure, but it’s nothing to sneeze at either.

I’m interested to see how — if — Apple’s competitors respond.

Ever since Amy Wang published these two paragraphs in the Washington Post, my Twitter timeline has been lit up with UI designers wanting to know how what is described here is possible:

Shortly after 8 a.m. local time Saturday, an employee at the Hawaii Emergency Management Agency settled in at the start of his shift. Among his duties that day was to initiate an internal test of the emergency missile warning system: essentially, to practice sending an emergency alert to the public without actually sending it to the public.

[…]

Around 8:05 a.m., the Hawaii emergency employee initiated the internal test, according to a timeline released by the state. From a drop-down menu on a computer program, he saw two options: “Test missile alert” and “Missile alert.” He was supposed to choose the former; as much of the world now knows, he chose the latter, an initiation of a real-life missile alert.

Today, the Honolulu Civil Beat published a photo of this menu. I’ve reproduced it here for ease of discussion:

Photo of Hawaii Emergency Management alert screen by Honolulu Civil Beat.
Hawaii Emergency Management alerts screen

Look at this list. I mean really look at it. The link that the operator clicked is marked “PACOM (CDW) – STATE ONLY”; the one that they should have clicked is marked “DRILL – PACOM (CDW) – STATE ONLY”.

  1. This list appears to be in no particular order. The link to initiate an internal test is not beside the one for a live, public alert, nor is grouped with other internal tests.

  2. The use of uppercase text is inconsistent. In some instances — “PACOM” and “CAE” — it is used for initialisms, but in others — “DRILL” and “TEST” — it is used for emphasis. In the case of the two links here, uppercase is used for both emphasis and an initialism.

  3. On a related note, uppercase text is harder to read than mixed-case text.

  4. Aside from the text itself, here are no visual clues in this list to differentiate a test alert from a live alert.

Without knowing how this system is built, it would be ridiculous to suggest they modernize it or create separate menus for test and live alerts. In fact, I think the simplicity of this menu is a strength, not a weakness. But there are some steps that I think the Emergency Management Agency could take to reduce the likelihood of this happening again:

  1. Reorder the list so that test alerts are grouped together, and clearly separated from live alerts.

  2. Clarify the use of uppercase words. Because government agencies love to use initialisms, that is, by default, the only instance in which uppercase words should be used. All other words should be in sentence or title case.

  3. Differentiate test and live alerts further. If it is not possible to change their colour, perhaps it is possible to add a symbol in front, even something as simple as three exclamation marks to indicate that the alert will be sent to the public. Test alerts should also be more clear; perhaps prefacing each one with something as simple as “Internal Only:” would make it easier to understand that those alerts won’t be public.

    I know I’m making it sound trivial to differentiate each kind of alert, but it isn’t — it needs to be something that’s clear in both a calm test-only environment and in an emergency.

  4. More clearly indicate the false alarm option, as it is neither a test nor an emergency live alert. It undoes a previous live alert, and should more clearly indicate that.

As I wrote above, I don’t know what is possible within the existing emergency system; even something as apparently simple as reordering the list may require many hours of work. But I don’t think the options above are entirely unreasonable and, in conjunction with requiring that more than one person sign off on issuing an alert, would add an extra layer of safety by reducing complexity in this UI.

Update: Marcel Honore of the Honolulu Civil Beat reports that the UI shown above is not, in fact, identical to the UI that an operator would see:

However, state officials now say that image was merely an example that showed more options than the employee had on the actual screen.

“We asked (Hawaii Emergency Management Agency) for a screenshot and that’s what they gave us,” Ige spokeswoman Jodi Leong told Civil Beat on Tuesday. “At no time did anybody tell me it wasn’t a screenshot.”

Honore obtained a different screenshot which, while prettier, still has the same problems as the example screenshot above.

Thanks to Kyle Dreger for pointing me to this update.

The Awl and the Hairpin announced today that, as of the end of this month, both will stop publishing. I’m gutted. Both have been a must-read for me every single day for years. They will be missed.

Update: Christina Warren:

As we saw with Gawker, DNA Info/Gothamist, and now The Awl sites, independent media needs support more than ever. These voices are too important to just disappear.

All of these publications have been shuttered for different reasons. None were, as far as I know, heavily buoyed by Facebook referral traffic. But with Facebook deemphasizing publisher pages you can expect to see more independent media organizations losing staff or shutting down from a raw decline in traffic. Independent publishers — this aspiring one included — need to find a more sustainable revenue source.

See Also: Writers published in the Awl nominate their favourite Awl pieces.

Dave Mark:

Over the weekend, someone started a thread asking why an artist’s album view in Apple Music has gotten so cluttered.

To see this for yourself, pick a relatively modern artist and check out their list of albums in the Music app. For example, fire up Siri and say:

Show me all the Bruno Mars albums

When the Bruno Mars page appears, scroll down to the Albums section and tap See All. Amongst the actual Bruno Mars albums, you’ll find a lot of singles and EPs. Way more singles and EPs than actual albums, in fact.

It gets worse than that — many artists list both clean and explicit versions of each release, which means that the Album view in Apple Music is often twice the apparent size.

Mark and Kirk McElhearn put the blame on the ID3 audio metadata standard, and that’s fair: ID3 doesn’t have a field to distinguish between LPs, EPs, singles, and other release types.

The iTunes Store worked around the ID3’s limitations by sorting releases by popularity — I presume — instead of reverse chronologically. Compare, for example, Kanye West’s albums on the iTunes Store and Apple Music. The iTunes Store fits one more release onscreen than Apple Music but, more importantly, everything shown on the iTunes Store is a full-length album; on Apple Music, all six releases shown at the top of the Albums screen are singles.1 I don’t know if sorting by popularity is translatable to Apple Music and its users’ listening patterns, but it is perhaps worth investigating.2

Of note, Spotify does not have this problem; it correctly separates albums and singles. I don’t know how they do this — manually, perhaps? — but it makes Apple Music look sloppy by comparison.


  1. And five of them are collaborations where West is only a featured artist on the track. ↥︎

  2. I’d also like to see separate clean and explicit releases consolidated in Apple Music, with a toggle at the bottom of the album page to show one or the other. Showing both and effectively treating them as separate releases just creates clutter. ↥︎

Alex Kantrowitz, Buzzfeed:

Facebook on Thursday introduced major changes to its News Feed that will prioritize content it hopes will spark meaningful conversations between friends while deprioritizing content from businesses, brands, and media. The move is widely expected to hurt publishers that rely on traffic from Facebook.

Fred Vogelstein interviewed Adam Mosseri of Facebook for Wired:

FV: Talk to me about like the evolution of this. What’s changed over the course of the past 18 months to make you feel like this is something worth doing?

AM: The biggest thing has been just the explosion of video. Video is a paradigm shift in a lot of different ways. We’ve done a lot to try and nurture it. We think video is going to continue to be a more and more important part about how people communicate with each other, and how publishers communicate with people.

But as video has grown on Facebook, it has changed the nature of how people interact with the platform in a lot of different ways. Video is, primarily, a passive experience. You tend to just sit back and watch it. And while you’re watching it, you’re not usually liking or comment or speaking with friends. So this change is, in part, a reaction to how the ecosystem has shifted around us.

Less than a year ago, Facebook was actively encouraging publishers to create video specifically for their platform. Entire companies laid off employees and began a “pivot to video”, to the extent that even that phrase became a cliché in publisher circles, but most of the videos were lightly edited ads that — unsurprisingly — failed to find an audience. In a handful of markets, Facebook already moved publishers’ posts to a secondary news feed, with predictable consequences.

It is absolutely critical for publishers to disconnect from their reliance upon major referrers like Facebook and Google. And, yet, I’m not sure that’s realistic for a lot of major media organizations. Referral traffic remains a massive source of visitors — as Casey Newton points out, however, visitors are not the same thing as an audience. And, as I wrote six months ago, I think it’s a mistake to write off changes by referral sources as the fault of the publisher for relying upon that traffic:

As companies like Facebook and Google increasingly dominate actual publishers for how users get their news, even creating proprietary formats like Instant Stories and AMP for preferential treatment, shouldn’t their practices be scrutinized to a greater degree? Is it really fair for the rug to be pulled out from under publishers’ feet when their primary referrer decides it’s convenient for their business model? Does it make sense for the future of the worldwide digital media economy to be decided by a few young men in California? To return to the argument against publishers’ reliance upon traffic sources like Facebook and Google, is it possible to build a successful new publication without them?

Publishers shouldn’t be reliant upon Facebook and Google sending them traffic,1 but that truth also abdicates the responsibility of large tech companies.

Joshua Topolsky of the Outline is optimistic that the hit to publishers from a lack of Facebook traffic won’t be as significant as the hit to Facebook from a lack of news posts:

Facebook, despite all its best intentions, is still just a dumb pipe — a thing that delivers, not the thing itself. The pipe must be filled up, yes, with stuff like groups you belong to and photos of new babies, yes with Messenger conversations and events and fundraisers. But information is currency, and what is valuable to most people is to know what the fuck is going on in the world and to try and understand it. That doesn’t go away because Facebook wants to keep its hands clean. It simply goes somewhere else. Even the market had a negative reaction to this news, stripping around $25 billion off the network’s market cap following the announcement. I don’t think that’s a fluke — I think Facebook doesn’t know what its product really is.

[…]

Frankly, any publisher relying on Facebook for survival fucked up. But there’s a flip side to this. There’s the opportunity for outlets willing to rely less on social networks to set their fate, publishers who have diversified their traffic sources, who have pushed back on Facebook’s News Feed carrots, who have built (or are building) brands that resonate with audiences beyond what can be bought or given. Value not gifted by Facebook could be a very good thing for publishers. (And yes, I get that I’m also talking about The Outline, which is fighting for its right to survive in a very uncertain landscape every single day.)

I hope this means a new dawn for good publications, and an awakening to build a dedicated audience instead of simply driving traffic.2 I also think that this unfairly excuses Facebook from building their business on publishers and media for years. A consequence Facebook would understand is if their active users dropped — unfortunately, even if you, I, and most of the people we know stopped using Facebook, their Borg-like dominance on the web is unparalleled. But we can make a difference in the fortunes of publishers: support them financially by subscribing.

Update: Borzou Daragahi:

Terrifying to think how much one rich man’s decision on what direction he wants to take his company will have an impact on people’s epistemological sense of self – how they perceive the world. It’s too much power.


  1. In fact, publications that are entirely dependent on traffic from Facebook or Google are, typically, nothing you’d actually want to read anyway. ↥︎

  2. Covering websites in ads that generate money solely based on the number of views and clicks likely has a significant role in this. ↥︎

Graham Cluley:

The Washington Post sheds some more light on what went wrong:

Shortly after 8 a.m. local time Saturday morning, an employee at the Hawaii Emergency Management Agency settled in at the start of his shift. Among his duties that day was to initiate an internal test of the emergency missile warning system: essentially, to practice sending an emergency alert to the public without actually sending it to the public.

Around 8:05 a.m., the Hawaii emergency employee initiated the internal test, according to a timeline released by the state. From a drop-down menu on a computer program, he saw two options: “Test missile alert” and “Missile alert.”

This sounds like terrible user interface design to me. Why have the genuine “Jeez Louise! Freak out everybody!” option slap-bang next to the harmless one labelled “Test the brown alert”?

Even though the menu option still required confirmation that the user really wanted to send an alert, that wasn’t enough, on this occasion, to prevent the worker from robotically clicking onwards.

How on Earth were those buttons next to each other? And why can just one person send an alert like this to millions of people? And, finally, why weren’t the local authorities authorized to send out a retraction of this alert for thirty-eight minutes?

Joel Spolsky:

Both Twitter and Facebook’s selfish algorithms, optimized solely for increasing the number of hours I spend on their services, are kind of destroying civil society at the same time. Researchers also discovered that the algorithms served to divide up the world into partisan groups. So even though I was following hundreds of people on social networks, I noticed that the political pieces which I saw were nevertheless directionally aligned with my own political beliefs. But to be honest they were much… shriller. Every day the Twitter told me about something that The Other Side did that was Outrageous and Awful (or, at least, this was reported), and everyone was screeching in sync and self-organizing in a lynch mob, and I would have to click LIKE or RETWEET just to feel like I had done something about it, but I hadn’t actually done anything about it. I had just slacktivated.

What is the lesson? The lesson here is that when you design software, you create the future.

The public awakening in the past year to the more toxic and unethical effects of Silicon Valley firms, generally, is long overdue. The tech industry should have done a better job of regulating themselves for years, but they now have an opportunity to make up for their delinquency. I worry that they are incapable of doing so, and could be answering to the current U.S. administration instead.

Update:Market forces” isn’t a good answer to any of this.

Brad Frost:

But ultimately, Facebook is a place you go to. You can decide whether you want to visit the restaurant, or just continue throwing their flyers in the recycling bin alongside the coupon-stuffed weekly circulars and junk mail.

Google is equally needy, but feels a lot more insidious than Facebook. Unlike Facebook, Google isn’t just a place you go. It’s built into the infrastructure of your life. It’s your house. It’s the roads and sidewalks you travel on. Google is a lot more infrastructural than Facebook, which is why breeches of trust feel a lot weirder and scarier.

Turns out that if you buy a smartphone that runs an operating system made by an advertising company that loves to scoop up as much user data as it can, it’s going to endlessly nag you to provide more information to that company. That’s not to say that anyone who buys an Android phone is an idiot for expecting otherwise; on the contrary, users’ expectations should guide Google’s actions.

Also, always remember that someone actually built this stuff. There are, of course, employees in every industry who hang their souls up when they walk into their office, but very few have the kind of power and responsibility of a global tech giant.

See Also:Death to Bullshit”, also by Frost.

Maya Kosoff, Vanity Fair:

The stagnation goes beyond C.E.S.’s scant diversity and casual sexism, extending to the products themselves, which feel like rehashed versions of the same technologies, packaged and presented in only slightly new ways. Year after year, the show produces more of the same from headlining companies: Internet-connected refrigerators (which have been around since 1998 but have failed to take off, despite their persistent presence on showroom floors); self-driving cars; and virtual- and augmented-reality technology. It’s telling that the most interesting thing that has happened so far this year was the show’s complete loss of power on Wednesday, which offered a brief, terrifying preview of the sort of Stone Age hysteria we can expect if the Internet of Things ever takes down the power grid.

Ben Bajarin published a decent piece today about Apple’s fading influence at CES. He has theories on why that may be, like Amazon’s Alexa devices dominating the smart speaker space, and a more mature consumer electronics market. But I have another theory: maybe CES is full of companies trying to carve their own little space with expensive gadgets that don’t work well and, ultimately, are of little relevance to what consumers will actually want or buy. Sure, there were plenty of products shown that work with Apple’s ecosystem — mostly HomeKit — but so much of what is shown at CES is just gadgetry for the sake of gadgetry. Does it matter how much Apple’s influence is felt at a showcase of stuff that’s mostly irrelevant?

Zac Hall, 9to5Mac:

When asked about the move to sell a third-party mesh system and the future of the AirPort line, an Apple spokesperson shared this with 9to5Mac:

People love our AirPort products and we continue to sell them. Connectivity is important in the home and we are giving customers yet another option that is well suited for larger homes.

Apple’s choice for that option is the Linksys Velop Whole Home Mesh Wi-Fi System which comes in two flavors: $350 for a 2-pack system or $500 for a 3-pack solution. The Tri-Band Wi-Fi system is rated to provide coverage for 2,000 square feet with each Node which can be configured from the Linksys iPhone and iPad app.

There are non-answers, and then there are Apple-grade non-answers. That statement confirms that WiFi is basically an expectation these days — duh — and that they are presently selling their AirPort lineup. More telling, though, is what they don’t say: there’s no confirmation that they’re even remotely interested in continuing to offer their own base station, which is remarkable even less commitment than they made to updating the Mac Mini.

In addition to this move, Mark Gurman’s reported in 2016 that Apple had disbanded the AirPort team, and I’ve heard thirdhand that no updates are planned.1 I’m convinced that the AirPort lineup is dead and will quietly be removed from Apple’s store and website in the not-too-distant future.

Update: Good question from Al Rumball-Smith:

I’ve been trying to work out what’s the future for Time Capsule then? iCloud?

I think something like Time Machine in the Cloud is a reasonable guess. I could also see more third-party routers supporting Time Machine via a USB-connected hard drive — apparently, some Netgear and Asus routers have done so for a while.

Update: A reader email reminded me that Apple took at least two months to patch their base station products to protect against a significant WiFi vulnerability. iOS and MacOS were updated within two weeks. I don’t know if the thirdhand information I have is right, of course, but the general thrust of the reports I’ve seen and moves Apple has made when it comes to their AirPort lineup strongly suggests that they’re not interested in the WiFi router market much longer.


  1. They haven’t even bothered to update the iOS app with support for the iPhone X’s display. ↥︎

Alexis C. Madrigal, the Atlantic:

It all started with an Instagram ad for a coat, the West Louis (TM) Business-Man Windproof Long Coat to be specific. It looked like a decent camel coat, not fancy but fine. And I’d been looking for one just that color, so when the ad touting the coat popped up and the price was in the double-digits, I figured: hey, a deal!

The brand, West Louis, seemed like another one of the small clothing companies that has me tagged in the vast Facebook-advertising ecosystem as someone who likes buying clothes: Faherty, Birdwell Beach Britches, Life After Denim, some wool underwear brand that claims I only need two pairs per week, sundry bootmakers.

[…]

Several weeks later, the coat showed up in a black plastic bag emblazoned with the markings of China Post, that nation’s postal service. I tore it open and pulled out the coat. The material has the softness of a Las Vegas carpet and the rich sheen of a velour jumpsuit. The fabric is so synthetic, it could probably be refined into bunker fuel for a ship. It was, technically, the item I ordered, only shabbier than I expected in every aspect.

Madrigal’s a smart guy, so I’m not sure I buy the idea that he thought he could get anything better than H&M quality for H&M-like prices. But it’s pretty incredible to me that almost anyone with a few hours to spare every day could conceivably run a convincing-looking boutique online with no held inventory, no unique products of its own, and little risk. This sort of thing fascinates me — partly because I find fast fashion brands generally objectionable, but also because of how inventive it is. The scheme Madrigal describes is the product of relatively accessible technologies that simply weren’t available not that long ago.

Younger and more alert shoppers have already cottoned on to this scheme, though, and are bypassing the Shopify storefront to shop directly from AliExpress. In the haul video genre on YouTube, there are over half a million results for AliExpress shopping sprees. For comparison, there are a little over a million results for each “H&M haul” and “Zara haul”, and less than 200,000 results for each “Abercrombie haul”, “Hollister haul”, and “Lululemon haul”.

See Also: I previously linked to Jenny Odell’s essay, “There’s No Such Thing as a Free Watch”.

Criteo:

Earlier this month, Apple launched a new version of its mobile operating system, iOS 11.2, which disables the solution that some companies in the advertising ecosystem, including Criteo, currently use to reach Safari users. As a result, we believe the projected 9%-13% ITP net negative impact on Criteo’s 2018 Revenue ex-TAC relative to our pre-ITP base case projections, communicated on November 1, 2017, is no longer valid.

We are focused on developing an alternative sustainable solution for the long term, built on our best-in-class user privacy standards, aligning the interests of Apple users, publishers and advertisers. This solution is still under development and its effectiveness cannot be assessed at this early stage. Should it not mitigate any ITP impact, we believe the ITP net negative impact on Criteo’s 2018 Revenue ex-TAC, relative to our pre-ITP base case projections, would become approximately 22%.

Alex Hern, the Guardian:

Internet advertising firms are losing hundreds of millions of dollars following the introduction of a new privacy feature from Apple that prevents users from being tracked around the web.

[…]

With [Criteo’s] annual revenue in 2016 topping $730m, the overall cost of the privacy feature on just one company is likely to be in the hundreds of millions of dollars.

It’s hard to feel sympathy for a company that earned its money by encouraging website owners to embed their JavaScript without requiring they disclose their tracking practices, forcing their way around my cookie preferences, and trying to use another loophole to bypass users’ privacy settings. I’m very okay with Criteo getting their just deserts.

An open letter to Google:

We acknowledge the problem of Web pages being slow to load, relative to alternative, proprietary technologies such as Facebook Instant Articles and Apple News. Publishers (especially in news media) have long faced difficult choices and poor incentives, leading to bad decisions and compromises, and ultimately to terrible user experiences.

Search engines are in a powerful position to wield influence to solve this problem. However, Google has chosen to create a premium position at the top of their search results (for articles) and a “lightning” icon (for all types of content), which are only accessible to publishers that use a Google-controlled technology, served by Google from their infrastructure, on a Google URL, and placed within a Google controlled user experience.

Malte Ubl of Google announced today that, later this year, Google will stop rewriting AMP pages to their own domain. That still doesn’t fix the critical dependency problems that the AMP letter raises. I co-signed this letter with dozens of others. I think it’s important to recognize the dominant position Google has on the web.

Vlad Savov, the Verge:

After an entire year of speculation about whether Apple or Samsung might integrate the fingerprint sensor under the display of their flagship phones, it is actually China’s Vivo that has gotten there first. At CES 2018, I got to grips with the first smartphone to have this futuristic tech built in, and I was left a little bewildered by the experience.

The mechanics of setting up your fingerprint on the phone and then using it to unlock the device and do things like authenticate payments are the same as with a traditional fingerprint sensor. The only difference I experienced was that the Vivo handset was slower — both to learn the contours of my fingerprint and to unlock once I put my thumb on the on-screen fingerprint prompt — but not so much as to be problematic. Basically, every other fingerprint sensor these days is ridiculously fast and accurate, so with this being newer tech, its slight lag feels more palpable.

The technology here is impressive, but it is an iteration on a security solution that has been eclipsed by accurate facial recognition that isn’t dependent on ambient lighting conditions. I know that, for the iPhone users who aren’t convinced by facial recognition, it’s only fair to compare this familiar technology against today’s version of Face ID and want to see both in a future iPhone model.

I don’t think that scenario is likely. There are shortcomings with Face ID today — it’s unreliable at very close range, some sunglasses don’t work with it, and it can’t recognize faces through facial coverings — but the next iPhone is likely to feature improvements to Face ID, not a duplicative authentication mechanism. From my limited perspective, it seems more efficient for Apple to use their engineering talent to make progress on Face ID rather than trying to integrate both.

Elizabeth Jones, AppleToolBox:

Is it just me or are those daily upgrade notifications for upgrading to macOS High Sierra annoying the bleep out of you? Every time I turn on my MacBook (2017,) it immediately starts up with that exasperating High Sierra notice to upgrade to High Sierra so I can “enjoy the latest technologies and refinements.” And it’s even popping up on my iMac (2015 with Fusion Drive,) that Apple itself recommends NOT updating to High Sierra. And I really DON’T want to upgrade to macOS High Sierra right now on any of my Macs!

Unfortunately, Apple is only supporting fixes and mitigations for Meltdown and Spectre in High Sierra, contra their original statement. MacOS updates have generally been less impactful since Yosemite,1 there are still lots of reasons why users may be reluctant to upgrade to a major new OS version. While Apple’s developer site displays a pie chart indicating iOS version market share, I can find no such official chart for MacOS market share. As of December, though, MacOS Sierra was still more widely used than High Sierra, and El Capitan isn’t that far behind, according to StatCounter. For serious security vulnerabilities, Apple should strongly consider issuing patches for previous widely-used system versions.


  1. And, just as importantly, have improved older hardware compatibility. ↥︎