Start Up: ageing Android, charging iOS, US guns by numbers, the nearest “Earth”, bitcoin bumps in Zimbabwe, and more

We can find out who owns this, but what do corporations own in England? Photo by Jonathan Rolande on Flickr.

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 11 links for you. Unsupported by pastors. I’m @charlesarthur on Twitter. Observations and links welcome.

How out of date are android devices? • Dan Luu

Luu is an ex-Googler:


People sometimes compare Android to Windows XP because there are a large number of both in the wild and in both cases, most devices will not get security updates. However, this is tremendously unfair to Windows XP, which was released on 10/2001 and got security updates until 4/2014, twelve and a half years later. Additionally, Microsoft has released at least one security update after the official support period (there was an update in 5/2017 in response to the WannaCry ransomware). It’s unfortunate that Microsoft decided to end support for XP while there are still so many XP boxes in the wild, but supporting an old OS for over twelve years and then issuing an emergency security patch after more fifteen years puts Microsoft into a completely different league than Google and Apple when it comes to device support.

Another difference between Android and Windows is that Android’s scale is unprecedented in the desktop world. The were roughly 200 million PCs sold in 2017. Samsung alone has been selling that many mobile devices per year since 2008. Of course, those weren’t Android devices in 2008, but Android’s dominance in the non-iOS mobile space means that, overall, those have mostly been Android devices. Today, we still see nearly 50 year old PDP-11 devices in use. There are few enough PDPs around that running into one is a cute, quaint, surprise (0.6 million PDP-11s were sold). Desktops boxes age out of service more quickly than PDPs and mobile devices age out of service even more quickly, but the sheer difference in number of devices caused by the ubiquity of modern computing devices means that we’re going to see many more XP-era PCs in use 50 years after the release of XP and it’s plausible we’ll see even more mobile devices around 50 years from now. Many of these ancient PDP, VAX, DOS, etc. boxes are basically safe because they’re run in non-networked configurations, but it looks like the same thing is not going to be true for many of these old XP and Android boxes that are going to stay in service for decades.

We’ve seen that Android devices appear to be getting more out of date over time. This makes it difficult for developers to target “new” Android API features, where new means anything introduced in the past few years. It also means that there are a lot of Android devices out there that are behind in terms of security. This is true both in absolute terms and also relative to iOS.


link to this extract

Just how fast is “faster wireless charging” in iOS 11.2? • BirchTree

Matt Birchler:


iOS 11.2 is currently in beta, and will be released to all iPhone and iPad users in the coming weeks, and one of the key features for iPhone 8/8 Plus/X owners is accelerated wireless charging. Previously, all wireless charging was limited to 5W, but this update will raise that limit to 7.5W. That’s a 50% increase in power on paper, but I had to know what the real world difference was.

As you can see from the graph above, the difference between wireless charging on my “fast charge” Samsung charging pad was slight. There is definitely a difference here, and if you’re already using wireless charging (and your pad supports it), then this is an undeniable win. However, if you were hoping that this would make wireless charging catch up to wired then you’re going to be very disappointed…

…Wired charging remains the fastest way to charge the iPhone in 2017, and it’s not even close. It’s popular to hate on the charger in the box, butthe stock iPhone charger gets the iPhone 8 Plus to 79% in 2 hours (68% faster) and up to 21% at the 30 minute mark (91% faster). That’s a pretty striking difference, and if speed is of the essence, it’s a much better way to get topped up fast.


Wonder if Apple’s AirPower will do any better.
link to this extract

The gun numbers: just 3% of American adults own a collective 133m firearms • The Guardian

Lois Beckett:


surveys show that gun ownership in America is actually highly concentrated. Only 22 to 31% of Americans adults say they personally own a gun.

Rates of personal and household gun ownership appear to have declined over the past decades – roughly two-thirds of Americans today say they live in a gun-free household. By contrast, in the late 1970s, the majority of Americans said they lived in a household with guns.

Most of America’s gun owners have relatively modest collections, with the majority of gun owners having an average of just three guns, and nearly half owning just one or two, according to a 2015 survey by Harvard and Northeastern researchers, which gave the most in-depth estimate of Americans’ current patterns of gun ownerships.

But America’s gun super-owners, have amassed huge collections. Just 3% of American adults own a collective 133m firearms – half of America’s total gun stock. These owners have collections that range from eight to 140 guns, the 2015 study found. Their average collection: 17 guns each.

After the Las Vegas shooting, officials said the killer had 23 guns in his hotel room, and another 19 at home. Some Americans asked, shocked, why one person purchasing so many guns had not set off any red flags.

Part of the answer is that owning more than 40 guns is actually fairly common in the United States: there are an estimated 7.7 million super-owners, which might make it difficult to flag a mass shooter building an arsenal from enthusiastic collectors and gun enthusiasts piling up different kinds of guns for hunting different kinds of game, a selection of handguns for self-defense, and various accessories for the popular, customisable military-style rifles that enthusiasts have compared to lethal Lego sets for grown men.


Easily overlooked that ownership isn’t evenly distributed.
link to this extract

The companies and corporate bodies who own a third of England & Wales • Who owns England?

Anna Powell-Smith:


On 7 November 2017, the Land Registry for England & Wales for the first time released details of 3.5 million land titles owned by UK corporate bodies – councils, UK companies, housing associations and more. Going by separate Land Registry figures we’ve seen for the acreages these bodies own, we can safely say that companies and the public sector own around a third of England and Wales.

And now, for the first time, I’ve mapped them.

To be precise, I’ve mapped 1.8 million of the 3.5 million land titles released by the Land Registry – all the ones that include postcode locations. The remaining 1.7 million are rather trickier to map – their land titles are somewhat vaguer in their descriptions, like “Land north of Stansfield Road, Wigan”. But I’m having a crack at mapping approximate locations for these, using an OpenStreetMap-based geocoder (it takes time). Watch this space!

We’ll also be analysing the complete dataset in the weeks and months to come, so check back for more blogs assessing things like land owned by housing developers, councils and airports.


Powell-Smith does amazing things with data.
link to this extract

The brutal fight to mine your data and sell it to your boss • Bloomberg

Drake Bennett on the legal fight between LinkedIn and HiQ, which used LinkedIn’s API to build its business:


The easier it gets to harvest and analyze information, the more actively that information has to be protected. That, Verrilli argues, is what LinkedIn is trying to do.

It’s LinkedIn, after all, not HiQ, that has the relationship with the members who have posted the information. “We’ve made promises in our privacy policy, and we have to work with regulators worldwide who hold us to our promises,” says Blake Lawit, LinkedIn’s vice president for legal. “We’re not under the radar, right? If we do something creepy with privacy, we’re going to hear about it from the FTC and the Irish Data Protection Commissioner and et cetera.” In other words, LinkedIn might be big and know lots about us, but with great power comes regulatory scrutiny, and with that comes a kind of responsibility.

If that argument is only somewhat reassuring, HiQ’s argument is effectively that we’re on our own, and that this is the price we pay for today’s internet. “There’s probably lots and lots of applications that might make someone feel a little queasy, right?” Gupta told Judge Chen. “But the thing is, we can’t sit here today and police every possible business model that some entrepreneur in Silicon Valley might come up with. It’s public information. It’s the marketplace of ideas. It’s the engine of our country’s growth.” The reason Google can put the entire internet at our fingertips is because, like HiQ, it scrapes public data. That includes LinkedIn pages, which is why they tend to be among the top results if you Google a noncelebrity (unlike HiQ, Google has LinkedIn’s explicit permission to collect data).


link to this extract

Closest temperate world orbiting quiet star discovered • European Space Observatory


A temperate Earth-sized planet has been discovered only 11 light-years from the Solar System by a team using ESO’s unique planet-hunting HARPS instrument. The new world has the designation Ross 128 b and is now the second-closest temperate planet to be detected after Proxima b. It is also the closest planet to be discovered orbiting an inactive red dwarf star, which may increase the likelihood that this planet could potentially sustain life. Ross 128 b will be a prime target for ESO’s Extremely Large Telescope, which will be able to search for biomarkers in the planet’s atmosphere.

A team working with ESO’s High Accuracy Radial velocity Planet Searcher (HARPS) at the La Silla Observatory in Chile has found that the red dwarf star Ross 128 is orbited by a low-mass exoplanet every 9.9 days. This Earth-sized world is expected to be temperate, with a surface temperature that may also be close to that of the Earth. Ross 128 is the “quietest” nearby star to host such a temperate exoplanet.

“This discovery is based on more than a decade of HARPS intensive monitoring together with state-of-the-art data reduction and analysis techniques. Only HARPS has demonstrated such a precision and it remains the best planet hunter of its kind, 15 years after it began operations,” explains Nicola Astudillo-Defru (Geneva Observatory – University of Geneva, Switzerland), who co-authored the discovery paper.


Red dwarf stars are dead, surely? But this is close-ish, astronomically.
link to this extract

Bitcoin surges in Zimbabwe after military seize power • South Africa Times

Robert Brand, Brian Latham and Godfrey Marawanyika:


Bitcoin climbed as much as 10% on Zimbabwe’s Golix exchange on Wednesday after the country’s armed forces seized power.

The price of the cryptocurrency in the Southern African nation jumped as high as $13,499, almost double the rate at which it trades in international markets, according to prices cited on Golix’s website.

Demand for Bitcoin in Zimbabwe has surged amid a shortage of hard currency. Golix processed more than $1m of transactions in the past 30 days, compared with turnover of $100,000 for the whole of 2016, according to data on the exchange’s website. Zimbabwe doesn’t have its own currency, with the government adopting the US dollar and South African rand, among others, as legal tender in 2009 after hyperinflation rendered the local dollar worthless.

Golix, an unregulated platform that also trades other cryptocurrencies, has been in operation since 2014. Prices for Bitcoin are set by supply and demand, according to Taurai Chinyamakobvu, co-owner of the exchange. Sellers are paid in US dollars deposited electronically, which can only be converted at a steep discount on the black market.


link to this extract

We knew Julian Assange hated Clinton. We didn’t know he was secretly advising Trump • The Intercept

Robert Mackey:


[Barrett Brown, who went to prison for posting a link to a Wikileaks dump of Stratfor documents because those included credit card details] was particularly outraged by an Oct. 26, 2016 message, in which Assange had appealed to Trump Jr. to let WikiLeaks publish one or more of his father’s tax returns in order to make his group’s attacks on Hillary Clinton seem less biased. “If we publish them it will dramatically improve the perception of our impartiality,” the Assange-controlled @Wikileaks account suggested. “That means that the vast amount of stuff that we are publishing on Clinton will have much higher impact, because it won’t be perceived as coming from a ‘pro-Trump’ ‘pro-Russia’ source, which the Clinton campaign is constantly slandering us with.”

As Brown pointed out in another tweet, it was all-caps exasperating that Assange was in this case “complaining about ‘slander’ of being pro-Trump IN THE ACTUAL COURSE OF COLLABORATING WITH TRUMP.”

The journalist, an Intercept contributor, whose work had been championed by WikiLeaks, also shared a link to a Reddit AMA conducted two days after the election in which WikiLeaks staff, including Assange’s longtime collaborator Sarah Harrison, had denied point-blank that they had collaborated with the Trump campaign.

“The allegations that we have colluded with Trump, or any other candidate for that matter, or with Russia, are just groundless and false,” the staffers wrote then. “We were not publishing with a goal to get any specific candidate elected.”

It is not surprising that Brown felt personally betrayed by Assange, since, as he explained on Facebook Tuesday night, “I went to prison because of my support for WikiLeaks.” Specifically, Brown said, the charges against him were related to his role in “operations to identify and punish members of the government and members of private companies that had been exposed by Anonymous hackers of my acquaintance, via email hacks, as having conspired to go after Assange, to go after WikiLeaks.”

That sort of activism, dedicated to making public secret wrongdoing, Brown argued, is very different from “colluding with an authoritarian presidential campaign backed by actual Nazis while publicly denying it.”


link to this extract

Russian Twitter accounts promoted Brexit ahead of EU referendum: Times newspaper • Reuters


The Times cited research from an upcoming paper by data scientists at Swansea University and the University of California, Berkeley, which it said showed accounts based in Russia had tweeted about Brexit in the days leading up to the June 23 vote.

The Times said most of the tweets seen by the newspaper encouraged people to vote for Brexit, although a number advocated remaining in the EU. It quoted Tho Pham, one of the paper’s authors, as saying “the main conclusion is that bots were used on purpose and had influence”.

The research tracked 156,252 Russian accounts that mentioned #Brexit, including one, Svetal1972 which posted 92 tweets between June 20 and 24, including one calling for Britain to “make June the 23rd our Independence Day”.

It said many of the messages appear to have come from automated accounts known as bots or from cyborg accounts which are heavily automated but have some human involvement.

In the June 23, 2016 referendum, 17.4 million votes, or 51.9% of votes cast, backed leaving the EU while 16.1 million votes, or 48.1% of votes cast, backed staying.


Did it make all the difference? Unlikely. Did it make no difference? Also unlikely. So how much difference did it make?
link to this extract

The best laptop ever made •

Marco Arment:


Apple has made many great laptops, but the 15-inch Retina MacBook Pro (2012–2015) is the epitome of usefulness, elegance, practicality, and power for an overall package that still hasn’t been (and may never be) surpassed.

Introduced in 2012, less than a year after Steve Jobs died, I see it as the peak of Jobs’ vision for the Mac.

It was the debut of high-DPI Macs, starting down the long road (which we still haven’t finished) to an all-Retina lineup. And with all-SSD storage, quad-core i7 processors, and a healthy amount of RAM all standard, every configuration was fast, capable, and pleasant to use…

…I recently returned to the 2015 15-inch MacBook Pro after a year away.

Apple still sells this model, brand new, just limited to the integrated-only GPU option (which I prefer as a non-gamer for its battery, heat, and longevity advantages), but I got mine lightly used for over $1000 less.

I thought it would feel like a downgrade, or like going back in time. I feared that it would feel thick, heavy, and cumbersome. I expected it to just look impossibly old. It didn’t.

It feels as delightful as when I first got one in 2012. It’s fast, capable, and reliable. It gracefully does what I need it to do. It’s barely heavier or thicker, and I got to remove so many accessories from my travel bag that I think I’m actually coming out ahead.

It feels like a professional tool, made by people who love and need computers, at the top of their game. It’s designed for us, rather than asking us to adapt ourselves to it. It helps us perform our work, rather than adding to our workload.

This is the peak. This is the best laptop that has ever existed.


I’m typing this on a 2012 model. Recently got the battery replaced; Apple cleaned the whole thing. Like having a brand new machine.
link to this extract

‘Unsafe and just plain dirty’: women accuse Vice of ‘toxic’ sexual-harassment culture • Daily Beast

Brandy Zadrozny:


In the summer of 2015, Phoebe Barghouty was 23 years old and had a new master’s degree in journalism from Stanford—but little other experience—when she landed what most of her peers would consider a dream job: associate producer at Vice’s Los Angeles bureau.

Though her job hadn’t technically started yet, her boss, then-Editor in Chief Jason Mojica, invited her to join the team at the L.A. Press Club Awards. After accepting an award for public service in journalism, the team from Vice—including Mojica and Kaj Larsen, the bureau chief who had hired Barghouty—celebrated with drinks. By the end of the night, Barghouty says a very drunk Larsen had brought up sex (musing about his chances with a group of “black girls” at the bar), asked her for a ride home, then passed out in her car.

“I had not even started work and he was being so inappropriate,” she remembers.

Things just got weirder.

Barghouty says within her first few weeks on the job, Larsen was asking her to meet him at his home in Venice Beach. She thought it was strange, but he was her boss so she complied. As she waited outside his house, she texted a friend her location—“like how you tell a friend before a Tinder date in case you get murdered”—when a shirtless Larsen walked up and told her to come wait inside his bungalow while he took a shower.


Got weirder still. Or creepy. Also shows this stuff is not limited to particular industries.
link to this extract

Errata, corrigenda and ai no corrida: none notified

Start Up: the fabulist smearing Moore’s accusers, Google back on Firefox, FaceID faces its fears, and more

On Twitter, nonsense about vaccines was a harbinger of what happened in 2016. Photo by VCU CNS on Flickr. (It’s an HPV vaccination.)

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 9 links for you. Use them wisely. I’m @charlesarthur on Twitter. Observations and links welcome.

Troll smearing Roy Moore’s accuser stole dead SEAL’s identity • Daily Beast

Betsy Woodruff, Ben Collins, Spencer Ackerman and Joseph Cox:


Umpire43, also known as Doug Lewis or DJ Lewis, has repeatedly invented stories in the past—particularly about his own background. Lewis said he was a 22-year veteran of the Navy, a pollster at Ipsos/Reuters, an expert on rigging voting machines, a source who was feet away from Reince Preibus, a man who speaks six languages, a beleaguered soul who needed time off after 9/11 when he saw Muslims “dancing on rooftops,” the owner of a polling company who claimed Trump had a sustained lead in California, and an actual baseball umpire with 50 years experience. Oh, and he worked at the American consulate in Calgary, where he claimed to obtain proof of a forged birth certificate for Ted Cruz’s father.

The Daily Beast spoke with each of the institutions and companies at which he claimed to be affiliated or employed. None of Umpire43’s employment or service claims are true, these organizations said.

Umpire43’s now-infamous allegation that “A family friend in Alabama just told my wife that a WAPO reporter named Beth offered her 1000$ to accuse Roy Moore,” posted last week, was deleted with the rest of his Twitter account Tuesday morning.


Wow. This is a big takedown; that account was responsible for all sorts of lies – always pro-Trump, anti-Democrat. They were picked up by idiot conservative sites which haven’t heard of fact-checking.

That the account is gone is a hell of a thing, but it won’t stop the idiot sites, and you can guess it will pop up again soon.
link to this extract

Google pays to put search engine back on Firefox browser in US • Bloomberg

Mark Bergen:


In a blog post, Mozilla said Firefox’s default search engine will be Google in the US, Canada, Hong Kong and Taiwan. The agreement recalls a similar, older deal that was scuttled when Firefox and Google’s Chrome web browser became bitter rivals. Three years ago, Mozilla switched from Google to Yahoo! Inc. as the default Firefox search provider in the US after Yahoo agreed to pay more than $300m a year over five years — more than Google was willing to pay.

The new Firefox deal could boost Google’s already massive share of the web-search market. When people use Firefox, Google’s search box will be on the launch page, prompting users to type in valuable queries that Google can sell ads against. But the agreement also adds another payment that Alphabet Inc.’s Google must make to partners that send online traffic to its search engine, a worrisome cost for shareholders.

It’s unclear how much Google paid to reclaim this prized digital spot. A Google spokeswoman confirmed the deal but declined to comment further, and Mozilla didn’t disclose financial details.


Bet it’s less than Yahoo paid. That was a stunning overbid.
link to this extract

Loot boxes, never-ending games and always-paying players • Rolling Stone

Jared Newman:


Years before Bioware added loot boxes to Mass Effect 3, the model gained steam in emerging markets like China, where lower disposable income and rampant piracy of full-priced software caused the free-to-play model to flourish.

“[Players] weren’t able to buy top-end PCs, they weren’t able to afford all these $50 or $60 games, and that’s why internet cafes were huge back then, and still are, in terms of just being able to sign on and play an online game,” says Daniel Ahmad, an analyst with Niko Partners. “Piracy was also huge back in the day, and the way around that was just releasing games for free, and then putting in content that people want to buy.”

Loot boxes proved especially lucrative, emerging for instance in a massive multiplayer game called ZT Online. A 2007 story in China’s Southern Weekly describes how players could purchase treasure chests, which promised a chance to earn the game’s best gear without grinding. It also rewarded the players who purchased the most chests, setting up a system where some players flushed their money away.

“It was effectively free-to-play with gambling mechanics in there,” Ahmad says.

While free-to-play PC games flourished in Asia, similar concepts started to take root in western markets, fueled by the rise of gaming on social media sites like MySpace and Facebook, and eventually on smartphones. A VentureBeat story from early 2009 said that Zynga was making at least $50 million per year, mostly from sales of virtual goods.


Loot boxes are basically gambling; they should be banned for any children’s game for the under 18s. But they aren’t. (If you like games, by the way, Ahmad is a worthwhile follow on Twitter.)
link to this extract

You probably don’t need to worry about someone hacking your iPhone X’s Face ID with a mask • TechCrunch

Taylor Hatmaker:


Just a week after the device’s release, Vietnamese research team Bkav claims to have cracked Apple’s facial recognition system using a replica face mask that combines printed 2D images with three-dimensional features. The group has published a video demonstrating its proof of concept, but enough questions remain that no one really knows how legitimate this purported hack is.

As shown in the video below, Bkav claims to have pulled this off using a consumer-level 3D printer, a hand-sculpted nose, normal 2D printing and a custom skin surface designed to trick the system, all for a total cost of US$150.

For its part, in speaking with TechCrunch, Apple appears to be pretty skeptical of the purported hack. Bkav has yet to respond to our questions, including why, if its efforts are legitimate, the group has not shared its research with Apple (we’ll update this story if and when we hear back). There are at least a few ways the video could have been faked, the most obvious of which would be to just train Face ID on the mask itself before presenting it with the actual face likeness. And it’s not like Apple never considered that hackers might try this methodology.


So people who are famous enough to feature in Madame Tussauds should worry? This is pretty daft. 3D printers and hand-sculpted noses. This is Mission Impossible territory.
link to this extract

Datasette: instantly create and publish an API for your SQLite databases • Simon Willison


I just shipped the first public version of datasette, a new tool for creating and publishing JSON APIs for SQLite databases.

You can try out out right now at, where you can explore SQLite databases I built from Creative Commons licensed CSV files published by FiveThirtyEight. Or you can check out, derived from the database of world political parties which illustrates some advanced features such as SQLite views.


Read-only, so people can’t hack into your system. Willison is very smart – in 2009 he was in charge of the systems crowdsourcing MPs’ expenses at The Guardian, and made it happen in lightning time.
link to this extract

Liberal arts and tech • Tech.pinions

Bob O’Donnell (who is a “liberal arts” graduate):


while no two liberal arts programs are the same, the one consistent thread across them is that they teach people to think critically, ask these essential why questions, and work through the implications and longer-term impact of ideas and concepts, particularly as they relate to people. Applying these kinds of human-centric principles to tech could make a profoundly important impact.

Consider, for example, where social media has brought us as a society. From a scientific and programming perspective, it’s clearly impressive to be able to not only link billions of people around the world and let them communicate with one another, but to use advanced computer science to create algorithms that can continuously feed each one of us with the kind of information that specifically interests each one of us (in theory, at least).

However, a liberal arts major familiar with works like Alexis de Tocqueville’s “Democracy in America,” John Mill’s “On Liberty” essay, or even the work of ancient Greek historians, might have been able to recognize much sooner the potential for the “tyranny of the majority” or other disconcerting sociological phenomena that are embedded into the very nature of today’s social media platforms. While seemingly democratic at a superficial level, a system in which the lack of structure means that all voices carry equal weight, and yet popularity, not experience or intelligence, actually drives influence, is clearly in need of more refinement and thought than it was first given.

Beyond these more philosophical debates, there are an increasing number of very practical concerns around the ethical application of technology in fields ranging from medicine to transportation to basic data analysis. Toss in the mind-numbing array of questions that arise from technologies like artificial intelligence (AI) and machine learning, and it’s clear that there’s a lot more discussion that needs to happen around how technologies get applied, rather than just how to build them.


link to this extract

June 2015: anti-vaxxers are using Twitter to manipulate a vaccine bill • WIRED

Renee DiResta and Gilad Lotan, writing in June 2015:


As early as Tuesday, the California State Assembly will vote on SB-277, a law that would ban the so-called personal belief exemption. School boards, medical associations, and community leaders support the law.

But a small group of vocal anti-vaxxers is fighting hard to keep it from passing. This group, which leverages the power of social media, has launched a full-scale attack on the bill as it travels through the legislature. Each day, leaders craft tweets and instruct followers to disseminate them. Several senators who voted in favor of the California legislation have found themselves receiving extensive attention from the group—one, Senator Hannah Beth Jackson, has been @-mentioned (often unfavorably) in a particular Twitter hashtag more than 2,000 times since casting her vote in favor of the legislation.

This anti-vax activity might seem like low-stakes, juvenile propaganda. But social networking has the potential to significantly impact public perception of events—and the power to influence opinions increasingly lies with those who can most widely and effectively disseminate a message. One small, vocal group can have a disproportionate impact on public sentiment and legislation. Welcome to “Anti-Vax Twitter.”


A harbinger, of course, for what would come in the future. One wonders now whether this was all ways to test how propaganda efforts might work on social media, or whether it was just spontaneous, led by idiots.
link to this extract

How the iPhone earned its security record • FT

Tim Bradshaw:


Apple’s security team, led by Ivan Krstić, has won increasing respect from researchers in the field over the past few years. Typically, as the volume and variety of a company’s devices on the market increases, the security can often deteriorate. With Apple, even after more than 1.2bn iPhones have been sold over 10 years, its security has been improving. 

iPhones and iPads “are legitimately the most secure phones and tablets out there”, says Rich Mogull, chief executive of Securosis, an independent security research and advisory firm. “I don’t know if I can put a timeline on when Apple’s culture changed, but it did,” he says. “They take security and privacy very seriously now and they are getting a little better with every release of hardware and software.”

One key ingredient is the Secure Enclave, an encrypted “coprocessor” in the iPhone’s A-series chips that was first introduced with the iPhone 5s in 2013. 

This was the “underpinning for a significant step forward in their security model”, says Pepijn Bruienne, research and development engineer at Duo Security. “They can embed the security architecture in at the silicon layer.”

…As well as requiring every new app submitted to the store to be reviewed by Apple’s staff before consumers are allowed to download it, the iOS operating system is much more restrictive than Google’s rival, Android, in what apps are able to do. 

“The app can’t just go on your phone and start requesting access to your location or contacts” without the user granting their permission, says Andrew Blaich, a researcher at mobile security specialist Lookout. There are also restrictions on reading text messages, overlaying ads and running in the background. “Apple have insulated themselves from a lot of the common attacks that we see on the Android platform day to day,” he says. 

As a result, in the fourth quarter of 2016 and first quarter of 2017, 47 in 1,000 of Android enterprise devices protected by Lookout encountered app-based threats, compared with only 1 in 1,000 iOS devices. 


link to this extract

Apple supplier eyes smart speakers with facial recognition • Nikkei Asian Review

Cheng Ting-Fang:


Apple HomePod maker Inventec Appliances said on Friday that it expects future voice assistant products to offer 3-D sensing features, including facial and image recognition.

“We see trends that engineers are designing smart speakers that will not only come with voice recognition but also incorporate features such as facial and image recognition,” President David Ho told reporters after the company’s earnings conference. “Such AI-related features are set to make people’s lives more convenient and to make the product easier to use.” He added, however, that he was unsure at the moment whether smart speakers with more AI features in the future would become a hit in the market.

Ho did not specify which product he was talking about, but analysts said he is likely referring to the next generation of Apple’s HomePod, the $349 voice-activated speaker that will compete with Amazon Echo and Google Home.

Inventec Appliances, a subsidiary of Taiwanese electronic contract manufacturer Inventec, currently monopolizes orders for the HomePod as well as AirPods, Apple’s wireless earbuds, according to analysts. It also makes smartphones for China-based Xiaomi, wearable products for America’s Fitbit and smart speakers for US-based Sonos and others.

Jeff Pu, an analyst at Yuanta Investment Consulting, said Apple could roll out HomePods with 3D-sensing cameras in 2019.


It’s nice to see the conception of bullshit rumours set out in so few paragraphs. Man speculates about the possibility of face and image recognition in “smart home assistants”. (Amazon already does this, by the way, in the Amazon Show.) Analyst suggests Apple could do it. Rumours take wing that next HomePod will do FaceID.

Nonsense on stilts which never asks quite how FaceID on a HomePod would (a) work (b) be useful.
link to this extract

Errata, corrigenda and ai no corrida: none notified

Start Up: the Android view of iPhone X (and bigger?), Facebook’s fake profile business, is Huawei stiffing Qualcomm?, and more

Let the machines do the translation. We might benefit. Or not. Photo by earlynovelsdatabase on Flickr.

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 12 links for you. Use them wisely. I’m @charlesarthur on Twitter. Observations and links welcome.

iPhone X: the Android Central review • Android Central

Daniel Bader really likes FaceID (Samsung’s versions don’t come close, he shows), but finds much of the rest in the hardware side a wash. But then you come to responsiveness – iOS is way ahead, he says – and apps:


I want to believe, now that we’re in 2017 and not 2012, that developers care as deeply about feature parity on Android, but they don’t: the best indie apps still don’t come to Android (although one can argue, and I’d agree in some cases, that the indie app scene is extremely vibrant on Android — just in a way that doesn’t make them much money); games arrive months late, if at all; and beloved products, especially camera-based networks like Instagram and Snapchat, lack specific features or optimizations that drive me crazy.

My banking app, for instance, brought Touch ID (and, thanks to transferrable APIs, Face ID) support to its iOS app two years ago; the Android version forces me to enter my password like a chump every time. My favorite writing app, Bear, has no intention of building an Android version, and my formerly favorite meal-planning app, Grocery King, hasn’t updated its Android app in over two years.

Of course, given that I spent the vast majority of my year with Android, I have come up with viable cross-platform alternatives — Google Docs is pretty good, and Mealime is great, too — but it still feels like Android apps play second fiddle to their iOS counterparts…


But he finds things he still prefers on Android:


…After spending any length of time with iOS, a few things really stand out to me: notifications are still much better on Android; the typing experience is more enjoyable on Android; using Android is much more flexible; and the variety of Android hardware is breathtaking.

Notifications are among the most critical details in any operating system today, and Android nailed it years ago and only continues to get better with every iteration. Google’s lead in this regard is so absolute it might as well as insurmountable. In contrast, I loathe dealing with notifications on the iPhone.


(Thanks Papanic for the pointer.)
link to this extract

KGI: Apple to launch 6.5in ‘iPhone X Plus’ and lower-priced 6.1in full-screen LCD model next year • Mac Rumors

Joe Rossignol:


Apple will launch a trio of new iPhone models in 2018, including 5.8-inch and 6.5-inch models with OLED displays and a 6.1-inch model with an LCD display, according to respected KGI Securities analyst Ming-Chi Kuo.

Kuo believes all three models will be equipped with a full-screen notched design and TrueDepth camera system like the iPhone X.

Kuo expects the 5.8-inch model to have 458 pixels per inch, suggesting the second-generation iPhone X’s display will likely continue to have a resolution of 1,125×2,436. He said the 6.5-inch model will have roughly 480 to 500 PPI, while the 6.1-inch model is estimated to have between 320 and 330 PPI.

In his latest research note, obtained by MacRumors, Kuo said the 6.1-inch model will have a lower-resolution LCD display and target the low-end and mid-range markets with an estimated $649 to $749 starting price in the United States.


So the tweener model is the LCD? One assumes the smaller SE will live on. Apple’s in no hurry to phase out LCD: the screens are way cheaper, with a better profit margin for it.

(Also, one notes that the rumour mill is well underway when most people haven’t decided about buying this year’s model. This is the point in the year when Apple is locking down designs for next year, so Kuo may be optimistic here.)
link to this extract

Inside the fake Facebook profile industry •

Jeff Yates:


There’s more than one way to lie in order to achieve success on social media. I’ve been covering fake news and online disinformation for three years now, and I thought I’d seen it all. But a Facebook profile by the name of Béatrice Boistard really gave me pause.

She was gorgeous, she was mysterious and she had created an online audience of several hundred thousand followers. The thing is, she also stole the identities of handicapped or sick people to get it.

So began the strangest investigation of my career, which catapulted me into the kaleidoscopic world of fake Facebook profiles, where nothing is real.

I first got acquainted with Béatrice’s profile in February 2016. She would regularly share pictures of amputated or bald people, asking her followers to write “amen” in the comments section. Why? Because “no one loves me since I got cancer”, or “my husband left me because I lost my legs.”

These posts invariably got thousands of likes, comments and shares on Facebook. It’s no surprise, then, that Béatrice’s page has over 671,000 followers. That’s significantly more than major Canadian news outlets such as the National Post and the Toronto Star and almost as many as the Globe and Mail…

…Surely, this must be about money, I thought. In fact, certain posts within the network send links to fraudulent websites, where you are asked to enter your credit card information. But I also believed that these fake accounts were used in sextortion schemes. The report by Corde sensible confirmed this.

This all suggests that this massive network is used primarily to attract men online and send them careening towards different fraudulent schemes. But by analyzing the network, I was able to determine that different accounts play different roles. All of them co-operate to create a huge trap that filters potential victims in order to find the most vulnerable targets.


This gets very weird. But it’s a rewarding read.
link to this extract

She warned of ‘peer-to-peer misinformation’; Congress listened • The New York Times

Sheera Frankel:


“We were monitoring closely to see when the [social media] companies gave misleading or partial answers [to the US Congress] so that we could follow up,” said Renee DiResta, 36, who became immersed in disinformation campaigns in her spare time outside of her job as a founder and head of marketing at Haven, a shipping technology company.

How a small group of self-made experts came to advise Congress on disinformation campaigns is a testament to just how long tech companies have failed to find a solution to the problem. For years, the informal group — about a dozen or so people — have meticulously logged data and published reports on how easy it was to manipulate social media platforms.

In 2016, they monitored thousands of Twitter accounts that suddenly started using bots, or automated accounts, to spread salacious stories about the Clinton family. They watched as multiple Facebook pages, appearing out of nowhere, organized to simultaneously create anti-immigrant events. Nearly all of those watching were hobbyists, logging countless hours outside their day jobs.

“When I put it all together and started mapping it out, I saw how big the scale of it was,” said Jonathan Albright, who met Ms. DiResta through Twitter. Mr. Albright published a widely read report that mapped, for the first time, connections between conservative sites putting out fake news. He did the research as a “second job” outside his position as research director at the Tow Center for Digital Journalism at Columbia University.

Senate and House staff members, who knew of Ms. DiResta’s expertise through her public reports and her previous work advising the Obama administration on disinformation campaigns, had contacted her and others to help them prepare for the hearings.


Diresta and Albright are two people to really watch: their outputs on this topic are always useful.
link to this extract

Missouri Attorney-general Josh Hawley launches investigation of Google • The Kansas City Star


Missouri Attorney General Josh Hawley’s office on Monday issued a subpoena to Google as part of an investigation into whether the tech giant is violating Missouri’s consumer protection and anti-trust laws.

The investigation delves into Google’s collection of data on users and whether Google, the world’s most popular search engine, has manipulated search results at the expense of competitors, according to a release from Hawley’s office.

“When a company has access to as much consumer information as Google does, it’s my duty to ensure they are using it appropriately,” said Hawley, a Republican who is mounting a campaign for U.S. Senate. “I will not let Missouri consumers and businesses be exploited by industry giants.”

Patrick Lenihan, Google’s spokesman, said in an email that the company has not yet received the subpoena. Lenihan said Google has “strong privacy protections in place for our users and continue to operate in a highly competitive and dynamic environment.”


He’s also looking at whether Google “misappopriated content from competitors”. The FTC looked into all this back in 2010 and dropped the case. I wonder how much Hawley fits into the Bannonite wing of the party, given that the latter wants Google declared a utility. But quite why he thinks it would be an electoral asset to run a campaign against a search engine that pretty much everyone uses is hard to know.
link to this extract

The offline solution to online hate • Demos Quarterly

Jamie Bartlett:


Social media platforms are right to get rid of content that’s illegal when they are informed about it – and perhaps they could streamline it a little more. But they cannot reasonably be expected to find every case of hate crime, nor proactively seek it all out. And there are some sorts of hateful content that do deserve the full weight of the law. But the truth is neither law nor tech will rid us of the problem.

Both of these approaches are in some sense lazy: a superficial effort to deal with a human problem through technocratic means. The only answer is a long-term, hard slog: the task of teaching society to be decent. The task of educating young people about the responsibilities of life online and what it’s like to get bullied. The task of parents raising their children to understand the value of civility – or in some cases, children teaching their parents.

After all, if offline and online crime is the same, perhaps the root causes are similar too: decades of research has found that anti-social behaviour offline is driven by complicated and overlapping causes, including poor parental supervision, low school achievement, anti-social parents, low family income, antisocial peers. In other words, deep rooted social problems that are far more complex than even the most sophisticated algorithm or CPS guidelines could ever fix. Until we deal with any of these issues, online bullying, hate and cruelty will continue to exist, and even the best intentioned law or tech-led approach will not solve it.


This feels a bit like “solve society, because social networks are fine”. That’s not exactly what he’s saying, but it does matter that social networks as presently set up try to generate feedback loops in behaviour; they don’t care what behaviour as long as there’s more of it, on their platform, where they can show ads in the meantime. (I know Jamie a little, from having shared a platform; he’s very knowledgeable and experienced on deep dark web stuff and what people get up to online.)
link to this extract

Android camp reportedly seeking to renegotiate royalties with Qualcomm • Digitimes

Cage Chao and Steve Shen:


The Android camp could follow the steps of Apple to temporarily suspend royalty payments to Qualcomm aiming to force the US-based chip vendors back to the negotiating table to work out “fairer” royalty schemes, according to industry sources.

In fact, Qualcomm revealed at its latest investors conference that a China-based smartphone brand has already discontinued royalty payments for the use of Qualcomm’s patented technologies. Qualcommm did not identify the China client, but it is believed it is Huawei as the vendor, according to the sources.

With Huawei’s smartphone shipments reaching about 150m units a year and with an ASP of US$300, Hauwei’s royalty payments could account for 5-10% of Qualcomm’s annual royalty income, the source noted.

Leveraging on its own base station technology and many related patents, Huawei has the bargianing chip to suspend payments to Qualcomm, the sources added.

Meanwhile, Samsung, possessing a wide range of mobile technologies and patents of its own, could also suspend its payments to Qualcomm, the sources said.


If Huawei and Samsung suspended payments, Qualcomm would be in big trouble. That would be the three largest smartphone makers, all in dispute with it.

Also: quite a burial of the lede, as they say. Huawei has stopped paying Qualcomm?! Huge if true.
link to this extract

Qualcomm rejects Broadcom’s takeover bid • WSJ

Ted Greenwald:


Qualcomm rejected Broadcom Ltd.’s unsolicited $105 billion offer, setting up a potentially hostile showdown between two giants of the chip industry over what would be the biggest technology takeover ever.

A combination of the two would create a huge company whose chips manage communications for consumer devices and appliances, phone-service providers and data centers.

In a statement Monday, Qualcomm’s board said the offer, which Broadcom submitted last week, dramatically undervalues the company and comes with significant regulatory uncertainty.

Broadcom said Monday it remains committed to the deal. “We continue to believe our proposal represents the most attractive, value-enhancing alternative available to Qualcomm stockholders,” Broadcom CEO Hock Tan said in a prepared statement.


Wonder how the Huawei/Apple/Samsung stuff feeds into this. Though generally, the feeling in the industry is that this would be a bad transaction. Qualcomm’s better off on its own.
link to this extract

Finding the right color palettes for data visualizations • Graphiq

Samantha Zhang:


Rather than diving in head first and creating our own color palette, we started by conducting some research on existing color palettes around the web. Surprisingly, we found that few are actually designed for complex charts and data visualizations. We identified several reasons as to why we couldn’t use existing color palettes:

Problem 1: Low Accessibility

Many of the color palettes we looked at were not designed for visualizations. Not only do they not vary enough in brightness, but they were often not created with accessibility in mind. Flat UI Colors is one of the most widely used color palettes out there, and it’s easy to see why: it looks great. But, as its name indicates, it’s designed for user interfaces. Those who are color blind may find it difficult to interpret a data visualization that uses the Flat UI palette:


Useful. I struggle to find good palettes for graph colours with the knowledge that colour-blind people will also want to look at them. (Via Sophie Warnes’s Fair Warning newsletter.)
link to this extract

Google will remove Play Store apps that use Accessibility Services for anything except helping disabled users • Android Police

Corbin Davenport:


For years, Android has allowed apps to modify the behavior of other applications, using Accessibility Services. While the intended purpose is for developers to create apps for users with disabilities, the API is often used for other functionality (to overlay content, fill in text fields, etc.). LastPass, Universal Copy, Clipboard Actions, Cerberus, Tasker, and Network Monitor Mini are just a few examples of applications heavily using this API.

While Accessibility Services can greatly extend the functionality of applications, they can potentially create a security risk. Once granted the right permissions, the API can be used to read data from other apps. Likely for this reason, Google has sent emails to app developers regarding the usage of Accessibility Services. The developer of BatterySaver received this message [telling him to explain to users how it was using Accessibility Services to help users with disabilities, on pain of removal from the Play Store]…

…Several other developers have told us they received this email, and there is a Reddit thread full of additional reports. This means many apps will have to severely degrade their functionality if they wish to remain on the Play Store, unless they can convince Google that users with disabilities benefit from them. Some applications, like LastPass, entirely rely on this API and can’t function without it.


Closing a security loophole, but could be a problem for apps like LastPass.
link to this extract

Software 2.0 • Medium

Andrej Karpathy on how data-taught self-coding networks will write software of the future:


Software 2.0 is not going to replace 1.0 (indeed, a large amount of 1.0 infrastructure is needed for training and inference to “compile” 2.0 code), but it is going to take over increasingly large portions of what Software 1.0 is responsible for today. Let’s examine some examples of the ongoing transition to make this more concrete:

Visual Recognition used to consist of engineered features with a bit of machine learning sprinkled on top at the end (e.g., SVM). Since then, we developed the machinery to discover much more powerful image analysis programs (in the family of ConvNet architectures), and more recently we’ve begun searching over architectures.

Speech recognition used to involve a lot of preprocessing, gaussian mixture models and hidden markov models, but today consist almost entirely of neural net stuff.

Speech synthesis has historically been approached with various stitching mechanisms, but today the state of the art models are large convnets (e.g. WaveNet) that produce raw audio signal outputs.

Machine Translation has usually been approaches with phrase-based statistical techniques, but neural networks are quickly becoming dominant. My favorite architectures are trained in the multilingual setting, where a single model translates from any source language to any target language, and in weakly supervised (or entirely unsupervised) settings.

Robotics has a long tradition of breaking down the problem into blocks of sensing, pose estimation, planning, control, uncertainty modeling etc., using explicit representations and algorithms over intermediate representations. We’re not quite there yet, but research at UC Berkeley and Google hint at the fact that Software 2.0 may be able to do a much better job of representing all of this code.

Games: Go playing programs have existed for a long while, but AlphaGo Zero (a ConvNet that looks at the raw state of the board and plays a move) has now become by far the strongest player of the game. I expect we’re going to see very similar results in other areas, e.g. DOTA 2, or StarCraft.


link to this extract

Fujitsu offloads PC division in joint venture with Lenovo • Canalys

Fujitsu is selling a 51% share in its PC division to Lenovo for US$224m, and another 5% to Development Bank of Japan for $22m:


Fujitsu insists it is not quitting the PC market, but instead sees this partnership as a way to strengthen its competitive position against larger rivals. With a focus on enterprise PCs only (apart from in Japan), Fujitsu’s global PC market share, excluding tablets, has fallen from around 1.9% in 2013 to 1.3% in 2016. The primary benefit for Fujitsu is the combined purchasing power that Lenovo brings with Intel, Microsoft and other component vendors, which will bring substantial pricing advantages.

For Lenovo, the attraction is different. Fujitsu brings some size benefits – combining Q2 shipment volumes would have given Lenovo a 21% market share (excluding tablets). But the real appeal lies in Fujitsu’s sizeable consumer business in Japan. Lenovo is already Japan’s PC leader through its joint venture with NEC, so adding Fujitsu significantly extends its leadership (assuming no objections from competition authorities) as Japan’s PC market struggles with growth. Fujitsu’s notebook manufacturing and R+D operations in Japan will move into the joint venture, improving efficiencies for Lenovo, which the company hopes will boost profitability. Unlike in the rest of the world, FCCL will own the sales and support organization and go-to-market operation in Japan. This means it will be responsible for large retail relationships and direct sales in Japan.


Fujitsu will still make servers and storage. But what a capitulation. It values the whole PC business at $439m; given that it’s probably loss-making, that will be “going concern” pricing.

link to this extract

Errata, corrigenda and ai no corrida: none notified

Start Up: Russia’s twitter troll, harassment figured, lives in virtual reality, Logitech relents, and more

How to get rich: go bankrupt in bitcoin. Photo by Francis Storr on Flickr.

You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 13 links for you. The week beckons. I’m @charlesarthur on Twitter. Observations and links welcome.

Russia Twitter trolls deflected Trump bad news • Associated Press

Ryan Nakashima and Barbara Ortutay:


Disguised Russian agents on Twitter rushed to deflect scandalous news about Donald Trump just before last year’s presidential election while straining to refocus criticism on the mainstream media and Hillary Clinton’s campaign, according to an Associated Press analysis of since-deleted accounts.

Tweets by Russia-backed accounts such as “America_1st_” and “BatonRougeVoice” on Oct. 7, 2016, actively pivoted away from news of an audio recording in which Trump made crude comments about groping women, and instead touted damaging emails hacked from Clinton’s campaign chairman John Podesta.

Since early this year, the extent of Russian intrusion to help Trump and hurt Clinton in the election has been the subject of both congressional scrutiny and a criminal investigation by special counsel Robert Mueller. In particular, those investigations are looking into the possibility of collusion between the Trump campaign and the Russians.

AP’s analysis illuminates the obvious strategy behind the Russian cyber meddling: swiftly react, distort and distract attention from any negative Trump news.

An exclusive AP analysis found that disguised Russian agents on Twitter rushed to deflect scandalous news about Donald Trump just before last year’s election while refocusing criticism on the mainstream media and Hillary Clinton’s campaign. (Nov. 9)

The AP examined 36,210 tweets from Aug. 31, 2015, to Nov. 10, 2016, posted by 382 of the Russian accounts that Twitter shared with congressional investigators last week. Twitter deactivated the accounts, deleting the tweets and making them inaccessible on the internet. But a limited selection of the accounts’ Twitter activity was retrieved by matching account handles against an archive obtained by AP.


This would be the election that Trump has been assured by Putin there was no meddling in?
link to this extract

Six in ten women say they’ve been sexually harassed by a man • YouGov


The latest outpouring of news stories about sexual harassment in the workplace reflects a common experience for many American women. In the latest Economist/YouGov Poll, 60% of women report having been sexually harassed by a man. A third of those women report the harassment occurred in the last five years. For most women, sexual harassment is serious and not unusual – and the problem isn’t getting better.

The recognition that harassment occurs is widespread. When women are asked what percentage of women they believe would say they have been harassed, the average response is 70%. Men are less likely to see the problem as that widespread: their average response is about 50%.

Still, most people, male and female, say sexual harassment in the workplace is a serious problem. But for women it is a greater concern: 78% of women say sexual harassment in the workplace is a serious problem today, and 33% of women say it is a very serious problem). 60% of men agree it is a serious issue, with 21% calling it very serious. 


link to this extract

What reality TV teaches us about Russia’s influence campaign • The New York Times

Farhad Manjoo:


We may never know for certain if Russia’s campaign to influence American society through social networks changed the course of history in big ways — if it altered the election results, say. But it is already clear that Russia’s efforts did change the world in countless small ways. A few dozen real Americans did protest that Saturday in Houston [having been prompted by fake Facebook pages aiming to foment conflict]. Videos of the protest show real emotion — people on opposite sides of the street screaming, swearing and truly angry to have to share the country with the bozos on the other side.

As I watched these videos recently, I had an epiphany about the Russia influence campaign. The Houston protest videos depicted a bunch of Americans duped into fighting each other in public, all at the whim of an unseen force that, through expert and surreptitious cajoling, had gotten them to lose control of themselves on camera. I’d seen this show many times before, and you probably have, too. It’s called “The Bachelor.”

And not just the “The Bachelor,” but every show like it. The Russians are running a reality show through Facebook and Twitter, and their contestants are all of us.

Over the past few days, I reached out to several reality show producers, asking them to compare the Russian digital influence campaign and the world of unscripted TV. The more they told me about reality shows, the more the metaphor seemed to explain Russia’s trolling campaign — how it worked, what it aimed to do and why campaigns like it will be so difficult to fight…

…Sarah Gertrude Shapiro, a former producer on “The Bachelor” who later created “Unreal,” a scripted show about the reality industry, said the key to manipulating contestants into acting a certain way was to “tap into their fears, passions and ego.”

On reality TV, producers can do that because they keep detailed dossiers on everyone on set. But guess what? Russian trolls had detailed dossiers, too — and they could consult them at scale. Using Facebook’s exquisitely detailed ad-targeting and viral propagation systems, trolls could create content that perfectly matched your fears, passions and ego.


(Thanks JC for the link.)

link to this extract

Third quarter sees closest ever gap between the top three smartphone brands in China • Counterpoint Research


Chinese market shipments slowed in Q3 2017 compared to a strong quarter last year. Q3 2016 was marked by robust growth by a number of brands including OPPO and Vivo. This year shipment growth was strong in the second quarter, resulting in many brands ending Q2 2017 with high inventory levels and leading to softer shipments in the third quarter. However efforts to correct the excess inventory meant that sales to consumers were strong in Q3 2017, growing by almost 12% year/year.

These factors resulted in a close competitive dynamic between the top three brands. Commenting on the analysis, James Yan, Research Director at Counterpoint Research said, “Compared to last year, where OPPO and Vivo were the fastest growing brands in China with volumes up 109% YoY and 78% YoY respectively, growth for these two brands has slowed down this quarter. Nonetheless, the two brands are still growing at a healthy rate and have closed on the previous quarter’s leader, Huawei. OPPO’s R11 was the bestselling model overall, which helped OPPO edge just ahead of Huawei during the quarter.”


Worth just noting what happens in the world’s single largest smartphone market. Oppo and vivo are owned by the same company; they took 37.5% of the market, and Huawei 18.6%. That’s over 56% going to two companies; Xiaomi and Apple accounted for 10.3% and 8.5% respectively. Everyone else – and that’s a lot of companies – were just over a quarter of the 100m market. One always expects the squeeze to come, but it always seems to be a few quarters away.
link to this extract

Mark Karpeles will end up taking $859m from Mt. Gox bankruptcy • Cryptocoins News

Joseph Young:


In 2014, Mt. Gox, once the largest bitcoin exchange in the world, filed for bankruptcy. At the time, Japanese creditors requested Mt. Gox to return the equivalent amount of their funds stored in bitcoin in Japanese yen. Since then, the price of bitcoin has risen 70-fold, and former Mt. Gox CEO Mark Karpeles is expected to take the majority of the profit from the bankruptcy proceedings.

In July, Karpeles attended a court hearing and pleaded not guilty to charges on money laundering and embezzlement, for his involvement in the loss of approximately one million bitcoins, which were worth $400 million during the period wherein the Mt. Gox bankruptcy was filed.

Over the past three years, from 2014 to 2017, the price of bitcoin has increased exponentially, from around $400 to $7,000, by 17-fold. During the investigation into the Mt. Gox bankruptcy from 2014 to 2016, 200,000 bitcoins were recovered and with that, Karpeles was requested to proceed accrediting creditors of Mt. Gox with the recovered bitcoins.

However, after a strange turn of events, it turns out that creditors of Mt. Gox, or former bitcoin traders on the Mt. Gox exchange, are set to be credited with Japanese yen equivalent to the value of bitcoin in 2014. 


This must be the strangest rags-to-riches-to-rags-to-riches stories ever.
link to this extract

All mixed up: I spent eight weird hours wearing Microsoft’s new headset • Tom’s Guide

Andrew Freedman did an unwise thing so you don’t have to:


When you boot up a Windows 10 Mixed Reality headset, you land in the Cliff House, a serene ranch flanked by a lake on one side and a mountain on the other. Birds chirp. You could stay awhile.

I thought I could work in the Cliff House. But I was wrong: It was 8 hours in hell.
Before I go any further, I should tell you not to try this at your workplace. Keeping yourself in virtual reality for that long at a time, especially without frequent breaks, can be taxing on your eyes and possibly even your mental health.

I started my day with our usual editorial meeting and then plugged in an Acer Mixed Reality Headset to my office-issued Dell XPS 15. Setup was a breeze, but then I had a decision to make. I needed to keep my mouse and keyboard if I wanted to maintain my normal workflow. We had the motion controllers in the lab, but I might hit my colleagues next to me and behind me. Since I like my colleagues, I opted for an Xbox One controller to move around the Cliff House. After all, I had to stay at my desk, facing forward.

I picked a spot in the virtual house with two walls (one in front of me, one behind me) and a beautiful lake view with a tree and some floating islands. It’s the closest I’ve ever come to a corner office. On both walls,

I started to open a few windows, and there was my first hiccup: You can use only a few apps from the Microsoft Store. To get around not having HipChat, I logged in through the Edge browser. But my bookmarks and tools are in Chrome, so I had to open up a virtual version of my desktop and place it on a wall. I also use TweetDeck, Firefox, Sublime Text, Photoshop and a number of other apps that don’t work natively in Windows Mixed Reality. Since I use three monitors at once, I placed them on both walls and turned as necessary to see them.

If this sounds tiring, that’s because it is.


It was terrible, overall.
link to this extract

The Pixel 2 XL has another screen issue: unresponsive edges • Engadget

Mariella Moon:


It looks like Google still isn’t done fielding complaints about the Pixel 2 XL’s display. While some users are experiencing premature screen burn-in and seeing a bluish tint, others are apparently having trouble with its responsiveness. Comments posted on the Pixel 2 community website have revealed that some units are having issues getting their phones to register touches near the edges of the screen. One poster even conducted a test and found that while the edges on his display can recognize swipes just fine, they can’t always recognize taps.


This device has gone from hero to zero in the matter of a month or so. People just can’t stop finding problems with it.
link to this extract

Second Life: the digital ruins of a forgotten future • The Atlantic

Leslie Jamison:


Gidge Uriza lives in an elegant wooden house with large glass windows overlooking a glittering creek, fringed by weeping willows and meadows twinkling with fireflies. She keeps buying new swimming pools because she keeps falling in love with different ones. The current specimen is a teal lozenge with a waterfall cascading from its archway of stones. Gidge spends her days lounging in a swimsuit on her poolside patio, or else tucked under a lacy comforter, wearing nothing but a bra and bathrobe, with a chocolate-glazed donut perched on the pile of books beside her. “Good morning girls,” she writes on her blog one day. “I’m slow moving, trying to get out of bed this morning, but when I’m surrounded by my pretty pink bed it’s difficult to get out and away like I should.”

In another life, the one most people would call “real,” Gidge Uriza is Bridgette McNeal, an Atlanta mother who works eight-hour days at a call center and is raising a 14-year-old son, a 7-year-old daughter, and severely autistic twins, now 13. Her days are full of the selflessness and endless mundanity of raising children with special needs: giving her twins baths after they have soiled themselves (they still wear diapers, and most likely always will), baking applesauce bread with one to calm him down after a tantrum, asking the other to stop playing “the Barney theme song slowed down to sound like some demonic dirge.” One day, she takes all four kids to a nature center for an idyllic afternoon that gets interrupted by the reality of changing an adolescent’s diaper in a musty bathroom.

But each morning, before all that—before getting the kids ready for school and putting in eight hours at the call center, before getting dinner on the table or keeping peace during the meal, before giving baths and collapsing into bed—Bridgette spends an hour and a half on the online platform Second Life, where she lives in a sleek paradise of her own devising.


Second Life still has 600,000 regular users.
link to this extract

Update: we will replace your Logitech Harmony Links • Logitech blog


We heard you and we want to make it right.

If you are a Harmony Link user, we will reach out to you between now and March 2018 to make arrangements to replace your Link with a free Harmony Hub, a product with similar app-based remote control features to Link, with the added benefit of controlling many popular connected home devices plus, it works with popular voice assistants. You can also contact us at to make arrangements for your replacement.

We understand that services are important to you. Because the certificate that’s expiring relates to security, we would be acting irresponsibly by continuing the service knowing its potential/future vulnerability.

Additionally, Harmony Link customers that have already redeemed their 35% discount on a Harmony Hub will also be refunded the full amount they paid for the replacement. Again, we will be in touch with you regarding these updates between now and March 2018, before your Harmony Link will no longer function.


This is good. But as so often, one is left asking “how did you think it was going to be ok just to abandon people?” (The detective work on why the Link is going to die is that it has a soon-to-expire https certificate bought from Equifax which can’t be replaced on a like-for-like basis – that is, a software update wouldn’t continue its functionality. Teach them to buy certs from third parties.)
link to this extract

It’s time to stop trusting Google search already • The Verge

Adi Robertson:


Even if [Google] search is overwhelmingly accurate, highlighting just a few bad results around topics like mass shootings is a major problem — especially if people are primed to believe that anything Google says is true. And for every advance Google makes to improve its results, there’s a host of people waiting to game the new system, forcing it to adapt again.

Simply shaming Google over bad search results might actually play into its mythos [of infallibility, which it plays to through voice assistants], even if the goal is to hold the company accountable. It reinforces a framing where Google search’s ideal final state is a godlike, omniscient benefactor, not just a well-designed product. Yes, Google search should get better at avoiding obvious fakery, or creating a faux-neutral system that presents conspiracy theories next to hard reporting. But we should be wary of overemphasizing its ability, or that of any other technological system, to act as an arbiter of what’s real.

Alongside pushing Google to stop “fake news,” we should be looking for ways to limit trust in, and reliance on, search algorithms themselves. That might mean seeking handpicked video playlists instead of searching YouTube Kids, which recently drew criticism for surfacing inappropriate videos. It could mean focusing on reestablishing trust in human-led news curation, which has produced its own share of dangerous misinformation. It could mean pushing Google to kill, not improve, features that fail in predictable and damaging ways. At the very least, I’ve proposed that Google rename or abolish the Top Stories carousel, which offers legitimacy to certain pages without vetting their accuracy. Reducing the prominence of “Popular on Twitter” might make sense, too, unless Google clearly commits to strong human-led quality control.


Google’s basic model comes straight from scientific papers’ impact measurement: the more papers quote a previous one, the more “impact” the paper has, and so the more important it is in the canon of science.

This was fine while search largely consisted of trying to find the authoritative White House site. But search has shifted, and Robertson makes excellent points: when everyone’s essentially falsifying their papers, what does impact mean, and should you still use it?
link to this extract

Let’s all take a breath: Trump is a joke, but Bush was worse • Vanity Fair

T.A.Frank (and the “Bush” is George W, not his father George HW):


Trump has proved to be generally dreadful. He’s entirely unsuited to his post. Still, as people look back in grief, you would think that we’d seen the realization of Matthew Yglesias’s prediction that “angry mobs will beat and murder Jews and people of color with impunity.” At the very least, we might be farther along down the road mapped out in the New York Review of Books by author Masha Gessen, who warned that Trump might appoint a crony like Rudy Giuliani to the Supreme Court, that he might use the justice system to punish his political opponents, and that journalists would fall in line rather than forfeit access.

In real life, Trump nominated a respectable Supreme Court justice, the justice system is ensnaring Trump’s own people more than any of his political opponents, and journalists have practically incorporated “resist” into their job description. Gessen’s alarm over encroachments on constitutional norms seems especially surreal in light of who actually did take us down such a road. It was George W. Bush who appointed a crony to the Justice Department and tried to do the same for the Supreme Court. It was Bush who tried to bend the justice system to partisan goals. It was under Bush that many journalists fell in line in order to maintain access.

Similar thoughts came to mind when I read The New York Times’s Michelle Goldberg calling the 2016 election an “apocalypse” and offering a long list of Trumpian misdeeds, most of which seem either exaggerated (Russia links, political prosecutions) or primarily indecent (nepotism, vulgarity). Meanwhile, by this time in the White House of George W. Bush, security failures had led to thousands of American deaths, armies were in Afghanistan, and we’d passed a piece of legislation authorizing roving wiretaps, sneak-and-peek warrants, and indefinite detention of non-citizens. “Enhanced interrogation,” Iraq, and the compensation of Wall Street’s worst actors for their losses was still to come. The nepotistic antics of King Donald and clown prince Jared seem minor by comparison.


link to this extract

Teen girl posed for eight years as married man to write about baseball and harass women • Deadspin

Lindsey Adler:


On Wednesday night, a woman named Erin tweeted a series of screenshots announcing that Schultz is not actually Ryan, a married father of two studying to become a pharmacist. Instead, Schultz is a 21-year-old college student in the Midwest, whose entire career as an aspiring baseball writer has been under a fraudulent byline.

Schultz began contributing to Baseball Prospectus’s local White Sox blog at the end of the 2016 season and wrote for BP South Side and BP Wrigleyville throughout the 2017 season. Additionally, Schultz wrote for the SB Nation sabermetrics site Beyond the Box Score throughout 2017.

People who knew Ryan Schultz online say that in retrospect, some of his behavior seemed odd, but no one expected that this moody White Sox fan from Missouri would actually be a teenage girl.

Schultz’s fraud was as true to the catfish genre as can be. She told the people who discovered she was not who she said she was that she assumed the identity because she felt as if she couldn’t write about baseball professionally as a woman, especially at the age of 13. As the deception went on, she couldn’t figure out how to get out of the middle of her web of lies.

Over time, Ryan formed serial relationships with women who use Twitter to talk about baseball and hockey. Some women told me that he would get drunk and berate them; others told me they felt emotionally abused and manipulated because he would imply that he’d hurt himself if they didn’t continue to talk to him. Ryan received nudes from at least two women I spoke with, one of whom said she did it because she was afraid he would hurt himself if she didn’t.


I’m OK with the posing to write about baseball, but..
link to this extract

Australia adds 107MW rooftop solar in October as 2017 heads for record year • RenewEconomy


Australian homes and businesses continue to install solar at an impressive clip, adding more than 100MW of rooftop PV capacity for the month of October, up from 97MW in September, and almost guaranteeing a record 1GW-plus total for the year.

In its latest monthly update, solar industry analysts SunWiz have charted another another “massive” month of small-scale solar installs, pushing the nation’s total installed PV capacity to 6.7GW, 6GW of which is made up of systems sized at 100kW and below.

According to the report, registrations in October shot up to 107MW, making it the best ever October, the fourth-highest ever level of registrations in a month, and more than double the volume of 22 months ago.

And with volumes this high, SunWiz notes, “it looks like we’re headed for the best ever Q4 AND best ever year,” with total registrations at 852MW for 2017, making the market “almost certain to eclipse 1GW of rooftop solar this year.”

One of the stars of the month for the PV market was commercial solar, with installations in the 10kW-20kW range outdone, in volume, by installs in the 75kW-plus range, as you can see in the table below.

SunWiz notes that the growth in volume occurred in every category, but was especially pronounced in the 6.3-8kW range and in the 75-100kW range.


Australia’s wholesale electricity consumption peaked at 210 terawatt-hours in 2007. For comparison, 100MW of solar will generate around 400GWh over the course of a year; an installed capacity of 6.7GW will generate about 29GWh.

Solar isn’t a baseline, but it can be a hell of an add-on.
link to this extract

Errata, corrigenda and ai no corrida: none notified

Start Up: Facebook the attention piranha, retail apocalypse ahoy!, Google says 250k logins stolen weekly, and more

Let’s just be careful out there that our external keyboard isn’t sending our keystrokes off to China. Photo by louisa_catlover on Flickr.

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 9 links for you. Also: Friday (local variations apply). I’m @charlesarthur on Twitter. Observations and links welcome.

Sean Parker says Facebook ‘exploits’ human psychology • Axios

Mallory Locklear:


Napster cofounder Sean Parker appears to have some regrets about the role he played in bringing social media to the world. Before speaking at an Axios event yesterday, he told reporters that he was now “something of a conscientious objector” on social media, according to Axios, and he shared a few thoughts on how he and others designed sites like Facebook to suck people in.

“When Facebook was getting going, I had these people who would come up to me and they would say, ‘I’m not on social media.’ And I would say, ‘OK. You know, you will be.’ And then they would say, ‘No, no, no. I value my real-life interactions. I value the moment. I value presence. I value intimacy.’ And I would say, … ‘We’ll get you eventually,” Parker said. And he added that the initial goals for companies like Facebook, which Parker served as the first president of, were to make sure users spent as much time on their sites as possible. Interactions such as likes and comments served to bring people deeper into the site, about which Parker said, “It’s a social-validation feedback loop … exactly the kind of thing that a hacker like myself would come up with, because you’re exploiting a vulnerability in human psychology.”


link to this extract

Google says hackers steal almost 250,000 web logins each week • CNN

Selena Larson:


For one year, Google researchers investigated the different ways hackers steal personal information and take over Google (GOOG) accounts. Google published its research, conducted between March 2016 and March 2017, on Thursday.

Focusing exclusively on Google accounts and in partnership with the University of California, Berkeley, researchers created an automated system to scan public websites and criminal forums for stolen credentials. The group also investigated over 25,000 criminal hacking tools, which it received from undisclosed sources.

Google said it is the first study taking a long term and comprehensive look at how criminals steal your data, and what tools are most popular.

“One of the interesting things [we found] was the sheer scale of information on individuals that’s out there and accessible to hijackers,” Kurt Thomas, security researcher at Google told CNN Tech.

Even if someone has no malicious hacking experience, he or she could find all the tools they need on criminal hacker forums.


Man, that’s a lot of stolen logins.
link to this extract

Twitter pauses account verifications after critics slam it for verifying Charlottesville rally organizer • TechCrunch

Sarah Perez:


Twitter today announced it’s pausing all account verifications – the process that gives public figures on Twitter a blue checkmark next to their names – while it tries to resolve “confusion” around what it means to be verified, the company says. The move comes shortly after a wave of criticism directed against the social network for verifying the account belonging to Jason Kessler, the organizer of the white supremacist rally in Charlottesville, Virginia in August that left one person dead.

The Daily Beast discovered that Kessler’s Twitter account had been given the preferred status indicated by the blue badge on Tuesday. When reached for comment, Twitter pointed reporters to its policies around account verification which explain the badge is awarded if an account is “of public interest.”

But the coveted blue checkmark is still hard to achieve for many noteworthy figures, and critics claimed that verifying a known white supremacist isn’t something that’s in the public interest.

Even Twitter doesn’t seem to understand its own rules on the matter, as it has withheld the checkmark before for controversial but influential accounts, including Julian Assange. It also has punished Twitter users by stripping them of verification, as it did with right-winger Milo Yiannopoulos last year, ahead of permanently banning him.


“Public interest” isn’t the test of verification. Or didn’t used to be. It’s whether you’re someone who might be in the public eye, and are who you say you are.

Isn’t it? Else what’s the point? Don’t biometrics verify the person? Twitter is descending into a mess, organisationally. It no longer seems to know quite what it stands for, or why it exists.
link to this extract

America’s ‘retail apocalypse’ is really just beginning • Bloomberg

Matt Townsend, Jenny Surane, Emma Orr and Christopher Cannon:


Making matters more difficult is the explosive amount of risky debt owed by retail coming due over the next five years. Several companies are like teen-jewelry chain Claire’s Stores Inc., a 2007 leveraged buyout owned by private-equity firm Apollo Global Management LLC, which has $2 billion in borrowings starting to mature in 2019 and still has 1,600 stores in North America.

Just $100m of high-yield retail borrowings were set to mature this year, but that will increase to $1.9bn in 2018, according to Fitch Ratings Inc. And from 2019 to 2025, it will balloon to an annual average of almost $5bn. The amount of retail debt considered risky is also rising. Over the past year, high-yield bonds outstanding gained 20%, to $35bn, and the industry’s leveraged loans are up 15%, to $152bn, according to Bloomberg data.

(Key: colour represents% of retail real estate loans that are delinquent by metro areas
Yellow 0-5%; orange 5-10%; red 10-25%; brown 25-53%)

Even worse, this will hit as a record $1 trillion in high-yield debt for all industries comes due over the next five years, according to Moody’s. The surge in demand for refinancing is also likely to come just as credit markets tighten and become much less accommodating to distressed borrowers.

Retailers have pushed off a reckoning because interest rates have been historically low from all the money the Federal Reserve has pumped into the economy since the financial crisis. That’s made investing in riskier debt—and the higher return it brings—more attractive. But with the Fed now raising rates, that demand will soften. That may leave many chains struggling to refinance, especially with the bearishness on retail only increasing.


Higher interest rates, even a little, will create big problems as this debt rolls over: stores will have to generate more money to pay the interest, at a time when the advantages for internet retailers will be growing.
link to this extract

MantisTek GK2’s keylogger is a warning against cheap gadgets (updated) • Tom’s Hardware

Lucian Armasu:


Multiple online user reports claim that the MantisTek GK2 mechanical keyboard’s configuration software is sending data to an Alibaba server. One of the reports even includes an analysis of the software’s traffic, which seems to include how many times keys have been pressed.

The MantisTek GK2 is a cheap RGB mechanical keyboard from China that costs half as much (or less) as the mechanical keyboards from better known companies. Multiple gadgets that come from China seem to have either poor security or privacy issues caused by collecting user data without consumers’ explicit permission. The MantisTek GK2 seems to be one of those products.

The main issue seems to be caused by the keyboard’s “Cloud Driver,” which sends information to IP addresses tied to Alibaba servers. Alibaba sells cloud services, so the data isn’t necessarily being sent to Alibaba, the company, but to someone else using an Alibaba server.

The data being sent—in plaintext, no less— has been identified as a count on how many times keys have been pressed.

The first way to stop the keyboard from sending your key presses to the Alibaba server is to ensure the MantisTek Cloud Driver software isn’t running in the background.

The second method to stop the data collection is to block the CMS.exe executable in your firewall. You could do this by adding a new firewall rule for the MantisTek Cloud Driver in the “Windows Defender Firewall With Advanced Security.”


“Yeah, just updating my firewall rules to stop it telling China what I type.” The update does point out that it’s only sending *how many* times the key was pressed – maybe to see key lifetimes or durability. But even so. Shouldn’t do, especially not without very explicit permission.
link to this extract

Warby Parker app uses iPhone X TrueDepth camera to find your ideal specs • Mac Rumors

Tim Hardwick:


Spectacles company Warby Parker recently updated its mobile app to include a novel implementation of Apple’s face recognition technology exclusive to the iPhone X.

The glasses app uses the smartphone’s front-facing TrueDepth camera to map the user’s face and create an ideal fit for a new set of frames.

Apple’s Face ID authentication works by projecting 30,000 dots on the surface of a person’s face, accurately mapping its curvature and unique features.

The camera’s sensors also capture the data in three dimensions, and it’s this technology in particular that the glasses app uses to recommend to the user a series of frames that it thinks will fit their facial structure.

The only failing of the app is that it doesn’t (yet) place the spectacles on the user’s face, Snapchat-style, to let the customer see what they look like wearing them.


Third-party app making a clever use of a new affordance. Makeup apps next, surely.
link to this extract

Social media has failed its self-driving test • Bloomberg

Leonid Bershidsky:


Both the creators of disturbing kids’ videos and fake news writers game the platforms. The tag-filled names of the videos are designed to exploit YouTube’s search algorithms, and that clearly works since the channels that run the content keep proliferating. The catchy headlines of the fake stories continue fooling Facebook’s supposedly sophisticated clickbait detection algorithms. During the recent congressional hearing on Russian meddling in the 2016 election, the platforms’ representatives were asked about fake accounts but couldn’t come up with any convincing answers about their efforts to purge them.

At least the tech platforms are beginning to recognize that, in order not to be gamed as easily and as often as today, they need more human eyes and human hands. But the hype they spurred by boasting about their intelligent algorithms has acquired a life of its own. I wouldn’t be surprised if a company testing autonomous vehicles took seriously a recent paper by a group of Massachusetts Institute of Technology and Carnegie Mellon University scientists describing something called the Moral Machine. The idea is to automate the ethical decisions that a human driver makes on the fly, even the toughest ones such as whether to hit a wall and kill the car’s passengers, including a young girl, or run over an athlete and his dog crossing the street on a red light. The researchers used a website to ask people about moral choices. The next step is to aggregate the data and have an AI-based algorithm figure out a decision that corresponding to the crowdsourced wisdom.

“The implementation of our algorithm on the Moral Machine dataset has yielded a system which, arguably, can make credible decisions on ethical dilemmas in the autonomous vehicle domain (when all other options have failed),” the researchers wrote. “But this paper is clearly not the end-all solution.”

Guess which parts of this sentence a tech company would throw away if it decided to implement the algorithm. My bet is on “arguably” and “clearly not the end-all solution.”


link to this extract

Even a novice hacker could breach the network hosting Kris Kobach’s bogus voter fraud program • Gizmodo

Dell Cameron:


To suggest that state officials involved with the program have been grossly negligent is simply too kind.

Kobach was appointed vice-chairman of President Donald Trump’s election integrity commission this year after Trump repeatedly and falsely suggested that between 3 and 5 million people voted illegally in the 2016 general election, ultimately costing him the popular vote. Since taking office, the Trump administration has been pushing to take Kobach’s flawed program nationwide. (As of this week, the commission is being sued by one of its own commissioners.)

Gizmodo has learned, however, that the records passing through the Crosscheck system have been stored on a server in Arkansas operating on a network rife with security flaws. What’s more, multiple sets of login credentials, which could be used by virtually anyone to directly access the Crosscheck system—as well the encrypted voter data it contains—have been compromised.

Our investigation into the program builds on the work of ProPublica, which last month published a report describing multiple security flaws plaguing Crosscheck’s operations. Documents obtained under state transparency laws by the anti-Trump group Indivisible Chicago revealed that Crosscheck had emailed Illinois election officials both the username and password to the program’s FTP server—credentials that Illinois neglected to redact before releasing the emails publicly.

The emails further revealed that participating states had submitted millions of voter files to the Arkansas server using an unencrypted file transfer protocol. Gizmodo has learned that while some of the data sets were encrypted prior to being transferred, the passwords to decrypt three year’s worth of voter files, belonging to every state participating in Crosscheck, have likewise been exposed.


link to this extract

Oppo, Xiaomi to adopt 3D sensors for smartphones in 2018 • Digitimes

Sammi Huang and Steve Shen:


China-based vendors Oppo and Xiaomi Technology will adopt 3D sensing solutions for smartphones to be launched in 2018, with such solutions to be developed by Himax Technologies via cooperation with Qualcomm and the sensor modules to be produced by Truly Opto-Electronics, according to industry sources.

The cooperation efforts by Qualcomm, Himax and Truly Opto-Electronics will help upgrade significantly the hardware specifications of high-end models rolled out by China-based smartphone vendors in the coming year further enhancing their competitiveness, said the sources.

The facial recognition solutions co-developed by Qualcomm, Himax and Truly Opto-Electronics are expected to enter volume production in March-April 2018 at the earliest, indicated the sources.

Meanwhile, China’s top smartphone vendor Huawei is reportedly cooperating with China-based Sunny Optical Technology to develop related 3D sensor solutions for its premier models, indicated the sources.


Oh, I thought the race was to develop fingerprint readers under the front screen? Guess that’s been abandoned now that Apple says it dropped that idea more than a year ago. The Android vendors do adequate catch-up, but given that Samsung has failed at face recognition, and that Apple had to buy a specialist company to do FaceID, I wouldn’t put a lot of store in this being a great – as in secure and fast – experience.

In 2015 Huawei, of course, was a couple of weeks ahead of Apple in showing off “Force Touch” in its phones, illustrated by weighing an orange. But it only worked with Huawei’s own apps, so it was essentially pointless. BY contrast, lots of third-party iOS apps have adopted it: all but one phone (the SE) that it now sells incorporates it.

Facial recognition will be patchy across Android devices; the implementations will be uneven and security variable. But they’ll be cheaper.
link to this extract

Errata, corrigenda and ai no corrida: none notified

Start Up: FaceID iPad?, Facebook’s creepy tracking, Pandora boxed in, inside the DNC hack, and more

Here’s your Logitech Harmony Link next year. Photo by Leo Reynolds on Flickr.

A selection of 12 links for you. Do not use for secret visits. I’m @charlesarthur on Twitter. Observations and links welcome.

Why social media users should never lower their guard • Time

Tim Bajarin:


While I’m highly selective about my interactions on Facebook and Twitter, with LinkedIn I tend to be more liberal about okaying requests to connect. I’ve reasoned that since LinkedIn is for business networking, the more people I network with, the better it is for my career and business relationships. I suspect that’s the feeling shared by the millions of others LinkedIn users who frequent the site for similar reasons.

But then I came across a report from SecureWorks, an Atlanta-based cybersecurity subsidiary of Dell (the computer company), titled “The Curious Case of Mia Ash: Fake Persona Lures Middle Eastern Targets.” According to the July 27 report, SecureWorks says it observed phishing campaigns targeted at Middle East and North Africa that delivered PupyRAT, the codename for a nasty bit of malware that targets Windows, Linux, OS X and Android systems, using a fake person named “Mia Ash.”

In short, this report reveals that a known Iranian hacker group called Cobalt Gypsy created the fake LinkedIn profile of a woman it dubbed Mia Ash and identified as a celebrated photographer. When I checked out Mia Ash’s profile, it looked like so many others I’ve scanned on both LinkedIn and other social media networks over the years.

The fake profile’s goal was to connect with individuals working in Middle Eastern companies, then trick users into opening a Word document using their company’s email in order to deliver the malware. The malware could then infect their company’s network and potentially allow malefactors entry into the network to steal information, or do who knows what else.

It turns out this wasn’t the first time Cobalt Gypsy had targeted LinkedIn users. Some years ago, the hacker group used agents posing as recruiters on the social networking service to lure their targets into downloading malware-laden job applications. Their goal was the same: to try and get users to open a Word document that used their company email addresses to deliver the payload. In this case, the fake LinkedIn persona was someone called “Timothy Stokes,” whose profile identified him as a recruiter for a well known company.


Yeah, trusting LinkedIn but not the others looks like something of a category error.
link to this extract

How the AR-15 became mass shooters’ weapon of choice • Rolling Stone

Tim Dickinson:


The AR-15 assault rifle was engineered to create what one of its designers called “maximum wound effect.” Its tiny bullets – needle-nosed and weighing less than four grams – travel nearly three times the speed of sound. As the bullet strikes the body, the payload of kinetic energy rips open a cavity inside the flesh – essentially inert space – which collapses back on itself, destroying inelastic tissue, including nerves, blood vessels and vital organs. “It’s a perfect killing machine,” says Dr. Peter Rhee, a leading trauma surgeon and retired captain with 24 years of active-duty service in the Navy.

Rhee is most famous at home for saving the life of Arizona Rep. Gabby Giffords after she was shot point-blank in the head with a handgun fired by a mass shooter in 2011. “A handgun [wound] is simply a stabbing with a bullet,” says Rhee. “It goes in like a nail.” With the high-velocity rounds of the AR-15, he adds, “its as if you shot somebody with a Coke can.”

Versions of the AR-15 have been the U.S. military’s standard-issue assault rifle in every war since Vietnam. But only in the past dozen years have semi-automatic models become a fixture of American life. Gun-makers – emboldened by Congress and cloaked in the Second Amendment – have elevated the AR-15 into an avatar of civilian manhood, independence and patriotism. In the process, this off-patent combat rifle has become an infinitely customizable weapon platform that now accounts for nearly one in five guns sold in America.

The federal government has deemed them “semi-automatic assault rifles” with magazine capacities that serve “no sporting purpose.” But the National Rifle Association now simply calls the AR-15 “America’s Rifle.”


Fascinating; your long read for the day. Also, it’s open source! The patent expired ages ago.
link to this extract

Double digit growth by 3 of top 5 vendors as global tablet shipments hit 45m units


Eric Smith, director – Tablets and Touchscreens said, “Global tablet shipments declined 5% annually from 46.9m units in Q3 2016 to 44.6m in Q3 2017, but grew 2% quarter on quarter from 43.7m in Q2 2017.

The global tablet market has reduced the high negative growth rates of the past couple of years and Apple just strung together two straight quarters of year-on-year growth. During Q3 2017, Huawei and Amazon also kept up their pace of strong gains in their respective corners of the Android market, while Lenovo bounced back to positive growth with good footing in the Android and Windows segments.

Windows tablet demand is experiencing a slump overall, compared to this time last year as consumer market pricing and marketing have failed to connect to consumers while enterprise demand is still swift for pricier 2-in-1 tablet form factors.”


IDC had broadly similar numbers earlier this week. What’s useful here is the insight that it’s Windows which isn’t going further in the tablet market. Also Samsung: it’s really struggling – can’t topple Apple, can’t compete with the low-priced whitebox Android tablets.
link to this extract

He solved the DNC hack. Now he’s telling his story for the first time • Buzzfeed

Jason Leopold:


One late morning in May 2016, the leaders of the Democratic National Committee huddled around a packed conference table and stared at Robert Johnston. The former Marine Corps captain gave his briefing with unemotional military precision, but what he said was so unnerving that a high-level DNC official curled up in a ball on her conference room chair as if watching a horror movie.

At 30, Johnston was already an accomplished digital detective who had just left the military’s elite Cyber Command, where he had helped stanch a Russian hack on the US military’s top leadership. Now, working for a private cybersecurity company, he had to brief the DNC — while it was in the middle of a white-knuckle presidential campaign — about what he’d found in the organization’s computer networks.

Their reaction was “pure shock,” Johnston recalled. “It was their worst day.”

Although the broad outlines of the DNC hack are now well-known, its details have remained mysterious, sparking sharp and persistent questions. How did the DNC miss the hack? Why did a private security consultant, rather than the FBI, examine its servers? And how did the DNC find Johnston’s firm, CrowdStrike, in the first place?

Johnston’s account — told here for the first time, and substantiated in interviews with 15 sources at the FBI, the DNC, and the Defense Department — resolves some of those questions while adding new information about the hack itself.


Johnston has a perfectly reasonable explanation for why the FBI wasn’t called in: it was the middle of the campaign, and it would have been sayonara servers. This, at a time when the FBI was still investigating Clinton’s use of a private email server. Imagine the mess the press would have made of that.
link to this extract

Pandora has lost $1bn in four years and is worth less than ever. Can it be salvaged? • Music Business Worldwide

Tim Ingham:


Pandora has posted net losses of $473.6m so far in a ‘transformative’ 2017 – which has seen it part ways with CEO Tim Westergren and other top execs, as well as the sale of its Ticketfly business to Eventbrite.

Although this figure includes a one-time Q2 goodwill impairment write down of $132m related to the net assets of Ticketfly (and other one-time expenses), it’s a very alarming indication of the firm’s long-term sustainability.

As is, it doesn’t have any sustainability.

It now looks almost guaranteed that Pandora, which lost $90m in Q4 2016, will post net losses in excess of half a billion dollars in this calendar year. Yikes.

The new regime at the top of Pandora includes some very smart and respected people – not least Greg Maffei, the recently-appointed Chairman of the company’s board.

Maffei is also Chairman of SiriusXM, which acquired an effective 16% stake in Pandora when it invested $480m in the flailing business back in June.

The question now is what Maffei and other newbies at the top of Pandora can do to transform a company which is haemorrhaging money, value and users at a frightening rate.

“Pandora is very much a business in transition, and there are tangible challenges,” admitted new CEO Roger Lynch (pictured, main) to investors last week.


It’s a business in transition to being shut down or sold. Streaming music is a fungible good now.
link to this extract

How Facebook figures out everyone you’ve ever met • Gizmodo

Kashmir Hill:


Facebook’s machinery operates on a scale far beyond normal human interactions. And the results of its People You May Know algorithm are anything but obvious. In the months I’ve been writing about PYMK, as Facebook calls it, I’ve heard more than a hundred bewildering anecdotes:

• A man who years ago donated sperm to a couple, secretly, so they could have a child — only to have Facebook recommend the child as a person he should know. He still knows the couple but is not friends with them on Facebook.

• A social worker whose client called her by her nickname on their second visit, because she’d shown up in his People You May Know, despite their not having exchanged contact information.

• A woman whose father left her family when she was six years old — and saw his then-mistress suggested to her as a Facebook friend 40 years later.

• An attorney who wrote: “I deleted Facebook after it recommended as PYMK a man who was defense counsel on one of my cases. We had only communicated through my work email, which is not connected to my Facebook, which convinced me Facebook was scanning my work email.”

Connections like these seem inexplicable if you assume Facebook only knows what you’ve told it about yourself. They’re less mysterious if you know about the other file Facebook keeps on you—one that you can’t see or control.


“Mysterious” isn’t the word I’d use; “creepy” feels more like it. The way it’s done is “shadow profiling”, using everything about you from everywhere, from people who just clicked “share contacts” on installing the app.
link to this extract

Apple is ramping up work on AR headset to succeed iPhone • Bloomberg

Mark Gurman:


Apple, seeking a breakthrough product to succeed the iPhone, aims to have technology ready for an augmented-reality headset in 2019 and could ship a product as early as 2020.

Unlike the current generation of virtual reality headsets that use a smartphone as the engine and screen, Apple’s device will have its own display and run on a new chip and operating system, according to people familiar with the situation. The development timeline is very aggressive and could still change, said the people, who requested anonymity to speak freely about a private matter.

While virtual reality immerses the user in a digital world, augmented reality overlays images and data on the real one. The applications for AR are endless, from a basketball fan getting stats while watching a game to a mechanic streaming instructions on how to fix a specific piece of equipment. Apple isn’t the only company working on the technology. Google, which drew derision for $1,500 smart glasses a few years ago, is developing a business-oriented variant. Startup Meta has developed a headset with a focus on education and medical uses.


2020? Why the hurry?
link to this extract

Apple is working on 2018 iPad redesign with facial recognition • Bloomberg

Mark Gurman:


Apple is working on a redesigned, high-end iPad for as early as 2018 that incorporates key iPhone X features such as slimmer edges and facial recognition, according to people familiar with the matter.

However, the new version of the tablet is unlikely to include an OLED screen, which provides more vivid colors and sharper clarity, the people said. They asked not to be identified talking about private product development.

At least one new iPad model with a screen size similar to the 10.5-inch iPad Pro is planned to include Face ID for unlocking the device, making payments, and sending animated emojis. The feature would replace the iPad home button that has come with a fingerprint scanner since 2014, one of the people said. The updated tablet is expected to be released later next year, a little more than a year after the last major iPad Pro upgrade, the people said.


Gurman’s supply-chain sources strike again. This will have been in the works since early this year. A couple of obvious questions: will the “notch” for the face recognition be on the long or the short side? And what specific stage of the manufacturing and testing stage has this iPad version reached that Gurman (or sources) has found out now?
link to this extract

Logitech to shut down “service and support” for Harmony Link devices in 2018 • Ars Technica UK

Valentina Palladino on EOL for the popular Harmony Link remote system (which lets you control home theatre and sound systems from a mobile app):


The only reason provided comes from a Logitech employee with the username Logi_WillWong, who explains in a response post from September 8, 2017 that Logitech will not be renewing a “technology certificate license” that expires in March. No details were provided about how this certificate license allows the Harmony Link to function, but it appears that without it, those devices will not work as promised. “The certificate will not be renewed as we are focusing resources on our current app-based remote, the Harmony Hub,” Logi_WillWong added, which seems to indicate that the shutting down of the Harmony Link system is a way to get more customers on the newer Harmony Hub system.

But customers are calling out Logitech for bricking a device that works perfectly fine for most of them, presumably in the hopes of forcing an upgrade to a new device. While out-of-warranty customers can get a discount on a Harmony Hub, according to an updated response posted yesterday on the Logitech support forum from Logi_WillWong, those still under warranty can receive a free Harmony Hub from Logitech as a replacement for their Harmony Link.

The Harmony Link system predates the Harmony Hub by a few years and allows users to control televisions, sound systems, and even VCRs and Blu-ray players from the free companion mobile app. The Harmony Hub expands on those capabilities by adding more IoT device support, making things like Roku players, Hue smart lights, and Sonos systems controllable via the remote app. The Harmony Link isn’t available to purchase from Logitech’s website anymore, and it’s listed as “discontinued by manufacturer” on Amazon.

The forced end-of-life of the Harmony Link is a harsh reminder that companies like Logitech have the power to make useful yet older devices obsolete for whatever reason they see fit.


Also a harsh reminder that companies like this ignore the press when it comes to explaining why they’re doing stuff.
link to this extract

There’s a trick that makes your iPhone X battery pretty much last forever • BGR

Zach Epstein:


By making some tweaks to your settings, you can configure the iPhone X so that it barely drains any battery. In fact, depending on your usage, you’ll be able to squeeze two days of battery life or perhaps even more out of your iPhone. The trick was noted by Twitter user Neil Hughes a few days ago, and we’ll explain how it works.

There are three things you need to do in order to make this happen. First, you’ll need to switch your screen to grayscale. In the Settings app, go to General > Accessibility > Display Accommodations. Tap Color Filters, then slide the toggle next to “Color Filters” to on. Now, tap the “Grayscale” option that appears below.

Next, you’ll want to invert the display’s colors. Still within Display Accommodations in the Settings app, tap Invert Colors and then slide the toggle next to “Classic Invert” to on.

Finally, on the main menu screen in the settings app, scroll down to Battery and enable “Low Power Mode” in the battery section. This will restrict background processes so that first- and third-party apps use as little energy as possible. Also, if you want to take things a step further, apply an all-white wallpaper to your home screen and lock screen. It’ll turn to pure black when you invert your colors, thus conserving power by preventing as many pixels as possible from lighting up.


Get yer white wallpaper from BGR. This reminds me of hypermilers trying to go the longest distance on a gallon of petrol. Though I’m not saying I might not try this for fun.
link to this extract

December 2016: why Snapchat’s design is deliberately confusing • Prototypr

Benjamin Bradall(who higher up in the story says “I’m 23 and will freely admit that I find Snapchat’s design confusing) writing in December 2016:


The launch of Snap’s new physical product, Spectacles, was what made me realize that the Snapchat app’s unlabeled press-it-and-see-what-happens UI is no mistake.
Just look at the vending machine:

A credit card reader and three buttons: hit one of those massive glowing pads to get a pair of Spectacles of that color. Like the Snapchat app, the machine has no text, no obvious instructions, no clarity. The machine assumes itself is a big enough deal that it’s your fault if you don’t know.
Same with the banner:

Snap pulled a Cloverfield on us for the Spectacles launch, teasing the Snapchat ghost with eyes on billboards around the US without being explicit with exactly what’s going on.

Tying this back to design, and the choices Snap’s making for its brand identify, it makes more sense the more you think about it.

Positioned as a rare place where your content is totally hidden from the public, the reasons for Snapchat’s arcane, exclusive design are clear. It’s not a social media platform in the same way that Facebook is, and that’s mainly because of its total lack of discovery.

What I mean by that is there is no way to find users without knowing their phone number or their handle: two pretty private pieces of information. And if/when your grandma does try to add you, you’ll still need to accept the request. Not that she will add you, because she’s not figured out how to get into the bloody thing.


November 2017: Snap, having been walloped by the easier-to-use Instagram Stories, say it’s redesigning its app to be easier to use.

Design is how it works. If it works.
link to this extract

With deletion of one wallet, $280M in Ethereum wallets gets frozen • Ars Technica UK

Sean Gallagher:


Digital currencies and the wallets that hold them have become an increasingly attractive target for digital pickpockets, resulting in millions of real dollars’ worth of lost currency. A $50 million heist of Ethereum currency last year exploiting weaknesses in the cryptocurrency’s underlying software threatened to break the Bitcoin competitor. But a new security bug in a popular Ethereum wallet platform has caused what amounts to a bank freeze on scores of high-value wallets. Today, Parity Technologies Ltd., the developer of cryptographic “wallets” for the digital currencies Bitcoin and Ethereum, announced that an “accidental” triggering of a bug affecting certain Parity wallets had broken them, making it impossible to transfer Ethereum funds out of them.

As a result, 1 million ETH have become frozen in wallets—roughly $280 million (US) worth of digital currency. Of that, about $90 million belongs to Parity founder and former Ethereum core developer Gavin Woods’ Initial Coin Offering (ICO) Polkadot, according to Tuur Demeester, editor in chief at Adamant Research.

The bug specifically affects multi-signature wallets created with a digital contract after July 20. Multi-signature wallets have cryptographic security measures that require multiple users to sign a transaction in order for it to be processed and approved—an approach that allows for escrow contracts to control payments from accounts belonging to a group.


link to this extract

Errata, corrigenda and ai no corrida: none notified

You can now sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

Start Up: car doom nears, a week of iPhone X, Snap swoons, another Pixel 2 XL woe, and more

The patent wrangle between Samsung and Apple over this feature is finally over. Who won? Photo by Oyvind Solstad on Flickr.

A selection of 13 links for you. That’s just the way it is. I’m @charlesarthur on Twitter. Observations and links welcome.

Bob Lutz: Kiss the good times goodbye • Auto News

Bob Lutz is a former vice chairman and head of product development at General Motors. He also held senior executive positions with Ford, Chrysler, BMW and Opel: I


t saddens me to say it, but we are approaching the end of the automotive era.

The auto industry is on an accelerating change curve. For hundreds of years, the horse was the prime mover of humans and for the past 120 years it has been the automobile.

Now we are approaching the end of the line for the automobile because travel will be in standardized modules.

The end state will be the fully autonomous module with no capability for the driver to exercise command. You will call for it, it will arrive at your location, you’ll get in, input your destination and go to the freeway.

On the freeway, it will merge seamlessly into a stream of other modules traveling at 120, 150 mph. The speed doesn’t matter. You have a blending of rail-type with individual transportation.

Then, as you approach your exit, your module will enter deceleration lanes, exit and go to your final destination. You will be billed for the transportation. You will enter your credit card number or your thumbprint or whatever it will be then. The module will take off and go to its collection point, ready for the next person to call.

Most of these standardized modules will be purchased and owned by the Ubers and Lyfts and God knows what other companies that will enter the transportation business in the future.

A minority of individuals may elect to have personalized modules sitting at home so they can leave their vacation stuff and the kids’ soccer gear in them. They’ll still want that convenience.

The vehicles, however, will no longer be driven by humans because in 15 to 20 years — at the latest — human-driven vehicles will be legislated off the highways.

The tipping point will come when 20 to 30% of vehicles are fully autonomous. Countries will look at the accident statistics and figure out that human drivers are causing 99.9% of the accidents.


link to this extract

One week with the iPhone X • Six Colors

Jason Snell:


I’m loving iPhone X in almost all the places I use it. The gestures are becoming second nature to me. But there’s one use case where it doesn’t really work: laying on a table. And it doesn’t work there for several reasons. The sizable camera bump makes the whole thing unstable. Facing straight up, the Face ID camera can’t see me, so I can’t unlock my phone without leaning way over the table or picking the phone up. And attention detection can’t detect me, so after 30 seconds the screen dims.

I hadn’t realized how much I left an iPhone unlocked on a table for a minute or two. The iPhone X is more aggressive about locking the phone (and dimming the display), and Face ID is no help. I suppose in the end, the phone will train me—but right now it’s one of the areas where my old way of using my iPhone no longer seems to apply.

I’m dissatisfied with the relegation of Control Center to the upper right corner of the screen. That corner is inaccessible to me when I’m using the phone one handed. (I can shimmy my hand around a bit and reach high spots on the screen, but the upper right corner is just too far away.) It’s all made me realize how often I used Control Center functions.

Perhaps Apple will add some feature to make Control Center more accessible. I like this idea from my pal Lex Friedman, who suggests that the optional Reachability shortcut (swipe down in the home indicator area) could be mapped to other functions, including Control Center, instead. Sounds like a great idea to me.

While we’re at it, I’d like the buttons on the lock screen to be customizable as well. It’s great that I can turn on the flashlight with one pleasant, haptic-filled 3D touch command from the lock screen. It’s great that I can activate the camera with a similar gesture (though it’s also redundant, since I can swipe from right to left to do the same thing). It would be even greater to drop a couple other commonly used Control Center features on there.


The Reachability/Control Center idea is a good one. (Reachability is off by default on the X, unlike previous phones since the 6.) I can see that the timer, say, or alarm might be useful on the lock screen.
link to this extract

The iPhone X is a user experience nightmare • Fast Company

Jesus Diaz:


You’re looking at a UX disaster, the result of eliminating what is probably the simplest, most intuitive form of navigation ever implemented in consumer electronics: the iPhone’s home button. The iPhone X replaces it with the mess above. This is bad news, because this interaction is a fundamental part of the user experience.

Joanna Stern’s review for the Wall Street Journal–which still concludes that, “Yes, There Are Reasons to Pay Apple $1,000”–documents what this means in detail: “[T]he lack of a home button means your thumb is about to turn into one of those inflatable waving tube-men outside the car dealership [. . .] you must master a list of thumb wiggles, waves and swipes [. . .] the other gestures, however, are buried. Many moves require almost surgical precision.” Heather Kelly, for CNN Money, adds her own experience: “To fill the void left by the Home button, the iPhone X has added new gestures (the different swipes you make with a finger). The process of learning them is a pain, and some of the new options are more work than before.” The Verge declared that “there’s a whole new system of gestures and swipes to learn and master, and many of them will be annoying to remember and difficult to perform with just one hand.”


Always good to include a piece that is categorically wrong. I’m fairly confident that if Steve Jobs could have introduced a phone without a home button, he’d have been over the moon.
link to this extract

Apple has finally won $120m from Samsung patent battle • The Verge

Jacob Kastrenakes:


After years of sparring in the courts, Apple has once and for all claimed victory over Samsung to the count of $120m. The Supreme Court said today that it wouldn’t hear an appeal of the patent infringement case, first decided in 2014, which has been bouncing through appeals courts in the years since.

The case revolved around Apple’s famous slide-to-unlock patent and, among others, its less-famous quick links patent, which covered software that automatically turned information like a phone number into a tappable link. Samsung was found to have infringed both patents. The ruling was overturned almost two years later, and then reinstated once again less than a year after that. From there, Samsung appealed to the Supreme Court, which is where the case met its end today.


Thank god that’s finished. But: there’s still another part, over $1bn (reduced to $400m) which returns to court in May.
link to this extract

Opinion: Broadcom will kill Qualcomm • PC Magazine

Sascha Segan:


“Broadcom” is actually a company called Avago, a spin-off of Hewlett-Packard that, in recent years, has spent as much time and energy buying, dismembering, cutting costs on, and selling off parts of other companies as it has inventing things. This has resulted in great financial performance, but not so much in the way of innovation. The company is run by Hock Tan, who Fortune describes as a “finance geek,” not an innovator.

This isn’t always a bad or a good thing, in the big picture. There’s a lot of consolidation going on in the chip industry right now, and if companies can come together in a way that preserves competition and improves their ability to create great products, I’d say more power to them.

But that’s not the general opinion of Avago among technology-focused analysts. On Twitter, Patrick Moorhead of Moor Insights and Strategy, who has three decades of knowledge about chips, says “Broadcom would slice, dice, destroy.”

Anshel Sag, also at Moor, says, “Buy. Chop up. Sell off. Raise prices. Rinse. Repeat.”

How about Ben Wood from CCS Insight? “Still astounded this has even got this far.”

Broadcom is also doing a suspiciously shifty thing right now in moving its nominal headquarters from Singapore to the US, possibly to avoid regulatory scrutiny over this deal. On Twitter, analyst Neil Shah of Counterpoint Research says this is a “smokescreen” where “core HR/finance” will still be controlled in Singapore.

Bloomberg quotes more Wall Street-esque analysts as saying that the Broadcom buy could smooth things over with Apple (Broadcom and Apple get along) and increase revenue. But there’s nary a word in there about innovation, merely about squeezing more milk out of the existing cows.

And even if Broadcom doesn’t want to sell off parts of Qualcomm, it may have to. The joined company’s control over Wi-Fi chipsets may be so great it would trigger antitrust scrutiny, dragging both companies down a rabbit hole as they try to shed whatever parts would maintain competition.


link to this extract

Mattress company shutters web publication, pivots to print • WSJ

Jack Marshall:


Mattress brand Casper is launching a print magazine and shuttering Van Winkle’s, the sleep-focused online publication it launched in 2015.

The company said its new magazine, titled Woolly, will be published multiple times a year and focus on themes including comfort, wellness and modern life. It will be bundled free with some Casper products and available for $12 per issue from Casper’s retail stores and website.

Companies have flocked to so-called content marketing in recent years in an attempt to align their brands with certain topics and issues without relying on straight-forward advertising. The tactic has become prevalent online, but some companies, such as Airbnb, have since taken the approach offline with their own branded print products.

But according to Casper, Woolly shouldn’t be viewed as marketing designed simply to drive mattress sales. Rather, it says it wants to use it as a vehicle to link the company to subjects it “believes in.”


Casper is behaving very strangely recently. Recall this article about “the war to sell you a mattress is an internet nightmare” from last month.
link to this extract

Experimenting to solve cramming • Twitter Engineering

Lucile Lu works on data science at Twitter:


When we make a decision as significant as changing the number of characters available for Tweeting, we need to know it’s right. How do we figure that out? Through a rigorous evaluation process to ensure any change — especially one at the core of who we’ve been for more than a decade – improves everyone’s experience on Twitter.

We started looking at making a change to our 140-character limit by performing a tweet-length analysis to validate our hypotheses that more characters for certain languages would lead to less frustration with Tweeting and fewer abandoned Tweets. We took caution and abundant pre-experiment preparation to ensure that we would get reliable data and meaningful results.

Usually, any new feature we test requires just one experiment. As discussed earlier, however, we were anticipating regional heterogeneity: people from different countries send Tweets at different lengths and would react differently since they’d be affected by the experiment in different ways. So, how do we control for this?


Still don’t like it, though the rationale for having longer tweets in different languages – to be equivalent to what you get from 140 in English – is fair.
link to this extract

Snap plunges again as revenue, user growth underwhelm • FT

Tim Bradshaw:


Snap’s difficult run as a young public company continued on Tuesday, after it posted revenues and user growth below Wall Street’s forecasts and was hit by a $40m charge related to unsold supplies of its Spectacles video-camera sunglasses.

As Snap’s stock plunged 18% in after-hours trading, Evan Spiegel, co-founder and chief executive, warned of further uncertainty ahead as it overhauls its app, in an attempt to win a broader audience.

Snapchat saw daily active users grow 17% year-on-year to 178m, adding just 5m in the last three months, with revenues up 126% year-on-year to $208m. On the plus side, Snap’s adjusted losses per share were 1 cent better than expected at 14 cents, and free cash outflows narrowed compared with a year earlier, even as net losses for the quarter more than tripled to $443m.

In response to growing pressure from investors, Mr Spiegel took the unusual step of laying out details of its product roadmap for the coming year — including a redesign to the main Snapchat app, which he said “will be disruptive to our business in the short term”.


Snap has a problem: Facebook can copy it faster and on a bigger scale. It looks a bit like Apple in the desktop days v Microsoft, except Apple could move faster than Microsoft. If Microsoft had been able to copy Apple’s ideas and implement them faster than Apple, there’d have been no iPhone – maybe even no iPod. (Muse on that.)
link to this extract

EU’s Vestager seeking details on Apple’s recent tax setup • Reuters

Foo Yun Chee:


European Competition Commissioner Margrethe Vestager, who issued the record back-tax bill against Apple in August 2016, said she wanted to make sure the company now complies with the bloc’s rules which ban unfair state aid.

“I have been asking for an update on the arrangement made by Apple, the recent way they have been organized, in order to get the feeling whether or not this is in accordance with our European rules but that remains to be seen,” Vestager told a news briefing at an international tech summit in Lisbon.

“We are looking into this of course without any kind of prejudice, just to get the information,” she said.

Vestager said her request preceded reports based on the “Paradise Papers” which showed that Apple shifted key parts of its business to Jersey as an offshore tax haven in a move to maintain a low tax rate. Apple has said no operations were moved from Ireland.


link to this extract

Why Google is in breach of the EC’s June 2017 prohibition decision • Foundem

Another forensic examination of the economics behind Google’s somewhat cynical “solution” to the EC’s ruling. I don’t understand why Margrethe Vestager allows Google to use such an obviously non-compliant solution, and why she doesn’t just hire Foundem to do the analysis of each proposal. Shivaun and Adam Raff have called every shot correctly in this long, sad saga – so far, they’re nearly a decade into Google abusing its control of search results to squash their company, which had a solution Google couldn’t compete with.
link to this extract

What Carter Page told House Russia investigators • Bloomberg

Billy House and Shannon Pettypiece:


Carter Page, a foreign policy adviser to Donald Trump’s 2016 presidential campaign, refused last week to give a congressional committee documents related to the Russia investigation because he said they might not all “match up” with information from earlier wiretaps that caught his conversations.

The House Intelligence Committee Monday night released a 243-page transcript of his lengthy appearance behind closed doors with its Russia probe. Page, who said he never met or spoke with Trump, said that he did have contact with Russian government officials during a July 2016 trip in Moscow. But he insisted he wasn’t doing so as a representative of the Trump camp. 

“Unfortunately, I am the biggest embarrassment surrounding the campaign,” he told the panel.


1) crowded field
2) wiretaps – plural??
link to this extract

CIA director met advocate of disputed DNC hack theory — at Trump’s request • The Intercept

Duncan Campbell and James Risen:


CIA director Mike Pompeo met late last month with a former US intelligence official who has become an advocate for a disputed theory that the theft of the Democratic National Committee’s emails during the 2016 presidential campaign was an inside job, rather than a hack by Russian intelligence.

Pompeo met on October 24 with William Binney, a former National Security Agency official-turned-whistleblower who co-authored an analysis published by a group of former intelligence officials that challenges the US intelligence community’s official assessment that Russian intelligence was behind last year’s theft of data from DNC computers. Binney and the other former officials argue that the DNC data was “leaked,” not hacked, “by a person with physical access” to the DNC’s computer system.

In an interview with The Intercept, Binney said Pompeo told him that President Donald Trump had urged the CIA director to meet with Binney to discuss his assessment that the DNC data theft was an inside job. During their hour-long meeting at CIA headquarters, Pompeo said Trump told him that if Pompeo “want[ed] to know the facts, he should talk to me,” Binney said.


Well, the FBI met a company which insisted that the Sony Pictures hack was an inside job. Didn’t change the FBI’s mind. I doubt Binney’s contra-rotating eyes had an impact on the FBI’s view of this case, but it got Trump to stop fulminating.
link to this extract

Some Google Pixel 2 XL oleophobic coatings are already wearing off, leaving behind smudges • 9to5Google

Ben Schoon:


As much as we love the Google Pixel 2 XL, saying it has a few issues is, unfortunately, an understatement. LG’s OLED display has been disappointing in more ways than one, with burn-in being the biggest sore point. Many can look past those issues, but now another one is popping up with problems regarding the oleophobic coating.

Almost every Android smartphone ships with an oleophobic coating on the glass, and the purpose of that is to help make fingerprints easier to get off of the display. In short, it keeps the oil from your fingertips from adhering to the glass, and it also makes water easier to wipe away.

Over time, this coating wears off, but it usually takes at least a few months or years of intense usage. On the Pixel 2 XL, however, some owners are having this come up within just a couple of weeks.

Reports on Google’s product forums as well as Reddit reveal that the oleophobic coating at least on some Pixel 2 XL devices may be pretty weak, rubbing off easily. While some cases are certainly worse than others, this doesn’t sound good for how this phone will age.


Oh man. (Also: almost every Android smartphone ships with an oleophobic coating?) Looks like LG isn’t going to get that contract renewed.
link to this extract

Errata, corrigenda and ai no corrida: none notified

You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

Start Up: what your kids are watching, BroadQual?, Russia’s early Trumpbots, Apple delights, and more

Art thefts have taken on a new form for the email age. Photo by AV Dezign on Flickr.

Tell a friend they can sign up to receive each day’s Start Up post by email. They’ll need to click a confirmation link, so no spam.

A selection of 8 links for you. Use them wisely. I’m @charlesarthur on Twitter. Observations and links welcome.

Broadcom offers $105bn for Qualcomm in landmark deal • Bloomberg

Ian King:


Broadcom Ltd. offered about $105bn for Qualcomm Inc., kicking off an ambitious attempt at the largest technology takeover ever in a deal that would rock the electronics industry.

Broadcom made an offer of $70 a share in cash and stock for Qualcomm, the world’s largest maker of mobile phone chips. That’s a 28% premium over the stock’s closing price on Nov. 2, before Bloomberg first reported talks of a deal. The proposed transaction is valued at approximately $130bn on a pro forma basis, including $25bn of net debt.

Buying Qualcomm would make Broadcom the third-largest chipmaker, behind Intel Corp. and Samsung Electronics Co. The combined business would instantly become the default provider of a set of components needed to build each of the more than a billion smartphones sold every year. The deal would dwarf Dell Inc.’s $67bn acquisition of EMC in 2015 – then the biggest in the technology industry.


Broadcom is so keen to do this that it doesn’t care whether or not Qualcomm’s current $47bn takeover bid for NXP completes or not. It wants Qualcomm anyway. Hard to see this sort of consolidation as good for the industry. But Singapore-based Avago, which reverse-qacuired Broadcom in 2016. is also moving its official headquarters to the US – which would make regulatory approval for the takeover a lot easier.

Trump supporters thought getting Broadcom to relocate was a coup. In fact it’s a way to erode the US’s supremacy in this chip space; the control of the unified company will rest outside the US.

Qualcomm, unsurprisingly, isn’t keen on this deal.
link to this extract

Galleries hit by cyber crime wave • The Art Newspaper

Cristina Ruiz, Anna Brady, Sarah P. Hanson and Julia Michalska:


Hackers are stealing large sums of money from art galleries and their clients using a straightforward email deception. The Art Newspaper has so far identified nine galleries or individuals targeted by this scam. They include Hauser & Wirth, the London-based dealers Simon Lee, Thomas Dane, Rosenfeld Porcini and Laura Bartlett and, in the US, Tony Karman, the president of Expo Chicago.

“We know a number of galleries that have been affected. The sums lost by them or their clients range from £10,000 to £1m,” says the insurance broker Adam Prideaux of Hallett Independent. “I suspect the problem is a lot worse than we imagine.”

The fraud is relatively simple. Criminals hack into an art dealer’s email account and monitor incoming and outgoing correspondence. When the gallery sends a PDF invoice to a client via email following a sale, the conversation is hijacked. Posing as the gallery, hackers send a duplicate, fraudulent invoice from the same gallery email address, with an accompanying message instructing the client to disregard the first invoice and instead wire payment to the account listed in the fraudulent document.


Old scam – many lawyers and their clients have already suffered at this – and it’s all for big money.
link to this extract

Russian Twitter support for Trump began right after he started campaign • WSJ

Mark Maremont and Rob Barry:


Kremlin-backed support for Donald Trump’s candidacy over social media began much earlier than previously known, a new analysis of Twitter data shows.

Russian Twitter accounts posing as Americans began lavishing praise on Mr. Trump and attacking his rivals within weeks after he announced his bid for the presidency in June 2015, according to the analysis by The Wall Street Journal.

A US intelligence assessment released early this year concluded the Kremlin developed a “clear preference” for Mr. Trump over his Democratic opponent Hillary Clinton, but cited December 2015 as the earliest suspected time that Russian social-media accounts advocated for Mr. Trump.

The earlier starting point of pro-Trump tweets highlights the breadth of the Russian effort to manipulate social media during the 2016 election. Kremlin-paid actors sowed division among Americans with fake pages and accounts, inflammatory postings and thousands of paid ads aimed at both liberal and conservative audiences, according to testimony before Congress last week.

The Journal analyzed 159,000 deleted tweets from accounts that Twitter identified to congressional investigators as operated by the Kremlin-backed Internet Research Agency.


I get a feeling that the journalists on the WSJ are trying to send a not-so-subtle message to their editor about his support for Trump.
link to this extract

Apple at its best • Stratechery

Ben Thompson found himself delighted – that’s the word – with the iPhone X, and reflects on where Apple’s sustainable advantage exists in the smartphone world of hardware and services:


smartphones are increasingly replacing PCs, but even then most use is additive, not substitutive. In other words, there is no reason to expect that the arrival of artificial intelligence means that people will no longer care about what smartphone they use. Sure, the latter may “recede into the background” in the minds of pundits, but they will still be in consumers’ pockets for a long time to come.

There’s a second error, though, that flows from this presumption of zero-summedness: it ignores the near-term business imperatives of the various parties. Google is the best example: were the company to restrict its services to its own smartphone platform the company would be financially decimated. The most attractive customers to Google’s advertisers are on the iPhone — just look at how much Google is willing to pay to acquire them — and while Google could in theory convince them to switch by keeping its superior services exclusive, in reality such an approach is untenable. In other words, Google is heavily incentivized to preserve the iPhone as a competitive platform in terms of Google’s own services; granted, Android is still better in terms of easy access and defaults, but the advantage is far smaller than it could be.

Apple, meanwhile, is busy building competing services of its own, and while its easy — and correct — to argue that they aren’t really competitive with Google’s, that doesn’t really matter because competition isn’t happening in a vacuum. Rather, Apple not only enjoys the cost of switching advantage inherent to all incumbents, but also is, as the iPhone X shows, maintaining if not extending the user experience advantage that comes from its integrated model. That, by extension, means that Apple’s services need only be “good enough” — there’s that phrase! — to let the company’s other strengths shine.


Worth it for the GIF he includes of unlocking for notifications. For all the (plentiful) moaning about Apple, In the past year, both AirPods and the iPhone X have really delighted people who begin using them expecting “just another” product. The iPhone X, in particular, has had rave reviews from customers.

What’s the last product you used that utterly delighted you?
link to this extract

Facebook estimates 60 million users may be fake: report • TheHill

Josh Delk:


Facebook estimates that around 60 million, or 2%, of its monthly average users may be fake accounts, according to a report from the company.

Many of the false accounts are used for spam, Facebook CFO Dave Wehner said in a recent investors call.

Separately, Facebook estimates that around 10% of its accounts are “duplicate” accounts, meaning they are accounts run by a user separate from their main account. This would amount to more than 200 million accounts.

Facebook, Twitter and Google testified before the Senate Intelligence Committee in a three-day session this week, providing investigators information on the efforts of foreign actors to meddle in U.S. politics. 

One of the investigators’ concerns, according to The New York Times, is the widespread use of “fake” social media accounts.

Twitter also reports that nearly 5% of its user base, or more than 16 million accounts, are fake “spam” accounts, Sean Edgett, the social media giant’s acting general counsel, said in testimony.


link to this extract

Something is wrong on the internet • Medium

James Bridle on the weird subculture within YouTube’s “Kids” space of knockoff and randomly-generated videos aimed at children:


A step beyond the simply pirated Peppa Pig videos mentioned previously are the knock-offs. These too seem to teem with violence. In the official Peppa Pig videos, Peppa does indeed go to the dentist, and the episode in which she does so seems to be popular — although, confusingly, what appears to be the real episode is only available on an unofficial channel. In the official timeline, Peppa is appropriately reassured by a kindly dentist. In the version above, she is basically tortured, before turning into a series of Iron Man robots and performing the Learn Colours dance. A search for “peppa pig dentist” returns the above video on the front page, and it only gets worse from here.

Disturbing Peppa Pig videos, which tend towards extreme violence and fear, with Peppa eating her father or drinking bleach, are, it turns out very widespread. They make up an entire YouTube subculture. Many are obviously parodies, or even satires of themselves, in the pretty common style of the internet’s outrageous, deliberately offensive kind…

…Here are a few things which are disturbing me:

The first is the level of horror and violence on display. Some of the times it’s troll-y gross-out stuff; most of the time it seems deeper, and more unconscious than that. The internet has a way of amplifying and enabling many of our latent desires; in fact, it’s what it seems to do best. I spend a lot of time arguing for this tendency, with regards to human sexual freedom, individual identity, and other issues. Here, and overwhelmingly it sometimes feels, that tendency is itself a violent and destructive one.

The second is the levels of exploitation, not of children because they are children but of children because they are powerless. Automated reward systems like YouTube algorithms necessitate exploitation in the same way that capitalism necessitates exploitation, and if you’re someone who bristles at the second half of that equation then maybe this should be what convinces you of its truth. Exploitation is encoded into the systems we are building, making it harder to see, harder to think and explain, harder to counter and defend against. Not in a future of AI overlords and robots in the factories, but right here, now, on your screen, in your living room and in your pocket.

Many of these latest examples confound any attempt to argue that nobody is actually watching these videos, that these are all bots. There are humans in the loop here, even if only on the production side, and I’m pretty worried about them too.


Something is definitely wrong, and YouTube’s utter laissez-faire attitude is a giant part of the problem. By treating anyone under the age of 18 as essentially the same – the sort of decision that would only be made by someone without children or without morals – it is seeding a deeply weird future. And by chaining videos together – so convenient! Just flag the unsuitable ones, kids, while we show you ads! – it deepens the rabbit hole.
link to this extract

Users exploit Twitter rule, post epic 30,000 character tweet • Daily Dot

Phillip Tracy:


The tweet, written in German, starts by introducing the two users who discovered the trick, “People! @Timrasett and @HackneyYT can override the character limit! You don’t believe us? Here is the approximately 35K character proof,” it reads. The rest is complete gibberish—one string of random numbers and character too long to even be a German word.

Eloquent or not, the post shows that it’s possible to publish a single tweet with more than 280 characters. Note, the tweet is actually “only” 30,396 characters, not 35,000. One of the tweet’s authors apologized, claiming Twitter showed them a different number.

So how did they do it? By exploiting a rule Twitter made in 2016 that links would no longer count in the 140-character limit. Yes, this is just one big web address with a URL code hidden deep in the large block of text. You can find it by opening up the tweet and searching for “.cc/”


And because Twitter de-obfuscates the URLs when you look at them (even though it stores them in its format in its database), you see a gigantic tweet. Personally, I just blocked the tweeters. Life’s too short.
link to this extract

Two popular conservative Twitter personalities were just outed as Russian trolls • Philadelphia Inquirer

Rob Tornoe:


Jenna Abrams was a popular figure in right-wing social media circles. Boasting nearly 70,000 followers, Abrams was featured in numerous news articles during the 2016 election, spotlighted by outlets as varied as USA Today, the Washington Post, the BBC, and Yahoo! Sports. Her tweet about CNN airing porn during Anthony Bourdain’s show (it didn’t) was reported by numerous outlets.

But Abrams never existed.

According to information released by House Democrats earlier this week, Abrams was one of more than 2,750 fake Twitter accounts created by employees at the Internet Research Agency, a “troll farm” funded by the Russian government based in St. Petersburg. In addition to the Abrams account, several other popular conservative social media personalities — @LauraBaeley, SouthLoneStar, Ten_GOP — were all revealed to be troll accounts. All have been deactivated on Twitter.

According to the Daily Beast, the agency developed a following around the Abrams account by offering humorous, seemingly non-political takes on pop culture figures like Kim Kardashian. The agency also furnished the fake account, which dates back to 2014, with a personal website, a Gmail account and even a GoFundMe page.

Once the Abrams account began to develop a following, the tone of its tweets shifted from pokes and prods at celebrities to divisive views on hot topics like immigration and segregation.

“To those people, who hate the Confederate flag. Did you know that the flag and the war wasn’t about slavery, it was all about money,” the Abrams account wrote in April of 2016. The tweet quickly went viral, earning rebukes from historian Kevin Kruse and Al Letson, the host of the Center for Investigative Reporting’s popular Reveal podcast.

Those rebukes only allowed Abrams initial message to spread even wider, which was the ultimate intention of Russia’s propaganda campaign — to sow dissension and increase the racial divide among America’s voting populace, revealing the world’s only superpower as a country in decline.


The other troll was a pro-Trump account, indistinguishable in its excitement from real-life American idiots wrongly excited about Trump.
link to this extract

Errata, corrigenda and ai no corrida: none notified

Start Up: animoji karaoke!, Twitter’s Russian errors, cheap VR, fake WhatsApps, and more

Even including this screwup, nuclear power is the least lethal major energy source. Oh yes. Photo by CMdRCoRd on Flickr.

You can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

A selection of 11 links for you. Use them wisely. I’m @charlesarthur on Twitter. Observations and links welcome.

Twitter sidestepped Russian account warnings, former worker says • Bloomberg

Selina Wang:


In 2015, researchers from the University of California at Berkeley approached Twitter, asking for help, [Leslie] Miley said. They had found that Twitter had a significant amount of fake accounts, but wanted more data to further their research. Three employees on the product safety and security team, including Miley, met with them. They declined to give the academics data, but the meeting made them curious.

Afterward, the employees ran an analysis on Twitter’s accounts. Miley [then a manager on the accounts team, responsible for the infrastructure handling logins] said he was stunned to find that a significant percentage of the total accounts created on Twitter had Russian and Ukrainian IP addresses. According to Miley’s recollections, he brought the information to his manager, who told him to take the issue to the growth team. Miley said that he doesn’t have records of the tallies. 

“When I brought the information to my boss, the response was ‘stay in your lane. That’s not your role’,” Miley said.

Miley said he advised the growth team to delete most of the accounts they had surfaced from Russia and Ukraine, since the analysis suggested that most were inactive or fake. The growth team didn’t take any action on the Russian and Ukrainian accounts after he presented the data to them, according to Miley.

Many pro-Trump bots that were active during the 2016 U.S. elections were long-dormant accounts, according to researchers. These profiles give the illusion that they’re legitimate, and not created for the sole purpose of spreading propaganda during a campaign, according to Samuel Woolley, research director of the Digital Intelligence Lab at Institute for the Future, a non-profit research organization.


What a mess. We’ve always known that the only number that mattered to Twitter was the number of accounts, but this is terrible.
link to this extract

A fake news writer reveals how he’s making money on Facebook • Mic

Jake Horowitz and Kendall Ciesemier:


The source attributed a fake quote in the headline of a story to the celebrity and distributed that story on Facebook. Eventually, the story was taken down after it was flagged by Snopes, a third-party fact-checking website that Facebook has enlisted to help flag fake news for removal.

However, the source said he still was able to make $20,000 before the story was taken down.

“The site had already made its money, we had already made our money. We could probably do that a dozen more times before Google and Facebook both would be like, ‘We’re now going to blacklist this website,’” the source said.

In a written statement to Mic, a Facebook spokesperson explained, “Our fact-checking partners diligently review items that are not apparent satire and are focused on the worst of the worst: hoaxes intended specifically to deceive.”

However, the spokesperson also admitted that it commonly takes Facebook more than three days before it is able to remove fake news stories.

“While we know that most of the impressions typically happen in the first day, and that we have missed many, we are getting better,” the spokesperson said.


Not really going to get ahead of this, are they.
link to this extract

How the web became unreadable • WIRED

Kevin Marks:


Typography may not seem like a crucial design element, but it is. One of the reasons the web has become the default way that we access information is that it makes that information broadly available to everyone. “The power of the Web is in its universality,” wrote Tim Berners-Lee, director of the World Wide Web consortium. “Access by everyone regardless of disability is an essential aspect.”

But if the web is relayed through text that’s difficult to read, it curtails that open access by excluding large swaths of people, such as the elderly, the visually impaired, or those retrieving websites through low-quality screens. And, as we rely on computers not only to retrieve information but also to access and build services that are crucial to our lives, making sure that everyone can see what’s happening becomes increasingly important.

We should be able to build a baseline structure of text in a way that works for most users, regardless of their eyesight. So, as a physicist by training, I started looking for something measurable…


He found it in contrast ratio (between type and background.)


For example: Apple’s typography guidelines suggest that developers aim for a 7:1 contrast ratio. But what ratio, you might ask, is the text used to state the guideline? It’s 5.5:1.

Google’s guidelines suggest an identical preferred ratio of 7:1. But then they recommend 54% opacity for display and caption type, a style guideline that translates to a ratio of 4.6:1.
The typography choices of companies like Apple and Google set the default design of the web. And these two drivers of design are already dancing on the boundaries of legibility.


link to this extract

How to make Animoji Karaoke with iPhone X • iMore

Rene Ritchie:


How do you make awesome Animoji Karaoke to share with your friends and all your socials? With these simple steps!

Animoji Karaoke is really more like Animoji or dubsmash or whatever it is we’re calling lipsync these days. You play some music, you move your mouth, and your iPhone X turns it into an animated emoji singing a song.

Let’s break it down.


It’s probably only going to be a thing for a week, if that – to be replaced by people lipsyncing Trump speeches, film extracts, books and so on – but it’s fun while it lasts. Here’s Bohemian Rhapsody:

(I think the trick is to move your mouth a lot.)

Has there ever been a phone which has created a genre like this so quickly? The thing’s been on sale since Friday.
link to this extract

Examining the malaise of bargain basement virtual reality • Anandtech

Ian Cutress went for a wander around IFA (back in September, but this is still valid), to look for the crappy VR headsets, to see how bad “bad” might be and still be on sale:


So here’s the thing: the Skyworth headset is essentially a smartphone under the hood that you can’t take out. So what makes it better than a Samsung VR headset where the smartphone can be attached / detatched? One would assume it’s a price thing, and we were told the price for the headset could be $400 to $800. I remarked that it was a pretty large range, but was told that the Skyworth headset is still a work in progress, with exact specifications to be decided later. If it was $400, it might be more palatable, but for $800 then it might be easier to go the smartphone route.

This brings us around to the problem with VR right now. Everyone wants in on the bandwagon, and in a keynote at the event it was pointed out that in order for triple-A style games and film studios to start making content in these new formats, there needs to be more potential sales out there. Current estimates put 500k headsets in the market (of varying degrees of power) with another 2.7 million by the end of 2017. No game studio or film studio, working on the next FPS or Avatar, will make a massive piece of content for only 3.2 million people – it needs to be in the hands of tens of millions to even start to make sense, and we won’t be at that point for a number of years.

All that being said, you have a choice – investing in a premium VR headset to be able to experience the best will cost $700+, in terms of the headset itself plus any extra hardware you need to power it. The easiest way to enter the VR space with some clout is the smartphone or all-in-one route, but that is still a hefty cost. Then there is a large, long gap to the segment of very basic all-in-one virtual reality headsets as shown at IFA this year.

For $100, or the rough price of the Samsung headset without any internal hardware, you get a basic quad-core Rockchip design with limited functionality. I’m half inclined to suggest that a bucket be provided as well, just in case nausea takes over. But it shows what a state VR is in, when the hardware is still so expensive. In order to get a base experience that can truly be called VR, such as with the Skyworth headset, it might be as much as a high-end smartphone anyway. For mass market adoption, the cost to enter has to be low, but not so low we’re scraping the barrel for basic frame rates.


I can’t see how VR gets to there from here.
link to this extract

The mystery of Apple’s one-time Services boost • Yahoo Finance

Evan Niu on the mysterious extra payment which appeared in Apple’s Services segment in the past quarter, ascribed to “a favorable one-time adjustment of $640 million due to a change in estimate based on the availability of additional supporting information”. Whaaat? It’s probably traffic acquisition payments from Google for being the default search on Safari and Siri on iOS:


[on the desktop] the rising popularity of Google’s Chrome browser over the past decade as it overtook Firefox directly undermined the need to occupy Firefox’s default search spot (which was long Mozilla’s primary revenue source); Yahoo! scored the default search spot in Firefox back in 2014. In other words, Chrome cannot displace Safari on iOS in the same way that it displaced Firefox on desktop [because you can’t change the default browser on iOS]. It’s also worth pointing out that Apple just switched Siri search from Microsoft Bing to Google too, which sounds an awful lot like a change in “partner agreements.”

This all comes just months after Bernstein analyst Toni Sacconaghi estimated that Google could end up paying Apple approximately $3bn this year in TAC, which gets booked into Apple’s services business. This revenue is “nearly all profit,” since Apple incurs virtually no cost in sending traffic to the search giant, which helps boost Apple’s overall gross margin. It could offset some other margin headwinds that Apple is currently facing, like the current memory pricing environment. Apple’s gross margin last quarter came in at 37.9%, near the high end of guidance.

Investors don’t have confirmation, but all signs point to that $640m adjustment coming from Google.


That’s a lot of money just for the Siri switch. And Apple now has millions of reasons not to allow people to switch default apps on iOS.
link to this extract

Smartphone shipments set third quarter record clocking 400m units, growing 5% annually • Counterpoint Research

Shobhit Srivastava:


According to the latest research from Counterpoint’s Market Monitor service, global smartphone shipments grew 5% YoY in Q3 2017.  Top 10 players now capture 75% of the market thereby leaving just a quarter of the market for the remaining 600+ brands to compete.

Commenting on the findings, Jeff Fieldhack, Research Director at Counterpoint Research said, “The global smartphone market continues to grow in single digits driven by growth in emerging markets. In such a scenario, we have seen key hardware differentiators proliferate to lower price points at much faster rates. For example, alternative aspect ratio 18:9 devices already penetrated sub $150 segment within two quarters of launch in the premium segment. This indicates how cut-throat the competition is within the industry.  Brands are striving for differentiation across price bands. In addition, the increasing share of leading brands is putting additional pressure on smaller brands which can lead to consolidation in some of the OEM-crowded regions going forward.”


That “quarter of the market” still amounts to 100m units for those 600 brands – an average of 167,000 each. There must be some tiny players out there, given that Sony, HTC, Google and even Nokia make up a few million each – reducing that to 90m units for 590-odd brands. Doesn’t change the maths much; a mean of about 152,000 each.

What’s really going to hurt them is the rise in the price of RAM. Big suppliers will be able to get lower prices; small ones will lose their price competitiveness, and likely their business.
link to this extract

In just three years Xiaomi has nearly taken over India’s booming smartphone market •

Sushma UN:


In the last year, Xiaomi’s market share in India has gone from just 6% to 22%, according to Hong Kong-headquartered market intelligence firm Counterpoint Research. That puts its share at par with South Korea’s Samsung, a leader for several years now. And as of September, three of the five most popular smartphones in India are from Xiaomi, Counterpoint said in a report on October 27.

It’s quite a coup considering Xiaomi entered India only in 2014 – Samsung has been around since 1995. The Beijing-based company has invested around $500 million in the Indian subcontinent in the last two years according to Bloomberg and intends to invest a similar amount between the next three and five years.

This stupendous success, analysts reason, is because of a strong supply chain and the company’s ability to sell value for money products in a very price sensitive market.

Since its entry into India, Xiaomi has stood out for its unique go-to-market strategy of selling only via e-commerce. It signed an exclusive partnership with e-tailer Flipkart and ran flash sales for new model launches, with the sales typically ending within seconds of opening. For instance, in a flash sale for the Redmi 1S model in September 2014, around 40,000 pieces were sold out in just 4.2 seconds.

This allowed the company to single-mindedly build capabilities around online retail, which now accounts for around 30% of India’s total smartphone sales. Most other brands have struggled with juggling online and offline sales, with many often failing to satisfy either set of customers.


Notice that Xiaomi is trying to do offline retail in China – which is comparatively expensive. India is a particular market though: very aware of technical specifications and “value for money”.
link to this extract

Over a million Android users fooled by fake WhatsApp app in official Google Play Store • The Register

Iain Thomson:


Once again Google’s Play Store has proved less than excellent at tackling malicious apps, after netizens found a fake version of WhatsApp that was good enough to fool over a million people into downloading it.

The rogue program was spotted by Redditors earlier today, and the software looks very much like the real deal. However, when opened, it appears to download and run the real WhatsApp Android client albeit with adverts wrapped around it, making a fast buck for whichever miscreant produced this dodgy imitation.

Fake on the left, legit on the right

“I’ve also installed the app and decompiled it,” reported DexterGenius.

“The app itself has minimal permissions (internet access) but it’s basically an ad-loaded wrapper which has some code to download a second apk, also called ‘whatsapp.apk.’ The app also tries to hide itself by not having a title and having a blank icon.”

The fake app, now removed from the official Play Store, appeared to be developed by WhatsApp Inc, the legit Facebook-owned maker of the messaging client. However, thanks to some Unicode trickery, a hidden space at end allowed this dodgy version to masquerade as a product of WhatsApp Inc, albeit with two bytes, 0xC2 0xA0, at the end forming an invisible space. In other words, it appeared to be a legit app from a real developer, but really it wasn’t.


The Play Store’s openness doesn’t work in anyone’s favour here. Manual checks on iOS apps bug developers. But is there an equivalent of this on the App Store?
link to this extract

Contrarily: out of all major energy sources, nuclear is the safest • Our World in Data

Hannah Ritchie:


Discussions with regards to energy safety often incite the question of: how many died from the nuclear incidents at Chernobyl and Fukushima? We addressed this question in a separate blog post. In summary: estimates vary but the death toll from Chernobyl is likely to be of the order of tens of thousands. For Fukushima, the majority of deaths are expected to be related to induced stress from the evacuation process (standing at 1600 deaths) rather than from direct radiation exposure.

As stand-alone events these impacts are large. However, even as isolated, large-impact events, the death toll stands at several orders of magnitude lower than deaths attributed to air pollution from other traditional energy sources—the World Health Organization estimates that 3 million die every year from ambient air pollution, and 4.3 million from indoor air pollution.15 As so often is the case, single events that make headlines overshadow permanent risks that result in silent tragedies.

Based on historical and current figures of deaths related to energy production, nuclear appears to have caused by far the least harm of the current major energy sources. This empirical reality is largely at odds with public perceptions, where public support for nuclear energy is often low as a result of safety concerns. This is shown in the chart below which measures the share of survey respondents in a given country who are opposed to nuclear energy as a means of electricity production. At a global level, opposition to nuclear energy stood at 62% in 2011.


As Lewis Wolpert was fond of saying, common sense isn’t, and science tends to reveal things which aren’t common sense. (“Common sense” suggests the sun revolves around the earth, for instance.)
link to this extract

Tablet market declines 5.4% in third quarter despite 4 of top 5 vendors showing positive year-over-year growth • IDC


The third quarter of 2017 (3Q17) closed with 40m tablets shipped globally, according to preliminary data from the International Data Corporation (IDC) Worldwide Quarterly Tablet Tracker. Growing demand for smartphones combined with the lengthening replacement cycle of tablets and strengthening position of traditional PCs left the tablet market in an awkward middle ground that it has not been able to escape. Growth in 3Q17 declined 5.4% year over year, marking the twelfth consecutive quarter of annual decline.

“There’s a penchant for low-cost slates and this holds true even for premium vendors like Apple,” said Jitesh Ubrani, senior research analyst with IDC’s Worldwide Quarterly Mobile Device Trackers. “However, many of these low-cost slates are simply long-awaited replacements for consumers as first-time buyers are becoming harder to find and the overall installed base for these devices declines further in the coming years.”

Meanwhile, growth in the detachable tablet market has been slower than expected as Apple and Microsoft are essentially the only two vendors supplying the category and other PC vendors champion the convertible PC form factor.

“In a recent IDC survey, owners of both convertibles and detachables stated they were far more inclined to recommend a convertible to another shopper than a detachable,” said Linn Huang, research director, Devices & Displays. “Market momentum has steadily shifted away from the latter towards the former over the course of this year. The 2017 holiday season may prove to be a critical crossroad for the detachables category.”


Apple’s iPad Pro and the Microsoft Surface (and some Samsung Tabs) are the only serious players in the “detachable” category; IDC doesn’t include “convertibles” here (which are PCs which have a twistable screen so they can be tablet-like).

More to the point: outside Apple, which has grown for the past two quarters, and up 10% in this quarter, the tablet market is falling away – down by 10% in this quarter. Cheap Chinese OEMs are quitting the market, which is likely saturated; Samsung does lots of “get a tablet with our phone” offers; Amazon sells them really cheap; and it’s hard to see Huawei and Lenovo making a handsome profit on them.
link to this extract

Errata, corrigenda and ai no corrida: none notified

Start Up: Lenovo’s PC problem, Bin Laden’s computer, Russian DNC hackers tracked, and more

This one wasn’t, but some would-be novels have had the first lines written by an AI. The results are.. different. Photo by sights set on Flickr.

A selection of 9 links for you. Also: Friday (day may vary according to location). I’m @charlesarthur on Twitter. Observations and links welcome.

Software matters in the world – Fog Creek Software • Medium

Anil Dash:


When our company, Fog Creek Software, was started as a little indie firm way back in 2000, we mostly saw bugs that way, too. We made a bug-tracking app and tried to help people make sure they were fixing what was wrong in their software.

While that was happening, our cofounder Joel Spolsky also wrote a lot about the culture of making software. Back then, at the height of the dot-com boom, it was seen as a bit eccentric to put as much focus on the human factors and ethical behavior as our founders did. But it helped us win fans, and some of those people tried out the various apps we built along the years, and we’ve been lucky to keep thriving as what feels like one of the last few independent tech companies that’s still relevant.

But we missed something important, too. Those ideas and insights about how to treat people, how to listen to customers (and to communities), and how to be thoughtful and responsible in creating technology were even more important than anything we built into our software. They were the first steps to trying to fix what we could now think of as “Big Bugs”. Little bugs were mistakes in the software. Big Bugs are when we exacerbate (or cause!) major problems in society.


link to this extract

Stop being so PC, Lenovo • Bloomberg Gadfly

Tim Culpan is pretty salty about Lenovo’s results and acquisition of 51% in Fujitsu’s PC business:


Apparently that loss on previous acquisitions wasn’t a lesson for the board, because they’ve just doubled down on PCs to the tune of at least 20.4bn yen ($179m), and as much as 30.7bn yen. The deal sees them join with with Fujitsu and Development Bank of Japan Inc., which will hold a 5% stake.

The joint venture will focus on the research, development, design, manufacturing and sales of client computing devices for the global PC market.

Spare me (and your shareholders)! Lenovo investing even more money in an anemic business is folly, and dressing it up as R&D looks like it might be intended to fool us.

I get why they want to go deeper into client computing: It’s the only division that’s capable of showing consistent growth and profitability. But we all know that this is an unhealthy addiction, because in the long term PCs are a dying business. Lenovo may be staying there because of old habits, or perhaps is driven by a need to report profits to shareholders every quarter; Lenovo’s client computing division remains the only unit capable of delivering profits.

This won’t be an easy addiction to kick. At least one rival, Dell Inc., went cold turkey and is trying to wean itself off the quarterly treadmill. Others have pivoted away from client devices.

A decade from now, Lenovo won’t be predominantly a PC company, because it will have shifted focus, or succumbed. It should start that process now, while it can.


It has lost money in smartphone for 15 straight quarters, and I can’t find evidence of its tablets making money. PCs are all it has.
link to this extract

15 things we learned from the tech giants at the Senate hearings • The Atlantic

Alexis Madrigal:


During three Congressional hearings spread over two days, we heard a lot of bluster from senators and pat answers from tech-company lawyers about the role their firms played in the 2016 election.

Scattered among all the questions, some new facts entered the public record. Here we attempt to catalog the important new information we learned. Some of the biggest disclosures came in the prepared testimony from Facebook, Twitter, and Google, as well as in the introduction from the ranking members of the Senate Intelligence Committee, Senator Richard Burr of North Carolina and Senator Mark Warner of Virginia.

1. Russian electoral disinformation reached 126 million people on Facebook and 20 million on Instagram. That’s 146 million total.

These topline numbers keep going up, and we hadn’t known that the influence campaign extended to Instagram. This information seems to have only reached the Senate committee in the last couple of days.

2. Most Russian advertising on Facebook was used to build up pages, which then distributed their content “organically.”

The $100,000 of advertising that has been a big focus of Congressional interest was used primarily to build audiences for a variety of Russian-linked pages. In other words, they paid to buy likes and build the distribution channels through which they would pump disinformation.


And plenty more like it.
link to this extract

Osama bin Laden’s compound computers held crochet lessons, viral YouTube videos, and sexy video games • The Verge

Adi Robertson:


The CIA-hosted archive includes hundreds of gigabytes’ worth of files, but its title indexes — for audio, documents, video, and images — are a lot more manageable. Agency director Mike Pompeo, who authorized their release, says the collection “provides the opportunity for the American people to gain further insights into the plans and workings” of al-Qaeda.

In addition to a mass of basic operating system elements and clearly terrorism-related material, they reveal some odd details about compound residents’ media diets. There are a few big-name films like Antz, Cars, and Resident Evil, which the CIA has withheld (alongside less prominent copyrighted videos) in case someone was planning to download a 174GB file to fish around for pirated media.

But beyond that, you can also find listings for a downloaded copy of the super-popular YouTube video “Charlie Bit My Finger;” as well as a video file called “Loosechange2” — likely a copy of the second edition of Loose Change, which argues that the September 11th attacks were masterminded by the American government, not bin Laden. You can even find a wealth of videos on crocheting baskets, baby socks, and beanie caps, among other things.


I guess that kills the stories about them not having really kill OBL. Or, um, maybe it doesn’t. Also includes Illuminati conspiracy theories. Guess it got boring there in Abbotabad. A marvellous resource for academics.
link to this extract

Prosecutors consider bringing charges in DNC hacking case • WSJ

Aruna Viswanatha and Del Quentin Wilber:


The Justice Department has identified more than six members of the Russian government involved in hacking the Democratic National Committee’s computers and swiping sensitive information that became public during the 2016 presidential election, according to people familiar with the investigation.

Prosecutors and agents have assembled evidence to charge the Russian officials and could bring a case next year, these people said. Discussions about the case are in the early stages, they said.

If filed, the case would provide the clearest picture yet of the actors behind the DNC intrusion. US intelligence agencies have attributed the attack to Russian intelligence services, but haven’t provided detailed information about how they concluded those services were responsible, or any details about the individuals allegedly involved.

The high-profile hack of the DNC’s computers played a central role in the US intelligence community’s assessment in January that “Russian President Vladimir Putin ordered an influence campaign in 2016 aimed at the US presidential election.” Mr. Putin and the Russian government have denied meddling in the US election.


There’s a sort of quiet war going on between the WSJ’s reporters and its editor. This story won’t get much – if any – coverage on Fox News, which has consistently run with a wild story about an insider hack. All the evidence anyone can gather suggests it wasn’t.
link to this extract

Russia hackers had targets worldwide, beyond US election • Associated Press

Raphael Satter, Jeff Donn and Justin Myers:


The hackers who disrupted the U.S. presidential election had ambitions well beyond Hillary Clinton’s campaign, targeting the emails of Ukrainian officers, Russian opposition figures, US defense contractors and thousands of others of interest to the Kremlin, according to a previously unpublished digital hit list obtained by The Associated Press.

The list provides the most detailed forensic evidence yet of the close alignment between the hackers and the Russian government, exposing an operation that stretched back years and tried to break into the inboxes of 4,700 GMail users across the globe — from the pope’s representative in Kiev to the punk band Pussy Riot in Moscow.

“It’s a wish list of who you’d want to target to further Russian interests,” said Keir Giles, director of the Conflict Studies Research Center in Cambridge, England, and one of five outside experts who reviewed the AP’s findings. He said the data was “a master list of individuals whom Russia would like to spy on, embarrass, discredit or silence.”

…In the United States, which was Russia’s Cold War rival, Fancy Bear tried to pry open at least 573 inboxes belonging to those in the top echelons of the country’s diplomatic and security services: then-Secretary of State John Kerry, former Secretary of State Colin Powell, then-NATO Supreme Commander, U.S. Air Force Gen. Philip Breedlove, and one of his predecessors, U.S. Army Gen. Wesley Clark.

The list skewed toward workers for defense contractors such as Boeing, Raytheon and Lockheed Martin or senior intelligence figures, prominent Russia watchers and — especially — Democrats.


List gathered by the security company Secureworks, covering the March 2015 – May 2016 period.
link to this extract

How the National Enquirer saw Donald Trump would get elected – Popbitch

A magnum opus on how the National Enquirer rose on the back of Mafia money (and Trump’s friend Roy Cohn) and then moved to Florida, and was doing down Trump’s opponents during the primaries in 2016 with exactly the electorate who would decide the election:


We started this whole story by pointing out that the non-profit organisation Leaders In Further Education (LIFE) had cancelled their proposed annual gala at Mar-A-Lago, and suggested that that could prove to be disastrous for Donald Trump’s presidency. Worse than his dismal approval ratings. Worse than the Russia investigation. Worse than any of the day-to-day setbacks and scandals he seems to be constantly embroiled with.

Why? Because the CEO and Chairwoman of LIFE is Lois Pope.

Maybe you remember her. We briefly alluded to her in Part Three. She’s the widow and executor of Generoso Pope Junior’s estate: the woman who sold the National Enquirer to form American Media, Inc. Now, to be crystal clear, it’s important to note that Lois Pope has never had an editorial position on the Enquirer, nor has she had any say in its corporate management since it was sold in 1988. She is in no way one of the Machiavellian puppetmasters of this operation.

And that’s exactly what makes her so useful.

Lois Pope is about the best canary we could possibly hope for in this coalmine. A Trump loyalist. A Florida resident. A mover in Tabloid Triangle society, married to the man who effectively created the industry. She lived and breathed this entire world every day for 35 years. She is about as tuned in to this whole scene as any person could possibly be – without actually being an integral part of it.

She is also a woman who has hosted more than twenty galas at Mar-A-Lago. At least two of them since Trump first said that Mexicans are rapists and murders (and some, we assume, are good people).

If even someone like Lois Pope has apparently given up hope with him, then it could be catastrophically bad news for Trump. As a barometer for what the Enquiring Mind is thinking, there are worse people to take under consideration than Gene Pope’s own wife.


A long read for you, but this is the kicker.
link to this extract

Mexico resort blackouts: TripAdvisor blocked warnings, tourists say • Journal Sentinel

Raquel Rutledge and Andrew Mollica:


Since July, when the Journal Sentinel began investigating the mysterious death of a Wisconsin college student in Mexico — and found widespread problems with tainted alcohol, derelict law enforcement and price gouging from hospitals — more than a dozen travelers from across the country have said TripAdvisor muzzled their first-hand stories of blackouts, rapes and other ways they were injured while vacationing in Mexico.

“To me it’s like censoring,” said Wendy Avery-Swanson of Phoenix, whose recent review of a Mexican resort — describing how she blacked out from a small amount of alcohol served at the swim-up bar — was removed from the website. 

“It wasn’t hearsay,” as TripAdvisor claimed, said Avery-Swanson, 52. “It actually happened to me.”

Massachusetts-based TripAdvisor touts its more than 535 million user reviews of hotels, restaurants and attractions around the globe. And company officials say it uses finely tuned software to detect fake reviews and has hundreds of employees dedicated to policing posts and ensuring “content integrity.”

 A Journal Sentinel investigation into the workings of the $1.5 billion company has found that it is what TripAdvisor does not publish that poses real problems for travelers.
The company’s policies and practices obscure the public’s ability to fully evaluate the information on its site. Secret algorithms determine which hotels and resorts appear when consumers search. Some hotels pay TripAdvisor when travelers click on their links; some pay commissions when tourists book or travel. 

An untold number of TripAdvisor users have been granted special privileges, including the ability to delete forum posts. But the company won’t disclose how those users are selected.

There’s no way to know how many negative reviews are withheld by TripAdvisor; how many true, terrifying experiences never get told; or for site users to know that much of what they see has been specifically selected and crafted to encourage them to spend.


The flip side of sites which let anyone post anything, and are thus open to spam.
link to this extract

A neural network tries writing the first sentence… • AI weirdness

Janelle Shane:


I decided to give a neural network examples of first sentences of novels, to see if it could generate some that might help writers get started. The main problem turned out to be finding enough examples of first sentences – ideally, I need thousands. I could only find a couple hundred of the most famous lines, and the neural network proceeded to do what it usually does when faced with too little data, which is to give up on trying to understand what’s going on, and instead just try to read it back to me word for word. Think of it like cramming for a test by memorizing instead of learning how to apply rules to solve problems.

Most didn’t make much sense, and/or were obvious mishmashes of famous lines. A few turned out to be maybe usable, probably by accident:


There was a man and he had seventy first sight.

It is a truth universally acknowledged, that a single man in possession of a good fortune must be in want of my life, fire of my loins.

4 Had come to America from Europe Privet Drive.

The snow is gone sometime, and you said, Why, and I said, To be with the darkness.

It was like the imagination.

It was a wrong number that struggled against the darkness.

It was a dark and stormy night; the swall of the gods?

The moon turned out to see me.

It was a wrong number four Privet Drive.

That’s good thinking: a bowl of the carriage’s parts.

The sky above the present century had reached the snapping point.



To some extent I think neural networks are inscrutable. Trying to understand why they do this stuff makes on feel like a dog trying to understand physics.
link to this extract

Before next week, you can sign up to receive each day’s Start Up post by email. You’ll need to click a confirmation link, so no spam.

Errata, corrigenda and ai no corrida: it has been pointed out that the photo captioned as a turtle yesterday was of a tortoise. We have fired the AI which chooses our pictures.