Start Up: ask a chatbot!, Alexa and manners, Brexit’s hedge fund profiteers, death of the phone, and more

Intel dominated the chip industry for years. What went wrong? Photo by Mark Sze 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.

Because WordPress sometimes decides to ignore the “category” tag for blogposts when I upload it, and because the correct category is essential for posts to go out by email, and because I expect computers not to change these things on their own and so didn’t check it, Monday’s post didn’t go out by email. You can find it here.

A selection of 10 links for you. Use them wisely. I’m @charlesarthur on Twitter. Observations and links welcome.

The laws behind ‘Fortnite’s’ PS4, Nintendo Switch woes • Variety

Trevor Ruben:


When Epic and Nintendo dropped Fortnite on the day of Nintendo’s E3 showcase, fulfilling the expectations of many and finally delivering unto the Nintendo Switch the world’s most popular game, a significant number of players were met with a second surprise. They encountered an unexpected roadblock when trying to load the in-game purchases they had made on another platform: “This Fortnite account is associated with a platform which does not allow it to operate on Switch.”

That platform was the PS4, and had that player loaded their account for even one second previously on a PS4, whatever purchases they made on that account and whatever progress they had achieved were now locked to that console, with no recourse. PS4 players cannot transfer their account to another video game console, nor completely disassociate their account with the PS4.

They were, and still are, stuck. One might ask if this kind of restriction is legal, considering the vast and possibly embarrassing amount of money some might have spent on the game. The answer to that is there is no answer. Our current laws simply fail to acknowledge the problem. In fact, our legal system exacerbates it by placing the rule-making in the hands of the profiteers…

…Sony’s only statement on the matter thus far, delivered to the BBC, is a thinly-veiled boast and something only a market-leader would feel it could get away with:

“With… more than 80 million monthly active users on PlayStation Network, we’ve built a huge community of gamers who can play together on ‘Fortnite’ and all online titles.”

“We also offer ‘Fortnite’ cross-play support with PC, Mac, iOS, and Android devices, expanding the opportunity for Fortnite fans on PS4 to play with even more gamers on other platforms,” adding, “we have nothing further to add beyond this at this point.”


Sony might think it’s being clever; but non-PS4 players look down on Sony players as “no-skins” who can’t communicate when you’re playing squad mode. (Ask a child.) This isn’t helping Sony at all.
link to this extract

Apple to unveil high-end AirPods, over-ear headphones for 2019 • Bloomberg

Mark Gurman and Debby Wu:


The Cupertino, California-based company is working on new AirPods with noise-cancellation and water resistance, the people said. Apple is trying to increase the range that AirPods can work away from an iPhone or iPad, one of the people said. You won’t be swimming in them though: the water resistance is mainly to protect against rain and perspiration, the people said.

Slated for 2019, the earbuds will likely cost more than the existing $159 pair, and that could push Apple to segment the product line like it does with iPhones, one of the people said. Apple is also working on a wireless charging case that’s compatible with the upcoming AirPower charger.

The company has also internally discussed adding biometric sensors to future AirPods, like a heart-rate monitor, to expand its health-related hardware offerings beyond the Apple Watch, another person said. The current AirPods will be refreshed later this year with a new chip and support for hands-free Siri activation, Bloomberg News reported.


Noise cancellation is good, water resistance useful. But we’ve known about the wireless case compatible with AirPower since Apple showed the promo video for AirPower in, er, September. Meanwhile…
link to this extract

Apple admits its computers are broken • The Outline

Casey Johnston:


While the the repair and replacement program covers costs and notes that Apple will repair both single keys as well as whole keyboards when necessary, it doesn’t note whether the replacements will be a different, improved design that will prevent the problem from happening again (and again, and again). Having become a one-woman clearinghouse for people complaining about these keyboards since I broke this story, I feel justified in saying that keyboard failures – dead keys, sticking keys, double-spacing spacebars – appear to happen early and often, and repairs do not permanently fix the issue. I also feel justified in saying that the design on offer as recently as February still presented the exact same issues as the design I purchased in the fall of 2016.

In the same vein, it is worth noting that, prior to the announcement of the program, repairs involved almost exclusively swapping out the entire top case of the keyboard. This process required shipping the computer out to one of Apple’s remote service centers, and then shipping it back either to the customer or the Apple store, a total turnaround time of about five days if the computer was brought directly to an Apple store in the first place.

Apple did not immediately return a request from this reporter for comments on whether repairs may now be done on site at stores to shorten the time customers must be without their computers; whether the keyboard design has changed such that a repair may eliminate the problem rather than prop up a faulty design; or whether Apple anticipates releasing updated hardware that is not so prone to failure at any point in the future. Perhaps their keyboards, too, are broken.


In similar vein: The Register’s take on the same topic, by the owner of one of those keyboards. My present laptop dates from 2012. Its keyboard is fine by me – all the mistakes are mine.
link to this extract

The Visual Chatbot • Letting neural networks be weird

Janelle Shane:


There is a delightful algorithm called Visual Chatbot that will answer questions about any image it sees. It’s a demo by a team of academic researchers that goes along with a recent machine learning research paper (and a challenge for anyone who’d like to improve on it), and its performance is pretty state-of-the-art, meant to demonstrate image recognition, language comprehension, and spatial awareness.

However, there are a couple of interesting things to note about this algorithm.

• It was trained on a large but very specific set of images.
• It is not prepared for images that aren’t like the images it saw in training.
• When confused, it tends not to admit it.

Now, Visual Chatbot was indeed trained on a huge variety of images. It can answer fairly involved questions about a lot of different things, and that’s impressive. The problem is that humans are very weird, and there are still many things it’s never seen. (This turns out to be a major challenge for self-driving cars.) And given Visual Chatbot’s tendency to react to confusion by digging itself a deeper hole, this can lead to some pretty surreal interactions.


So Ms Shane unleashes a few pictures on Visual Chatbot. Oh, you know – Darth Vader v Obi-wan Kenobi (“a person standing in a doorway of a train station”) – and takes it from there. Hugely entertaining. And you can play too!
link to this extract

Consent, data-driven inequities, and the risks of sharing administrative data • Powered by Data

Lorraine Chuen:


[Welfare rights organizer, and author of the 2018 book “Automating Inequality” Virginia] Eubanks highlights how administrative data-sharing has already facilitated a new form of “automated inequality”. She points to the Allegheny Family Screening Tool (AFST) as a case study in how data-driven tools can further profile poor communities and communities of colour.  The AFST is a tool meant to help child welfare staff identify and prioritize the most “at risk” children in Allegheny County, Pennsylvania. The tool links data between twenty-nine different administrative data sources from the county’s Department of Human Services (DHS), including data on whether families have accessed or interacted with mental health services, child protective services, correction systems, drug/alcohol services, and more. This linked administrative data is fed into an algorithm used to flag which cases need intervention from General Protective Services—which often looks like separating a child from their family.

Unfortunately, many of the variables used to predict abuse in the model are simply measures for poverty (e.g. use of the SNAP nutrition assistance program), or reflections of systems that disproportionately affect poor & racialized communities (e.g. juvenile probation). The DHS also holds less data on affluent families—who are afforded more privacy simply by accessing mental health and drug treatment programs that are private, rather than public. Eubanks also points out the frustrating and heartbreaking paradox of parents being seen as greater risks to their children through the algorithm when they access public services to try and improve their situation.


Yes: the algorithms think that trying to help your children means you’re trying to harm your children.
link to this extract

End of the line: our guide to the death of the telephone • The Guardian

Rhik Samadder:


I remember an hour-long argument with the phone company when I moved into my flat, on being informed that I had to have a landline if I wanted the internet. I felt like a bald man being sold pomade. The company advised me I could call friends and family cheaply if I signed up to special packages. I can call them for free using the internet, I replied. Besides, I’m not calling any friends or family, unless I need a kidney or a place to stay.

The landline’s primary use is on TV, as a signifier you’re watching a period drama, ie anything set in 1995. They were so inefficient they bordered on surreal. Upon picking up, you never knew who would be on the other end: the National Lottery or Beryl from down the road or his Holiness the Pope. Weirdly, you were expected to identify yourself, though they had called you. That’s because manual dialling led to a lot of miscalls. Large portions of the day were spent convincing strangers that you weren’t Darren, didn’t know Darren, and were sure he was sorry for what he’d done. Relics of a time when we remembered phone numbers, a disproportionate number of calls to landlines are probably from people needing bail. The only funny way to use a landline in recent years was to send a text message to one, and have a robotic voice read it aloud in a way that was guaranteed to unsettle your mother.

Landlines are solely for older relatives who haven’t got to grips with mobiles. Having said that, it’s possible they’ll make a comeback, in the same way the streaming age saw the resurgence of record players and vinyl. Imagine dialling a friend on a rotary phone, which takes about 20 minutes if you don’t make any mistakes. Imagine taking time out of your overstimulated, hectic day to do that. Quite nice, no? With that curly, twirlable wire tethering you to one spot, and their lack of screen, the practical limitations of the landline could see it become a mindfulness tool, encouraging us to sit and you know, really talk. Could – but almost certainly won’t. These days landlines are cordless, and come with Caller ID, and are really just mobile phones that never finished their degree.


A hilarious, but also true, piece which also looks at video calls, voicemail and more.
link to this extract

The case against teaching kids to be polite to Alexa • Fast Company

Mike Elgan:


Today’s toddlers are the first generation to grow up without any memory of the world before ubiquitous artificial intelligence devices in homes. Parents are justifiably concerned about how these gadgets affect their children. One concern is manners. According to the UK research organization Childwise, children almost never say “please” or “thank you” to virtual assistant appliances (unlike adults, who often do).

Parents aren’t happy. But at least two companies are trying to help: Amazon and Google.

In April, Amazon introduced a politeness feature for its Alexa virtual assistant, along with a colorful line of Echo Dot devices just for kids. The manners feature, called Magic Word, is part of FreeTime, a wider range of child-specific features and content. It’s designed to encourage children to say “please” and “thank you” when speaking to Alexa assistant. After consulting outside child development experts, Amazon decided on positive reinforcement, with no “penalty” when a child is rude. For example, when a child says “please” in a request, Alexa might respond with “Thanks for asking so nicely.” Alexa replies to “Thank you” with “You’re welcome” or something similar. But if a child doesn’t say “please” or “thank you,” there’s no consequence.

An Amazon spokesperson told me that parents had requested help with reinforcing polite speech when their kids talk to Alexa. The company says it’s “still super early days” with the Magic Word feature, and expects to make future improvements based on customer feedback.


Count me among the group that doesn’t say please.
link to this extract

Uber test car driver streamed Hulu before fatal crash • Consumer Reports

Jeff Plungis and Keith Barry:


The Tempe police report says distraction was a factor in the crash that killed the pedestrian, Elaine Herzberg.

During Vasquez’s ride in the Uber vehicle, which was recorded on video inside the vehicle as part of the testing, she looked down 204 times, mostly in the direction of the lower center console near her right knee, according to the police report. She was looking down for 5.2 of the final 5.7 seconds prior to the crash, the report says.

A log of Vasquez’s account provided by the video-streaming service Hulu, under a search warrant, showed that “The Voice” was streaming on her account in the final 43 minutes of the drive and that the streaming ended at 9:59 p.m., the approximate time of the collision, the police report says. 

The police concluded that the crash wouldn’t have occurred if Vasquez had been paying attention to the roadway, and indicated that she could be charged with vehicular manslaughter. Details from the police report were published Thursday by the Arizona Republic, Reuters, and other media outlets.


In which case what’s the point of it being “self-driving”? It’s the limitations that make this pointless. You couldn’t trust it on motorways, side roads, at night. In which case there’s no point having it. Self-driving systems have to be really, really good, or else not employed at all, because driver inattention will be a thing, and accidents will keep happening.
link to this extract

Intel and the danger of integration • Stratechery

Ben Thompson:


TSMC [founded in 1987! on the promise that it wouldn’t compete with its customers to design chips, only make them] got better, in large part because it had no choice: soon its manufacturing capabilities were only one step behind industry standards, and within a decade had caught-up (although Intel remained ahead of everyone). Meanwhile, the fact that TSMC existed created the conditions for an explosion in “fabless” chip companies that focused on nothing but design. For example, in the late 1990s there was an explosion in companies focused on dedicated graphics chips: nearly all of them were manufactured by TSMC. And, all along, the increased business let TSMC invest even more in its manufacturing capabilities.

This represented into a three-pronged assault on Intel’s dominance:

• Many of those new fabless design companies were creating products that were direct alternatives to Intel chips for general purpose computing. The vast majority of these were based on the ARM architecture, but also AMD in 2008 spun off its fab operations (christened GlobalFoundries) and became a fabless designer of x86 chips.
• Specialized chips, designed by fabless design companies, were increasingly used for operations that had previously been the domain of general purpose processors. Graphics chips in particular were well-suited to machine learning, cryptocurrency mining, and other highly “embarrassingly parallel” operations; many of those applications have spawned specialized chips of their own. There are dedicated bitcoin chips, for example, or Google’s Tensor Processing Units: all are manufactured by TSMC.
• Meanwhile TMSC, joined by competitors like GlobalFoundries and Samsung, were investing ever more in new manufacturing processes, fueled by the revenue from the previous two factors in a virtuous cycle.


When you consider the victory that modularisation has wrought in the right-hand part of that image, you have to marvel at how Apple has managed to navigate the rapids to get to where it is. Every company has to integrate to a degree; the question is how much, and when to stop/start. At Intel, it seems to have continued just that bit too long because the money was so good.
link to this extract

Brexit’s big short: how pollsters helped hedge funds beat the crash • Bloomberg Business

Cam Simpson:


These hedge funds [which had bought early access to private polling data on Brexit] were in the perfect position to earn fortunes by short selling the British pound. Others learned the likely outcome of public, potentially market-moving polls before they were published, offering surefire trades.

Hedge fund managers, of course, try to beat the market by getting the best information they can. For exit polling data, that’s a tricky business. Pollsters have always sold surveys to private clients, but UK law restricts them from releasing exit-poll data before voting ends. While some of the practices discovered by Bloomberg fall into a gray area, the law is clear: It would have been a violation if, prior to the polls closing, “any section of the public” had gotten the same data the pollsters sold privately to hedge funds.

One person with questions still to answer is [Nigel] Farage, a former commodities broker who also went to work for a London currency trading company after he moved into politics. He twice told the world on election night that Leave had likely lost, when he had information suggesting his side had actually won. He also has changed his story about who told him what regarding that very valuable piece of information.

Bloomberg’s account is based in part on interviews over seven months with more than 30 knowledgeable current and former polling-company executives, consultants and traders, nearly all of whom spoke only on the condition they not be named because of confidentiality agreements. Pollsters said they believed Brexit yielded one of the most profitable single days in the history of their industry. Some hedge funds that hired them cleared in the hundreds of millions of dollars…


Farage suddenly has a lot of questions to answer – about this, and about Russian influence in the Brexit vote.
link to this extract

Errata, corrigenda and ai no corrida: none notified

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.