Start Up No.1,092: Facebook launches Libra, driving v self-driving cars, how we changed dogs, Mazda dumps touchscreens, and more

Some of the bits are going out – and being reclaimed. CC-licensed photo by charlene mcbride 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 10 links for you. Sufficient votes to stay. I’m @charlesarthur on Twitter. Observations and links welcome.

Self-driving cars have a problem: safer human-driven ones • WSJ

Christopher Mims:


If you buy one of many new makes and models of car today, you might be surprised to find that, as a standard feature, it can do something your previous car couldn’t: It will take over when it thinks you’re making a mistake.

In the coming years, many cars will do more than that, even driving mostly by themselves, at least on highways. And not just luxury models such as the latest Audi A8 or Cadillac CT6, but something as mainstream as a Nissan Rogue.

Some of this technology has been in development for years, but the newest versions of it—with advanced object recognition, radar-and-laser detection and lightning-fast artificial intelligence—were created for autonomous cars. Many tech entrepreneurs have argued that fleets of robo-taxis would convince us to abandon personal car ownership in favor of “transportation as a service.” Some of them have predicted these robot cars will start populating U.S. roads within the next two years.

But the paradox of how this evolution is playing out is that technology developed to give us driverless vehicles from the likes of Tesla and Alphabet’s Waymo could actually delay their adoption.

When car makers put these incremental tech advances in human-driven cars, they pre-empt one of the fully self-driving car’s supposed advantages: safety. These new systems marry the best machines capabilities—360-degree sensing and millisecond reflexes—with the best of the human brain, such as our ability to come up with novel solutions to unique problems.


Maybe we’ll just never quite get there; maybe it’ll be an asymptotic, Zeno-style approach rather than a big bang.
unique link to this extract

☠️☠️☠️ PixelPirate MillionDollarHomepage-As-A-Service (MDHPAAS) ☠️☠️☠️

Domain Research Group:


Avast ye maties and welcome aboard the SS PixelPirate. Every day, domains linked from the MillionDollarHomepage expire.

With the help of PixelPirate’s revolutionary MillionDollarHomepage As A Service (MDHPAAS) technology, our band of robot scallywags scour the data seas, always on the lookout for domains prematurely sent to Davy Jones’ Locker.

Capture the domain, own the Pixels.


People still visit the Million Dollar Homepage? I’d have thought this is more like a scheme by its creator to get people to go back. (Though he has since moved on to much better things.)
unique link to this extract

Dogs’ eyes have changed since humans befriended them • The Atlantic

Haley Weiss:


We connect profoundly with animals capable of exaggerating the size and width of their eyes, which makes them look like our own human babies and “hijacks” our nurturing instincts. Research has already demonstrated that humans prefer pets with more infantlike facial features, and two years ago, the authors of this latest study showed that dogs who made the facial movement enabled by the RAOL and LAOM muscles—an expression we read as distinctly humanlike—were more likely to be selected for adoption from a shelter than those who didn’t.

We might not have bred dogs for this trait knowingly, but they gained so much from having it that it became a widespread facial feature. “These muscles evolved during domestication, but almost certainly due to an advantage they gave dogs during interactions with humans that we humans have been all but unaware of,” Hare explained.

“It’s such a classically human system that we have, the ways we interact with our own infants,” says Angie Johnston, an assistant professor at Boston College who studies canine cognition and was not involved with the study. “A big theme that’s come out again and again in canine cognition and looking at the domestication of dogs is that it seems like they really just kind of dove right into our society in the role of being an infant or a small child in a lot of ways. They’re co-opting existing systems we have.”


Quite how a muscle like that evolves is really puzzling: domestication is relatively recent (40,000-20,000 years ago) which doesn’t seem to give it much time. And it has to spread through the whole species.

unique link to this extract

Why Mazda is purging touchscreens from its vehicles • Motor Authority

Bengt Halvorson:


It wasn’t a decision that was hastily made, according to company officials. However, as they started studying the effects of touchscreens on driving safety (and driving comfort), it soon became clear what the priorities should be with this completely new system that makes its debut in the 2019 Mazda 3.

It started out by looking at actual times—the times spent looking away from the road to make a screen selection, and the time needed to refocus the eyes on something close versus the road ahead—and decided that it needed to home in on factors that reduced that time.

“Doing our research, when a driver would reach towards a touch-screen interface in any vehicle, they would unintentionally apply torque to the steering wheel, and the vehicle would drift out of its lane position,” said Matthew Valbuena, Mazda North America’s lead engineer for HMI and infotainment.

“And of course with a touchscreen you have to be looking at the screen while you’re touching…so for that reason we were comfortable removing the touch-screen functionality,” he added.

The head-up display that top trims of the Mazda 3 get is now projected onto the windshield.


Pretty obvious really that touchscreens don’t offer tactile feedback; the distraction factor is very high.
unique link to this extract

Congress must act on regulating deepfakes • OneZero

Mutale Nkonde:


We are 17 months away from the 2020 Presidential Election and the rise of deepfake videos should concern us all. The indictment of 13 members of Russia’s Internet Research Agency (IRA) by Special Counsel Robert Mueller for online African-American voter suppression highlights how American anti-blackness was weaponized against the country as a whole. Mueller’s team showed how Russian operatives created 30 pro-black Facebook and Instagram sites with names like “blacktivst” and “woke blacks,” which thanks to advertising recommendation algorithms reached 1.2 million people.

During the deepfakes briefing process, Russia expert Nina Jankowitcz debunked the idea that disinformation campaigns were built solely on lies. She explained that Russian disinformation campaigns build trust with their target audiences by introducing verifiable facts into public debate. In the case of the IRA they bought Facebook ads highlighting Hillary Clinton’s support of the 1994 crime bill, linking this legislation to the criminalization of the African-American community, and reinforcing this argument with clips of her describing young black boys as “super predators.” That much was true, but they also weaved in inspirational quotes, funny memes and provided a space to vent online for African-Americans frustrated with their presidential options. Once messages expressing dissatisfaction with both 2016 candidates began accumulating, the IRA began to release ads suggesting members of these pro-Black communities should not vote for Hillary Clinton because she may still hold these views. That part wasn’t true, but it did fit into the narrative laid out by this otherwise trustworthy community.

Now imagine if nefarious actors could create deepfake video of a presidential candidate saying the N-word, or appearing in blackface? That would potentially and unfairly end their run. That’s why we need to stop the spread of deepfake video before it is used to interfere with the 2020 election.


My prediction: deepfakes will be a notable part of the 2020 US presidential election. Facebook will dither about them.
unique link to this extract

Solar, wind, batteries to attract $10 trillion to 2050, but curbing emissions long-term will require other technologies too • BloombergNEF


Deep declines in wind, solar and battery technology costs will result in a grid nearly half-powered by the two fast-growing renewable energy sources by 2050, according to the latest projections from BloombergNEF (BNEF). In its New Energy Outlook 2019 (NEO), BNEF sees these technologies ensuring that – at least until 2030 – the power sector contributes its share toward keeping global temperatures from rising more than 2 degrees Celsius.[1]

Each year, NEO compares the costs of competing energy technologies through a levelized cost of energy analysis. This year, the report finds that, in approximately two-thirds of the world, wind or solar now represent the least expensive option for adding new power-generating capacity. Electricity demand is set to increase 62%, resulting in global generating capacity almost tripling between 2018 and 2050. This will attract $13.3 trillion in new investment, of which wind will take $5.3 trillion and solar $4.2 trillion.


Encouraging. But of course there’s still that “will require new technologies too” coda.
unique link to this extract

Time for Clockwise • Greylock Perspectives

John Lilly:


I met the Clockwise founders Matt Martin, Mike Grinolds, and Gary Lerhaupt early in 2018 and was immediately taken with them, their passion for helping us get control of our time, and their new approach to doing it — not by creating a new calendar app, but by using machine learning to make the calendars we already have work better. The approach would take a deep mix of product, UX, machine learning and systems thinking to make work, and that’s precisely what Matt, Mike & Gary demonstrated. This was a team I wanted to be in business with, building a technology that needed to exist in the world.

Clockwise makes a product and supporting technology that actually gives us time back. They’ve been heads down over the past couple of years building their first product — connect it to your own calendar and it figures out how to optimize your days to give you back meaningful chunks of time in whole blocks. For old school nerds (🙋 ♂️) , it will remind you of a disk defragmenter:

You can see from the animation above that Clockwise can figure out which meetings are movable (like weekly 1–1s) and which aren’t (like staff meetings), and can rework your weekly calendar to give you back time to think and time to work.


$10m for this? But Lilly says that the power emerges when you use it across the whole organisation, so that it can maximise multiple peoples’ time, and offers some examples.

Except… what if you’re in an organisation which just has lots of ad-hoc meetings? Or maybe newspapers are atypical organisations.
unique link to this extract

These influencers aren’t flesh and blood, yet millions follow them • The New York Times

Tiffany Hau:


Everything about Ms. Sousa, better known as Lil Miquela, is manufactured: the straight-cut bangs, the Brazilian-Spanish heritage, the bevy of beautiful friends.

Lil Miquela, who has 1.6 million Instagram followers, is a computer-generated character. Introduced in 2016 by a Los Angeles company backed by Silicon Valley money, she belongs to a growing cadre of social media marketers known as virtual influencers.

Each month, more than 80,000 people stream Lil Miquela’s songs on Spotify. She has worked with the Italian fashion label Prada, given interviews from Coachella and flaunted a tattoo designed by an artist who inked Miley Cyrus.

Until last year, when her creators orchestrated a publicity stunt to reveal her provenance, many of her fans assumed she was a flesh-and-blood 19-year-old. But Lil Miquela is made of pixels, and she was designed to attract follows and likes.

Her success has raised a question for companies hoping to connect with consumers who increasingly spend their leisure time online: Why hire a celebrity, a supermodel or even a social media influencer to market your product when you can create the ideal brand ambassador from scratch?

That’s what the fashion label Balmain did last year when it commissioned the British artist Cameron-James Wilson to design a “diverse mix” of digital models, including a white woman, a black woman and an Asian woman. Other companies have followed Balmain’s lead.


And no chance of them collapsing out of a taxi drunk. Although… isn’t that publicity, of which there is no bad form?
unique link to this extract

China’s unmanned store boom ends as quickly as it began • Nikkei Asian Review

Hiroshi Murayama:


In Shenzhen’s neighboring city of Guangzhou, i-Store, the first local unmanned convenience chain, has also been closing stores one after another. It had three left at the end of March, down from a peak of nine, according to a local newspaper.

Last December,, China’s second-largest online retailer, announced it would suspend its smart shelf business — small unmanned shops the size of train station kiosks. In July 2018, unveiled a plan to open 5,000 of them in office buildings and other places in major cities by the end of the year, only to withdraw the plan six months later.

What went wrong?

The difficulty of selling fresh groceries in stores without staff is one major obstacle.

Industry experts say a convenience store in a major Chinese city like Beijing needs to generate at least 5,000 to 6,000 yuan in sales per day to remain viable. A significant chunk of those sales come from boxed lunches, ready-made fresh meals, desserts and other products with limited shelf lives.

In Japan and China alike, the gross margin on processed food, which lasts longer, is about 25%, while that on fast food and fresh groceries stands at 40% to 50%. In other words, the higher the ratio of fresh food at a convenience store, the more stable the business becomes.

Many of the companies that attempted to run unmanned convenience stores appear to have lacked such knowledge.


Robots not as good at running shops as humans? There’s hope for us yet.
unique link to this extract

Facebook’s Libra will give billions access to cryptocurrency • CNBC

Salvador Rodriguez:


Facebook on Tuesday unveiled its long-rumored digital coin called Libra that will become available to users in the first half of 2020. The open-source digital currency has been under development by Facebook over the past year, but it will be managed by a nonprofit association with support from a variety of companies and organizations.

“Libra is a major validation of cryptocurrencies and blockchain technology to be the financial infrastructure of the future,” said Garrick Hileman, head of research at Blockchain, which makes a cryptocurrency wallet. Libra “could be one of the most significant and positive events in cryptocurrencies’ history as billions of Facebook users could join the ecosystem we’ve been building over the last decade.”

Many in the blockchain space say they believe Libra will leverage Facebook’s more than 2.7 billion monthly users to finally bring cryptocurrencies into the mainstream.

“Worst case scenario, Facebook crypto could become the gateway drug to introduce people to the greater crypto ecosystem,” said Roneil Rumburg, CEO of Audius, a blockchain-based music streaming service. “Best case, their stablecoin is sufficiently decentralized to enable interesting censorship-resistant use cases and is still usable by a mainstream audience.”


I don’t think it’s going to bring cryptocurrencies into the mainstream; it might bring Libra into the mainstream, but it’s colossally well funded compared to many others, and has the infrastructure behind it. That to me makes it worrying more than anything else.
unique link to this extract

Errata, corrigenda and ai no corrida: none notified

6 thoughts on “Start Up No.1,092: Facebook launches Libra, driving v self-driving cars, how we changed dogs, Mazda dumps touchscreens, and more

  1. re. smart cars: I think the main use case for now is people who wouldn’t use regular cars: seniors, … And in this case, smart cars don’t need to go as fast as legacy cars. Even I would be happy with a slower car, if it meant I got that time back to actually do stuff, let alone if it meant I could move around at all.
    That’s where disruption theory helps a bit: you’re not trying to replace at the high end a good driver’s nice ride, but at the low end a semi-blind grandma’s fed-up nephew and his Lada.

    As usual there’s a strong commenter bias. All those able-bodied and -minded adults only thinking of their own use case.

    • That’s a really good point. Definitely an older generation who might have the money but not the physical capacity (and let’s include disabled drivers too), these could be an excellent alternative. The disruption point is spot on.

  2. re. Car touchscreens: those have the same issue as smartphones and tablets: a confusing UI that’s not that discoverable and rather unwieldy. My Canadian brother rents car every summer when he comes over, usually gets something midrange w/ a touchscreen… finding stuff on there is a whole adventure. Even basic stuff such as the GPS is confusing enough we usually use our phones.
    A touchscreen saves a whole lot of money, but current UIs are bad, and, above all, nonstandard. The problem would mostly go away if everybody used if not the exact same layout, et least the same basic design principles.

  3. Engineers I spoke to at a certain US company that has a lot of off-road capability (I think you can guess who, but as they weren’t on the record, probably shouldn’t spell it out) told me that in their internal analysis, cars with automatic frontal collision systems have 50% fewer accidents than those without. This is huge!

    On the downside I’ve noticed since these systems turned up, visibility in a lot of vehicles seem to be worse (Mazda CX-5 for example). I asked the Mini dealer why they didn’t have blind spot monitoring and the answer was “you don’t need it, you can see out the window.” I thought he was being snarky at first but he was right. Visibility is terrific in my Countryman. Less so in the SUV.

    • More generally Tesla’s PR on Autopilot safety is being contested.
      It’s really really important to check sources and methodology. Maybe newer cars have fewer crashes than older cars. Maybe new fewer than second-hand, maybe expensive less than cheap, maybe the prevention collision sells more in some markets (urban, young parents, women, higher spenders…) that have fewer crashes. Maybe it’s not active in the most accident-prone circumstances (parking lots…).
      I’ve done a bit (way too much) of Marketing Statistics. a) you can always find a way to make them say whatever you want b) if you want unimpeachable results, it’s excruciating work similar to Shcroedinger’s cat: the mere fact you add and advertise the feature makes the results not comparable: “safety-conscious people who buy our safety-enhanced product have fewer accidents”. Who’d have thunk ?

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.