Facial recognition is in more and more places (like this airport gate). Maybe it’s the next big thing? CC-licensed photo by Delta News Hub 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. So let’s get started. I’m @charlesarthur on Twitter. Observations and links welcome.
Facial recognition creeps into everything at CES 2020 • CNET
Alfred Ng:
»
Konami Gaming, a slot machine maker, wants to weave facial recognition into its one-armed bandits. During a visit to its Las Vegas headquarters to hear more about its plans, I quickly discovered what the world would be like if facial recognition is everywhere.
“Hello, Alfred,” said a measured, robotic voice, startling me. It came from a kiosk called “Biometrics Welcome Console” positioned right next to the door of the conference room where my meeting was held. The kiosk knew who I was because Konami had set up a profile for me, using a public photo from my CNET bio without telling me. The facial recognition tagged me before I’d even said hello to the Konami team members in the room.
I looked at the screen showing the photo the kiosk took of me when I walked in. The camera had caught just my eyes and nose. Still, the facial recognition software calculated it detected me with 60.5% accuracy.
“Any picture you use online can be used to identify you already,” Sina Miri, Konami’s vice president of innovation and strategic research and design, told me. Konami had also set up profiles of my colleagues at the visit, again without telling them.
«
I think Ng is correct about what’s happening here: facial recognition is going to be in everything, and everywhere. Our homes and cars will recognise us, things in the street will recognise us. It will be the scene from Minority Report. The next big thing isn’t a positive thing like augmented reality spectacles that inform you about the world; it’s facial recognition everywhere, and we might not be in control of it as we are our smartphones. (Though even those leak data like mad.)
unique link to this extract
Why does the IEA keep getting renewables wrong? • Unearthed
Lauri Myllyvirta:
»
The IEA [International Energy Agency] argues that the New Policies Scenarios are not predictions or forecasts, simply assessments of ‘where today’s policy ambitions are likely to take the energy sector’. However, global policy ambition on solar did not increase ten-fold from 2005 to 2010, the ability of the technology to deliver did. Providing information on how much solar capacity current and expected policies are likely to deliver is exactly the job of these scenarios and, so far, they have been worse than useless in that job.
Forecasting the future is very hard, while picking faults in the work of an agency that puts out detailed scenarios every year is easy. However, the blatant underestimation of renewables goes well beyond normal shifts, surprises and misjudgments that are to be expected in any attempt to assess the future. If you keep making the exact same assessment for 15 years and are wildly wrong every time, you go back and assess the premises for your assessments. If something is happening in the real world that your models fail to capture, you improve your models. Anything else is not a good-faith effort to look at energy sector trends.
«
This is from 2017. Guess what? The IEA (no relation to the right-wing shadily funded Institute of Economic Affairs) has continued getting it wrong. On its own that wouldn’t be a problem, but big banks and investment companies use the IEA’s forecasts to decide where to put their money. That keeps open the coal plants that should be shut down because they’ll be uneconomic, and prevents the investment in renewables that we need.
unique link to this extract
State support helped fuel Huawei’s global rise • WSJ
Chuin-Wei Yap:
»
Tax deductions and exemptions helped Huawei save up to $25 billion in income, value-added and other taxes in at least the past decade, the Journal estimated. Responding to the estimate, a Huawei spokesman said the company is globally tax-compliant.
In his remarks at the conference, Mr. Li said local officials began waiving or reducing levies on Huawei, including income and value-added taxes, in the early 1990s.
Financial support helped the company undercut rivals. In 2010, the European Commission found that Chinese modem exporters including Huawei had benefited from subsidies, according to a confidential report reviewed by the Journal. The commission cut short its probe after the complainant prompting it reached a “cooperation agreement” with the company. Huawei denied receiving such subsidies.
Besides subsidies, Huawei since 1998 has received an estimated $16 billion in loans, export credits, and other forms of financing from Chinese banks for itself or its customers, the Journal found.
China’s state-controlled banking system underpins cheap loans that lower costs for Huawei and its customers to buy its products on credit. State lending facilities for Huawei were among the largest in history.
«
The WSJ puts the total subsidies at $75bn over its life. Not surprising that it has been able to undercut Nokia, Alcatel and the other network equipment companies in bids over the years; and that has a flywheel effect – you get more contracts, and your rivals aren’t getting them.
But is it really unfair, when China wanted to be able to control its destiny in the telecoms market?
unique link to this extract
Researchers: Are we on the cusp of an ‘AI winter’? • BBC News
Sam Shead:
»
Robot Wars judge Noel Sharkey, who is also a professor of AI and robotics at Sheffield University, told the BBC that he likes the term “AI autumn” – and several others agree.
…In 2014, Nick Bostrom, a philosopher at Oxford University, went one step further with his book Superintelligence. It predicts a world where machines are firmly in control. But those conversations were taken less and less seriously as the decade went on. At the end of 2019, the smartest computers could still only excel at a “narrow” selection of tasks.
Gary Marcus, an AI researcher at New York University, said: “By the end of the decade there was a growing realisation that current techniques can only carry us so far.”
He thinks the industry needs some “real innovation” to go further. “There is a general feeling of plateau,” said Verena Rieser, a professor in conversational AI at Edinburgh’s Herriot Watt University. One AI researcher who wishes to remain anonymous said we’re entering a period where we are especially sceptical about AGI.
“The public perception of AI is increasingly dark: the public believes AI is a sinister technology,” they said.
For its part, DeepMind has a more optimistic view of AI’s potential, suggesting that as yet “we’re only just scratching the surface of what might be possible”.
“As the community solves and discovers more, further challenging problems open up,” explained Koray Kavukcuoglu, its vice president of research. “This is why AI is a long-term scientific research journey.
“We believe AI will be one of the most powerful enabling technologies ever created – a single invention that could unlock solutions to thousands of problems. The next decade will see renewed efforts to generalise the capabilities of AI systems to help achieve that potential – both building on methods that have already been successful and researching how to build general-purpose AI that can tackle a wide range of tasks.”
«
Two states. Eight textbooks. Two American stories • The New York Times
Dana Goldstein:
»
The books The Times analyzed were published in 2016 or later and have been widely adopted for eighth and 11th graders, though publishers declined to share sales figures. Each text has editions for Texas and California, among other states, customized to satisfy policymakers with different priorities.
“At the end of the day, it’s a political process,” said Jesús F. de la Teja, an emeritus professor of history at Texas State University who has worked for the state of Texas and for publishers in reviewing standards and textbooks.
The differences between state editions can be traced back to several sources: state social studies standards; state laws; and feedback from panels of appointees that huddle, in Sacramento and Austin hotel conference rooms, to review drafts.
Requests from textbook review panels, submitted in painstaking detail to publishers, show the sometimes granular ways that ideology can influence the writing of history.
A California panel asked the publisher McGraw-Hill to avoid the use of the word “massacre” when describing 19th-century Native American attacks on white people. A Texas panel asked Pearson to point out the number of clergy who signed the Declaration of Independence, and to state that the nation’s founders were inspired by the Protestant Great Awakening.
All the members of the California panel were educators selected by the State Board of Education, whose members were appointed by former Gov. Jerry Brown, a Democrat. The Texas panel, appointed by the Republican-dominated State Board of Education, was made up of educators, parents, business representatives and a Christian pastor and politician.
«
You might think: America’s a big place, it’s as big as Europe, there are going to be differences. But where you have divergent pictures of a history of a single nation, you’re going to create differences in how people view the country. That will then count when it comes to picking politicians and voting on laws.
unique link to this extract
India’s top court says indefinite Kashmir internet shutdown is illegal • Reuters
Sankalp Phartiyal and Fayaz Bukhari:
»
India’s Supreme Court said on Friday that an indefinite shutdown of the internet in Kashmir was illegal, rebuking the government for the communications lockdown imposed after it withdrew the Muslim majority region’s autonomy in August.
Internet suspensions can be imposed only for “temporary duration” and an indefinite suspension violated India’s telecoms rules, the court said in an order published on its website.
It also ordered authorities to review all such curbs in Kashmir immediately.
Authorities must consider immediately allowing the functioning of essential internet services such as for hospitals and limited e-banking in regions where internet cannot be restored right away, the court added.
“Freedom of Internet access is a fundamental right,” Supreme Court justice N. V. Ramana said.
«
New fundamental right? The Indian government has been quietly turning into a very authoritarian one, though.
unique link to this extract
LG aims to turn around mobile unit • Korea Times
Baek Byung-yeul:
»
A senior LG Electronics executive said Thursday that the company’s long-time money-losing smartphone division will turn a profit by the end of 2021; however, he didn’t elaborate how.
“LG Electronics mobile business is going to be profitable by 2021. I can say we can make that happen as LG Electronics will expand our mobile lineup and steadily release new ones attached with some wow factors to woo consumers,” the company’s chief executive Kwon Bong-seok told reporters in a press conference on the sidelines of this year’s technology exhibition, here.
Regarding the specifics on how, the CEO didn’t delve into more but only reiterated LG Electronics’ plan to expand the phone lineup, which he believes is possibly a plus factor to improve LG’s competitiveness in the already saturated smartphone market.
«
LG’s mobile division has lost money for more than three straight years now. So when the new CEO says it’s going to become profitable, I say
Facebook’s PR feels broken • The Margins
Ranjan Roy:
»
to summarize, Andrew Bosworth, longtime Facebook exec, wrote a long, reflective internal post on Facebook’s role in the 2020 election:
So was Facebook responsible for Donald Trump getting elected? I think the answer is yes, but not for the reasons anyone thinks. He didn’t get elected because of Russia or misinformation or Cambridge Analytica. He got elected because he ran the single best digital ad campaign I’ve ever seen from any advertiser. Period.
In a section that got a lot of attention, he continued:
»
I find myself thinking of the Lord of the Rings at this moment. Specifically when Frodo offers the ring to Galadrial and she imagines using the power righteously, at first, but knows it will eventually corrupt her. As tempting as it is to use the tools available to us to change the outcome, I am confident we must never do that or we will become that which we fear.
«
I’m not a big LOTR person, and will let Gizmodo cover the accuracy of his reference, but how does Facebook possibly let this enter the national conversation? One of their most longtime, loyal leaders is directly saying they have the power to sway national elections. It is their decision, and their decision alone, to resist the temptation to “change the outcome”!
This is the very definition of a need for regulation. By its own admission, the company is acknowledging its unnatural power. In the memo, Boz clarifies he’s liberal in his politics, but the issue is not Facebook and its purported ties to the right. The issue is simply its size. An individual, for-profit corporation should not get to decide whether democracy will work.
To continue on the communications breakdown, Boz posted an explanation on Facebook, where he advertises the post as an organizational, internal call-to-debate. But while it’s great to have a safe space for internal, organizational debates, it’s still hugely concerning when that internal debate is whether we should all have a free and fair election in the U.S.
«
Roy and Manjan produce a consistently good newsletter (and it’s free). This dissects the whole Facebook debacle particularly well.
unique link to this extract
A lazy fix 20 years ago means the Y2K bug is taking down computers now • New Scientist
Chris Stokel-Walker:
»
Parking meters, cash registers and a professional wrestling video game have fallen foul of a computer glitch related to the Y2K bug.
The Y2020 bug, which has taken many payment and computer systems offline, is a long-lingering side effect of attempts to fix the Y2K, or millennium bug.
Both stem from the way computers store dates. Many older systems express years using two numbers – 98, for instance, for 1998 – in an effort to save memory. The Y2K bug was a fear that computers would treat 00 as 1900, rather than 2000.
Programmers wanting to avoid the Y2K bug had two broad options: entirely rewrite their code, or adopt a quick fix called “windowing”, which would treat all dates from 00 to 20, as from the 2000s, rather than the 1900s. An estimated 80% of computers fixed in 1999 used the quicker, cheaper option.
“Windowing, even during Y2K, was the worst of all possible solutions because it kicked the problem down the road,” says Dylan Mulvin at the London School of Economics.
Coders chose 1920 to 2020 as the standard window because of the significance of the midpoint, 1970. “Many programming languages and systems handle dates and times as seconds from 1970/01/01, also called Unix time,” says Tatsuhiko Miyagawa, an engineer at cloud platform provider Fastly.
Unix is a widely used operating system in a variety of industries, and this “epoch time” is seen as a standard.
The theory was that these windowed systems would be outmoded by the time 2020 arrived, but many are still hanging on and in some cases the issue had been forgotten.
«
Bing loses out to DuckDuckGo in Google’s new Android search engine ballot • The Verge
James Vincent:
»
EU citizens setting up Android devices from March 1 will be given a choice of four search engines to use as their default, including Google. Whichever provider they chose will become the default for searches made in Chrome and through Android’s home screen search box. A dedicated app for that provider will also be installed on their device.
The “choice screen” is being introduced by Google following an antitrust ruling from the European Union last March. Google was fined a record $5bn by EU regulators, who said the company had to stop “illegally tying” its search engine and browser to its mobile OS.
The search engines shown to new users will vary for each EU country, with the selection decided based on a “fourth-price” auction system. Each provider tells Google how much it’s willing to pay the company every time a user selects their product as the default. The three highest bidders are then shown to users, with the chosen provider paying Google the amount offered by the fourth-highest bid. This process is repeated every four months.
All this means that the choices Google will show to users don’t necessarily reflect a search engine’s popularity in that country. Rather, it shows how much the provider is willing to pay for users. This might explain why Microsoft’s Bing only appears as an option in the UK — a country where the revenue from search ads is likely to be higher than lower-GDP nations.
When Google announced the auction system last August, rival search providers were not happy. Eric Leandri, CEO of privacy-focused search engine Qwant, said it was a “total abuse of [Google’s] dominant position” to “ask for cash just for showing a proposal of alternatives.” Gabriel Weinberg, CEO of DuckDuckGo, said the auction system was a “pay-to-play auction” that meant “Google will profit at the expense of the competition.”
«
As one of the commenters points out, the money from the auction shouldn’t go to Google – it ought to go to a charity. Or it could go to the EC, or to rival search providers. Whichever; it doesn’t make sense for Google to be rewarded for abusing its position, which the EC decision clearly says it was doing.
unique link to this extract
Errata, corrigenda and ai no corrida: none notified
The Overspill: when there's more that I want to say 
How do Huawei’s subsidies compare to a) others in the industry (I still remember when the French state killed microcomputer inventor Micral by merging it with heavyweight Bull to create a “national champion” that not-promptly-enough failed) b) subsidies to other “national champion” corps (Airbus/Boeing…).
I’m having the same issue about subsidies as about spying. They’re doing it, and its evil. But everybody else is doing it too, and I’m unsure how much worse the Huawei version is, if at all. For all the hand-wringing about fake news and bias, I think this issue is being treated the same clickbaity way. And with no context, again: in the end, the only safe software is open-source software. That point never gets made in this stories.