Tuesday, September 25, 2018

"Market Forecasting: A Sensitive Practice at the Heart of Neoliberal Capitalism"

HT up front: Professor Pasquale who quotes:
“In the absence of theoretical and empirical justifications, analysts base their work on what they refer to as ‘market feeling’ – a technique that builds of affect, tacit knowledge, and experience – to make sense of market developments.”
Combine that intro with an Alfred Cowles reference in the second paragraph and I'm in.
(actually the 'absence of theoretical and empirical justifications' line would have been enough, Cowles is the cherry on top)  

From Economic Sociology and Political Economy (emphasis in original):
Since the emergence of modern financial markets, financial analysts have played a critical role in producing visions of “the economy” and its future development. As experts, they analyze market developments and predict future scenarios that enable other financial market participants to speculate on the rise or fall of stock prices, the success or failure of particular investment products, and the growth or decline of entire national economies. The substance of the analysts’ valuation and forecasting practices is, however, heavily disputed among economists. In neoclassical economic theory, the assumption that markets are informationally efficient has challenged the legitimacy of the work of financial analysts since the establishment of the efficient market hypothesis as a central paradigm in the mid 1960s. Alternative schools of thoughts – such as new institutional or behavioral economics – have criticized this paradigm. However, they have also argued that the degree of uncertainty, which is inherent to financial markets, makes prediction impossible.
Empirically, this critique has been around even longer. In 1933, economist Alfred Cowles published an article that tested the attempt to forecast stock market prices. After having analyzed thousands of stock market predictions from 16 financial service agencies, Cowles came to the conclusion that “[s]tatistical tests of the best individual records failed to demonstrate that they exhibited skill, and indicated that they more probably were results of chance.” Such results have been confirmed repeatedly. In the UK in 2012, for example, Orlando, a ginger cat that tapped over the pages of the Financial Times to select stocks, outperformed a group of financial professionals.
The lack of justification of the forecasting practices financial analysts deploy brings up an important question: Why do financial analysts exist at all? How do they manage to maintain their role as experts in the market? And how they cope with the uncertainty and lack of theoretical and empirical foundations of their practices?

I spend two years in the financial analysis department of an internationally operating bank, where I was given research permission to follow the financial analysts’ work practices on a daily basis. In my book
Stories of Capitalism: Inside the Role of Financial Analysts (University of Chicago Press, 2018), I illustrate how, in the absence of theoretical and empirical justifications, analysts base their work on what they refer to as “market feeling – a technique that builds of affect, tacit knowledge, and experience – to make sense of market developments.

In the first weeks of my fieldwork, a financial analyst who coached me, told me how I could learn to do financial analysis. He advised me to take some time getting a “feeling” for how markets work. “This takes a lot of time,” he explained, “but basically, you just have to observe the market and read financial newspapers and the reports of other analysts.” The analyst then stared off into space, groping for words. After a while, he said, “You know, it’s not just about observing and reading, it’s about…” He did not finish his sentence since he could not put into words how one should develop that feeling for the market he was talking about. “You know, it’s about…,” he made a gesture as if he was touching a very smooth fabric to check whether it was made of silk. “That feeling,” he continued, “is what differentiates a good analyst from a bad one.”

Weeks later, another analyst allowed me to sit next to him when he valuated a company’s stock in order to come up with a forecast. To come up with a company’s “target price,” that is an estimated future price of a stock, this analyst first looked at the facts and figures presented in the company’s quarterly financial statements. After looking at the numbers depicted on the statement, he entered them into the bank’s internal computer program. He was, however, not happy with the target prices proposed to him by the models he used. He looked at them and then told me that the numbers support his overall feeling. Looking at the numbers a second time, he sighed, turned to me, and said, “You know what, I’ll take the most bullish target price and adjust the projections on the overall market development a little. After that, I’ll have a target price that truly reflects my feeling about the future development of this particularly promising stock.”

Combining such market feeling with calculative practices allows financial analysts to create persuasive narratives of the future of the market. These narratives become visible in the way analysts explain future market developments. Moreover, they are inscribed into the aesthetics of charts, tables, and figures analysts produce to visualize past, current and potential future market developments. Once these narratives are created, they are circulated among the banks’ stakeholders and clients. Sometimes, the narratives also become part of broader public discourse – often through the help of newspapers and TV stations that give financial analysts a platform to share their opinions on economic developments....

I've mentioned that we are rather fond of Yale's Robert Shiller. He seems to have "a pretty good feel" (see above) for this stuff. Here's his his September 4, 2007 call:
Some U.S. Housing Markets could drop 50%- Shiller

In addition to publishing "Irrational Exuberance" in March, 2000 with the NASDAQ hitting its then-all time closing high of 5048 (subsequent low 1114, how's a 78% decline grab ya?) he is the keeper of the Cowles Commission records.* From one of our Forecasting Equity Returns posts:
A subject near and dear to my heart. I may be the only person I've ever met who read every page of "The Cowles Commission's Common Stock Indexes 1871-1937".
[you must be a blast at parties -ed]
Well, in addition to Common Stock Indexes  we've posted some of Cowles' other stuff including a quick hit in January 2008, after the August 'ought-seven' quant-quake but before things got really ugly in September '08:

"Can Stock Market Forecasters Forecast?" is the title of a paper by one of my heroes, Alfred Cowles III.

It appeared in Vol.1, No. 3 of Econometrica, after having been read to a joint meeting of the Econometric Society and the American Statistical Society.

Mr. Cowles answer to the question?
"It is doubtful."
December 31, 1932.

[this is also the paper linked in the body of the ES&PE story but our link went to an ungated version, should one be inclined to peruse]

I'll have more on Mr. Cowles work later today. In the meantime, here's The Cowles Foundation for Research in Economics at Yale University.
The Cowles Foundation continues the work of the Cowles Commission for Research in Economics, founded in 1932 by Alfred Cowles at Colorado Springs. The Commission moved to Chicago in 1939 and was affiliated with the University of Chicago until 1955. In 1955 the professional research staff of the Commission accepted appointments at Yale and, along with other members of the Yale Department of Economics, formed the research staff of the newly established Cowles Foundation.

Amazon's Treasure Trove of Data From The HQ2 Submissions (AMZN)

From Axios:
When Amazon invited cities to compete for its second global headquarters a year ago today, it got reams of data from the 238 entrants — enough to learn details of the cities' future plans that a lot of their residents don't even know about.
Why it matters: The information effectively provided Amazon with a database chock full of granular details about the economic development prospects of every major metropolitan area in the United States (and some in Canada). For a rapidly-expanding tech behemoth like Amazon, that database could help it make expansion decisions that go way beyond the new headquarters.
Companies have conducted site searches in the past, but none have come close to the scale of the Amazon HQ2 search. That's because Amazon is "not just looking for HQ2," says Joe Parilla of the Brookings Institution. "They're looking for where they're going to put the next data center, the next logistics center, the next R&D facility."
  • Amazon's warehouses are within 20 miles of 31% of the U.S. population, while Walmart — its competitor — owns stores within 20 miles of 98% of the population, says Cooper Smith, an industry analyst at Gartner L2.
  • "Given they are siting new facilities like mad, this is a huge gift provided by taxpayers," Richard Florida, a professor at the University of Toronto, tells Axios.
Much of the quantitative data that Amazon picked up from cities is publicly available, Parilla says. What matters is the qualitative data cities offered up — they let Amazon in on their wildest dreams.
  • The sort of details that might be in a typical HQ2 application include plans for new train stations or shopping complexes — information the city's own residents wouldn't have, he says....

Hurricane Watch: "Multiple Threats in the Atlantic, and a Super Typhoon in the Pacific"

First up, a quick look at the 2018 hurricane season compared to the 1981 - 2010 average using Accumulated Cyclone Energy, this version via Colorado State University.

The Pacific has seen above average activity this year while the Atlantic—our focus because of the insurance/reinsurance/cat bond/ILS aspect—has been almost exactly average, as can be seen in the graphic below the table. Click here for the interactive versions of both.

Current ACE vs Historical

And from Wunderground's Cat 6 blog (also on blogroll at right during the season):
September 24, 2018, 1:40 PM EDT
Storms are being classified and declassified at a snappy pace in the Atlantic, as several weak systems have been fighting off dry air and wind shear. We may yet see one or more of these systems strengthening as the week unfolds—and there is no question about the ferocity of Super Typhoon Trami in the Northwest Pacific (see below).
Figure 1. Saharan Air Layer (SAL) analysis for 11 am EDT September 24, 2018. Dry air (orange colors) was affecting ex-Tropical Storm Kirk, Subtropical Storm Leslie, 98L, and the remains of TD 11. Image credit: University of Wisconsin/CIMSS.
The Caribbean’s invisible wall of protection that has been present during much of the hurricane season of 2018 bashed up Tropical Storm Kirk overnight, reducing the storm to a wave of low pressure without a surface circulation. Kirk entrained some dry air overnight, and this disrupted the storm enough that it could not maintain its closed circulation, due to its fast forward motion to the west at over 20 mph. Kirk’s demise may also have been aided by the passage of a suppressed Kelvin wave, as explained in a tweet by Michael Ventrice of The Weather Company.

Satellite images early Monday afternoon showed that Kirk was growing better organized, with low-level spiral bands rebuilding and a surface circulation re-developing. Wind shear was moderate, near 10 knots, and the sea surface temperatures (SSTs) were very warm, near 29°C (84°F). It would not be a surprise to see Kirk regain tropical storm status later on Monday. In its Tropical Weather Outlook issued at 2 pm EDT Monday, the National Hurricane Center gives Kirk a 50% chance of redevelopment into at least a tropical depression through Friday....MUCH MORE

Maersk-backed TheOceanCleanup Puts Its System to Final Tests

Following up on September 10's "Maersk-backed TheOceanCleanup Have Launched their Contraption to Clear the 'Great Pacific Garbage Patch'".

The reason both of our headlines mention Maersk is not because we think multi-billion dollar corporations need a free plug but because the ship they contributed to the effort is such an impressive piece of equipment. It can do anything that TheOceanCleanup needs it to do.

Maersk Launcher (Photo: Maersk Supply Service) 
From World Maritime News:
Following the launch from the San Francisco Bay on September 8, the Ocean Cleanup’s System 001 traveled 350 nautical miles to commence the Pacific Trials. 

The trials will last approximately two weeks and are a crucial step before the System 001 heads for the Great Pacific Garbage Patch to clean the ocean plastic.

The system consists of a 600-meter-long U-shaped floating barrier with a three-meter skirt attached below. It is designed to be propelled by wind and waves, allowing it to passively catch and concentrate plastic debris in front of it.
“There are many tasks that must be accomplished during the Pacific Trials, but we have identified five main goals for the coming two weeks. Confirming these objectives will provide us with the understanding if the system is up for the challenge it’s set to face in the patch. Should we encounter any issues, it is much easier to tow the system back to shore from here than it would be all the way from the patch,” the Ocean Cleanup said in an update.

“We want to ensure the system can withstand the conditions of the ocean before we tow it all the way to the patch. We will perform checks at the end of the trials for damage. If we observe severe damage, we will then assess if this can be remedied on or offshore.”
The checklist
The first step is U-shape installation, which has been completed successfully. It is the first time the system is in its intended shape, which is needed to conduct many tests.
This required connecting the closing lines, which help the system maintain its shape and prevent the system from flipping inside out.

It took approximately two days to arrange the system in its operational configuration. With the aid of the Maersk Launcher and the Megamaid, the closing lines were connected from the inside of System 001....MORE
The two most recent posts at TheOceanCleanup's Updates site:

Research 25 September 2018
Remote Sensing of Ocean Plastics

General, Testing 19 September 2018
Pacific Trials - The Gateway to the Great Pacific Garbage Patch

"Norwegian gas exports to [the rest of] Europe jump"

This is positive for both buyer and seller. However...the onus is on Norway to find the stuff. They know the reserves are out there but the easy-to-recover reserves were found 20-30 years ago and current targets are a lot more challenging.

And the buyers had better realize, as the Poles have, that in addition to the BTUs or therms you are also buying a bit of supply insurance with your slightly higher prices.
It's probably better to do your next-decade negotiations in Norwegian than in Russian.

From LNG World News:
Gassco, the company running the Norwegian gas infrastructure, exported 36.8 billion cubic meters of natural gas during the May-August period hitting a new peak for summer deliveries. 
The operator noted this was a small rise from the same period of 2017, which was a record year for gas exports.

“The basis for these high deliveries is naturally European and British demand for Norwegian gas,” said Gassco CEO Frode Leversund.

More gas is usually exported in the winter months when demand for space heating increases, but this summer’s sales have almost matched a normal winter level.

Gassco transported 117.4 billion scm through the pipeline systems from Norway to continental Europe and the UK during 2017. This was clearly the highest volume sold in 40 years of Norwegian gas exports, and nine times Norway’s annual hydropower output....MORE

"As God Is My Witness I Thought Turkey's Could Fly"

I seem strangely drawn to scams this year—I mean more so than usual—and if you throw in Mount Everest you have my attention.
From FT Alphaville:

The ICO behind the tragic Everest stunt is now “airdropping” tokens from rockets
Remember ASKfm, the Ukraine-based social network that was encouraging people to climb Mount Everest in search of crypto tokens, while not telling them that a man who was guiding the climb, Lam Babu Sherpa, had died on the descent?
Unbelievably, we received an email with the following subject line on Sunday:
ASKfm kicks off a new stunt - this time without killing a sherpa
There is, truly, no shame in crypto publicity.
ASKfm's PR representative told us they thought the email title might “grab [our] attention” which, we must say, it did.
Then we got to the body of the email (emphasis theirs):
ASKfm's airdrop is quite literal as the ASQ Protocol - part of the ASKfm network - has taken its token into the troposphere with its most recent airdrop for users and the wider crypto community. 
ASQ has launched a rocket into the skies above Scotland loaded with a Ledger Nano S that was dropped somewhere in the countryside. The launch vehicle, operated by Edinburgh-based space technology firm Skyrora, reached Mach 1.45 and a maximum altitude of over 6 kilometers before unloading its precious cargo into the world. 
That's right -- ASKfm have once again gone for the making-a-metaphor-literal promo stunt.
The thinking behind burying the tokens at the top of Everest was that they would then be at “the closest starting point to reach the moon” -- a reference to the “to the moon” rallying cry of crypto HODLers worldwide....

The headline is from a U.S. television show popular in the 1980's, "WKRP in Cinncinnatti" which some of our octo- and nona-genarian readers may recall. This short clip has one of the better popular homages to the Hindenburg disaster, not that their are all that many, but something Alphaville also brought up in yesterday's "Beware the Hindenburg Omen?".

If interested see also Saturday's "A scam on the roof of the world".

Monday, September 24, 2018

Shipping: "CMA CGM Says Global Sulphur Cap to Cost Customers $160 Per Container on Average"

That's serious money, say on a 10,000 TEU average voyage, $1.6 million per trip?
Jeez, no wonder the industry is freaking out.

From gCaptain:
French shipping giant CMA CGM on Monday said the 2020 global sulphur cap regulation will cost an estimated $160 per twenty-foot container or equivalent unit (TEU) to be paid by its customers in the form of a fuel surcharge.

CMA CGM, the world’s fourth-largest container shipping line by overall TEU volume, is the latest of the major ocean carriers to reveal how it intends to handle the International Maritime Organization’s low sulphur fuel regulation entering into force beginning January 1, 2020. The regulation aims to reduce the environmental impact of the shipping industry by lowering the sulphur content in the fuel burned by ships to a maximum 0.5%. Ships can also comply by installing exhaust gas scrubbers or by switching to low sulphur alternative such as LNG.

To comply with the impending regulation, CMA CGM says has opted to favor the use low sulphur bunker fuel, but it will also invest significantly in LNG fuel for its future newbuildings and order several scrubbers for some of its ships....MORE
Perhaps time to revisit Shipping: CEO of Third Largest Fleet Says "We're All Going to Go Bust".

See also:
Top Norwegian Oil Analyst Quitting DNB to Pursue 2020 Low Sulphur Fuel Rule Riches

"Maersk Asks Customers to Pay for $2 Billion Low Sulphur Fuel Bill Through New ‘Bunker Adjustment Factor’"

Shipping: "Hapag-Lloyd Expects USD 1 Bn in Extra Fuel Costs from 2020 Sulphur Cap"

Shipping: "Testing Begins On First Product Tanker Vessel Utilising Wind Propulsion Technology"

Shipping: "More than 2000 vessels will get scrubbers before 2020"

Shipping: "Why CMA CGM ordered 'game changing' 22,000 TEU LNG-powered containerships"

And many more, it's a pretty big deal.

‘We Are All Accumulating Mountains of Things’ How online shopping and cheap prices are turning Americans into hoarders

Laurene Powell Jobs has been accumulating as well. She bought the lot next to her Florida house for $8 million. a place in San Francisco's Russian Hill neighborhood for $16 million, the house next door to her Malibu compound for $8 million, and the biggie: 29.4 acres in the heart of Silicon Valley to add to her 16 East Palo Alto acres.

In addition she purchased a majority stake in The Atlantic last year.
All of which leaves her net worth at $20.6 billion according to Forbes' real-time estimate, Sept 24.

And speaking of the Atlantic, here's their Aug. 21 story:

How online shopping and cheap prices are turning Americans into hoarders
It’s easier than ever to buy things online. It’s so easy that Ryan Cassata sometimes does it in his sleep. Cassata, a 24-year-old singer-songwriter and actor from Los Angeles, recently got a notification from Amazon that a package had been shipped to his apartment, but he didn’t remember buying anything. When he logged onto his account and saw that a fanny pack and some socks were on the way, he remembered: A few nights back, he had woken up in the middle of the night to browse—and apparently shop on—Amazon.

He shops when he’s awake, too, buying little gadgets like an onion chopper, discounted staples like a 240-pack of gum, and decorations like a Himalayan salt lamp. The other day, he almost bought a pizza pool float, until he remembered that he doesn’t have a pool. “I don’t really need most of the stuff,” he tells me.
Thanks to a perfect storm of factors, Americans are amassing a lot of stuff. Before the advent of the internet, we had to set aside time to go browse the aisles of a physical store, which was only open a certain number of hours a day. Now, we can shop from anywhere, anytime—while we’re at work, or exercising, or even sleeping. We can tell Alexa we need new underwear, and in a few days, it will arrive on our doorstep. And because of the globalization of manufacturing, that underwear is cheaper than ever before—so cheap that we add it to our online shopping carts without a second thought. “There’s no reason not to shop—because clothing is so cheap, you feel like, ‘Why not?’ There’s nothing lost in terms of the hit on your bank account,” Elizabeth Cline, the author of Overdressed: The Shockingly High Cost of Cheap Fashion, told me.

Shopping online also feels good. Humans get a dopamine hit from buying stuff, according to research by Ann-Christine Duhaime, a professor of neurosurgery at Harvard Medical School. “As a general rule, your brain tweaks you to want more, more, more—indeed, more than those around you—both of ‘stuff’ and of stimulation and novelty —because that helped you survive in the distant past of brain evolution,” Duhaime wrote in a Harvard Business Review essay last year. Online shopping allows us to get that dopamine hit, and then also experience delayed gratification when the order arrives a few days later, which may make it more physiologically rewarding than shopping in stores....MUCH MORE

Oh Great, Now Anak Krakatau (Child of Krakatoa) Is Erupting

Actually it's not a big deal (ye0. I wanted an excuse to link to MAGMA Indonesia (Multiplatform Application for Geohazard Mitigation and Assessmen)
First up, from the Jakarta Post, September 24:

Flights not yet affected by Mt. Anak Krakatau eruption, authorities say
 Anak Krakatau is seen from Japanese helicopter carrier Kaga in the Indian Ocean on Sept. 22. (Reuters/Kim Kyung-Hoon)
Authorities say flights have not been affected by the eruption of Mount Anak Karakatau in the Sunda Strait, South Lampung regency, Lampung.

The Meteorology, Climatology and Geophysics Agency (BMKG) reported that the volcano erupted over 44 times over the weekend,  with thick smoke billowing from the volcano’s crater at a height of 200 meters from the crater's peak.

The agency also recorded 253 incidences of seismic activity, prompting local authorities to warn tourists and residents to steer clear of the volcano and its surrounding areas.

BMKG spokesperson Hary Djatmiko told The Jakarta Post that as of Monday morning, the volcano was still spewing volcanic ash into the air.

“Based on satellite imagery, the volcanic ash has reached 7,000 feet and is moving west to Sumatra island,” he said on Monday....MORE
Yeah, a volcano inside a water-filled crater caldera inside a volcano.
Here's what Indonesia has to deal with on the volcano and earthquake front, (screenshot from MAGMA):


Here's MAGMA's interactive site. The last time i checked there was a 5.0 earthquake going on, apparently not a big deal.

And the September 23 warning.

"Drivers for Uber, Lyft are earning less than half of what they did four years ago, study finds"

The MIT study from February was mistaken and garnered our standard correction:

You Trusted Me and I Failed You

When a journalist accidently misquotes a source

Regret the error.

So you'll understand a bit of hesitance linking to this piece.
Oh who am I kidding, if a story shows a hole in the Ubesters business plan we'll probably run with it.

From Dow Jones' MarketWatch:

Drivers earned 53% less in 2017 than they did in 2013 
Drivers for online platforms including Uber and Lyft are making less than half of what they did four years ago, even as more and more people are drawn into working for them.

A new report from the JPMorgan Chase Institute, based on payments directed to 2.3 million families, showed that average monthly platform earnings dropped considerably — by 53% — between 2013 and 2017.

These drivers made $783 per month in 2017 versus $1,469 in 2013.

In the first quarter of 2014, nearly half of drivers who work for on demand platform such as Uber and Lyft, earned $900 a month or more. However during the first quarter of 2018, less than 25% of drivers were able to earn more than $900, the report found.

“These declines in monthly earnings among drivers may reflect the fact that the growth in the number of drivers could have put downward pressure on hourly wages; they may also reflect a potential decline in the number of hours drivers are driving,” the report stated.

“Regardless of whether the drop in earnings was caused by a fall in wages or hours or both, it indicates that driving has become less and less likely to replace a full-time job over the past five years, as more drivers have joined the market.”...MORE
Attentive readers, i.e. those not distracted by the correction .gif, have probably spotted at least two errors in methodology and/or presentation of the study but as I said, we'll run with it.

"News Site to Investigate Big Tech, Helped by Craigslist Founder"

How did  Craig come up with $20 mil. to give away?
Kids, now do you understand how profitable classified advertising used to be for newspapers?
And Craigslist is, for the most part, free.

Enough digression, on to investigation.

From the New York Times, Sept. 23:

The Markup, dedicated to investigating technology and its effect on society, will be led by two former ProPublica journalists. Craig Newmark gave $20 million to help fund the operation. 
When the investigative journalist Julia Angwin worked for ProPublica, the nonprofit news organization became known as “big tech’s scariest watchdog.” 

By partnering with programmers and data scientists, Ms. Angwin pioneered the work of studying big tech’s algorithms — the secret codes that have an enormous impact on everyday American life. Her findings shed light on how companies like Facebook were creating tools that could be used to promote racial bias, fraudulent schemes and extremist content. 

Now, with a $20 million gift from the Craigslist founder Craig Newmark, she and her partner at ProPublica, the data journalist Jeff Larson, are starting The Markup, a news site dedicated to investigating technology and its effect on society. Sue Gardner, former head of the Wikimedia Foundation, which hosts Wikipedia, will be The Markup’s executive director. Ms. Angwin and Mr. Larson said that they would hire two dozen journalists for its New York office and that stories would start going up on the website in early 2019. The group has also raised $2 million from the John S. and James L. Knight Foundation, and $1 million collectively from the Ford Foundation, the John D. and Catherine T. MacArthur Foundation, and the Ethics and Governance of Artificial Intelligence Initiative.
Ms. Angwin compares tech to canned food, an innovation that took some time to be seen with more scrutiny.

“When canned food came out, it was amazing,” said Ms. Angwin, who will be the site’s editor in chief. “You could have peaches when they were out of season. There was a whole period of America where every recipe called for canned soup. People went crazy for canned food. And after 30 years, 40 years, people were like, ‘Huh, wait.’

“That is what’s happened with technology,” Ms. Angwin said, calling the 2016 election a tipping point. “And I’m so glad we’ve woken up.”

The site will explore three broad investigative categories: how profiling software discriminates against the poor and other vulnerable groups; internet health and infections like bots, scams and misinformation; and the awesome power of the tech companies. The Markup will release all its stories under a creative commons license so other organizations can republish them, as ProPublica does....

News You Can Use: "Google’s AI to detect toxic comments can be easily fooled with ‘love’"

From The Next Web:
A group of researchers has found that simple changes in sentences and its structure can fool Google’s perspective AI, made for detecting toxic comments and hate speech. These methods involve inserting typos, spaces between words or add innocuous words to the original sentence.

The AI project, which was started in 2016 by a Google offshoot called Jigsaw, assigns a toxicity score to a piece of text. Google defines a toxic comment as a rude, disrespectful, or unreasonable comment that is likely to make you leave a discussion. The researchers suggest that even a slight change in the sentence can change the toxicity score dramatically. They saw that changing “You are great” to “You are fucking great”, made the score jump from a totally safe 0.03 to a fairly toxic 0.82.
Simple changes in the sentence fools Google AI
This clearly denotes that the toxicity score is probably not the best measure to identify hate speech. Last year, another study found that inserting spaces and making typos reduced the toxicity score drastically. Google has improved its AI since then to detect these changes. But it’s not perfect, the researchers presenting the latest study said if someone introduced a word like ‘love’ in these sentences the score took a plunge....MORE

Why Are People Linking to a Four-Year Old Story On Anti-Vaxxers?

The story is from the Atlantic:
Wealthy L.A. Schools' Vaccination Rates Are as Low as South Sudan's
And is dated:
Sep 16, 2014
Its source is The Hollywood Reporter story "Hollywood's Vaccine Wars..." which itself is dated
9/10/2014 9:00 am PDT 
Since last week it has been linked in a half dozen places.

None of the linkers mention the date and none of the linkers dig further into the story with updates on the effects of the California immunization law which actually is a story:
Medical Xpress: "Vaccine opt-outs dropped slightly when California added more hurdles"
KPDB: "Medical Exemptions Up As Vaccination Rates For Kindergartners Decline In State, San Diego County"
and a bit spookier from The Sacramento Business Journal:
ICAN vs. HHS: Key Legal Win Recasts Vaccine Debate

What's up?
You've got record measles outbreaks in Europe and mumps in the U.S. who (WHO?) needs old news?

"How Jim Chanos Uses Cynicism, Chutzpah — and a Secret Twitter Account — to Take on Markets (and Elon Musk)"

From Institutional Investor, September 17, 2018:

The LeBron James of short-selling talks Ponzinomics.
It’s a sweltering, 95-degree August day in Manhattan, but Jim Chanos — fresh off a two-week holiday, rocking a sharkskin suit — is pumped: Elon Musk had once again called a hero of the Thai cave rescue a pedophile.It’s 1:30 in the afternoon. Chanos bolts through the door to his office building on West 55th Street, grabs the journalist waiting for him, and, on the elevator ride to his eighth-floor office, fills her in on the latest news.

“Musk is at it again. He’s doubling down on his pedophile comment,” Chanos says, amusement noticeable in his voice. Minutes earlier the CEO of Tesla Motors — who is already under investigation by the Securities and Exchange Commission for a tweet about taking the company private — had returned to an earlier tweetstorm in which he accused the British cave diver of being a “pedo guy”.

“You don't think it’s strange he hasn’t sued me?" Musk tweeted at 12:30 p.m. on August 28. Tesla shares immediately began to slide, and within hours the diver’s lawyer reportedly informed Musk that he would, indeed, be sued.

Recent months have seen one bizarre event after another in the saga of the electric-automaker and its iconic boss, and Chanos has been watching it all. The 61-year-old founder of hedge fund Kynikos Associates announced his short against Tesla in the fall of 2015 and has been nursing losses ever since — for despite a raft of bad news and the increasingly controversial, and potentially illegal, behavior of Musk, Tesla’s stock, while off significantly from its highs, has stubbornly refused to crash.

The Tesla short made for a tense working vacation for Chanos, who had left New York for his annual trek to the Greek island of Mykonos on August 9. That was just two days after Musk tweeted that he was considering taking Tesla private — with the now-infamous “funding secured” line — and Wall Street went into a frenzy trying to see if it could happen.

Chanos was skeptical.

“I thought immediately it was not true, because I know Elon Musk at this point,” he said during a two-hour interview that touched on everything from hedge funds and short-selling to politics and the history of fraud, then, eventually, got back to social media and Musk. “The way it came out during market hours, right after the Saudi passive investment announcement [the Saudi sovereign wealth fund has reportedly taken a 5 percent stake in Tesla] — it was so irresponsible, number one, and was not clearly vetted, because no one who vetted that would’ve let him put that out.”

Although Kynikos’s two senior portfolio partners, Chuck Hobbs and David Glaymon, were monitoring the stock from New York, Chanos couldn’t stay away from the news, even in Mykonos. “It was breaking so fast,” Chanos says. As the stock rallied, he sold more shares short.
Finally, at 11:00 p.m. on Friday, August 24, Musk ended the suspense with a statement on Tesla’s blog. He had decided against going private.

Tesla has been the biggest short in the market for much of this past year, with nearly $10 billion wagered that its stock will fall. Chanos is hardly the only one who takes that view. Kynikos now runs less than $2 billion, and with individual short positions capped at 5 percent each for risk management purposes, Tesla represents at most about $100 million of the firm’s total short bets, which it currently has on 65 companies.

Win or lose, the Tesla short won’t make much of a difference to Kynikos’s bottom line. But for Chanos — who claims he’s also advising much larger “pools of capital” on their Tesla shorts — there’s more than money at stake.

Kynikos is the lone short-selling hedge fund of any size — and the only one that that has been in business since 1985.

That’s no mean feat, as the stock market has risen almost 1,500 percent since Kynikos’s debut. Chanos has lived through at least three bear markets — the 1987 crash, the bursting of the dot-com bubble in 2000, and the global financial crisis of 2008 — and he’s waiting patiently for the next one.
“I’m not saying it’s the top of the market, but on the other hand, bear markets haven’t been outlawed,” he quips.

Chanos, of course, is already a legend. He will go down in Wall Street history for predicting the demise of Enron Corp., whose collapse resulted in a wave of prosecutions and the imprisonment of top executives — the kind of harsh penalties that have not been seen since.
But that was in 2001 — 17 years ago.

Since the financial crisis year of 2008, Kynikos’s shorts overall have been bleeding red ink, and investors have bailed. Kynikos has lost almost three quarters of its assets since the end of 2008. This year alone, its funds had fallen between 9 percent and 19 percent through July (net of fees), depending on the fund, according to a report to investors Institutional Investor has seen that has not been previously reported on.

“It has been one giant short-squeeze market,” he moans.

Take Tesla. To Chanos it represents a market euphoria last witnessed during the dot-com craze. Although Tesla is unprofitable and loaded with debt, its market cap rivals that of General Motors. A rush to meet production targets on its midprice Model 3 sedan this year has been accompanied by defective cars, unhappy employees-turned-whistleblowers, and a stream of exiting executives — not to mention ever-stranger outbursts by a CEO under enormous pressure to meet the projections he promised shareholders. The current stock price of about $299 per share (above Chanos’s average cost of $250) is based on ambitious plans for the future. But what most people are missing, says Chanos, is that Tesla has quit making the capital investments required to realize those ambitions. It can no longer afford to do so.

But facts don’t seem to matter. As Chanos reminds us, referring to President Trump and his supporters, we are living in “a post-truth environment.” That translates, he argues, into a mood in which investors are also willing to suspend disbelief.

“If we don’t hold our leaders to that standard,” muses Chanos, “then why should we hold managements to that standard? I think that’s part of where we are now.”
Chanos, who jokes that he is considered an anarchist on Wall Street for his liberal political views, is not prone to proselytizing. He doesn’t write letters to the Securities and Exchange Commission or publish long treatises on his short theses. Though he believes the political environment has an impact on the market, he never counts on the SEC to take down the bad guys. “They’re archaeologists, not detectives,” he scoffs.

Chanos can appear world-weary and somewhat guarded — but, as the saying goes, in every cynic beats the bleeding heart of an idealist. “There’s more than a little of the crusader in him,” says longtime friend Jim Grant, founder and editor of Grant’s Interest Rate Observer. “He would like to clean up Wall Street. He would like to improve the quality of corporate reporting. He would like to rid Wall Street of the scoundrels and clean up corporate management.”

That might explain what happened earlier this summer....

HT: Barry Ritholtz but we don't have a specific post. So here's The Big Picture.

Luxembourg's ^#@*&! Space Agency and Fund

This is a couple weeks old but we've been keeping track of  the goings-on in the Grand Duchy for quite a while (some links below) and thought this should be on the blog.

From Space News, September 13:

Luxembourg establishes space agency and new fund
The government of Luxembourg continued work to expand its role in the global space economy Sept. 12 by formally establishing a national space agency, a move designed in part to ensure the effort continues after an upcoming election.

In a ceremony in Luxembourg City, Étienne Schneider, deputy prime minister and minister of the economy, formally announced the creation of the Luxembourg Space Agency. The agency will be led by Marc Serres, previously the head of space affairs at the Ministry of the Economy.

Unlike traditional national space agencies, which support spacecraft missions and scientific research, the Luxembourg Space Agency will focus primarily on building up the country’s space industry as well as supporting education and workforce development.

Schneider noted that Luxembourg’s recent efforts, most notably the SpaceResources.lu project to attract companies working in the nascent space resources field, had led to 20 countries establishing a presence in the country. “All this is why it’s so important to me to launch today this Luxembourg Space Agency in order to professionalize our approach to this new community,” he said.

Serres said that the agency will work with a wide range of other organizations, both within the government and the private sector, to meet the agency’s goals. “The agency will be well-equipped to support industry in their daily challenges, and it leads to the most favorable environment for this sector to continue to grow,” he said, describing its four “strategic lines” as expertise, innovation, skills and funding.

That last item will include a new fund for supporting space companies. Schneider announced that the space agency will work with other government agencies and the private sector to establish the Luxembourg Space Fund, valued at 100 million euros ($116 million). The fund, according to a government statement, will “provide equity funding for new space companies with ground-breaking ideas and technology.”

Only part of the new fund will involve government money. “It will be a public-private partnership, where the government will take a share of 30 to 40 percent,” Schneider said....MORE
In more mundane Luxembourg news:
EU says McDonald's, Luxembourg tax deal not illegal

Also on the Luxembourg Channel:

Goldman Sachs on Asteroid Mining: As If Luxembourg Wasn't Insufferable Already
Luxembourg, with their #3 in the world GDP per capita (PPP) and their Jean-Claude Juncker and...
Insufferable. It's like Bono and Elevation Partners. Besides being Bono, they ran $90 mil. to $1.5 billion in Facebook.
Billion with a 'B'. As in Bono....
[re the Goldman bit]
...It's a 24 minute podcast, something one could listen to while in transit but did anyone bother to tell me about it?
No. That's why we're getting to it a year late, despite my avowed interest in becoming the world's first trillionaire.
And ignoring that Alchemist's Fallacy thing and all.

Here's ZeroHdge from April 19th (yes, they're a year late as well but how does that help moi?): 
The World's First Trillionaire Will Be A Space Miner 
Huh, I must have been crabby that day.
Anyhoo, onwards:

Dammit! Luxembourg Is Not A Microstate!
From Brilliant Maps:

The map above shows how big Luxembourg is compared to Singapore, Andorra, Malta, Liechtenstein, San Marino, Monaco, and the Vatican.

Luxembourg is 2,586.4 km2 (998.6 sq mi) with a population of 576,249, making it one of the world’s smallest states; 168th by size or 164th by population.

However, it’s still bigger than Singapore, Andorra, Malta, Liechtenstein, San Marino, Monaco, and the Vatican combined!

Here’s how big they are respectively:...MORE
Crabby that day as well.

Previously on the wonder that is Luxembourg:
What's the Scam? Why Did Deloitte Set Up Their Art & Finance Practice In Luxembourg?
Luxembourg’s Asteroid Mining Plan
Luxembourg Invests €25 million in Asteroid Mining 
Luxembourg’s Bid to Become the Silicon Valley of Space Mining
Luxembourg's New Space Mining Law Is Basically "Finders, Keepers"
"Oligarchs and Orchestras: Inside Luxembourg’s Secretive Low-Tax ‘Fortress of Art’ Warehouse"
Uh Oh: "Bail-In Blues: Luxembourg Warns of Investor Flight from Europe"
Luxembourg and Switzerland are two of the "Banking-assets-an-order-of-magnitude-bigger-than-GDP" powerhouses....
Short The Swiss (and Luxembourg)
Maybe Liechtenstein too. Never much cared for Doha either. And then there's...
errrmmm, excuse me.... 

Luxembourg-based Rare Earth Company Hoping to Mine in South Africa by 2014 Does Oversubscribed IPO in Toronto (FRO.tsx)

FREFF 0.012  0.00

For some reason posting on Luxembourg seems to bring out the worst in me.

Finland's Dry Bulk Carrier ESL Shipping Claims First LNG-Powered Handysize, First Northern Sea Route Transit

And speaking of Russia and natural gas...
From the Barents Observer, Sept. 21:

World’s first LNG-fueled bulk carriers cross Northern Sea Route
The two brand new 160-meter long vessels that are making it through the Russian Arctic shipping route produce less than half of normal vessels’ carbon dioxide emissions.  

 The "Haaga" sails from Japan to Sweden through the NSR. Photo: ESL Shipping
The 25,600 dwt sister ships «Haaga» and «Viikki» are getting their maiden voyage through some of the toughest waters on earth. The two ships built in China for Finnish company ESL Shipping will make it to their home port in Finland through the Northern Sea Route.

«They are the world’s most environmentally friendly bulk carriers, and their passage via the Northern Sea Route is a concrete indication of the impact of climate change - but also of new business opportunities,» company ESL Shipping says.

According to information from the Northern Sea Route Administration, the «Haaga» on the 14th September sailed into the eastern part of the Arctic shipping route and was on 21st September anticipated to pass archipelago Novaya Zemlya with course for Norwegian waters.

The ship is built by the Jinling shipyard in Nanjing, China, and loaded raw material cargo in Japan before it set course for the Arctic. It is estimated to arrive in the Swedish port of Oxelösund around 1st October.

According to company Managing Director Mikki Koskinen, the ship had Russian icebreaker assistance when it crossed the East Siberian Sea. Ice data from the Russian Arctic and Antarctic Research Institute show that the remaining parts of the NSR now are practically all ice-free.
The second ship, the «Viiki», was delivered by the yard on 4th September and is expected to start its Arctic voyage in about two weeks....MORE

Natural Gas: "EU Taxpayers Aren't Funding Nord Stream 2 So Why The Hubbub"—Gazprom

Interesting pitch from Sputnik:
Sweet, innocent, a bit like putting this guy in the role of the ingénue:


EU Taxpayers Not Funding Nord Stream 2, Political Tension Surprising - Gazprom
EU taxpayers are not funding the Nord Stream 2 gas pipeline, Russia's Gazprom Export Director General Elena Burmistrova has stated, stressing that the political tensions surrounding the project were surprising.
"We really feel quite a political tension, not to mention our biggest infrastructure project such as Nord Stream 2. For me, it is quite a surprise because I see a lot of debates in this respect. But at the same time, all of us understand that it is a 100 percent commercial project and it is not paid by taxpayers of the European Union, it is a 100 percent commercial capital that was invested there," Burmistrova said during a session at the Gastech conference in Barcelona.
Some of the European countries have repeatedly expressed their concerns over the project, while others, such as Germany and Austria, have welcomed the new pipeline.

Nord Stream 2 AG is owned by Public Joint Stock Company Gazprom, while its investors include France's Engie, Austria's OMV AG, UK-Dutch Royal Dutch Shell, and Germany's Uniper and Wintershall. The pipeline is expected to bring gas from Russia to Germany via the Baltic Sea.

The project has been welcomed by some countries in Europe, such as Germany and Austria, and opposed by others, including Ukraine, which may suffer transit revenue drops if Nord Stream 2 becomes operational.

Energy Ties With China Will Not Affect European Supplies
Russian energy giant Gazprom's supply of gas to China, including via western route, will not affect the volume of the company's deliveries to Europe, Burmistrova stated.

Earlier in September, an official from China's National Energy Administration (NEA) said that Beijing intended to increase imports of natural gas from Russia and Kazakhstan in the near future. Also this month, Russian energy company Rosneft and Chinese National Oil and Gas Corporation (CNPC) agreed on broader cooperation in the area of exploration and extraction.
"I would not say it would affect our relationship with Europe. It has always been our natural customer, a partner as well, geographically. We are lucky to be just in the middle. China is our new partner and we are happy that our project is finally alive and we can talk about new options. It will be different fields and I do not think it will somehow influence our European supplies," Burmistrova said.
In 2014, Gazprom and China's CNPC signed a framework agreement on the delivery of 38 billion cubic meters of natural gas annually. The agreement stipulates the construction of appropriate infrastructure for the gas deliveries....MORE
Re: the pic in the intro, if you're thinking "I recognize the face but can't quite recall the name", it's Russia's Sergey Shoigu.
I was thinking of him this weekend after his U.S. counterpart, Secretary of Defense Mattis said “What I want our adversary to know is please work with our State Department. You really don’t want to work with me. That’s our message.”

Sunday, September 23, 2018

Wow: Anatomy of an AI System (everything that goes into Amazon's Echo)

This is a stunning piece of research.
note: this version of the map is via the Verge, HT up fron to Professor Crawford's twitter feed which points up a giant wall-sized copy at the V and A.

From NYU's Kate Crawford:

The Amazon Echo as an anatomical map of human labor, data and planetary resources
By Kate Crawford 1 and Vladan Joler 2


Download map in pdf format here


A cylinder sits in a room. It is impassive, smooth, simple and small. It stands 14.8cm high, with a single blue-green circular light that traces around its upper rim. It is silently attending. A woman walks into the room, carrying a sleeping child in her arms, and she addresses the cylinder.
‘Alexa, turn on the hall lights’

The cylinder springs into life. ‘OK.’ The room lights up. The woman makes a faint nodding gesture, and carries the child upstairs.

This is an interaction with Amazon’s Echo device. 3 A brief command and a response is the most common form of engagement with this consumer voice-enabled AI device. But in this fleeting moment of interaction, a vast matrix of capacities is invoked: interlaced chains of resource extraction, human labor and algorithmic processing across networks of mining, logistics, distribution, prediction and optimization. The scale of this system is almost beyond human imagining. How can we begin to see it, to grasp its immensity and complexity as a connected form? We start with an outline: an exploded view of a planetary system across three stages of birth, life and death, accompanied by an essay in 21 parts. Together, this becomes an anatomical map of a single AI system.

Amazon Echo Dot (schematics)
Amazon Echo Dot (schematics)


The scene of the woman talking to Alexa is drawn from a 2017 promotional video advertising the latest version of the Amazon Echo. The video begins, “Say hello to the all-new Echo” and explains that the Echo will connect to Alexa (the artificial intelligence agent) in order to “play music, call friends and family, control smart home devices, and more.” The device contains seven directional microphones, so the user can be heard at all times even when music is playing. The device comes in several styles, such as gunmetal grey or a basic beige, designed to either “blend in or stand out.” But even the shiny design options maintain a kind of blankness: nothing will alert the owner to the vast network that subtends and drives its interactive capacities. The promotional video simply states that the range of things you can ask Alexa to do is always expanding. “Because Alexa is in the cloud, she is always getting smarter and adding new features.”

How does this happen? Alexa is a disembodied voice that represents the human-AI interaction interface for an extraordinarily complex set of information processing layers. These layers are fed by constant tides: the flows of human voices being translated into text questions, which are used to query databases of potential answers, and the corresponding ebb of Alexa’s replies. For each response that Alexa gives, its effectiveness is inferred by what happens next:
Is the same question uttered again? (Did the user feel heard?)

Was the question reworded? (Did the user feel the question was understood?)

Was there an action following the question? (Did the interaction result in a tracked response: a light turned on, a product purchased, a track played?)
With each interaction, Alexa is training to hear better, to interpret more precisely, to trigger actions that map to the user’s commands more accurately, and to build a more complete model of their preferences, habits and desires. What is required to make this possible? Put simply: each small moment of convenience – be it answering a question, turning on a light, or playing a song – requires a vast planetary network, fueled by the extraction of non-renewable materials, labor, and data. The scale of resources required is many magnitudes greater than the energy and labor it would take a human to operate a household appliance or flick a switch. A full accounting for these costs is almost impossible, but it is increasingly important that we grasp the scale and scope if we are to understand and govern the technical infrastructures that thread through our lives.


The Salar, the world's largest flat surface, is located in southwest Bolivia at an altitude of 3,656 meters above sea level. It is a high plateau, covered by a few meters of salt crust which are exceptionally rich in lithium, containing 50% to 70% of the world's lithium reserves. 4 The Salar, alongside the neighboring Atacama regions in Chile and Argentina, are major sites for lithium extraction. This soft, silvery metal is currently used to power mobile connected devices, as a crucial material used for the production of lithium-Ion batteries. It is known as ‘grey gold.’ Smartphone batteries, for example, usually have less than eight grams of this material. 5 Each Tesla car needs approximately seven kilograms of lithium for its battery pack. 6 All these batteries have a limited lifespan, and once consumed they are thrown away as waste. Amazon reminds users that they cannot open up and repair their Echo, because this will void the warranty. The Amazon Echo is wall-powered, and also has a mobile battery base. This also has a limited lifespan and then must be thrown away as waste.

According to the Aymara legends about the creation of Bolivia, the volcanic mountains of the Andean plateau were creations of tragedy. 7 Long ago, when the volcanos were alive and roaming the plains freely, Tunupa - the only female volcano – gave birth to a baby. Stricken by jealousy, the male volcanos stole her baby and banished it to a distant location. The gods punished the volcanos by pinning them all to the Earth. Grieving for the child that she could no longer reach, Tunupa wept deeply. Her tears and breast milk combined to create a giant salt lake: Salar de Uyuni. As Liam Young and Kate Davies observe, “your smart-phone runs on the tears and breast milk of a volcano. This landscape is connected to everywhere on the planet via the phones in our pockets; linked to each of us by invisible threads of commerce, science, politics and power.” 8


Our exploded view diagram combines and visualizes three central, extractive processes that are required to run a large-scale artificial intelligence system: material resources, human labor, and data. We consider these three elements across time – represented as a visual description of the birth, life and death of a single Amazon Echo unit. It’s necessary to move beyond a simple analysis of the relationship between an individual human, their data, and any single technology company in order to contend with with the truly planetary scale of extraction. Vincent Mosco has shown how the ethereal metaphor of ‘the cloud’ for offsite data management and processing is in complete contradiction with the physical realities of the extraction of minerals from the Earth’s crust and dispossession of human populations that sustain its existence. 9 Sandro Mezzadra and Brett Nielson use the term ‘extractivism’ to name the relationship between different forms of extractive operations in contemporary capitalism, which we see repeated in the context of the AI industry. 10 There are deep interconnections between the literal hollowing out of the materials of the earth and biosphere, and the data capture and monetization of human practices of communication and sociality in AI. Mezzadra and Nielson note that labor is central to this extractive relationship, which has repeated throughout history: from the way European imperialism used slave labor, to the forced work crews on rubber plantations in Malaya, to the Indigenous people of Bolivia being driven to extract the silver that was used in the first global currency. Thinking about extraction requires thinking about labor, resources, and data together. This presents a challenge to critical and popular understandings of artificial intelligence: it is hard to ‘see’ any of these processes individually, let alone collectively. Hence the need for a visualization that can bring these connected, but globally dispersed processes into a single map.
Extractive operations


If you read our map from left to right, the story begins and ends with the Earth, and the geological processes of deep time. But read from top to bottom, we see the story as it begins and ends with a human. The top is the human agent, querying the Echo, and supplying Amazon with the valuable training data of verbal questions and responses that they can use to further refine their voice-enabled AI systems. At the bottom of the map is another kind of human resource: the history of human knowledge and capacity, which is also used to train and optimize artificial intelligence systems. This is a key difference between artificial intelligence systems and other forms of consumer technology: they rely on the ingestion, analysis and optimization of vast amounts of human generated images, texts and videos.
...MUCH MORE (an incredible amount of work went into this)

Why the World Only Has Two Words For Tea

From Quartz:

Tea if by sea, cha if by land: Why the world only has two words for tea
With a few minor exceptions, there are really only two ways to say “tea” in the world. One is like the English term— in Spanish and tee in Afrikaans are two examples. The other is some variation of cha, like chay in Hindi.

Both versions come from China. How they spread around the world offers a clear picture of how globalization worked before “globalization” was a term anybody used. The words that sound like “cha” spread across land, along the Silk Road. The “tea”-like phrasings spread over water, by Dutch traders bringing the novel leaves back to Europe.
The term cha (茶) is “Sinitic,” meaning it is common to many varieties of Chinese. It began in China and made its way through central Asia, eventually becoming “chay” (چای) in Persian. That is no doubt due to the trade routes of the Silk Road, along which, according to a recent discovery, tea was traded over 2,000 years ago. This form spread beyond Persia, becoming chay in Urdu, shay in Arabic, and chay in Russian, among others. It even made its way to sub-Saharan Africa, where it became chai in Swahili. The Japanese and Korean terms for tea are also based on the Chinese cha, though those languages likely adopted the word even before its westward spread into Persian.

But that doesn’t account for “tea.” The Chinese character for tea, 茶, is pronounced differently by different varieties of Chinese, though it is written the same in them all. In today’s Mandarin, it is chá. But in the Min Nan variety of Chinese, spoken in the coastal province of Fujian, the character is pronounced te. The key word here is “coastal.”

The te form used in coastal-Chinese languages spread to Europe via the Dutch, who became the primary traders of tea between Europe and Asia in the 17th century, as explained in the World Atlas of Language Structures....MORE
If interested see also:
The Great British Tea Heist: Or How England Stole the Secret, Discovered a Fraud and Created the Modern World

"Christopher Hitchens And George Orwell’s Ironclad Rules for Making a Good Cup of Tea"

Watch Out Mary Poppins: The World's First Tea Brewing System Utilizing Machine-Learning Algorithms Has Received Pre-Launch Seed Funding (plus a Princess Rap Battle)

The recent price action looks like a cartel is in place.
From Index Mundi:


It's Been Four Years Since We Posted "Deep Learning is VC Worthy" (and what's coming)

Let's see how it turned out.
August 21, 2014
From recode:

Nervana Raises Second Round This Year, as Silicon Valley Bets Big on Deep Learning
Nervana Systems, another player in the suddenly hot “deep learning” space, has closed its second round of capital in the last four months.

The San Diego startup said it raised $3.3 million in Series A funding led by DFJ, which comes on top of a $600,000 seed round in April.

DFJ’s Steve Jurvetson will take a seat on the company’s board as part of the latest investment. Allen & Co., AME Cloud Ventures and Fuel Capital also participated.

Deep learning is a form of artificial intelligence that researchers have credited with recent leaps in areas like speech recognition and image search. That has sparked growing interest in Silicon Valley, with Google, Facebook and Twitter making notable acquisitions or hires in recent months and various prominent players betting their own money on the space.
As Re/code explained in an earlier piece:
Deep learning is a form of machine learning in which researchers attempt to train computer algorithms to spot meaningful patterns by showing them lots of data, rather than trying to program in every rule about the world. Taking inspiration from the way neurons work in the human brain, deep learning uses layers of algorithms that successively recognize increasingly complex features — going from, say, edges to circles to an eye in an image.
Notably, these techniques have allowed researchers to train algorithms using unstructured data, where features haven’t been laboriously labeled by human beings ahead of time. It’s not a new concept, but recent refinements have resulted in significant advances over traditional AI approaches.
Nervana is aiming to distinguish itself in the nascent field by focusing on building hardware optimized for deep learning software — and vice versa....MORE
 On August 9, 2016 Intel purchased Nervana for a rumored $408 million.

Here's Semiconductor Engineering, September 11, 2018 with the story that spurred this stroll down Memory Lane:

Intel’s Next Move
Gadi Singer, vice president and general manager of Intel’s Artificial Intelligence Products Group, sat down with Semiconductor Engineering to talk about Intel’s vision for deep learning and why the company is looking well beyond the x86 architecture and one-chip solutions.

SE: What’s changing on the processor side?
Singer: The biggest change is the addition of deep learning and neural networks. Over the past several years, the changes have been so fast and profound that we’re trying to assess the potential and what we do with it. But at the same time, you also need to step back and think about how that fits in with other complementary capabilities. That’s part of the overall transition.

SE: What really got this going was a recognition that you could develop algorithms with machines rather than by hand, right?
Singer: The original approach was from the 1960s, and it went dormant until [computer scientist Geoffrey] Hinton and others found a better way to deal with multiple layers effectively in the early 2000s. The big breakthrough, when deep learning was recognized as a major computational force, occurred a couple of years ago. That was when ImageNet showed you can reach near-human accuracy with image recognition. We started to see great results on speech recognition. Around 2015 and into 2016, results began to look promising enough to be a major change factor. At that time, the world was basically flat, at least in terms of images. It was relatively simple images and simple, direct speech. Most of the effort was proving things were possible with deep learning so you could reach some level of accuracy or some set of results. In terms of the way to create and prove models, the main architectures were CPUs and GPUs. The way to do the problem before that was C++, like some of the predecessors to Caffe, and with proprietary environments such as CUDA. It required a lot of expertise and effort in building the compute architecture, as well as in the deployment. In terms of who was involved, if you look at the technology in the field today, those were the early adopters.

SE: What’s changed since then?
Singer: Over the last few years, we’ve seen the coming of age of deep learning. The data itself has become much more complex. We’ve moved from 2D to 3D images. We’re working with Novartis, which is looking at 3D microscopic images of cells, trying to identify potentially malignant cells. The images themselves are 25 times more complex in terms of data, but what you’re identifying is a more refined model.

SE: Where does Intel fit in with these architectures. One of the big problems with AI and deep learning is they’re changing quickly, so you need a very flexible architecture. What does Intel plan here?
Singer: In the past, the problem statement was clear. You knew what you needed for a graphics chip or a CPU chip two or three years out, and companies competed on having the best solution for a known problem. Deep learning is a space where companies compete based on who best understands the problem as it evolves. You need an architecture that is able to understand and foresee trends, and be ready for what is coming when it’s out there in the market in full production and deployment—not when it’s being designed and tested....