Claus.dahl

The Intercept reports on several news media who are selling special services at the national political conventions — meetings, cocktail parties, and more. The services are corrosive. Some are explicitly corrupt, “…they make explicit the inevitable failure of the distinction between “paid” and “earned” content.”making explicit the inevitable failure of the distinction between “paid” and “earned” content.

The less controversial services are corrosive because they let the media take money from the people they cover. Having spent a few decades as a marketing communications guy, I can promise you that in every business considering these offers, the conversation includes someone saying, “It doesn’t matter if no one comes to the cocktail party. It’d still improve our relationship with the publication.” Why? Because it’s a way to pay the journal money. That’s corrosive.

Larry Lessig points out that it’s not much different from news organizations tuning their coverage to their ratings. But such tuning at least caters to perceived piopular interest. These new services let an organization or candidate buy coverage despite a decided lack of public interest. It is worse than buying ads because the news media have traditionally had a “Chinese wall” between the advertising and editorial departments. This has been a fairly effective way of protecting editorial content from the direct influence of the marketing needs of the journal, even though the wall is sometimes breached, and Time Magazine has shamefully torn it down.

Once the media started letting companies pay for phony news coverage, they pretended to honor the breach by distinguishing “earned” and “paid” content. “Earned content” is coverage provided by media of events they think are newsworthy. “Paid content” is, well, paid content. Non-sleazebag companies and their PR reps expect media to mark paid content as paid for. Edelman, the world’s largest independent PR company, created ethical guidelines that not only say that the paid content must be well marked, but that Edelman will have its own Chinese wall between the processes by which earned content is pitched (“Yo, I have a client who’s invented a time travel machine. Wanna an interview? How’s yesterday for you?”) and the negotiations that result in the placement of paid content. (Disclosure: I had a tiny hand — Trump-sized — in drafting those guidelines.)

That’s better than nothing, but paid content still makes me queasy. Companies are willing to pay for content precisely because it looks like real coverage and thus tends to be taken more seriously than obvious ads. This erodes the phenomenological line between news and ads, which is bad for democracy and culture. Indeed, “the point of paid content is to erode the line. ”the point of paid content is to erode the line.

But letting candidates pay for interviews takes this to a whole new level. This is what The Intercept says:

Sponsors who pay $200,000 are promised convention interviews with The Hill’s editorial staff for “up to three named executives or organization representatives of your choice,” according to a brochure obtained by The Intercept. “These interviews are pieces of earned media,” the brochure says, “and will be hosted on a dedicated page on thehill.com and promoted across The Hill’s digital and social media channels.”

The Hill says the resulting interviews will be earned media. Suppose the interview is stupid, boring, self-serving and non-newsworthy? If it weren’t, the client wouldn’t be paying for it. But The Hill is promising it’s going to run anyway because the client paid them $200,000. That is the very definition of paid content. So, by calling it “earned content,” The Hill can only mean that the article will not be marked as paid content, even though that is precisely what it is.

This corrupts the already corrosive practice of accepting paid content. It is disgraceful.

The post Corrupting "Earned media" appeared first on Joho the Blog.

Data for Change, Computational Economy, Programming Language Usability, and Against Dashboards

1. Teach Programming to Analyse Data for Change -- What I imagine is a project which explicitly uses all the tools students have learned over their years of high school to advocate for change in their communities. (via Cory Doctorow)
2. The Computational Engine of Economic Development -- how can we measure the bits and flops of an economy?
3. The Usability of Programming Languages -- lecture notes, slides, and readings.
4. Real-Time Dashboards Considered Harmful -- If there’s nothing that can be done in response to a report, does it need to be reported in that manner?

Continue reading Four short links: 21 June 2016.

It's a throwaway line in a longer talk and we probably shouldn't make too much of it, but I will anyway.

In five years time Facebook "will be definitely mobile, it will be probably all video," said Nicola Mendelsohn, who heads up Facebook's operations in Europe, the Middle East and Africa, at a conference in London this morning. Mark Zuckerberg, Facebook's CEO, has already noted that video will be more and more important for the platform. But Mendelsohn went further, suggesting that stats showed the written word becoming all but obsolete, replaced by moving images and speech.

"The best way to tell stories in this world, where so much information is coming at us, actually is video," Mendelsohn said. "It conveys so much more information in a much quicker period. So actually the trend helps us to digest much more information."

Maybe this is coming from deep within the literacy bubble, but:

Text is surprisingly resilient. It's cheap, it's flexible, it's discreet. Human brains process it absurdly well considering there's nothing really built-in for it. Plenty of people can deal with text better than they can spoken language, whether as a matter of preference or necessity. And it's endlessly computable -- you can search it, code it. You can use text to make it do other things.

In short, all of the same technological advances that enable more and more video, audio, and immersive VR entertainment also enable more and more text. We will see more of all of them as the technological bottlenecks open up.

And text itself will get weirder, its properties less distinct, as it reflects new assumptions and possibilities borrowed from other tech and media. It already has! Text can be real-time, text can be ephemeral -- text has taken on almost all of the attributes we always used to distinguish speech, but it's still remained text. It's still visual characters registered by the eye standing in for (and shaping its own) language.

Because nothing has proved as invincible as writing and literacy. Because text is just so malleable. Because it fits into any container we put it in. Because our world is supersaturated in it, indoors and out. Because we have so much invested in it. Because nothing we have ever made has ever rewarded our universal investment in it more. Unless our civilization fundamentally collapses, we will never give up writing and reading.

We're still not even talking to our computers as often as we're typing on our phones. What logs the most attention-hours -- i.e., how media companies make their money -- is not and has never been the universe of communications.

(And my god -- the very best feature Facebook Video has, what's helping that platform eat the world -- is muted autoplay video with automatic text captions. Forget literature -- even the stupid viral videos people watch waiting for the train are better when they're made with text!)

Nothing is inevitable in history, media, or culture -- but literacy is the only thing that's even close. Bet for better video, bet for better speech, bet for better things we can't imagine -- but if you bet against text, you will lose.

“I’ve told people what they need to hear to get their money in and it’s been very difficult.”

“Which is a lie.”

“Yes. I’ve told some things that were not true.”

“It’s a lie. It’s a lie. It’s a material lie.”  

THE MARIJUANA START-UP Ebbu LLC held its 2015 Christmas party at a stylish townhouse in downtown Denver.

As guests arrived, they selected from edibles and vape pens fanned out on cocktail tables. Hired lovelies dressed for Burning Man twirled silently through the crowd. At one point, a man ran into the middle of the room and shouted, “Everybody in here get the fuck out!” — before security showed him the door. The most committed stoners colonized a windowless basement decorated with a pirate flag. It still felt like a work party, with significant others making small talk around a messy taco bar. 

For more than two years, Ebbu has promised to unveil a revolutionary line of marijuana products called “Feelings” — Energy, Create, Chill, Giggle and Bliss. Marijuana’s flaw, the company believes, is that it’s unpredictable. Customers want to know how a product will make them feel, and they want it to work every time. Ebbu’s goal is to position consistent marijuana products as the pot equivalent of the Intel microchips that power nearly everything that emerges from Silicon Valley. Imagine: “Ebbu inside” stamped on the packaging of edibles and other manufactured pot products across the country.

The company’s co-founders, Michael “Dooma ” Wendschuh, 39 (above, left) and Jon Cooper, 40 (above, right), met over a decade ago in Los Angeles, where Dooma had co-founded a production company, and Cooper worked in movie finance. Neither of them are heavy pot smokers. They formed Ebbu in 2013, and soon emerged as leaders among a wave of new pot entrepreneurs — technocrats drawn to cannabis more for the potential profits and the excitement of pioneering a new industry than personal devotion to the plant. Before Ebbu had a product to sell, the company was featured in outlets like The Economist, Bloomberg, and the Fox Business Channel. Dooma “saw a problem that nobody seemed to be addressing,” The New Yorker website reported last May. During occasional TV spots and on the conference circuit, Dooma, Ebbu’s public face, personified the image that legal marijuana businesses hope to project: smart, sophisticated and law-abiding.

At the party, Dooma stood above the snowy patio, sober, and called the crowd to attention. He is angularly thin, with short, spiky hair and deep-set eyes, so that he sometimes resembles a very well dressed prisoner of war; that night he wore a sport coat, a conservative striped tie and a Quebec flag belt buckle the size of a saucer. “There are a lot of fucking people here, which is awesome!” He cried out. “One year ago this was in my apartment.”

Before Ebbu had a product to sell, the company was featured in outlets like The Economist, Bloomberg, and the Fox Business Channel.

In November, Ebbu had topped $100,000 in monthly sales for the first time. It had about 30 employees and contractors. Dooma had stopped the party to award dog tags to the year’s new hires who had held on to their jobs. “We’re in combat every day at Ebbu,” Dooma said. “What we’re trying to do is the hardest thing you can imagine.” Each dog tag had an employee’s name and the motto, “Lead by example.”

Dooma scanned the crowd. “I know you’re doing it because you believe, ‘cause we don’t pay you guys shit. It’s about the equity. It’s about the brand that we’re building…It’s about creating something that’s going to be bigger than Facebook or Google. We are going to make a product that is going to be in every bar, every restaurant, every hotel and every household around the world!”

Two weeks later, Dooma had left Ebbu and the company he’d devoted years to build had plunged into a crisis.

THE BUDWEISER OF WEED

A year earlier, Dooma and Cooper had invited me to “embed” with Ebbu. At the time, the company had minimal name recognition and no sales, but it stood out as one of the few marijuana companies both determined to advance science and willing to break federal law. Between then and January 2016, I sat in on well over 100 company meetings and calls, including many with investors, partners and other outside parties. I spoke freely with the staff and had a free pass to spend time at the office. The plan was to tell the story of a company working to solve marijuana industry’s most exciting problem, while the rules for how marijuana companies operate are still very much in flux.

After Dooma left, the story became far more complicated. I obtained evidence that he would, on occasion, “stretch the truth” with prospective investors about the company’s finances, exaggerate product development, and that his fundraising efforts may not have been in full compliance with Colorado’s marijuana laws. Dooma and Cooper have had a falling out and neither will discuss what transpired.

In the months following Dooma's departure, I tried to piece together the story of what went wrong for a talented entrepreneur who sought to dominate America’s wildest industry.

“We’re in combat every day at Ebbu... What we’re trying to do is the hardest thing you can imagine.”

Legal marijuana is among the fastest growing industries in the country, but it’s not yet big business. In 2015, industry-wide sales totaled $5.4 billion, the overwhelming majority of it in three states: Washington and Colorado, which have legalized sales to all adults, and California, where doctors’ notes are available online. By comparison, the illegal pot market is worth roughly $40 billion and domestic alcohol sales top $200 billion annually.

In November, five states, including California, are likely to vote on recreational legalization, which could expand the market by tens of millions of potential customers overnight. Ambitious cannabis entrepreneurs like Cooper and Dooma didn’t set out to replace the local dealer. Their goal is to reinvent this stigmatized plant as an essential part of modern life, as ubiquitous, in its way, as our smartphones.

It’s hard to overstate the potential. While smoking it works well enough, marijuana is a chemically fascinating material that has been exiled from academic and corporate research for decades. “Nobody understands this plant,” Dooma said. “In order to build a business, we have to do the very basic scientific research to fundamentally understand what this plant is capable of.”

The state legalization experiments have set off the economic frenzy known as the Green Rush. But like other economic frenzies, from the California Gold Rush to dot-com bubble, a tiny elite is bound to do way better than everyone else. In Colorado alone, hundreds of companies manufacture edibles, vaping oil and other marijuana products that are largely indistinguishable from each other. Many of these companies agree with Ebbu that controlling a product’s effects, medical or recreational, or convincing customers that you can, will be central to determining who becomes Facebook and who becomes Friendster. Some companies approach this question with genetics or refined agricultural techniques. Ebbu aims to pull the plant apart and then reconstitute the relevant chemicals into products.

The legal obstacles companies face might be more daunting than the scientific problem. In Colorado, generally regarded as the world’s most mature marijuana market and an example studied by other states, the industry’s legal framework is just over three years old and individual rules change often, with potential implications far beyond the state’s borders.

If Colorado companies follow state laws, they’re unlikely to be raided and shut down by the federal government —  during the Obama administration, at least — but they operate in unmapped terrain. For marijuana companies, bedrock business needs like fundraising, intellectual property protection, selling in other legal states and simply opening a bank account involve layers of complexity that companies in other industries don’t have to think about. Colorado “has the most stringent regulations where I’m least likely to go to jail. I don’t want to go to jail,” Dooma said. “That means that I’m also swallowing some of these really ridiculous rules.”

With the legal landscape still fractured and uncertain, Ebbu hopes to get a head start: right now it can simultaneously research the plant and develop new products while corporate America—everyone from Big Pharma to Big Tobacco -- salivates on the sidelines. Green Rush entrepreneurs love to compare the current moment to the Internet’s early days. Dooma, with characteristic hyperbole, considers that modest. “Maybe two billion people in the world use the Internet,” he said. “There has got to be close to seven billion who need some kind of psychoactive. Marijuana can be better than all of them, but it's not right now. It needs to change in order to reach its potential.”

I WANT EVERYTHING, AND I WANT IT NOW

Before he entered the pot business, Dooma’s L.A. production company helped create video games. He and his partner “were young and insanely confident,” Joshua Rubin, a writer who worked for them, said. “They knew how to talk people into things. They knew how to get people excited about ideas.” In the mid-Aughts, a colleague gave Dooma a copy of Alamut, a Slovenian novel based on the life of the 11th century Persian warlord, Hassan-i Sabbah. In the book, Sabbah lives by the credo, “Nothing is true, everything is permitted,” and controls an elite band of soldiers — known to history as the Assassins — by feeding them “little balls” of hashish. The book helped inspire Dooma and his partner to write Assassin’s Creed. Video games in the ensuing Ubisoft franchise have sold more than 78 million copies and grossed over $3 billion. A movie, starring Michael Fassbender, Marion Cotillard and Jeremy Irons opens in December.

A Princeton grad with a film degree from USC, Dooma, then known as Mike, smoked his first joint with some buddies on the roof of a Miami hotel during his Star Trek-themed bar mitzvah. Still “super afraid” of drugs, Dooma tried not to inhale.  In more than a year, I remember seeing him consume marijuana only once, during a meeting for work purposes. I never saw him take any other illegal drug, but he met his first wife at Burning Man and has claimed, “When I think back on my life, the most important moments, every single one of them, involved a psychoactive substance.” (He’s including caffeine.)

Dooma, a vegetarian and light drinker, has a far-flung network of interesting and accomplished friends and also knows his way around the national parks. He gets along so well with his parents that he brought them to Burning Man one year. They had a wonderful time.

At work, he presents a different face. In meetings, his default tone is brusque with a withering edge. He said he prefers not to hire people with ski racks on their cars because he thinks they’ll play hooky whenever it snows. He respects a few competitors, but says things like, “If you look at the jokers who are making the real money in this industry it can’t be that hard.” He got in the habit of Googling new contacts with terms like “fraud,” “scam” and “SEC” [Securities and Exchange Commission.] When interviewing a job candidate, he characterized his management style as, “What I want is everything and I want it now.”

He once berated an employee for bringing the wrong pastries to an investor meeting. “I got so mad,” Dooma said later while driving his dinky Honda hybrid. “I yelled at him in front of a lot of people, which I shouldn't have done.” He then brought up a female employee, “I am such an asshole to [her] and she tries so hard. [She] has this incredible ability to, like, bounce back when I say the most condescending and offensive things to her. She doesn't even get perturbed.” For a while, Dooma had a daily note on his calendar reminding him not to be an asshole. (Through a spokesman, Dooma said he doesn’t remember this and suggested that it was posted by an employee.)

By 2010, Dooma had a exciting career while, around him, hundreds of medical marijuana dispensaries had opened in L.A. Dooma saw that discerning users varied their intake to focus, relax or sleep, but the available products didn’t always deliver. “People who really love cannabis are willing to forgive the problems of the plant,” he said years later. “They think that like six times out of ten having a pretty good experience is good enough, whereas the average consumer would be furious with a six out of ten situation.” It wasn’t until later, after Colorado and Washington State voted to legalize in 2012, that Dooma called his old friend Cooper, who had returned to the Denver area where he grew up. “Hey Coop,” Dooma said. “What do you think of marijuana?” 

"I know you're doing it because you believe, 'cause we don't pay you guys shit. It's about the equity."

Cooper, who had built a career as a tech executive and entrepreneur,  didn’t think much about it, but he invited his friend to visit. Cooper, Ebbu’s CEO, has the beard and stocky build of a young Henry VIII. He’s more approachable than Dooma. He laughs easily, cheers for the Broncos and named his son Che.

As Colorado prepared for recreational sales, Dooma and Cooper searched for investment opportunities, but nothing struck them as a billion dollar idea. Grow operations needed capital, but the pair figured cannabis would soon be a commodity crop, like wheat or soybeans. Retail dispensaries looked like another low-margin slog, as sexy as running liquor stores. Dooma and Cooper decided that the big winners would be trusted brands, the Cokes or Budweisers of weed. Companies with similar ideas already touted their products’ flavor or all-natural ingredients. But no one uses cannabis because it’s baked inside a tasty cookie. “People make purchasing decisions in this industry based on how they think it's going to make them feel,” Dooma said. The existing companies were “all branding the wrong thing.”

Marijuana gets people high, but “high” is an inadequate description of the drug’s effects. Pot can feel euphoric or zonk users into “couch lock” for hours. It can induce laughing fits, paranoia or an urge to cuddle. Some users consider it essential for exercise, sleep, and, yes, work. Underground growers have long sought to engineer highs by crossbreeding familiar plants, but in an illegal market, their skill was worth little more than street cred.

When customers enter a dispensary, the buds displayed in glass jars beneath the counter are advertised as sativa, indica or hybrid. The French naturalist Jean-Baptiste Lamarck first distinguished between the tall cannabis sativa plants found in Europe and the bushier cannabis indica that grew in India in 1783. Today sativa is said to pep users up, while indica induces a mellower “in-da-couch” high.  But most experts believe that years of untracked commerce and aggressive crossbreeding have stripped the classification of any real scientific basis, at least at the dispensary counter. “I would strongly encourage the scientific community, the press and the public to abandon the sativa-indica nomenclature,” Dr. Ethan Russo, a neurologist and medical director of a Los Angeles company focused on cannabinoid receptors, said in a recent journal interview.

 The thousands of strain names like Alaskan Ice, Heisenberg Kush, Chem’s Sister and Amnesia Haze confuse matters further. The Web site Leafly, which has backing from Peter Thiel’s Founder’s Fund [disclosure: a Pando investor], says the canonical strain Blue Dream, a sativa-dominant hybrid, carries a balance of “full-body relaxation with gentle cerebral invigoration.” Who, after a few hits, would disagree? But savvy buyers know that what one dispensary calls Blue Dream, will be stronger or produce a different nature of experience than the Blue Dream sold next door, because neither is the real Blue Dream, if there even is such a thing.

“Even the word ‘strains’ is not typically applied to plants,” Jonathan Page, CEO of the cannabis testing company Anandia Labs, and an adjunct botany professor at the University of British Columbia told me. “In roses, or petunias, or corn you have cultivars or varieties of plants that are actually registered with a government body.” With cannabis, of course, there’s no registration or documentation. 

Cindy Orser, chief scientific officer at testing company, Digipath Labs, said that when it comes to strains “truth in advertising doesn’t exist in this industry.”

During one of Dooma’s visits to Colorado, he and Cooper were talking in the living room — sober, Cooper said. Pot had never been hard to come by in Colorado, especially in Boulder where Cooper went to college, but he had an uneven history with it. “We have all these people who are coming to the state, and they want to try marijuana, and they just want to giggle,” he said to Dooma. “What do they take?” Remembering one of his best times with weed, he then asked, “What if we made a product that was called ‘Giggles?’”

Before Ebbu launched, Cooper put together a survey to learn about their potential customers. The most intriguing responses, he found, came from former users who had given it up. “You hear a whole wide range of excuses, and you can probably name them just as well along with me,” he said. “‘I have kids.’ ‘It doesn’t fit into my lifestyle now.’ ‘I had a bad experience.’” These same people had had bad experiences with alcohol but they continued to drink. Cooper and Dooma suspected that predictable marijuana products could bring those former users back. And these customers, Cooper said, “actually have money.”

Cooper now thinks a lot about marijuana, and how he can establish Ebbu as a trusted American brand. Ebbu, he said in a meeting is not about getting high, it’s about enhancing life experiences, whether that means enjoying a party, taking a nap or painting a picture. “There’s going to be an interesting question to figure out, related to this, or maybe it’s kind of the undertone of the whole thing, which is the experience of life,” he said, sounding a bit enhanced himself. Then he recovered, “We need to have a really good answer about why psychoactives are a good thing in people’s lives.”

THE END OF THE TOUR

Ebbu occupies a chalet-style building in the Rocky Mountain foothills near the town of Evergreen, Colo. Visitors enter a cramped office furnished with all the care and charm of a police station. There’s no ping-pong table, no free snacks, no kegerator. Some employees park their laptops wherever they find space at a large wooden table in the main room. Behind a locked door, where a factory and lab contains more than $1 million worth of equipment, a team of white-coated scientists are manipulating chemicals found in marijuana to create the five “Feelings”: Energy, Create, Chill, Giggle and Bliss.

In April 2015, the facility was still under construction when the co-founders met near Dooma’s standing desk to dissect everything that wasn’t working. Dooma scribbled on a window in red marker that Ebbu had no quality control, couldn’t meet deadlines and a pricey machine — a liquid chromatography mass-spectrometer — didn’t appear to work for their purposes. Cooper, seated in the small cluttered room, added that they had no way to scale up production until the factory opened, still months away. The packaging wasn’t finalized for a product they hoped to launch within weeks. “Do you know we have 20 employees right now?” Dooma asked, with a joyless laugh. “It sure doesn’t feel like it.”

The founders were sorting through their concerns over the loud scraping of an unseen worker — “That’s not annoying at all,” Dooma said — when the bookkeeper, an affable guy named Ben, arrived with a letter. The bank had apparently learned that Ebbu is in the marijuana business. “E-Trade has decided to exercise its discretion to close the above referenced account,” it read. Dooma and Cooper had 30 days to find a new bank for their investor’s money — the company raised $2 million in 2014 — even though none in the state openly worked with marijuana companies.

As Cooper and Ben huddled around a computer to go over expenses, Dooma sat slightly removed, leg bouncing, eyes on his phone. “Dooma we need to catch up,” he suddenly read aloud. “All tours are on hold until you and I speak about my role with Ebbu.” It’s from the landlord who was renting the company lab space in Denver while Ebbu built its own. When prospective investors landed from out of town, Dooma showed them the rented lab; he thought its bubbling beakers and gleaming machines would convince investors to dig deep. The landlord, a powerful figure in the industry, knew this, and hoped that by taking away one of Dooma’s main tools for raising money, he could maximize his stake in the company.

“Why the fuck is [Dooma’s  assistant] telling these people that we’re doing tours?” Dooma snapped, and stomped off to find her. She departed the company soon after. (It’s not clear whether she quit or was fired.) “We’re so fucked,” he said as he returned. “We’re so fucking fucked.”

Over the course of 2015, Cooper and Dooma collaborated on strategy and business development. Cooper oversaw operations including the construction of Ebbu’s headquarters and factory and building a sales team. Dooma focused on fundraising. His central task for the year was leading Ebbu's effort to secure $9 million in venture capital. 

Raising money is a peculiar challenge for marijuana companies, so anything that threatened to disrupt his efforts was an issue. “Sales is a very delicate dance,” he told me. “It's a lot about creating an illusion about what this company is and what the investment is going to be like and what your experience is going to be like with the investment.”

Almost none of the VC funds that most start-ups hit up will invest in weed. This narrows the pool mainly to rich folks who are willing to risk stigmatization, or, perhaps even prison time, to join the Green Rush. In Colorado, only two-year state residents can own equity in marijuana companies.   (The state is phasing out this requirement.) As a result, Ebbu could only offer out-of state investors the relatively boring opportunity of interest-paying “notes.” And note holders have to get fingerprinted and pass a background check for the privilege. Even Dooma, who moved to Colorado for Ebbu, was not eligible to officially own equity in the company until January 2016. He was technically a consultant, though his LinkedIn page lists him as "Co-founder, Co-CEO."

“We’re so fucked,” he said as he returned. “We’re so fucking fucked.”

Even once the company has raised money, it had to wait for the bank it found to quietly accept its business to release it. In the fall, Ebbu was so short on cash that senior executives volunteered to defer their pay, while almost $5 million languished inaccessible.

In January 2015, I sat in as Dooma called some of the company’s investors to ask for more money. Outside, a blizzard had whited-out the roads. “Life is fucking great,” a California tech entrepreneur said through the speakerphone. "Up and to the right, up and to the right."

The conversation stalled after Dooma acknowledged that Ebbu didn’t have any revenue. “I love being invested in you no matter what you’re doing,” the investor said. “I know you’re a beast and you’re going to kill it.” Still, he declined to invest further. 

The second investor wasn’t ready to invest more either, and Dooma opened the third call with some small talk. “I can’t imagine what your ex-wife looks like,” he said, “because [your daughter] looks like they cloned you.”

The ex-wife “looks like a Russian hooker,” the investor said. He wanted to invest more in Ebbu once he got through a temporary cash crunch. “Within two weeks I should have full access to what is quote unquote mine.”

Dooma’s professional life was an endless cycle of calls and travel to raise the money, but he sometimes displayed an unsteady grasp of the company’s finances. In early October, for example, he told a prominent investor visiting Ebbu that sales were expected to be $100,000 in September and $240,000 in October. He repeated the latter figure on a panel at an industry conference in Denver several days later. In fact, the company wouldn’t surpass $100,000 in monthly sales until November.  (Through his spokesman, Dooma said he was never provided weekly sales numbers and what he offered in these meetings were his best estimates.)

During the same meeting, Dooma alluded to selling purified cannabinoids—chemicals found in marijuana—for $2,000 a gram. Some of the rarer chemicals likely sell for these prices, but Ebbu has yet to sell purified cannabinoids at any price. (Through a spokesman, Dooma said he believed a deal had previously gone through.) To one prospective investor, and at least once in front of me, he talked about companies that sell purified cannabinoids for far higher, more than $100,000 per gram. Dooma likely arrived at this price by taking the $150 that a scientific supply company might charge for one milligram quantities for use as laboratory references and multiplying it by 1,000 — milligrams to grams. Ebbu has never sold these products either and is unlikely to in the future, a company spokesman said. Experts I contacted said the small number of labs which need them typically buy tiny quantities; one company said about 20 milligrams of a given purified cannabinoid per year.

Ebbu sells pre-filled vape pen cartridges under the brand name Fancy, as well as hash, and potent extracts that are vaporized, or “dabbed,” for a fast-acting, intense high. In June 2015, Dooma told a reporter from Fortune Magazine that it cost $2 to produce these products, which were not on the market at the time. (Through his representative, Dooma said he was misquoted). In the October investor meeting, the number changed slightly. “This is a very high margin business,” Dooma said. “It's about $35 a gram wholesale for just the extract. About $4 a gram to produce.” (Through his spokesman, Dooma said his statements were based on financial models that showed what Ebbu's margins would be when the company was producing products at scale.)

At the time, Ebbu was paying more like $11 to produce a gram, Cooper told me later, worried that if the wrong numbers got out his customers would demand lower prices. “If you haven’t noticed, I’m the number guy in the relationship,” he said with rare exasperation. (Cooper said he did not read the Fortune article until January 2016.)

The lab and factory opened over the summer and Ebbu shipped its first vape pens. Cooper had to recall the whole first batch after they leaked. Hiccups like this are standard as companies adapt manufacturing practices to the unfamiliar raw material.

Ebbu’s extracted marijuana products brought in more than $300,000 in revenue by the end of the year. But Dooma and Cooper didn’t start the company to sell top shelf weed. Dooma described the distinction with a produce analogy: Ebbu’s current offerings are like premium orange juice, which needs to come from premium oranges; but vitamin C companies can extract their product from mediocre fruit. Feelings, he told investors, will be superior products derived from non-premium plants.

That pitch helped Dooma raise $7.9 million in 2015 — short of the initial $9 million goal, but an impressive haul for the industry. Despite his fundraising success, it remains unclear, as someone leaving one of Dooma’s Manhattan pitches put it, whether the creation of the Feelings  — Giggle, Bliss and the rest — is “a fairy tale or if it’s real.”

 GIGGLE AND BLISS

The weed arrives at Ebbu’s factory in bags, stuffed into boxes and plastic tubs. Companies like Ebbu that make “marijuana-infused products” typically buy “trim,” the excess foliage cut away from the buds or “flower.” Wholesale trim can sell for about $500 a pound; flower, which is better for smoking, can cost four times as much. Before growers ship an order, they must fill out a manifest specifying the amount and name of the product being shipped, the date, the vehicle, the estimated time of departure and arrival and the worker accompanying it. When it arrives, an Ebbu employee weighs it on a digital scale in the garage-like receiving room and then it’s catalogued in a state government database. RFID chips track all legal marijuana in Colorado “from seed to sale,” and every worker who handles it must first pass a state background check. Plenty of Colorado’s 1,000-plus licensed grows don’t have bank accounts, so delivery drivers often leave Ebbu with thousands of dollars in cash.  

It’s not well understood why pot produces — or seems to produce — such different experiences. Marijuana’s potency is largely determined by the concentration of delta-9-tetrahydrocannabinol, or THC, often called the plant’s psychoactive ingredient. But THC is only one of scores of compounds known as cannabinoids found in the plant. Cannabis also contains terpenes, the chemicals responsible for its unmistakable smell, and there’s evidence that they modify the high as well. (Fruits, conifers and many other plants also contain terpenes.) Ebbu thinks that specific ratios of cannabinoids and terpenes will produce the promised “Feelings” of Energy, Create, Chill, Giggle and Bliss. These chemicals, Cooper said, are “our flour and our sugar and our butter.”

First, Ebbu extracts the oil. The company uses ethanol (drinking alcohol) to strip away the leaves and flowers, leaving a golden, THC-rich viscous. Marijuana oil, like high-fructose corn syrup, is more versatile and less distinct than the plant it comes from. It can be vaped on its own, or deployed as the active ingredient — “Ebbu inside” — in edibles and pseudo-medical products like transdermal patches, pills and tinctures. A blend of marijuana and coconut oils, a few sellers claim, can heighten female orgasms when spritzed on their genitals a few minutes prior to sex.

As legalization spreads, many believe flower will shrink into a niche market for mavens who care about their pot’s appearance, bouquet and ancestry. Already in Colorado, more than 30 percent of cannabis retail sales go to factory-produced items, according to the data firm BDS Analytics.

Creating the Feelings, Ebbu believes, will require separating marijuana oil into purified cannabinoids. An aficionado might be able to namedrop 10 of the chemicals, but so far there’s only consumer interest in two: THC and cannabidiol (CBD), which is associated with the plant’s medicinal properties. Ebbu’s largest chromatography machine, known in-house as the Novasep, for the French manufacturer, is a custom-made, two-ton metallic puzzle of dials, gauges and piping that promises to separate marijuana oil into more cannabinoids than currently have names. It cost Ebbu more than $750,000. For years, underground growers have had an incentive to maximize the THC content in their plants. The Novasep represents a big bet by Ebbu that there’s unrealized commercial value in the lesser-known cannabinoids. In a private video, Dooma told investors that the Novasep, “could very easily make us a million dollars a day.”

Before Ebbu can manufacture the Feelings, it has to invent the recipes. In his lab at the back of the factory, Brian Reid, Ebbu’s chief science officer, uses a process called high-throughput cell screening to test how cannabinoids and terpenes interact with cells. Reid, a Ph.D. in biochemistry who previously worked at the University of Colorado and in the biotech industry, has a trim beard and a soft-spoken good-humor. With an 8-pronged pipette the shape of a hair pick, he filled the wells on an opaque rectangular plate with tiny drops of a clear fluid containing tens of thousands of cells “engineered to overexpress” the main cannabinoid receptor known to bond with THC. Across the room, a hulking liquid handling robot diluted droplets of three distinct cannabinoid solutions laid out on another plate. One solution contained THC, another CBD and the third equal amounts of THC and CBD. 

According to both stoner lore and the scientific literature, CBD is supposed to moderate THC’s effects. But when Reid fed the plates into a $400,000 refrigerator-sized machine called a FLIPR, the solution containing both THC and CBD generated more cell activity than the two single-cannabinoid solutions combined. Reid explained that cannabinoids don’t act on the body like hyper-specific conventional medicines, but rather by way of a phenomenon known as the entourage effect, in which the chemicals in marijuana work in concert to amplify each other. (Dr. Russo, the neurologist, has compared it to a symphony.) This is what makes understanding individual cannabinoids a central challenge for any company seeking to harness the plant’s potential. “Cannabinoids are fairly ‘dirty,’ pharmacologically speaking,” Reid said. “They hit a number of receptors in the cells with varying affinities and effects.”

When Cooper and Dooma thought up Giggle and the other Feelings, they knew far more about what they could sell than what was scientifically feasible. Reid’s job is to perform tens of thousands of assays, varying the ratios, compounds and targets, until he lands on potential formulas for the Feelings. He said he can also test his own ideas, such as combinations that might mitigate side effects like the munchies or red eye. (Similar principles, Ebbu believes, could lead to products with specific medical benefits.) “We're going to really create a baseline of how each one of these components works at the cellular level,” Reid said. “It's really only known for a few of them.”

Reid’s work resembles the drug discovery process at pharmaceutical companies, but unlike drugmakers, Ebbu can conduct cellular research and human testing simultaneously while also generating revenue from selling similar products. For FDA regulated pharmaceutical companies, the process is more linear: Drug discovery, then additional lab research, then clinical trials and the payday only arrives, if it arrives, after years.

Ebbu can’t credibly claim that the Feelings work as promised unless it has data. But the logistics of testing intoxicants are tricky. Marijuana is a federally illegal drug and there are no established procedures or guidelines for testing it on humans. The company has had to think about things like how old the subjects should be and their familiarity with cannabis. Should it provide a way for subjects to get home, or, alternately, should they take samples home and answer questions online?

Throughout 2015, it was never clear to me which of the innumerable vape pens in the factory contained pure strain extracts—products available in every dispensary --  and which, if any, were manufactured prototypes of the Feelings.  When a TV crew from Fusion visited Ebbu in March 2015, they taped a segment where three people received vape pens and then tried to guess which Feeling they had sampled. Cooper, who was not involved, believes that the samples were marijuana oil extracted from specific strains — products available in every dispensary — not precise formulas that Ebbu had developed. Dooma’s spokesman did not respond to requests for comment on what was provided for the Fusion shoot.

Through his spokesman, Dooma said the company had prototypes of Feelings by the second quarter of 2015. Through a spokesman, Cooper said it did not have formulated products by then. Either way, by the time Ebbu began formal testing this year, Dooma was gone.

A LIE, A MATERIAL LIE

On December 29th, Cooper and Dooma told me separately that big changes would come in the New Year. (Neither remembers saying so.) Dooma looked unhappy, while Cooper was his usual amiable self. Two nights later, Dooma threw a New Year’s Eve party that several employees attended. The next day he and his wife drove to Aspen where they had dinner with at least one Ebbu investor who said business didn’t come up. Dooma looked relaxed, the investor said. Perhaps for the first time.

The following Monday afternoon, the first workday of 2016, Cooper told the staff that Dooma was on an indefinite leave of absence.  Dooma’s workspace had been cleared out. The news surprised almost everyone in the company. According to one Ebbu employee, speaking on the condition of anonymity, Dooma’s “personality was tough, but his expectation levels are the reason why we are where we are.” Dooma’s “the visionary,” the employee continued. “I can't believe that somebody put that much time and energy and effort into this business and they just dismissed him the way they did.”

In the weeks after, both Dooma and Cooper refused to clarify what happened, though Dooma told me, "It's the worst situation I've ever been in in my life." Then, in mid-February, I received a call from someone who identified themself as a “concerned citizen,” who told me that Dooma had misrepresented Ebbu’s revenue and profit margins, along with the amount Dooma personally invested, to make the company more attractive to prospective investors.

Not all of this was news. Dooma had misstated margins and revenues in front of me. A few days after the 2015 Christmas party, a coveted investor visited Ebbu’s headquarters. Afterward, Dooma sent an email to the team to congratulate them on a well-executed tour. A scientist who took part in the discussions with the investor replied to him: “I would like to share one thing that I have told you before: I do not like it if we change our wording as such that we are being dishonest. Tonight this happened a few times” including once, “when you were acting like we had prototypes for Feelings at this moment… I emailed you previously to ask you please not to say this and I was disappointed that you never replied.” When we first met in November 2014 Dooma told me that Ebbu had prototypes of all five Feelings. 

I also learned that the co-founders met in late December and discussed Dooma’s status at the company. Without Cooper’s knowledge, Dooma recorded the conversation and then, according to the concerned citizen, inadvertently uploaded the recording to a company shared drive. I was first sent 17 clips totaling about 15 minutes of audio, which Dooma's spokesman characterized as a selectively edited recording of a much longer conversation. I knew who was arranging the leak and later received a copy of the entire recording, which lasted about one hour and 45 minutes. (Ebbu will not confirm the tape's authenticity.)

In the audio, Dooma attempted to convince Cooper that the two of them should “protect” themselves by signing an agreement that would make it difficult to fire them as a result of what Dooma had told prospective investors. He may have recorded the conversation to catch Cooper saying something he could be held to later. 

Throughout the recording, Dooma repeatedly appealed to their long friendship. He also accused Cooper of trying to push him out. “Angry employees, missed emails…lies to investors that’s a case. Those are the things you say in court. It’s a case.”

At the time, Dooma was a month away from being a two-year Colorado resident, which would have ultimately allowed him to own a significant stake in Ebbu. Dooma suggests that Cooper wants to force him out, in part to take what would have been Dooma’s. “If you get rid of me then you own a lot more of this company,” Dooma said. “It is not rocket science and now the round is raised, the money is in the bank and you don’t need me.” Cooper dismissed the accusation, and says that Dooma’s false assertions with investors put the company at enormous risk. “You tell people that you’ve put a million dollars into this company,” Cooper says. “That’s material. You never put a million dollars into this company.” Misled investors have a tendency to sue.

Dooma replies, “I’ve told people what they need to hear to get their money in and it’s been very difficult.”

“Which is a lie.”

“Yes. I’ve told some things that were not true.”

“It’s a lie,” Cooper says. “It’s a lie. It’s a material lie.”

Dooma also implicated his co-founder: “You’ve been in enough investor meetings to know that I will occasionally stretch the truth on small things where I can’t get caught. You know. That’s what I do. It’s something that you do. Everyone does it to some extent.” 

“I don’t know if that’s true,” Cooper responded.

Elsewhere in the recording, Dooma tells Cooper, “It’s not false that I didn’t invest. I got fucking paperwork saying I invested… I’m on record as a note holder in Ebbu for $240,000.”

It’s not clear how much of the $240,000 was actually Dooma’s. In a “custodian agreement” dated September 13, 2013, Dooma agreed to receive $140,000 “to make certain investments as directed by investor.” As best I could determine without endangering anyone’s livelihood, the signer is the brother of one of Dooma’s college friends who works for a global bank. According to a document, apparently executed September 19, 2013, which I obtained, Dooma then invested $140,000 in Ebbu. If this transaction went through, it could have served as a way for someone who didn’t want their name associated with marijuana to invest in a marijuana company.

In October 2013, Dooma’s future wife signed a similar agreement to receive $100,000. A document dated two days later was prepared to confirm a $100,000 investment in Ebbu, and names both her and Cooper as signatories, though Cooper hasn’t signed it.

Through his spokesman, Dooma said he invested a total of $249,000 in the company, including a $100,000 note executed in September, a month before his future wife accepted custody of the same sum. Dooma’s spokesman did not respond to requests to clarify the source of the $140,000 investment.

In part to prevent illegal marijuana operations from gaining control of the Colorado market, the licensing process requires executives, in this case Cooper, to sign documents under penalty of perjury disclosing everyone who has a current or pledged financial interest in the company. Ebbu received its state license in March 2015. If any of the transferred $240,000 made it into Ebbu’s coffers, the company had an obligation to disclose the names of the original investor or investors. Cooper said neither Dooma’s college friend nor anyone with his last name has ever been an Ebbu note holder. Cooper also said he is not aware of Dooma’s wife ever being a note holder and Dooma’s camp said she never was. Dooma’s spokesman did not respond to specific questions about the contents of the audio.

In the recording, Cooper said that Dooma’s embellishments have imperiled Ebbu, “I don’t think you understand the severity of what’s been going on and what’s happening.”

“It’s more severe for me than it is for you,” Dooma responded.

“It could be severe for everyone who’s associated with the company,” Cooper said. “If stuff got out, if all this stuff was publicized…it could affect every single person in the company.”

“Correct,” Dooma said.

“Right?”

“Oh Yeah.”

“Everyone out of a job, everything under.”

“Correct.”

NOBODY UNDERSTANDS THIS PLANT

Ebbu’s quest to understand the mercurial plant continues. The Ebbu investors I spoke to seemed unconcerned about Dooma’s exaggerations and it appears that in many cases he told prospective investors accurate figures. "In any private deal,” one investor said, “verbal numbers are going to be corrupted and it's up to the individual investor to do their due diligence.” Others who sat for Dooma’s fundraising pitch say he was always candid about the risks Ebbu investors faced. An early Ebbu funder, Darren McCammon, said he remains confident in the company’s direction. “I would be more upset if I had lost Coop,” he said.

Since Dooma’s departure, the company has established an oversight committee. Many Ebbu staffers are probably still unaware of the recording Dooma made. In March, someone claiming to be an Ebbu employee sent me an anonymous note saying there has been a purge of Dooma’s loyalists. “They're wiping any possible memory of the founder,” the employee wrote. “But we all know what they did to him.”

Through his spokesman, Cooper said the company has made inquiries to the company's investors, “so that we can understand what occurred and take any remedial measures."  The round is now closed at $7.9 million, the same amount raised before Dooma left; Dooma is no longer invested in the company. Cooper said operations are stronger than ever. The company is working on a product line called Ratio that will contain specific ratios of THC and CBD. Several competitors already sell similar products, but it’s a step toward the Feelings, which the company now says will be out in early 2017. Perhaps the strongest indication that Ebbu can recover and thrive is that in the audio, Cooper said everything a responsible CEO should, without knowing he had to.  

For as long as I knew him, Dooma always presented money as a secondary motivation. His orations about providing “safer” alternative to alcohol sometimes rang hollow, but his insistence that Ebbu would change the world never did. Joshua Rubin, who co-wrote Assassin’s Creed 2, told me Dooma “is excited by opportunity, to do things. He wants to go for the ride.”

In a forgiving light, then, his actions were likely what he thought Ebbu’s outsize ambitions required given Colorado’s restrictions on fundraising.

When we last spoke, in mid-February, before his recording and the investment documents surfaced, he sounded upbeat and said he’s starting a new marijuana venture, one that would surpass Ebbu. “I’m on a mission,” he said. Next time, though, it won’t be in Colorado.

Alex Halperin is a freelance reporter in Denver. He writes the newsletter WeedWeek.

Photography by Morgan Rachel Levy

People are doing amazing things with motion capture these days. (via colossal)

Little Data, Decentralised Web, Testing Distributed Systems, and Mastering Programming

1. Building Image Classification Models Using Very Little Data -- just a few hundred or thousand pictures from each class you want to be able to recognize.
2. Decentralised Web Summit (New York Times) -- see also Kevin Marks' notes.
3. Readings in Distributed Systems Testing -- crowdsourced on Twitter.
4. Mastering Programming (Kent Beck) -- Here are ways effective programmers get the most out of their precious 3e9 seconds on the planet.

Continue reading Four short links: 8 June 2016.

As a child, Danica McKellar played Winnie Cooper on The Wonder Years. After the show was over, McKellar had difficulty breaking away from other people's perceptions of her. But in college, she discovered an aptitude for mathematics, went on to have a theorem named after her -- not because she was famous but because she'd helped prove it -- and forged a new identity. (via @stevenstrogatz)

I have a problem lately: When I look in the mirror, I see a left-wing extremist. I’m uneasy about my strengthening belief that Free Enterprise is gonna ruin everything good unless we take a knife to its testicles first.

I think we need to:

1. Tax the crap out of the 1% [disclosure: I’m one],

2. stamp out most forms of high-leverage financial speculation,

3. introduce ruthless transparency such that any asset whose beneficial ownership cannot be established where legally appropriate is subject to summary confiscation,

4. adopt a zero-tolerance posture on business crime, with jail time regularly administered for significant financial misdeeds, in rough proportion to the size of the takings, and

5. roll out a universal basic income to deal with the inevitable decline in the proportion of humans “enjoying” full employment.

I think we can all agree that these are extreme measures. And inside, I feel like this reasonable mild-mannered guy. But there’s overwhelming evidence that we’re facing extreme problems; read on.

Taking these steps would require strong unpleasant language and make many powerful and persuasive people unhappy. At the moment, the critical mass to make these things happen isn’t there. But it feels to me like it’s steadily less and less not-there.

Readings

I’ll spare you paragraphs of my own discourse: As a political rhetorician, I’m a fine software geek. But there’s some awfully good, awfully strong things being said out there; here are some:

Unnecessariat by “Anne Amnesia” is messy, disorganized, data-dense, and darkly beautiful, a raw scream out of the raw parts of Middle America. Out-take: “Lets be honest- Clinton doesn’t give a shit about me. When Clinton talks about people hurt by the economy, she means you: elite-educated white-collar people with obvious career tracks who are having trouble with their bills and their 401k plans. That’s who boomed under the last president Clinton, especially the 401ks. Me, or the three guys fighting two nights ago over the Township mowing contract, we’re nothing. Clinton doesn’t have an economic plan for us. Nobody has an economic plan for us. There is no economic plan for us, ever. We keep driving trucks around and keep the margins above gas money and maybe take an odd job here or there, but essentially, we’re history and nobody seems to mind saying so.”

On Revolutionary Attitude by Craig Murray is way more British and polished than Unnecessariat but perhaps even angrier: “Society is so obviously broken to the disadvantage of the many, that to indulge those who, from self-interest or media brainwashing or nostalgia, support the status quo is not helpful.”

Americans Don’t Miss Manufacturing — They Miss Unions, from Ben Casselman at stats-wonk site FiveThirtyEight, is unsurprisingly data-heavy: “a third of production workers — non-managers working on factory floors and in related occupations — earn so little that their families receive some form of public assistance such as food stamps or the Earned Income Tax Credit”.

Finally, On Some Issues, Moral Appeals Don't Seem To Work by John Judis at big-league US-liberal blog TPM, is short but focused on a point I want to expand on: While you can rally round and move the needle on social-liberal projects (in this case the US Deep South’s Neanderthal anti-sexual-minority laws), nobody is willing to do much about the poor, who are getting crapped on at least as comprehensively as trans people.

Anne Amnesia echoes this: At some point starting in the Seventies, the gay community woke up and wouldn’t take it any more, and they’ve won a lot of their fights. The powers that be realized, first, that it was stupid and indefensible to discriminate against people for being born that way; and second (more important) it wouldn’t cost them anything to stop discriminating.

Which is why, obviously, it’s easier to address the issues of smaller factions like the LGBTQers, than the big race and gender power issues, which you can’t touch without major economic dislocation.

I wonder, now that the social-progressive program is largely (thank goodness) scoring wins, whether we might direct some of that awesome energy to economic-progressive ends.

Questions

“Are you a Marxist?” I don’t think so, and I’ve even read Marx. While he said smart things about class conflict and talked about “workers alienated from the means of production” (translation: a lot of jobs suck), Marx didn’t say much useful about how to fix the problem, and was appallingly data-illiterate (see Piketty). Anyhow, even if he was right, there isn’t a proletariat any more.

“But if you cripple the Free Market, won’t that plunge us into poverty?” It might; the free market is one of Homo sapiens’ greatest inventions. But it’s profoundly artificial, relies on the Rule of Law and a functioning court system and a central bank and all sorts of other public apparatus.

And it’s here to serve us, not the other way round. Guess what: If the executive class were whacked back to the single-digit millions they accumulated a few decades back, instead of all the zeroes they accumulate now, there’d still be lots of people knocking themselves out to build companies and be the boss.

Let me quote Piketty: “There is no statistically significant relationship between the decrease in top marginal tax rates and the rate of productivity growth in the developed countries since 1980.”

I remember having lunch with some Deutsche Bank execs in 2009, with lots of financial-crisis blood still on the floor. They were feeling sorry for themselves: “These politicos don’t get the market, they want to regulate the hell out of us, make the banking system run like the civil service.” Hm, that doesn’t sound obviously worse than the way the banking system runs today. But it was the cloud-castle sense of entitlement that, all these years in the distant rear-view, still makes me angry.

“Won’t technology create enough jobs to replace the ones it’s destroying, like it always has?” Maybe, but I don’t think so. In particular, I think self-driving vehicles are going to blow a hole in employment for the relatively-unskilled that may never heal.

“But this century has seen massive poverty reduction in Asia, are you against that?” No, and I hope that the Indians manage to repeat the China trick. But I don’t see any reason why the emergent Asian middle class isn’t gong to end up whipsawed just like Middle America’s is, once the manufacturing boom eases off.

“But what about intersectionality?” Seems to me that the nastiness I observe in the world is way better explained by old-fashioned economic class interests than by any intersectional theory I’ve read. Lots of socially-progressive Silly Valley titans are working diligently to disempower workers by turning them into “independent contractors”.

Next?

Well, that’s the problem. I’ve presented my best policy ideas above but, like I said, I’m a server-side Net geek not a political economist. Anyone know any political parties with ideas like those? If I were American, I’d be in the Sanders/Warren camp. In Canada, I usually vote NDP, but without much joy.

I think the ground is fertile. I think the “conventional wisdom” which sustains the current finance-centric rentier economy is thought wise by fewer and fewer. I think the path from here to something saner will have messy and ugly parts. But I’m increasingly sure that our current path, as a society and species, is unsustainable.

• What I am using AWS Lambda for
  Good set of notes on using AWS Lambda.
  (tags: aws lambda orchestration )
• Shit I’m a Geek / The joy of Piezoelectricity | leafcutterjohn.com
  Leafcutter John's notes on piezos and preamps
  (tags: audio diy piezo microphones contactmics )

Jobloss Prediction, Transparent Displays, Friendly Machine Learning, and Rethinking Knowledge

1. OECD: Automation Unlikely to Destroy Large Numbers of Jobs -- In contrast to other studies, we take into account the heterogeneity of workers’ tasks within occupations. Overall, we find that, on average across the 21 OECD countries, 9% of jobs are automatable. The threat from technological advances thus seems much less pronounced.
2. Transparent Monitors Available -- Though transparent OLED displays are much beloved of sci-fi movie directors, the main reason for this seems to be that it allows the camera to interact better with actors in a situation that is notoriously hard to film. Both Samsung and Planar seem to be struggling for practical applications of see-through displays that have no touch capability, outside of the exhibition sector. (via Slashdot)
3. Aerosolve -- AirBnB open source a machine learning library designed from the ground up to be human friendly. This library is meant to be used with sparse, interpretable features such as those that commonly occur in search (search keywords, filters) or pricing (number of rooms, location, price). It is not as interpretable with problems with very dense non-human interpretable features such as raw pixels or audio samples.
4. Rethinking Knowledge in the Internet Age (David Weinberger) -- That wouldn’t mean that we have given up on the project of knowing our world. Instead, it would mean that that project has changed shape—from content to networks that mix ideas and sociality. That’s already happening. Knowledge is becoming what happens when links connect differences and people.

Continue reading Four short links: 19 May 2016.

Self-Driving Uber, Shenzhen for Startups, Share FDA Docs, and Hackable Ethereum Contracts

1. Uber has Self-Driving Car on Pittsburgh Streets -- The National Highway Traffic Safety Administration has said it could have self-driving car guidelines ready by July. California, Nevada, Utah, Arizona, North Dakota, Michigan, Tennessee, Florida, and Washington, D.C., have enacted autonomous vehicle legislation. Pennsylvania lawmakers and state Department of Transportation officials are working on similar laws.
2. Startups Find Mecca in Shenzhen -- "We’ve got a saying here that a week in Shenzhen is like a month anywhere else," said Hax partner Duncan Turner. "Teams come here for three and half months and basically get a year's worth of R&D done, which really sets them ahead."
3. StartupFDA -- a repository on GitHub where medical device startups are encouraged to open source and share their 510(k) submissions with each other. The idea is to demystify the process and show how much variation there can be in a successful submission for FDA clearance. (via The Macro)
4. Ethereum Contracts are Going to be Candy for Hackers -- Dan Mayer cites research showing industry average bugs per 1,000 lines of code at 15-50 and Microsoft released code at 0.5 per 1,000, and 0(!) defects in 500,000 lines of code for NASA, with a very expensive and time-consuming process. My review of Ethereum Smart Contracts available for inspection at dapps.ethercasts.com shows a likely error rate of something like 100 per 1,000, maybe higher.

Continue reading Four short links: 20 May 2016.

The Atari Domain, Designing Experiments, AI Challenge, and Future of Coding

1. Asynchronous Methods for Deep Reinforcement Learning -- The best performing method, an asynchronous variant of actor-critic, surpasses the current state-of-the-art on the Atari domain while training for half the time on a single multi-core CPU instead of a GPU. Furthermore, we show that asynchronous actor-critic succeeds on a wide variety of continuous motor control problems as well as on a new task involving finding rewards in random 3D mazes using a visual input. Two things: "on the Atari domain" (12-year-old me would love to know this is the state of the art in AI measurement) and "actor-critic" (12-year-old me would be very familiar with the idea of an internal critic).
2. A First Course in Design and Analysis of Experiments (PDF) -- textbook on designing experiments. Golden.
3. Vindinium -- an Artificial Intelligence programming challenge. You have to take the control of a legendary hero using the programming language of your choice. You will fight with other AI for a predetermined number of turns, and the hero with the greatest amount of gold will win. See earlier comment about 12-year-old me.
4. The End of Code (Wired) -- “I was just having a conversation about that this morning,” says tech guru Tim O’Reilly when I ask him about this shift. “I was pointing out how different programming jobs would be by the time all these STEM-educated kids grow up.” Traditional coding won’t disappear completely—indeed, O’Reilly predicts that we’ll still need coders for a long time yet—but there will likely be less of it, and it will become a meta skill, a way of creating what Oren Etzioni, CEO of the Allen Institute for Artificial Intelligence, calls the “scaffolding” within which machine learning can operate. Future of coding (AI? formal methods? provably secure?) is super interesting to me. Send me links if there's something you see that interests you: @gnat on Twitter. kthx.

Continue reading Four short links: 18 May 2016.

wired.com - The Friday afternoon news dump, a grand tradition observed by politicians and capitalists alike, is usually supposed to hide bad news. So it was a little weird that Elon Musk, founder of electric c...

You are probably equipped with ears as sharp and precise as the world’s top sound professionals. What you lack, then, is training.

In instrumental music education, the very first thing you do is teach ear training and keyboard skills. And you do that with students who are, frankly, terrible. But anyone can learn perfect pitch (or something close to it, at least). You couple those ears with an understanding of theory, and you’re able to navigate common musical practice.

Yet in audio, I’ll bet the vast majority of people making music and working in sound have never had an equivalent training. And yet the essentials of sound, and tools like compression and EQ, are just as dependent on training your ears. You need to couple that training with a sense of theory.

Right in your browser, listen to equalization by band and amount with actual recordings.

And music software developers are starting to realize that the more you know, the more you’re likely to use their tools.

iZotope had already added extensive education to their offerings – teaching you how mastering works being a great way of getting more mastering customers. But now, they’re doing something much further-reaching, which they call Pro Audio Essentials.

And it’s not just lessons – it’s also a game.

You start with tutorial content. At the helm for iZotope is their education director, Jonathan Wyner, who has a Grammy nomination under his belt. (It’s for work on a “Best
Classical Crossover Album” nominee, in case you’re wondering. See also his article on mastering.)

Video tutorials cover the basics.

If you’re entirely new to this stuff, you’ll start at the very beginning – like what equalization is. You can get deeper into stuff that you might be a little shaky on if it’s been a while since you studied audio (like the difference between dBFS, Peak Level, and RMS). Those videos are all on-demand, so you can skip them if you’re not interested.

What I suspect almost everyone might be interested in is the Explore, Practice, and Challenge tools.

A series of tests lets you practice and challenge yourself – starting with equalization as a topic.

These let you play with tools like equalization directly in your browser window, and hear the results. Then, you can actually test and train your hearing. So, with equalization, this means the ability to try to detect difference in amounts of attenuation (in dB), and specific frequency bands.

In Challenge mode, you can not only test yourself, but actually try to make targets, matching existing files by ear.

And yeah, that could be a fun exercise even if you’ve got a fair bit of experience.

Quiz your ears.

It’s kind of a wonderful reversal of current trends, if you think about it. We’ve got plug-ins that purport to match EQ for you; cloud tools that are supposed to replace the mastering engineer. In this case, you’ve got something better: you can actually train your own ears so you can take control of getting the results you want. And since the ability to hear amplitude and frequency matters in creative work, too, that means the power to make your own choices as you produce, too.

So far, lessons on equalization, compression, and digital audio basics are all available as videos. There’s a practice and challenge mode for equalization (the others aren’t done yet).

It really isn’t an advertisement for iZotope products. It’s really something you could apply to anything; it’s just likely to build your hearing confidence in such a way that might make you a better customer.

It seems like a great idea. And it’s terrific to watch this industry wake up to education as being part of what they do. (Ableton’s efforts with Loop and a book on creative strategies also spring to mind.) I expect that trend will spread.

Give it a go, for free:

https://pae.izotope.com/

The post Learn audio skills as a game, free, with your ears as guide appeared first on cdm createdigitalmusic.

Back in March, I filed a Freedom of Information request with the FBI asking if the agency had ever wiretapped an Amazon Echo. This week I got a response: “We can neither confirm nor deny...”

Read more...

There’s a simple antidote to the endless circular “pop music was better in the old days” debate: actually listen to the songs. A new interactive tool lets you do just that. But warning: it may not be as pleasurable to turn back time as you think.

We all know how this goes – music used to be better, and then accelerating in the digital / Millennial-driven age of the past decade or so, it got much worse.

The problem is that your memory of pop music past is already filtered to what you like, whereas a listen of the current top pop hits isn’t. That means that when we recall the great music of the past, while we will invariably think of some chart-toppers, we’re also likely removing a bunch of stuff that didn’t stand the test of time.

Now, I can explain that, but it’s much easier to hear it. And an interactive Flash-based tool, calling on streaming music, does just that.

http://polygraph.cool/history/

There’s a little subtle editorializing by the creators. Apart from hoping this will shut you up about music in the old days, the player by default whisks you to 1997. That’s two decade ago – so if music has gotten steadily worse, it should be at least marginally better than what you’ve got now – but it takes you straight to Hansen’s “MMMbop.” No amount of 90s nostalgia is likely to help you forgive that number, or at least you can’t rank it terribly high as far as lyrical sophistication.

Some other trends are clear. Musical genre and instrumentation have changed – literally, pop music now is basically unrecognizable from pop music in 1958. The year I was born, 1978, is dominated by disco. The constant is change itself, if anything, more than it is something we might measure in quality. Loudness wars are real, too, but we might recall that this is partly the impact of corporate-owned radio trying to sound as loud as possible, combined with technological advancements in both processing and listening devices (for better or for worse).

Nor is any of this linear. Cue up 2015, and you’re treated to Mark Ronson’s “Uptown Funk!” – a clear stylistic throwback.

Another significant change is how much more dominant non-white and female artists are in recent music. It’s significant to note the impact of payola and out and out racism on the charts from decades past – another reason why I think, whatever musical merits they might have, complaints about today’s music need to be judged with some suspicion.

Even having made these arguments before, though, I’m struck listening by how much unremarkable music is on the older charts. Some stuff is simply terrible; other tracks are fine but dated or bland. Musical quality may not always progress, and no one who believes in independent taste is going to celebrate the commercial music industry to begin with. But one thing we do have is some benefit of distance in judging music.

Oh, and by the way, I also find that even as I’ve defended recent pop, the pop music on charts from the past few years is often actually pretty darn good. It’s louder, it’s more digital, to be sure – but it also shows some bravery in production and sound and there’s some good writing. Basically, whatever the year, pop charts have some cheese and some substance, and you have to do your own filtering. But isn’t that what you’d expect in a mass media popularity contest? Wouldn’t you expect that sometimes the masses do latch on to things you’d like some of the time, and at other times don’t?

Sure enough, the folks at Polygraph have other experiments hooking into Spotify and so on that explore these issues.

Most significant is a set of charts that compares chart performance from when a track was released to its play count on Spotify today. Some of those results are puzzling – I’m not sure why Spotify fans prefer the Bee Gees over Gloria Gaynor, exactly. Others are telling: Nirvana’s “Smells like teen spirit” is simply a lot more popular than it was when it came out.

And, as usual, looking at the Spotify results – there’s no accounting for taste.

http://poly-graph.co/timeless/

And these tools should tell you the other fundamental difference between Now and Then: now we have a lot more data. We have actual listening data, thanks to streaming, and not only sales. We have the ability to instantly listen to any song, ever – as this tool demonstrates. And I think we have so much data, in fact, that it’s almost impossible to predict what the impact of data will be on the business.

Tony Visconti got a little worked up at this year’s South by Southwest, fashioning a dystopian science fiction story:

Taking place 10 years in the future, the story revolves around a senior A&R expert at the planet’s only remaining major record label. (It’s called The Universe.) Businesspeople around the world sync their schedules by taking drugs that block their circadian rhythms, letting them control their sleep schedules; the label releases just one single a week, recorded by the winner of a lottery and crafted by the label’s employees. The A&R employee spends entire days listening to Jimi Hendrix records and dreaming of the good ol’ days. When he begs his boss to sign a talented street musician, he’s told signing artists based on talent is too risky. Distraught, the employee commits suicide by leaping from the balcony of his Sydney condo. Visconti started getting choked up as he read the story’s final lines.

David Bowie’s producer is terrified by the music industry’s ‘downward spiral’ [The Verge]

While I have indescribable respect for the man, it seems while he may think he’s some sort of combination of Philip K Dick and Arthur Miller … he should possibly stick to the music business.

It’s unclear how much real input Mr. Visconti has for this vision. For one thing, even his story is revealing – it’s fully rooted in the old music business, since it ignores the fact that recording and releasing and even promoting music now takes far less risk than it once did. It may get the big pop side of the equation right, but it ignores plenty of other trends. And no matter how talented he may be, Visconti lacks a certain humility in failing to recognize the fact that his own evaluation is effectively self-centered on the model of his own career. (Well, they did ask him to speak, so he’s certainly entitled to that perspective – and we should listen – but we should also assume it won’t paint a complete picture. Or produce great sci-fi.)

The real question I have: with access to data like Spotify’s, how long will something like a Billboard chart – or even the pop moniker – remain relevant? Is it great pop music that’s endangered, or old-fashioned charts? My money is on the latter. In the meantime, we’ll keep listening.

Oh, and the one thing that is definitely dead? Your productivity. Now that you have those links. Sorry. (Yes, what will kill music will actually be interesting interactive infographics, because we’ll all fail to get work done. I need some of those circadian-rhythm destroying drugs about now.)

Previous rant on the topic:
Music is getting worse, unless, of course, it isn’t

The post Listen to how pop hits have evolved over the years appeared first on cdm createdigitalmusic.