• Sunday Photo Gallery: Life in Death

    Today is a travel day. We’re flying to Dublin for a couple of days, and then back home. So here’s something to entertain you while I’m offline.

    Rebecca Louise Law is a London artist who specializes in installations constructed primarily of dried plants and flowers. Her largest piece to date, “Life in Death,” is on display at Kew Gardens, and it’s really quite lovely. It takes up a single large room and consists of hundreds of strands of flora hanging from the ceiling. Visitors can stroll through the room—a limited number at a time—and to my surprise they have no problem with taking as many photographs as you want.

    The lighting is fairly harsh, but the dried flowers and the wooden walls soften it into warm tones. Here’s a selection of photos:

  • Quote of the Day: The EPA Declares War on Elitist Clickbait

    The EPA under Donald Trump has suddenly reversed course on regulating a raft of chemicals. The New York Times asked them for comment and got this in return:

    “No matter how much information we give you, you would never write a fair piece,” Liz Bowman, a spokeswoman for the E.P.A., said in an email. “The only thing inappropriate and biased is your continued fixation on writing elitist clickbait trying to attack qualified professionals committed to serving their country.

    Before joining the E.P.A., Ms. Bowman was a spokeswoman for the American Chemistry Council.

    Do they make these folks attend a “How To Talk Like Donald” class before they join the administration? Is the entire government going to become Duck Dynasty? Do I even need to ask these questions?

    At this point, I think James Mattis is the only person in the Trump administration who has still managed to avoid embarrassing himself. How long before Trump targets him and forces the issue?

  • Weekend Video

    A few days ago I mentioned that District Line trains are open for their entire length, which means you can watch the train swinging in and out of curves. It’s kind of mesmerizing.

    And naturally I want to share. So here’s a video of the entire 1:12 minute trip from Victoria to St. James’s Park. It’s kind of shaky because that’s how it goes holding cheap equipment with one outstretched hand. But that just adds a touch of realism anyway. Watch the poles go back and forth!

  • Condolencegate Shows, Yet Again, That Donald Trump Is a Terrible Human Being

    This whole row over Donald Trump’s condolence call to Myeshia Johnson is, possibly, the most Trumpian controversy ever. I was inclined to ignore it at first, since Trump’s actual conversation with Johnson didn’t strike me as all that bad. The fact that he was apparently winging it on the call is typical Trump, but a non-hostile reading of what he said suggested to me that he expressed himself a little poorly but nothing much more.

    But he couldn’t just leave it at that, could he? Of course not. He’s Donald Trump. So he blurted out something about Obama and other presidents not making condolence calls. This is a typical off-the-cuff Trump lie: something vaguely plausible that he invents on the spot to defend himself. Naturally, it prompted the press to check into this, and they reported that Trump was lying. They also uncovered the fact that he had promised $25,000 to a father he had called but had never delivered it. Then, instead of leaving bad enough alone, he decided to call congresswoman Rep. Frederica Wilson a liar, even though her account of the call to Myeshia Johnson was perfectly correct.

    Then, because everyone close to Trump eventually pays a price, Trump roped chief of staff John Kelly into the whole thing, claiming that Obama never called Kelly when Kelly’s son was killed in Afghanistan. Kelly was then forced to offer a self-indulgent attack on everyone, and especially on Rep. Wilson, who he said had acted badly at a building dedication in 2015. Video of the event showed that Wilson had done none of the things Kelly said. Was he lying? Did he just misremember? Who knows.

    But naturally no one backed down. That’s verboten in Trumpland, even in the face of video evidence. So of course Sarah Huckabee Sanders defended Kelly, saying that it was all but unpatriotic to even question the word of a 4-star Marine general.

    The key thing here is that all of this was completely avoidable. All of it. Trump just had to provide a vague answer to the condolence call question on Monday and none of this would have happened. It’s 100 percent a result of the fact that Trump can’t abide even the tiniest hint of criticism and apparently has no control over his own response. No matter how dumb and self-destructive it is, he will automatically lash out and then continue to escalate things forever. Quite aside from his political beliefs, it’s yet another confirmation that Trump is simply a terrible human being.

    For now, the only result is the character assassination of a congresswoman and the descent into the muck of Trump’s chief of staff. It’s horrible, but we’ll all live through it. Next time, however, we might not be so lucky.

  • Friday Super Cat Blogging – 20 October 2017

    For two weeks I’ve seen no cats in London close enough to photograph. Last night it started raining and I figured the jig was up. No cats.

    Then, this morning, we decided to go up to Portobello Road and browse the antiques. We got off the bus and Marian suddenly motioned for me to slow down. The proximate cause was this little cutie-pie of a cat who came over and let us scratch her head while she smooched around the gate:

    A minute later, I saw a shop across the street that announced itself as a Pet Boutiqué & Cafe Lounge. Was this one of those cafes that has cats roaming around? No, but it did have this magnificent Grumpy Cat wannabe lounging on one of its pink stools:

    Two cats this week! But we’re not finished yet. Remember that gray cat from last week who was prowling around Edwardes Square Park but disappeared into the foliage before I could get a picture? On our way home, we saw him again. He clearly didn’t want to make a new friend, but he was kind enough to pose for a glamour shot before he hightailed it away:

    Not done yet! We got home, and around dusk the tiger-stripe cat showed up again. He was friendly and came up to me, but then I made the huge mistake of picking him up. That was OK, but then I walked into the house and he went crazy. So this is the best picture I got:

    And there’s more! It turns out the white cat was watching the whole thing:

    After two weeks of drought, it was suddenly pouring cats. So why not put up one more? Last night, thinking that it was time to give up on London cats, I decided to photoshop a picture of my mother’s cats onto some London scenery. Since I went to all that trouble, I might as well put it up. Here are Luna and Lilly keeping a close eye on the pigeons in Trafalgar Square.

  • The Economy Sure Looks Headed for a Fall

    It’s not just e-commerce that’s partying like it’s 1999. The stock market is doing it too:

    Those looking for reason to worry don’t have to search far. There’s a potential war with North Korea and ongoing drama in Washington D.C. But stepping aside from politics, the market has tended to drop when just about everyone thinks the good times will never end. Some people think we’re hitting “peak giddiness” now. Consider these stats:

    63 percent of Americans believe the stock market will be higher a year from now…This is the highest level ever recorded by the survey…60 percent of market experts are bullish and only 15 percent are bearish…The P/E ratio, a widely watched gauge of how expensive the market is, has topped 21 for the Dow…It was 20 heading into the 1987 crash.

    Venture capital looks giddy. The stock market looks giddy. Housing prices look giddy. I’m cautious by nature, so maybe it’s best to ignore me. And on the positive side, we don’t have a huge debt bubble right now that could turn a recession into a great recession. But even an ordinary recession would be painful. Color me one of the worriers.

  • E-Commerce Is Partying Like It’s 1999

    Consider the following business model: you enter into long-term leases for raw office space and build out the interiors with minimal, communal designs. Then you rent out the space for as little as a month to startups and others who aren’t sure how long they’ll need it.

    Sounds reasonable. But there’s a problem: if you call yourself a real-estate company you’re boring. That’s no good. However, if you call yourself a—well, something else, venture capital riches can be yours. The Wall Street Journal explains:

    [CEO Adam Neumann] has said WeWork is neither a real-estate company nor a tech company. The “We Generation,” as he calls it, craves sharing and collaboration rather than isolated offices. “They’re coming to us for energy, for culture,” he said at an event this summer.

    He talks of “space as a service,” a play on the concept of software as a service, in which a provider makes software available to users as they need it over the internet. He calls the company a “platform”—like a computer operating system—from which it can sell other services such as insurance or software.

    Buzzwords! But the chart on the right, from the Journal, shows just how effective this can be.

    And it’s not just real estate that can benefit from this. There are dozens of mundane businesses that have gotten investors excited by insisting that they can disrupt old-school industries by appealing to digitally-obsessed millennials:

    Venture capitalists and mutual funds have poured billions into companies claiming they can upend traditional industries whether through the use of technology or their unique appeal to millennials. Startups in the business of selling meal kits, mattresses and razors have received tech-like valuations based on the idea their rapid growth can continue for years.

    Mattresses? Yes indeed, and here comes the big segue. The “mattress in a box” business is booming, allegedly because it appeals to millennials who wouldn’t be caught dead in a department store or a mattress retailer. But guess what? It turns out that these folks are using boring, old, borderline corrupt marketing to attract business.

    Have you ever wondered about the sites that come up if you google a product? www.microwaves-ranked.com. www.top-ten-sofas.com. www.best-bicycles.com. Are there really that many people around who are obsessed with all these different things? Yes indeed. And they’re obsessed because they get paid to be obsessed. David Zax tells the story of online mattress marketing in Fast Company this month:

    I wanted to learn how Derek Hales had gotten into mattress reviewing, so I called him up in Arizona….In 2012, Derek messaged Samantha Niezwaag, a math teacher, on ChristianMingle.com….They got married in May of 2014.

    ….The young husband and wife needed a new mattress, but were shocked by the prices at the local mattress store….One of Derek’s coworkers told him about a two-year-old Phoenix-based company called Tuft & Needle, which sold its queen-size mattress directly to consumers online for just $600….A few weeks later, in September of 2014, Derek spotted an opportunity. He registered the domain Sleepopolis-Mattress-Reviews.com and threw together a quick website comparing his experiences with Tuft & Needle and Casper (he eventually migrated his content to Sleepopolis.com, which he had also registered). A week later, Derek and Samantha posted a positive video review of their Casper on YouTube.

    ….The question of just how much money Derek made off Sleepopolis interested everyone I spoke to….All told, these numbers suggested Derek may have been making as much as $2 million per year by 2016….Derek had made millionaires among the new mattress entrepreneurs–and he himself was one of them.

    Oh yes, you need to read the whole thing. As you might guess, internet review sites are not entirely neutral in their loyalties. The result has been lawsuits, buyouts, payouts, personal feuds, and more. The story is genuinely fascinating.

    So what ties these two stories together? Basically, that it’s 1999 all over again. Take a standard, boring old business. Slap it on the internet. Use standard, boring old high-pressure sales techniques. Insist that your boring old high-pressure sales techniques are actually new and innovative and will power your company through exponential growth forever. Wait for the money to roll in.

    How long will it last this time? Beats me. But beware of internet entrepreneurs bearing tall tales.

  • Wisconsin Shows How To Do Voter Suppression Right

    Ari Berman has a big piece in our current issue about the effect of Wisconsin’s persistent campaign to suppress the black vote. Because I am the kind of person I am, I’m going to skip the personal anecdotes and go straight to the numbers:

    The state, which ranked second in the nation in voter participation in 2008 and 2012, saw its lowest turnout since 2000….Turnout fell only slightly in white middle-class areas of the city but plunged in black ones.

    ….The impact of Wisconsin’s voter ID law received almost no attention….After the election, registered voters in Milwaukee County and Madison’s Dane County were surveyed about why they didn’t cast a ballot. Eleven percent cited the voter ID law and said they didn’t have an acceptable ID.

    ….In Wisconsin, the intent of those who pushed for the ID law was clear. On the night of Wisconsin’s 2016 primary, GOP Rep. Glenn Grothman, a backer of the law when he was in the state Senate, predicted that a Republican would carry the state in November, even though Wisconsin had gone for Barack Obama by 7 points in 2012. “I think Hillary Clinton is about the weakest candidate the Democrats have ever put up,” he told a local TV news reporter, “and now we have photo ID, and I think photo ID is going to make a little bit of a difference as well.”

    Why does photo ID make a difference? Because blacks and other minorities lack them in larger numbers than whites. If you require photo ID to vote, you put a bigger hurdle in place for blacks, and that naturally suppresses the black vote. Republicans are keenly aware of this, and after the 2004 election that prompted them to begin a nationwide campaign to require photo ID to vote. You can read all about that campaign in my 2012 piece, “The Dog That Voted.”

    But perhaps you’re still not convinced? After all, blacks voted in unusually large numbers for Barack Obama, so it’s only natural that black turnout declined when he wasn’t on the ballot. And that’s true. But check out this chart:

    One of these lines is not like the others. I’ve drawn it in firecracker red just to make sure you see it.

    Pennsylvania, Michigan, and Wisconsin are, of course, the three famous “firewall” states that Hillary Clinton lost by a whisker. In two of them, which lacked new photo ID laws, black turnout was down slightly. Nationally, black turnout was down by seven points, returning to 2004 levels.

    But in Wisconsin, black turnout was down a stunning 29 points. This is not only because Wisconsin passed a very strict photo ID law, but because it made sure to go above and beyond in enforcing it illegally. Molly McGrath, the head of voter outreach for the Wisconsin ACLU, saw this firsthand:

    She had secretly recorded the DMV employees to show that the state was not complying with a court order to distribute voter IDs within a week to people…who did not have access to their birth certificates or other required documents.

    A few weeks earlier, US District Judge James Peterson, who oversaw the implementation of the voter ID law, had found that Wisconsin’s process for issuing IDs was a “wretched failure” that “has disenfranchised a number of citizens who are unquestionably qualified to vote.” Eighty-five percent of those denied IDs by the DMV were black or Latino, he noted in his ruling.

    ….Wisconsin assured the court that such a safety net was in place….The month prior, the Wisconsin Elections Commission had issued a similar release titled “Free Photo ID for Voting Now Available With One Trip to DMV.”

    But McGrath was skeptical. She enlisted her parents, who visited 11 DMVs across the state over the next two weeks to test what would happen to voters like Moore who did not have a birth certificate and wanted to get an ID. In recordings of those encounters, DMV workers said it would “take quite a while” to get the credentials needed to vote, and that it was “hard to predict” when that would be. Only 3 of the 11 DMVs confirmed they would issue a voter ID in a week or less, as the court had ordered.

    The lesson of Wisconsin is pretty simple: Voter suppression can work if (a) an election is close and (b) you’re really dedicated to making it work. Both of these things happened in Wisconsin in 2016, and Hillary Clinton lost the state.

    UPDATE: A couple of notes. First, Michael McDonald of the United States Elections Project says the chart above is flawed. It comes from a Brookings report based on Census data, but it turns out there may be some problems with the Census numbers. I’ve reached out to the author of the Brookings article to see what he has to say about this.

    In any case, please note that the chart is mine, not Ari Berman’s. His article stands or falls based on what he wrote, not on my commentary.

  • Lunchtime Photo

    Along one stretch of Regent’s Canal, there’s a strip of pavement that parallels the towpath opposite the boats. Apparently, the folks who have mooring rights are also allowed to set up little sitting areas on the patch across from their houseboat. These setups revealed a lot of character about their owners, and I’m sorry I didn’t take more pictures of them.

    But I took a few, and this is one of them. It’s one of the simplest and most austere, but on a sunny day it offers a nice respite from the boat itself.

  • Artificial Intelligence Continues to Barrel Ahead

    Lee Jae-Won/AFLO via ZUMA

    I have a long piece in the current issue of the magazine about the way long-term trends in artificial intelligence are likely to cause mass unemployment in the future. Because of this focus, I included only a few brief examples of the current state of AI research along with details about how it works.

    And it’s a good thing, since that stuff is obsolete already. For example, I wrote a sentence or two about Google’s DeepMind computer and how it was able to become the best Go player in the world years before anyone thought it could. But now it’s even better, and this has some lessons for us. Here is Christina Bonnington in Slate:

    On Monday, researchers announced that Google’s project AutoML had successfully taught itself to program machine learning software on its own….On Wednesday, in a paper published in the journal Nature, DeepMind researchers revealed another remarkable achievement. The newest version of its Go-playing algorithm, dubbed AlphaGo Zero, was not only better than the original AlphaGo, which defeated the world’s best human player in May. This version had taught itself how to play the game. All on its own, given only the basic rules of the game. (The original, by comparison, learned from a database of 100,000 Go games.) According to Google’s researchers, AlphaGo Zero has achieved superhuman-level performance: It won 100–0 against its champion predecessor, AlphaGo.

    ….Early AlphaGo versions operated on 48 Google-built TPUs. AlphaGo Zero works on only four. It’s far more efficient and practical than its predecessors. Paired with AutoML’s ability to develop its own machine learning algorithms, this could seriously speed up the pace of DeepMind’s AI-related discoveries.

    This highlights some the things I briefly mention in my article. We have a tendency to think of AI primarily in terms of raw hardware power, and there’s no question that this is important. Full AI will simply never be possible until we have cheap, energy-efficient computing platforms with roughly the computing power of the human brain.

    But “computing power” is a combination of hardware and software. And hardware is a combination of CPU speed, custom chip design, and massive parallellism. In this case, the new AlphaGo machine has become a dozen times more efficient not because Intel has come out with a faster CPU, but via the use of better software, which is executed on Google’s custom Tensor Processing Unit chips.

    This is why I’m so confident that computing power will continue to double every couple of years, just as it has for the past half century. Standard CPUs aren’t likely to keep doubling in raw speed, but they’ll get smaller, cheaper, and more energy efficient. Combine this with better algorithms, better use of parallellism, and custom AI processing chips (which are in their infancy right now), and effective computing power is likely to continue to grow exponentially.

    We’re also in the infancy of making use of AI to help build better AI. Right now this is extremely limited, but that’s the way everything starts. In another decade, AI-assisted AI development is likely to be yet another factor keeping development on an exponential curve. This means it won’t be long before AI starts to get good at tasks that we currently pay human beings to do.

    This is going to put a lot of people out of work starting in about a decade or so, and this won’t be a rerun of the Industrial Revolution. Millions of people will be out of work for good, since by definition, any new jobs created by the transition to AI can also be done by AI. This is something we should be thinking hard about. But as I was researching my article, I was disappointed that even now, when the future of AI seems to be barreling toward us in a way that’s hardly deniable anymore, very little thought is going into this.

    So what can we do? I have a few ideas, but mostly I’m hoping that my piece sparks some more serious discussion. You can join in when it appears online.