Please Look at My Metal Credit Card

Please Look at My Metal Credit Card
Credit-card makers are ditching plastic in favor of something with much more … plunk.
By Amanda Mull
Nov 29 2022
https://www.theatlantic.com/technology/archive/2022/11/metal-credit-cards-status-symbol/672296/

Although it may be difficult to imagine a universe in which George Clooney needs a little help charming women, that’s the case in Up in the Air, the 2009 movie in which he plays a frequent-flying HR consultant in charge of executing mass layoffs. In a Dallas hotel bar, he flirts with a comely business traveler played by Vera Farmiga, needling her over her preferred rental-car loyalty program; soon, the two are comparing mileage goals and flinging their respective stacks of bonus-rewards credit cards down next to their drinks. Eventually, Clooney seals the deal with a rare American Airlines ConciergeKey card, rendered in matte graphite among all the shiny plastic. Farmiga picks it up, complimenting its weightiness. “This is pretty fucking sexy,” she marvels. They retire to his hotel room.

During the 2000s, a metal credit card could have that effect on a person. In 2004, American Express swapped plastic for titanium in its invite-only, unlimited-spending Centurion Card, and one of the most successful credit-card marketing gambits in banking history was born—or, perhaps more accurately, was finally realized. After its plastic introduction in 1999, the Centurion Card—or the Black Card, in popular parlance—became a status symbol known far outside its rarefied clientele, largely thanks to countless namechecks in rap hits by artists including Lil Kim, Jay-Z, Lil Wayne, and Kanye West. Within just a few years, the card’s legend had grown to such mythic proportions—aided by the fact that almost no one had ever seen one in person—that it was somehow widely believed to be made of metal already.

In the time since, metal credit cards have become not only a reality, but a mundanity. Once limited to products like the Centurion that require proof of high net worth and a history of lavish spending, the cards are now available to pretty much anyone with passable credit. Even Venmo, the cash-swapping app, is enticing people to use their balance like a bank account with a metal debit card in pink or black. As a marketing play, the cards are brilliant. But they’re also an object lesson in the life cycle of the consumer status symbol. When everyone’s special, no one is.

Metal credit cards may have begun as markers of extreme wealth, but they were spawned by something far more pedestrian: consumer-loyalty programs. Frequent-flier miles are the most famous of these programs, but they’re everywhere now—hotels, clothing brands, electronics retailers, fast-food chains. They’re especially popular at the top of the glutted credit-card market, where people with good credit and a relatively high income need to be tempted to open and use new cards, even though doing so tends to be expensive and annoying. Promises of free plane tickets, iPhones, and points-accrual multipliers on dining and gas purchases can be enticing perks, but after a while, all the benefits of opening a new card can start to sound the same. Credit-card companies have tried to come up with different strategies to stand out, especially because these usual perks tend not to be part of the everyday user experience; you might cash in for a free plane ticket or an iPhone upgrade once every year or two, but those eventualities are hardly a constant reminder to pluck that card out of your wallet over all the others.

Enter metal. Many people in the credit-card industry point to 2016 as the year that metal cards went wild, thanks to the launch of the Chase Sapphire Reserve Card. The card was itself an upgrade from an existing—merely Preferred—product, and it came with a hefty $450 annual fee at the time of launch in addition to its promises of fast-accruing, easily redeemable points. The shopping public couldn’t get enough of it, according to Nick Ewen, the director of content at the travel-rewards website The Points Guy. So many people applied (Ewen among them) that Chase ran out of metal and had to mail temporary plastic cards. Ewen said that although he believed much of the card’s appeal was in the big bonus-points offer for new accounts and the company’s well-liked rewards program, the metal card wasn’t exactly unrelated to its success. “At the time, it was still enough of a novelty that when you would go and pay for something with the Chase Sapphire Reserve, you would get comments from the waiter or the cashier,” he told me. Elizabeth Crosta, a vice president of communications at American Express, told me that this is referred to in the industry as the plunk factor—a heavier card is more satisfying to plunk down on the table after dinner. It lands with more authority.

That kind of response to a card launch turned heads, Ewen said, and it didn’t take long before most issuers’ fanciest publicly available cards were metal. And then their next-fanciest. American Express, which had long kept metal cards for Centurion high rollers, began issuing less exclusive metal Platinum Cards in early 2017; in 2018, its Gold Cards also made the switch, with a limited-edition rose-gold option for early adoptees. For a long time, the credit-card industry looked at nonfunctional tweaks to the card itself—a college or sports-team logo, for example—primarily as a way to market mid-tier products to people with mediocre credit. When the Chase card became a massive hit, it was suddenly clear that affluent people, too, are delighted by the prospect of a special little card.

Unlike team-logo cards, though, metal cards aren’t meant to signal fandom or allegiance—they’re meant to signal status, and not just of the airline variety. Keeping meticulous track of points balances and bonus offers can pay real dividends when it’s time to redeem those rewards, but a card needs more than that to entice people whose hobbies don’t typically involve spreadsheets. When cleverly branded, credit cards have always made for status symbols so potent that they easily tip over into the absurd, or even parodic—the costume designer Lizzy Gardiner wore a dress made out of gold American Express cards to the 1995 Academy Awards. Lots of people are willing to shell out for things that project wealth and discernment to others. This is the principle on which the entire high-end-fashion industry is based, and metal cards are maybe most accurately described not as a financial tool, but as a luxury accessory.

[snip]

How the Supreme Court Is Erasing Consequential Decisions in the Lower Courts

[Note:  This item comes from friend David Rosenthal.  DLH]

How the Supreme Court Is Erasing Consequential Decisions in the Lower Courts
By Lisa Tucker and Stefanie A. Lindquist
Nov 29 2022
https://www.nytimes.com/2022/11/29/opinion/supreme-court-decisions-vacated.html

The Supreme Court is increasingly setting aside legally significant decisions from the lower courts as if they had never happened, invalidating them in brief procedural orders. The pace of these actions has increased in the past 22 months, neutralizing important civil rights and civil liberties decisions.

Reasoned opinions by the federal appeals courts on issues ranging from voting rights to Donald Trump’s border wall have been wiped from the books, leaving no precedent for the lower federal courts to follow. Legally, it is as if these decisions by the appeals courts, one rung below the Supreme Court, had never existed. The Supreme Court’s final, unilateral exercises of power in these cases have gone largely unreported.

The Supreme Court has broad discretion to decide which cases it hears and accepts only a small number each term. Appeals court decisions that the justices choose not to review typically remain in effect in the judicial circuit where they were decided, operating as precedent to be followed by that appeals court and the courts below it.

But when the Supreme Court vacates a lower court decision, that decision is erased, and any subsequent litigation on the issue must begin anew. Since January 2021, soon after Amy Coney Barrett took her seat on the Supreme Court, expanding the conservative majority, the court has relied on a 1950 decision to vacate 13 politically and legally significant decisions issued by federal appeals courts. That case, United States v. Munsingwear, held that when a case becomes moot during an appeal — meaning there is no longer a continuing controversy — the justices can vacate the decision with an order known as a vacatur.

The recent flurry of Munsingwear vacaturs is sharply at odds with the pace of past court practice. Since a 1994 Supreme Court decision scaled back the use of the Munsingwear precedent, the court has cited the case only 48 times to vacate lower court decisions, according to our research. Slightly more than half of those actions have occurred in just the past five years, after the appointment of Justice Neil Gorsuch, which maintained a 5-to-4 conservative majority. More than a quarter have come in the past 22 months, with the conservatives holding a 6-to-3 majority.

In most of the cases, the vacated federal appeals court opinions were written by judges appointed by Democratic presidents. (In some of the cases, the Biden administration sought the vacatur, probably because the lower court decided against the government and a Supreme Court decision affirming that ruling could restrict a future administration’s options.)

In 12 of the 13 lower court rulings vacated by the justices in the past 22 months, the court erased decisions that seemed to align with progressive values and objectives:

• Two that granted standing to plaintiffs suing Mr. Trump for violations of the emoluments clauses by profiting from his hotels and restaurants.

• One that held that Mr. Trump could not block critics on his Twitter account.

• One that allowed states to be sued directly under the Voting Rights Act.

• Two that allowed Pennsylvania election boards to count undated ballots or to extend mailing deadlines for ballots during the pandemic.

• Two that tossed out work requirements approved by the Trump administration for beneficiaries to receive Medicaid.

• One that held that the House of Representatives had standing to sue the executive branch for violations of its appropriations power in connection with Mr. Trump’s border wall.

• One that prohibited the Trump administration from returning asylum seekers to Mexico under its Migrant Protection Protocols.

• One that allowed the House Judiciary Committee to obtain redacted grand jury materials in its impeachment case of Mr. Trump.

• One upholding a preliminary injunction against an executive order issued by the governor of Tennessee during the pandemic blocking abortion procedures for three weeks.

In another case during the same period, an appeals court upheld an executive order by Gov. Greg Abbott of Texas to prohibit most abortions at the beginning of the pandemic. The order had expired and been replaced, and the court agreed with Planned Parenthood that the case was moot and vacated it.

The most recent vacatur involved a Pennsylvania election dispute that the losing candidate asked the court to vacate. The U.S. Court of Appeals for the Third Circuit had allowed the local election board to count mail-in ballots that did not comply with a state law requiring voters to write the date on the ballot’s outer envelope, on the grounds that the omissions were not material to determining the voter’s qualifications. (One of us, Lisa Tucker, consulted on strategy in this case with her husband, Adam Bonin, a lawyer who had argued against vacating it.)

[snip]

It’s not your imagination: Shopping on Amazon has gotten worse

It’s not your imagination: Shopping on Amazon has gotten worse
Everything on Amazon is becoming an ad
By Geoffrey A. Fowler
Nov 26 2022
https://www.washingtonpost.com/technology/interactive/2022/amazon-shopping-ads/

A previous version of this article incorrectly reported that the “Highly rated” label could be purchased. In fact, Amazon selects who is featured under that label. This article has been corrected.

Amazon is the first app many of us think about to buy things online. But is it actually a good place to go shopping?

When you search for a product on Amazon, you may not realize that most of what you see at first is advertising. Amazon is betraying your trust in its results to make an extra buck.

Let me show you.

We’ll search Amazon for “cat beds.”

Here are the results. Next we’ll put an orange highlight on all of the ads. Believe it or not, that’s pretty much everything you see.

There’s an ad for a brand called Bodiseint at the top. Underneath are three results that paid their way to the top of the “cat beds” listing.

They’re not even very relevant: On the left is a product featuring a photo of dog — yes, a dog — for one of Amazon’s own brands.

On the right is a luxury cat condo that costs $389.

Scrolling to the second screen, we finally start to see non-ads.

These are the first products that were actually chosen because they’ve got the best combination of price and quality.

But the real results don’t last long.

Scroll to the next screen, and it’s all ads again.

Here’s a set of listings labeled “Highly rated,” but don’t be fooled: These aren’t the highest-rated cat beds on Amazon. These are also just ads.

Scroll again, and this screen has even more ads.

These three, under the heading “Top rated from our brands,” are all for Amazon’s own products. (Wait, why is there another dog photo?)

Keep on scrolling, and the ads keep coming — even if they’re repeats.

On these first five screens, more than 50 percent of the space was dedicated to ads and Amazon touting its own products.

This isn’t just a cat problem: The first page of Amazon results includes an average of about nine sponsored listings, according to a study of 70 search terms conducted in 2020 and 2021 by data firm Profitero. That was twice as many ads as Walmart displayed, and four times as many as Target.

Amazon might feel unbeatable for service, fast shipping and easy returns. But as a place to find products, it’s becoming a tacky strip mall filled with neon signs pointing you in all the wrong directions.

Amazon founder Jeff Bezos owns The Washington Post, but I review all tech with the same critical eye.

One of the great promises of the internet is that we can get access to more information and make better choices. Amazon has pioneered a kind of online advertising business that feeds us sponsored information that can cloud our choices. We the users want honest online shopping experiences and some common-sense limits on ads that are designed to deceive.

I call it the “shill results” business. Even when they contain a tiny disclaimer label — as do Amazon’s — these kinds of ads can be misleading because they fill up spaces people have every reason to expect to contain trustworthy, independent information.

What’s worse, many other apps and online marketplaces are following Amazon’s lead. Shill results now crowd Apple and Google’s smartphone app stores — search for an app used for couple’s therapy, and you’ll get an ad for a dating app. (Seriously.) Food-delivery apps shill eggnog and whipped cream when you’re just looking for milk. As I’ve written before, Google search has a big shill results problem, too, though it’s as much about ads as pointing you back to Google’s own products.

Amazon has turned shill results into its next big thing. After selling $31 billion in ads last year, Amazon became the third-largest online ad company in the United States, trailing only Google and Facebook. Some brands and sellers love Amazon ads because they show up right at the moment you’re making a purchase — though others tell me ads have become an extra Amazon tax they have to pass on to customers.

Amazon insists they’re actually a good thing for us. “We are dedicated to providing customers with a world-class shopping experience, including working hard every day to ensure the ads they see are useful, informative, and help make shopping a little bit easier,” said spokesman Patrick Graham.

But in my experience, Amazon’s ads are often not useful, not informative and can make shopping a little bit harder. If you are searching for a cat bed, you have an expectation that Amazon will show you the cat beds that are most useful for you. Not $389 cat beds. Not the pet bed Amazon makes the most money from. Not a weird knockoff.

Let me be clear: Advertising isn’t necessarily bad. When it’s done well, ads can inform us about new products and help new businesses get a foot in the door. It pays the bills for much of the internet — including this news website.

“Right now, consumers are tolerating ads pretty well overall on Amazon, despite the number of them,” said Andrew Lipsman, a principal analyst at market research firm Insider Intelligence. But he warns there could be a tipping point: “There is a very clear tension between advertising and customer experience.”

Amazon told me an internal study found 89 percent of U.S. customers are pleased with results pages. I would like to invite them to run the survey again after showing customers their results with all my orange highlights on the ads.

How everything on Amazon became an ad

The Amazon we experience today is pretty much the opposite of how Amazon used to work.

Even as recently as 2015, Amazon’s results pages were filled with actual results, ranked by relevancy to your search. I found an archive of pages and marked one up compared with the same search today.

Here’s how the Amazon search results for “4k tv” looked in 2015, left, and again in 2022, with the ads marked in orange. There were no ads visible in this much of the first results screen in 2015. (Baymard Institute/Amazon/Washington Post illustration)
What happened? Back in the 2000s, when we started learning to buy all kinds of things online, Amazon was investing heavily in a new kind of shopping science: personalization and recommendations.

Amazon’s mission was to marry up everything it knew about its products with everything it knew about you to help you make the best choices. “The store radically changes based on customer interests, showing programming titles to a software engineer and baby toys to a new mother,” Amazon researchers wrote in an academic paper published in 2003.

This is probably how most of us imagine Amazon still works. But today advertisers are driving the experience.

Story continues below advertisement

Amazon’s focus has shifted from “trying to find ways to delight consumers with great recommendations, personalization and discovery to building better advertising technology,” says industry analyst Juozas Kaziukenas of research firm Marketplace Pulse, who has written about how everything on Amazon is becoming an ad.

Amazon also now uses search results to push its own in-house products. An investigation from The Markup exposed how Amazon results list its own brand and exclusive products ahead of others with higher ratings.

Sure, Google and Facebook are chock full of ads, too. But on Amazon, we’re supposed to be the customers, not the eyeballs for sale. We’re paying Amazon to buy a product, not to mention probably also paying for a membership in its Prime two-day-shipping product.

The reason they get away with it is that busy shoppers can’t easily detect how they’re now promoting the products that are best for Amazon’s bottom line. “Consumers can either ignore ads or assume that the advertised product is good enough,” said Kaziukenas. “Part of the reason why those ads on Amazon and Google work so well is because it’s near impossible for them to perfectly determine the best search results.”

[snip]

Universal flu vaccine may be available within two years, says scientist

[Note:  This item comes from friend David Rosenthal.  DLH]

Universal flu vaccine may be available within two years, says scientist
Vaccine against all strains of virus hailed as major step in protecting against potentially devastating flu pandemic
By Ian Sample
Nov 25 2022
https://www.theguardian.com/society/2022/nov/25/universal-flu-vaccine-may-be-available-within-two-years-says-scientist

A universal flu vaccine that protects against all strains of the virus could be available in the next two years, according to a leading scientist.

An experimental vaccine based on the same mRNA technology used in the highly successful Covid jabs was found to protect mice and ferrets against severe influenza, paving the way for clinical trials in humans.

Prof John Oxford, a virologist at Queen Mary University in London, who was not involved in the work, said the vaccine developed at the University of Pennsylvania could be ready for use the winter after next.

“I cannot emphasise enough what a breakthrough this paper is,” Oxford told the BBC’s Radio 4 Today programme. “The potential is huge, and I think sometimes we underestimate these big respiratory viruses.”

Researchers have been working on universal flu vaccines for more than a decade, but the latest breakthrough, published in Science, is seen as a major step towards a jab that could help protect humans from a potentially devastating flu pandemic.

Seasonal flu vaccines, which protect against up to four strains of the virus, are updated every year to ensure they are a good match for flu viruses in circulation. The new vaccine is designed to prime the immune system against all 20 subtypes of influenza A and B, potentially arming the body to tackle any flu virus that arises.

The world last experienced a flu pandemic in 2009 when a virus that jumped from pigs to humans spread around the world. While that outbreak was far less lethal than health officials feared, the 1918 flu pandemic demonstrated how dangerous new strains could kill tens of millions of people.

Giving people a “baseline” level of immunity against the full range of flu strains could lead to far less illness and fewer deaths when the next flu pandemic happens, said Dr Scott Hensley, a researcher on the team in Pennsylvania. Experiments in mice and ferrets found that the mRNA flu vaccine provoked high levels of antibodies that were stable for several months and protective against the virus.

While the results from the animal tests are promising, clinical trials are needed to see whether the vaccine protects humans in the same way without causing problematic side-effects. The vaccine raises questions for regulators around whether to approve a shot that could protect against viruses with pandemic potential, but which have not yet actually emerged.

“This vaccine has only been tested in animals to date and it will be important to investigate its safety and efficacy in humans,” said Dr Andrew Freedman, a reader in infectious diseases at Cardiff University. “It does seem a very promising approach to the goal of producing a universal flu vaccine as well as vaccines that protect against multiple members of other viral families such as rhino- and corona-viruses.”

[snip]

Using AI to study 12 years of representation in TV

[Note:  This item comes from friend Desire Banse.  DLH]

Using AI to study 12 years of representation in TV
“See It, Be It: What Families are Seeing on TV” study measures representation, inclusivity in the most popular TV shows of U.S over the past 12 years.
By Komal Singh
Nov 18 2022
https://blog.google/technology/ai/using-ai-to-study-12-years-of-representation-in-tv/

The show is a popular medical drama, and two police officers have just rushed a victim to the hospital. Orderlies lift the patient onto a gurney and dash into the ER. Nurses hook up IVs. A doctor takes command, shining a light into the victim’s eyes, barking orders.

Now hit pause. Look closer: Who plays the doctor and who plays the orderlies? Are the cops men or women? Is the victim light or dark-skinned? Who gets the most screen time? Who gets to speak, and who doesn’t?

See It, Be It: What Families are Seeing on TV is a new study that analyzes trends in the screen and speaking time of the visually presenting attributes of the characters — gender, skin tone and age — in scripted television over the last 12 years. The report — led by the Geena Davis Institute on Gender in Media(GDI), in partnership with Google Research as the technology provider, and the Signal Analysis and Interpretation Laboratory (SAIL) at the University of Southern California (USC) as the academic advisor — examines representation in Nielsen’s top 10 scripted U.S. TV shows for each season from 2010 to 2021.

Some of the findings from the report

The report analyzed 440 hours of programming, from hit comedies and political dramas to romances and supernatural sci-fi, to reveal insights like:

• The screen time of female characters is rising, but the male-female gap remains. In 2021, male characters still received about 16 percent more screen time than females.
• The screen time gap between characters with light and dark skin tones narrowed from 81% in 2010 to 55% in 2021. But a large gap still remains. Characters with medium and dark skin tones saw screen time increases of 8 and 9 percentage points, respectively, from 2010 to 2021.
• Older men but younger women get more screen time. The most common age group on screen for male characters appears to be 33 to 60; for female characters, it appears to be 18 to 33. Women over the apparent age of 60 still receive less than 1 percent of all available screen time.
• Speaking time for female characters of dark skin tone rose the most. But they are still the group least likely to speak when shown on screen. The speaking time increased at an average rate of 1.2 % per year, but female characters with dark skin tone are speaking only 16 % of the time.
A first-of-its-kind study using advanced technology

The study, a unique partnership between the GDI, USC and Google Research, is the first collaboration of its kind using advanced AI models at this scale. And the partnership was forged from these organizations recognizing how their missions could complement each other.

GDI was founded in 2004 by Academy Award-winning actor Geena Davis, and received a Governor’s Award Emmy in 2022 in recognition of its work to promote gender balance and foster inclusion, and reduce negative stereotyping, in family entertainment. Madeline Di Nonno, President & CEO of GDI notes: “Although our findings demonstrate progress over the past decade for gender and screen time, there are still significant gaps in gender, age and skin tone and much more work to do.”

Google Research’s MUSE (Media Understanding for Social Exploration) project builds AI-enabled technology designed to understand patterns in how people are portrayed in mainstream media, and to inspire more equitable content. That makes the ongoing partnership with the GDI a natural fit.

To reveal the insights in this study, 440 hours of footage were processed in less than 24 hours. More than 100 frames were processed each second, for a total of over 12 million second-by-second face appearances analyzed using machine learning models. This kind of analysis would be impractical to do manually.

Building off our earlier study of representation in media

Our earlier joint study of gender equity in Hollywood movies was the first use of AI to effectively study representation in media. For this TV study, Dr. Shri Narayanan and his team at SAIL once again provided its deep expertise to co-analyze the data with GDI. As Dr. Shri tells us, “Computational media intelligence not only allows us to gain nuanced insights about the stories we tell using media, at scale and in detail never possible before, but offers new tools for creating inclusive and equitable experiences.”

As the technology evolves, the MUSE project continues to add and improve signals. Google has open-sourced a more inclusive skin tone scale, based on the Monk Skin Tone (MST) Scale developed by Dr. Ellis Monk of Harvard University. Our latest study involves one of the first applications of a skin tone classifier based on MST Scale to study representation patterns in a socio-technical system.1

[snip]

Study finds a striking difference between neurons of humans and other mammals

Study finds a striking difference between neurons of humans and other mammals
Human neurons have fewer ion channels, which might have allowed the human brain to divert energy to other neural processes.
By Anne Trafton
Nov 110 2022
https://news.mit.edu/2021/neurons-humans-mammals-1110

Neurons communicate with each other via electrical impulses, which are produced by ion channels that control the flow of ions such as potassium and sodium. In a surprising new finding, MIT neuroscientists have shown that human neurons have a much smaller number of these channels than expected, compared to the neurons of other mammals.

The researchers hypothesize that this reduction in channel density may have helped the human brain evolve to operate more efficiently, allowing it to divert resources to other energy-intensive processes that are required to perform complex cognitive tasks.

“If the brain can save energy by reducing the density of ion channels, it can spend that energy on other neuronal or circuit processes,” says Mark Harnett, an associate professor of brain and cognitive sciences, a member of MIT’s McGovern Institute for Brain Research, and the senior author of the study.

Harnett and his colleagues analyzed neurons from 10 different mammals, the most extensive electrophysiological study of its kind, and identified a “building plan” that holds true for every species they looked at — except for humans. They found that as the size of neurons increases, the density of channels found in the neurons also increases.

However, human neurons proved to be a striking exception to this rule.

“Previous comparative studies established that the human brain is built like other mammalian brains, so we were surprised to find strong evidence that human neurons are special,” says former MIT graduate student Lou Beaulieu-Laroche.

Beaulieu-Laroche is the lead author of the study, which appears today in Nature.

A building plan

Neurons in the mammalian brain can receive electrical signals from thousands of other cells, and that input determines whether or not they will fire an electrical impulse called an action potential. In 2018, Harnett and Beaulieu-Laroche discovered that human and rat neurons differ in some of their electrical properties, primarily in parts of the neuron called dendrites — tree-like antennas that receive and process input from other cells.

One of the findings from that study was that human neurons had a lower density of ion channels than neurons in the rat brain. The researchers were surprised by this observation, as ion channel density was generally assumed to be constant across species. In their new study, Harnett and Beaulieu-Laroche decided to compare neurons from several different mammalian species to see if they could find any patterns that governed the expression of ion channels. They studied two types of voltage-gated potassium channels and the HCN channel, which conducts both potassium and sodium, in layer 5 pyramidal neurons, a type of excitatory neurons found in the brain’s cortex.

They were able to obtain brain tissue from 10 mammalian species: Etruscan shrews (one of the smallest known mammals), gerbils, mice, rats, Guinea pigs, ferrets, rabbits, marmosets, and macaques, as well as human tissue removed from patients with epilepsy during brain surgery. This variety allowed the researchers to cover a range of cortical thicknesses and neuron sizes across the mammalian kingdom.

The researchers found that in nearly every mammalian species they looked at, the density of ion channels increased as the size of the neurons went up. The one exception to this pattern was in human neurons, which had a much lower density of ion channels than expected.

The increase in channel density across species was surprising, Harnett says, because the more channels there are, the more energy is required to pump ions in and out of the cell. However, it started to make sense once the researchers began thinking about the number of channels in the overall volume of the cortex, he says.

In the tiny brain of the Etruscan shrew, which is packed with very small neurons, there are more neurons in a given volume of tissue than in the same volume of tissue from the rabbit brain, which has much larger neurons. But because the rabbit neurons have a higher density of ion channels, the density of channels in a given volume of tissue is the same in both species, or any of the nonhuman species the researchers analyzed.

“This building plan is consistent across nine different mammalian species,” Harnett says. “What it looks like the cortex is trying to do is keep the numbers of ion channels per unit volume the same across all the species. This means that for a given volume of cortex, the energetic cost is the same, at least for ion channels.”

Energy efficiency

The human brain represents a striking deviation from this building plan, however. Instead of increased density of ion channels, the researchers found a dramatic decrease in the expected density of ion channels for a given volume of brain tissue.

The researchers believe this lower density may have evolved as a way to expend less energy on pumping ions, which allows the brain to use that energy for something else, like creating more complicated synaptic connections between neurons or firing action potentials at a higher rate.

“We think that humans have evolved out of this building plan that was previously restricting the size of cortex, and they figured out a way to become more energetically efficient, so you spend less ATP per volume compared to other species,” Harnett says.

[snip]

The Near Future of AI is Action-Driven

The Near Future of AI is Action-Driven
…and it will look a lot like AGI
By John McDonnell
Nov 15 2022
https://jmcdonnell.substack.com/p/the-near-future-of-ai-is-action-driven

In 2022, large language models (LLMs) finally got good. Specifically, Google and OpenAI have led the way in creating foundation models that respond to instructions more usefully. For OpenAI, this came in the form of Instruct-GPT (OpenAI blogpost), while for Google this was reflected in their FLAN training method (Wei et al. 2022, arxiv). Flan’s which beat the Hypermind forecast for MMLU performance two years early: 

But the best is yet to come. The really exciting applications will be action-driven, where the model acts like an agent choosing actions. And although academics can argue all day about the true definition of AGI, an action-driven LLM is going to look a lot like AGI.

Let’s act step by step

Famously, LLMs often perform better at question-answering tasks when prompted to “think step by step.” (Kojima et al. 2022, arxiv). But they can do even better if they’re given external resources, or what I call external cognitive assets. The ReAct model puts these pieces together (Yao et al. 2022, arxiv). ReAct takes three steps iteratively: Thought (about what is needed), Act (choice of action), and Observation (see the outcome of the action). The actions often make use of cognitive assets like search. The authors give an example below:

Here, the main actions the model uses above are Search and Finish. The authors also provided  Lookup action that can search within the text of the article. In 1a/1b we can see the LLM doesn’t do very well answering the question on its own, because it doesn’t have enough information. In 1c we see that simply giving the algorithm a search function is not sufficient because the model seems to forget what it is doing by the time it hits the end. The authors’ approach of including both actions and specific thought steps is able to search deeply and find the answer.

Answering factual questions is useful, but for real-world tasks, providing LLMs with an array of tools can enable a much broader array of powers. Imagine a customer service LLM able to do React-style chaining. We might give it actions SearchDocs for information about the service and policies, SearchClientFacts to look up particular facts about a client, PythonCode to run python programs, ManualEscalation to submit cases to a human for approval, MessageClient to send the client messages, and Finish to indicate completion. I’ve fabricated a possible client assistance transcript below with React-style formatting:

Question: Client is asking if they can cancel their meal kit order.

Thought 1: I need to know our policy for cancelling orders.
Act 1: SearchDocs[Meal kit cancellation policy]
Obs 1: Meal kits can be cancelled up until 3 days before their ship date, after which manual escalation is necessary.

Thought 2: I need to know what the customer’s ship date is.
Act 2: SearchClientFacts[NextShipDate]
Obs 2: ‘2022-12-03’

Thought 3: I need to know if that is 3 days from now.
Act 3: PythonCode[`from datetime import datetime; (datetime.strptime(‘2022-12-03’, ‘%Y-%m-%d’) – datetime.now()).days`]
Obs 3: 2

Thought 4: This means the cancellation will require a manual escalation. 
Act 4: ManualEscalation[Client needs shipment cancelled.]
Obs 4: Case has been escalated to a human service provider.

Thought 5: I should tell the client what I have done.
Act 4: MessageClient[“This order is happening in 2 days. That means we need to escalate to a manual cancellation. Please wait for a customer support person to reach out to you.”]
Act 4: Message sent successfully

Thought 5: This means I am done.
Act 5: Finish[]

We can now make agents that, on the basis of natural language questions or instructions, can perform actions in a wide action space. The potential is not limited to CX or sales any particular business workflow. Any set of behaviors where an agent manipulates symbols in a computer can in principle be performed with this schema. And it looks an awful lot like AGI.

This is not 10 year tech. It may be possible right now with off-the-shelf tools. But to make it work we need to set up the right feedback loops.

The importance of training

If you were to attempt the above CX example in the latest version of GPT-3, it’s unlikely that you would get consistently good behavior. The LLM needs to understand the power of its own tools and it needs to know what kinds of outcomes we the user desire.

There are many tools for doing this. Google Brain recently achieved SOTA question-answering performance via instruction tuning (arxiv). The secret to OpenAI’s 002-text-davinci model seems to be attributable to a combination of instruction tuning and Reinforcement learning from Human Feedback (RLHF, blogpost), wherein humans rate the success of a given prompt. The ReAct paper takes advantage of a method developed by Zelikman et al. (2022, arxiv) to bootstrap a large number of “valid” chains for training by using LLMs to generate them. I suspect that the very best results will come from actual reinforcement learning where a system can actually be trained to produce better results as measured via a metric of interest. For example, consider a model generating marketing copy that could be trained on conversion rate data to produce copy with consistently higher conversion.

Implementing a powerful AI feedback loop.

Schematically, the stack for a successful action-driven deployment will look something like this:

In the middle we have the “out of the box” foundation model. Prompts are engineered to ask for outcomes, passed into the LLM, and an output is received. Many of the recent “LLM” startups take advantage roughly of this toolset. Instruct GPT has made OpenAI’s version of this stack particularly easy to get started with.

On the left we have the External Cognitive Assets that can supercharge a model’s power. These can be any function that takes text as an input and provides text as an output, including searches, code interpreters, and chats with humans.

Finally on the right we have the task oriented training that’s needed to make this work well. This is the hard part. Some techniques, like instruction tuning, seem fairly straightforward to implement. RLHF is harder and involves tuning a PPO algorithm (OpenAI post). RL will particularly benefit from proprietary datasets, especially usage logs.

Some startups will become very successful creating powerful feedback loops: Solving a customer pain point (maybe bootstrapping by starting with something very simple), collecting data about how to solve that better, training their models to be more consistent, and iterating. This is roughly what a moat will look like in AI, at least for now. But as the agents get more domain-general, the spaces that can be automated and offerings that are possible will expand.

[snip]

Is the world’s richest person the world’s worst boss? What it’s like working for Elon Musk

Is the world’s richest person the world’s worst boss? What it’s like working for Elon Musk
Elon Musk’s track record as a boss is an endless scroll of impulse firings, retribution, tone-deafness on race — and the impregnation of a subordinate.
By Russ Mitchell
Nov 14 2022
https://www.latimes.com/business/story/2022-11-14/elon-musk-toxic-boss-timeline

In the last two weeks, thousands of Twitter employees have gotten a small taste of what it‘s like to work for Elon Musk: the out-of-nowhere firings, the threats and the bluster, the pubescent jocularity, the day-to-day uncertainty and the urgent demands to work through the night.

If there’s such a thing as a warm and cuddly boss, Musk has long been the opposite to his employees, who now number more than 100,000. He burns through executives with the heat of a battery fire. He takes criticism personally, even when it’s a matter of worker or customer safety. He’s been known to fire people on a whim. Since buying Twitter, his public image is shifting fast, from self-described techno-king to unpredictable court jester and human tornado.

Because Musk makes new employees sign tough nondisclosure agreements, and because he’s developed a reputation for exacting retribution on those who cross him, we’ll never know all the stories.

But there’s plenty in the public record. Personal attacks. Union busting. A casual attitude toward factory floor injuries and other health concerns. A dismissive approach to workplace racism. And an allegation involving a horse and sexual favors.

Anger management

Musk’s short fuse is legend. If something’s wrong and it’s important to Musk, employees have learned to avoid his presence if possible.

Back in 2017 Musk was looking for someone to blame after his plans for cutting-edge automation at Tesla’s Nevada battery factory began chewing up factory productivity.

According to Wired writer Charles Duhigg, in a story called “Dr. Elon & Mr. Musk,”the CEO was apoplectic, trying to figure out what was wrong and who was to blame when he summoned a young engineer over to assist him.

“Hey, buddy, this doesn’t work!” he shouted at the engineer, another employee told Duhigg. “Did you do this?”

“You mean, program the robot?” the engineer said. “Or design that tool?”

“Did you f— do this?” Musk asked him.

“I’m not sure what you’re referring to?” the engineer replied apologetically.

“You’re a f— idiot!” Musk shouted back. “Get the f— out and don’t come back!”

Tim Higgins’ 2021 Tesla book, “Power Play,” offers several different looks at how Musk vents his anger. “Musk’s fury caused several executives to leave the company, Higgins wrote, including Peter Rawlinson, the executive leading the development of the Model S, who left Tesla to found the electric-car company Lucid Motors.

When asked about his temper, Musk has said he doesn’t do rage firings, but provides “clear and frank” feedback.

Toxic avenger

A pattern: If Musk perceives he’s been crossed, he does more than seethe — he seeks retribution.

Musk’s quickness to lash out when wounded was on display in the defamation trial that resulted after a British diver suggested that a mini-submarine developed by Musk to rescue youth soccer players trapped in a cave wouldn’t work and Musk responded by calling the diver a pedophile. (Musk won the trial.)

More than once, he’s displayed the same spirit of retaliation toward employees who raised issues at one of his companies. 

After Model S engineer Cristina Balan emailed Musk about what she saw as serious safety issues in product design, she was escorted to what she thought would be a face-to-face with him. Instead, she was led into a security room and fired.

For years Balan has been trying to sue Tesla for defamation. But Tesla lawyers have been able to keep Balan’s evidence from being presented to a judge.

Another whistleblower, Martin Tripp, moved to Hungary to escape the wrath of Musk after the news site Insider ran a story about excessive scrap waste at Tesla’s battery factory in 2018. Private investigators hired by Musk to identify the source named Tripp, a factory employee.

Tripp was fired. Tesla said he stole company data. Musk later called a reporter to say he’d heard Tripp was on his way to the factory with a gun. The local sheriff’s department later said, no, he was miles away in Reno, with no gun and no evidence he had one.

Accidents happen

Twitter employees don’t work with dangerous, heavy machinery, like factory workers do. Good thing, based on numerous reports about the safety culture at Tesla over the years.

In May 2017, The Times detailed the safety record at Tesla’s Fremont, Calif., factory. Tesla’s injury incident rate topped that of some industries commonly associated with especially dangerous work, including sawmills and slaughterhouses. Tesla did not dispute the numbers but said that it was “learning how to be a car company” and that “what matters is the future.”

The injury rate did improve. But in 2018, the public radio investigative reporting program “Reveal” alleged that Tesla was leaving injuries off the books.

One way Tesla lowered its injury numbers, according to “Reveal,” was by denying ambulance service to some injured factory workers who requested it. Medical staff were told not to call 911 without management permission.

“The electric car maker’s contract doctors rarely grant it, instead often insisting that seriously injured workers — including one who severed the top of a finger — be sent to the emergency room in a Lyft,” “Reveal” said, quoting five former medical clinic employees at Tesla’s Fremont auto assembly plant.

Lockdown showdown

With COVID-19 ripping through the nation in March 2020, and counties across California ordering “shelter in place” lockdowns, Musk defied the orders and kept the plant open.

Under official pressure, Musk temporarily closed the plant. In May he announced the plant would reopen. Employees could stay home, he said, but wouldn’t be paid.

County public health head Erica Pan and other officials told him it wasn’t safe. The factory would reopen, he said. If officials didn’t like it, they could arrest him. He called Pan “unelected & ignorant” and deemed stay-at-home orders “fascist.” He threatened to move Tesla headquarters from California to Texas. And, in 2021, he did.

Health officials said at least several hundred Tesla factory workers were infected by COVID.

Union, jacked

Tesla’s factories are union free, but when workers in Fremont tried to organize, Musk cracked down hard. Tesla later was cited by the National Labor Relations Board for repeatedly violating U.S. labor law, including the firing of a union leader and forcing workers to remove clothing with messages that supported the union. The company was also ordered to remove a Musk tweet threatening the disappearance of employee stock options should a union vote prevail.

Boundary issues

Most companies discourage executives from fraternizing with staff too intimately because of the messy conflicts it can create. That did not stop Musk from secretly having twins with a top executive at his brain implant company Neuralink, according to an Insider report.

Shivon Zilis, the 36-year-old executive, told Neuralink managers that the twins were born through in vitro fertilization, according to Reuters, and that she didn’t have a romantic relationship with her 51-year-old billionaire boss. Musk has said underpopulation is one of the biggest threats to civilization.

Zilis has continued working as Neuralink’s director of operations and special projects.

In another strange episode, Musk found himself tethered in what’s become known as the horse-for-sex scandal.

Business Insider also broke that story, which revolved around a lawsuit filed by a woman who said she was hired to provide massage services to the world’s richest human.

She alleged she was summoned by Musk aboard his Gulfstream G650ER private jet for a “full body massage” and Musk showed her his erect penis, then “touched her” and “offered to buy her a horse” in return for sex. SpaceX paid the woman $250,000 in a legal settlement.

[snip]

The Pandemic Broke the Flu … Again

The Pandemic Broke the Flu … Again
COVID knocked flu, RSV, and other respiratory diseases out of whack. When will they be back to normal?
By Katherine J. Wu
Nov 23 2022
https://www.theatlantic.com/health/archive/2022/11/flu-rsv-covid-next-winters/672252/

In the Northern Hemisphere, this year’s winter hasn’t yet begun. But Melissa J. Sacco, a pediatric-intensive-care specialist at UVA Health, is already dreading the arrival of the one that could follow.

For months, the ICU where Sacco works has been overflowing with children amid an early-arriving surge of respiratory infections. Across the country, viruses such as RSV and flu, once brought to near-record lows by pandemic mitigations, have now returned in force, all while COVID-19 continues to churn and the health-care workforce remains threadbare. Most nights since September, Sacco told me, her ICU has been so packed that she’s had to turn kids away “or come up with creative ways to manage patients in emergency rooms or emergency departments,” where her colleagues are already overwhelmed and children more easily slip through the cracks. The team has no choice: There’s nowhere else for critically ill kids to go.

imilar stories have been pouring in from around the nation for weeks. I recently spoke with a physician in Connecticut who called this “by far the worst spike in illness I’ve seen in 20 years”; another in Maryland told me, “There have been days when there is not an ICU bed to be found anywhere in the mid-Atlantic.” About three-quarters of the country’s pediatric hospital beds are full; to accommodate overflow, some hospitals have set up tents outside their emergency department or contemplated calling in the National Guard. Last week, the Children’s Hospital Association and the American Academy of Pediatrics asked the Biden administration to declare a national emergency. And experts say there’s no end to the crisis in sight. When Sacco imagines a similar wave slamming her team again next fall, “I get that burning tear feeling in the back of my eyes,” she told me. “This is not sustainable.”

The experts I spoke with are mostly optimistic that these cataclysmic infection rates won’t become an autumn norm. But they also don’t yet fully understand the factors that have been driving this year’s surge, making it tough to know with certainty whether we’re due for an encore.

One way or another, COVID has certainly thrown the typical end-of-year schedule out of whack. Respiratory viruses typically pick up speed in late fall, peak in mid-to-late winter, and then bow out by the spring; they often run in relay, with one microbe surging a bit before another. This year, though, nearly every pathogen arrived early, cresting in overlapping waves. “Everything is happening at once,” says Kathryn Edwards, a pediatrician and vaccinologist at Vanderbilt University. November isn’t yet through, and RSV has already sent infant hospitalizations soaring past pre-pandemic norms. Flu-hospitalization rates are also at their worst in more than a decade; about 30 states, plus D.C. and Puerto Rico, are reporting high or very high levels of the virus weeks before it usually begins its countrywide climb. And the country’s late-summer surge in rhinovirus and enterovirus has yet to fully abate. “We just haven’t had a break,” says Asuncion Mejias, a pediatrician at Nationwide Children’s Hospital.

Don’t miss what matters. Sign up for The Atlantic Daily newsletter.

Previous pandemics have had similar knock-on effects. The H1N1-flu pandemic of 2009, for example, seems to have pushed back the start of the two RSV seasons that followed; seasonal flu also took a couple of years to settle back into its usual rhythms, Mejias told me. But that wonky timetable wasn’t permanent. If the viral calendar is even a little more regular next year, Mejias said, “that will make our lives easier.”

This year, flu and RSV have also exploited Americans’ higher-than-average vulnerability. Initial encounters with RSV in particular can be rough, especially in infants, whose airways are still tiny; the sickness tempers with age as the body develops and immunity builds, leaving most children well protected by toddlerhood. But this fall, the pool of undefended kids is larger than usual. Children born just before the pandemic, or during the phases of the crisis when mitigations aplenty were still in place, may be meeting influenza or RSV for the first time. And many of them were born to mothers who had themselves experienced fewer infections and thus passed fewer antibodies to their baby while pregnant or breastfeeding. Some of the consequences may already have unfurled elsewhere in the world: Australia’s most recent flu season hit kids hard and early, and Nicaragua’s wave at the start of 2022 infected children at rates “higher than what we saw during the 2009 pandemic,” says Aubree Gordon, an epidemiologist at the University of Michigan.

In the U.S., many hospitals are now admitting far more toddlers and older children for respiratory illnesses than they normally do, says Mari Nakamura, a pediatric-infectious-disease specialist at Boston Children’s Hospital. The problem is worsened by the fact that many adults and school-age kids avoided their usual brushes with flu and RSV while those viruses were in exile, making it easier for the pathogens to spread once crowds flocked back together. “I wouldn’t be surprised,” Gordon told me, “if we see 50 to 60 percent of kids get infected with flu this year”—double the estimated typical rate of 20 to 30 percent. Caregivers too are falling sick; when I called Edwards, I could hear her husband and grandson coughing in the background.

By next year, more people’s bodies should be clued back in to the season’s circulating strains, says Helen Chu, a physician and an epidemiologist at the University of Washington. Experts are also hopeful that the toolkit for fighting RSV will soon be much improved. Right now, there are no vaccines for the virus, and only one preventive drug is available in the U.S.: a tough-to-administer monoclonal antibody that’s available only to high-risk kids. But at least one RSV vaccine and another, less cumbersome antibody therapy (already being used in Europe) are expected to have the FDA’s green light by next fall.

Even with the addition of better tech, though, falls and winters may be grueling for many years to come. SARS-CoV-2 is here to stay, and it will likely compound the respiratory burden by infecting people on its own or raising the risk of co-infections that can worsen and prolong disease. Even nonoverlapping illnesses might cause issues if they manifest in rapid sequence: Very serious bouts of COVID, for instance, can batter the respiratory tract, making it easier for other microbes to colonize.

A few experts have begun to wonder if even milder tussles with SARS-CoV-2 might leave people more susceptible to other infections in the short or long term. Given the coronavirus’s widespread effects on the body, “we can’t be cavalier” about that possibility, says Flor Muñoz Rivas, a pediatrician at Baylor College of Medicine. Mejias and Octavio Ramilo, also at Nationwide, recently found that among a small group of infants, those with recent SARS-CoV-2 infections seemed to have a rougher go with a subsequent bout of RSV. The trend needs more study, though; it’s not clear which kids might be at higher risk, and Mejias doubts that the effect would last more than a few months.

Gordon points out that some people may actually benefit from the opposite scenario: A recent brush with SARS-CoV-2 could bolster the body’s immune defenses against a second respiratory invader for a few days or weeks. This phenomenon, called viral interference, wouldn’t halt an outbreak by itself, but it’s thought to be part of the reason waves of respiratory disease don’t usually spike simultaneously: The presence of one microbe can sometimes crowd others out. Some experts think last year’s record-breaking Omicron spike helped punt a would-be winter flu epidemic to the spring.

Even if all of these variables were better understood, the vagaries of viral evolution could introduce a plot twist. A new variant of SARS-CoV-2 may yet emerge; a novel strain of flu could cause a pandemic of its own. RSV, for its part, is not thought to be as quick to shape-shift, but the virus’s genetics are not well studied. Mejias and Ramilo’s data suggest that the arrival of a gnarly RSV strain in 2019 may have pushed local hospitalizations past their usual highs.

[snip]

Time Is Running Out for the Leap Second

Time Is Running Out for the Leap Second
To the world’s timekeepers, the leap second is a kludge, a bane, a pain in the little hand. Now they’re proposing to ditch it. Will our days ever be the same?
By Alanna Mitchell
Nov 14 2022

https://www.nytimes.com/2022/11/14/science/time-leap-second.html

Roughly every four years, an extra day gets tacked onto the end of February, a time-keeping convention known as the leap year. The practice of adjusting the calendar with an extra day was established by Julius Caesar more than 2,000 years ago and modified in the 16th century by Pope Gregory XIII, bequeathing us the Julian and Gregorian calendars.

That extra day is a way of aligning the calendar year of 365 days with how long it actually takes Earth to make a trip around the sun, which is nearly one-quarter of a day longer. The added day ensures that the seasons stay put rather than shifting around the year as the mismatch lengthens.

Humanity struggles to impose order on the small end of the time scale, too. Lately the second is running into trouble. Traditionally the unit was defined in astronomical terms, as one-86,400th of the mean solar day (the time it takes Earth to rotate once on its axis). In 1967 the world’s metrologists instead began measuring time from the ground up, with atomic clocks. The official length of the basic unit, the second, was fixed at 9,192,631,770 vibrations of an atom of cesium 133. Eighty-six thousand four hundred such seconds compose one day.

But Earth’s rotation slows ever so slightly from year to year, and the astronomical second (like the astronomical day) has gradually grown longer than the atomic one. To compensate, starting in 1972, metrologists began occasionally inserting an extra second — a leap second — to the end of an atomic day. In effect, whenever atomic time is a full second ahead, it stops for a second to allow Earth to catch up. Ten leap seconds were added to the atomic time scale in 1972, and 27 more have been added since.

Adding that extra second is no small task. Moreover, Earth’s rotation is slightly erratic, so the leap second is both irregular and unpredictable. Fifty years ago, those qualities made inserting the leap second difficult. Today the endeavor is a technical nightmare, because precise timing has become integral to society’s highly computerized infrastructure.

“What was before just a way of measuring the flow of time is today essential for transportation, location, defense, finance, space competition,” said Felicitas Arias, former director of the time department of the International Bureau of Weights and Measures, known as B.I.P.M. from its French name and based outside Paris. “Time is ruling the world.”

The process of squaring these two time scales has become so unruly that the world’s time mavens are making a bold proposal: to abandon the leap second by 2035. Civilization would wholly embrace atomic time; and the difference, or tolerance, between atomic time and Earth time would go unspecified until timekeepers come up with a better plan for reconciling the two. A vote, in the form of Resolution D, is expected on Nov. 18 at a meeting in Versailles of the Bureau’s member nations.

“From a technical point of view,” said Patrizia Tavella, the current director of B.I.P.M.’s time department, “all the colleagues all over the world agree that we have to do something.”

Farewell, heavens

If the resolution passes, it would sever the timekeeping of atoms from the timekeeping of the heavens, probably for generations to come. The change would be indiscernible for most of us, in practical terms. (It would take a few thousand years for atomic time to diverge as much as an hour from Earth time.)

But the second is a huge amount of time in the technology of the internet. Cellphone transmissions, power grids and computer networks are synchronized to minuscule fractions of a second. High-frequency traders in financial markets execute orders in thousandths and even billionths of a second. By international law, data packages related to these financial transactions must be time-stamped to that fine level of precision, recorded and made traceable back to Coordinated Universal Time, the universally agreed-upon standard managed by the timekeepers at the B.I.P.M.

Every additional leap second introduces the risk of confusion: that some digital networks won’t implement the change correctly, won’t know precisely what time it is with regard to the other systems, and will fail to synchronize properly. The leap second is a dollop of potential chaos in a soufflé that demands precision.

For that reason, discarding the leap second has wide support from nations across the world, including the United States. The result of the vote is not a foregone conclusion, however. The fate of the leap second has long been the stuff of high diplomatic drama, designated one of just four “hot topics” at the B.I.P.M. Getting Resolution D on the agenda has involved more than two decades of study, negotiation and compromise to resolve the issue.

“It should have happened 20 years ago, and if not for political maneuvering, it probably would have happened 20 years ago,” said Judah Levine, a physicist at the National Institute of Standards and Technology, or NIST, in Boulder, Colo. He is co-chair with Dr. Tavella of the B.I.P.M. committee that discusses hot topics, and he helped draft the resolution.

Russia, for instance, has tried to delay a shift away from the leap second because doing so would require extensive alterations to its GLONASS satellite system, which incorporates the extra second. As a result, the resolution has been phrased to postpone any change until 2035. The United Kingdom, historically and emotionally tethered to the astronomical standard, enshrined in Greenwich Mean Time, has been reluctant to commit publicly.

The fate of the leap second is more than just the fate of the leap second. At stake is Coordinated Universal Time, the international standard for timekeeping, which the continued existence of the leap second is slowly undermining.

Coordinated Universal Time, or U.T.C., is tenderly constructed from readings made by atomic clocks kept at national laboratories around the world. These clocks tick off, or “realize,” their best seconds and send the measurements to the B.I.P.M. There, timekeepers painstakingly assemble the readings — averaging, weighting, adjusting for discrepancies — into an ideal second for everyone everywhere to agree on and employ, occasionally adding leap seconds as needed. This assembly process takes time. And so once a month the Bureau publishes the perfect time in the form of a newsletter, called Circular T, that tells each national clock how much it diverges from the international standard, to help it improve its aim the following month.

Coordinated Universal Time is the world’s official time scale, and will continue to be whether or not it incorporates leap seconds. Global time zones are described in reference to it. (New York time currently is U.T.C. minus five hours.) And the beating heart, the second, is the most important in the constellation of standard measurements overseen by the B.I.P.M., alongside the meter (length), kilogram (mass), kelvin (temperature), candela (intensity of light), ampere (electric current) and mole (amount of substance).

The idea, formalized a century and a half ago by national signatories to an international treaty called the Meter Convention, is that each unit of measurement should be identical everywhere in the world; one meter in Spain is precisely one meter in Singapore. The seven standard units are integral to fair commerce, reproducible science and reliable technology. The second is extra-special because it underpins all the other units except the mole. For instance, the meter is defined as the distance light travels in a vacuum during one-299,792,458ths of a second, and the kilogram was recently redefined in terms of the second.

In addition, the second is tethered to a time scale, or flow of seconds. A key tenet of modern life is that not only must the unit of time be identical no matter where it is measured, so must the flow of seconds of which the one is a part.

But the leap second is putting that tenet at risk. The kludge is so technically difficult for digital technology to incorporate that other, ersatz methods of timekeeping — unofficial, but free of leap seconds and easier to implement — have begun to displace U.T.C., according to a recent article in the journal Metrologia. To supporters of Resolution D, removing the leap second from U.T.C. would make the standard time scale friendlier to modern digital technology, at least in the century following 2035. Coordinated Universal Time would still be universal, just not coordinated with Earth time.

“There is this problem we want to stop, which is this proliferation of pseudo time scales, because they are not time scales in the metrological sense,” Dr. Arias said.

[snip]