Anyone who thinks we shouldve drafted him thinks real life is madden, end of discussion. Real people arent video game characters. Having ABYSMAL pre draft interviews really does mean something, especially when youre own college coaches throw you under the bus, 99% of yall didnt understand how bad that really was lmao HAHAHA Imagine that asshole on a Eberflus lead bears team vs literally the best possible landing spot. A winning team (super bowl winning btw), veteran leadership and he still acts like a little bitch.
Leading AI models can fix broken code, but they're nowhere near ready to replace human software engineers, according to extensive testing [PDF] by OpenAI researchers. The company's latest study put AI models and systems through their paces on real-world programming tasks, with even the most advanced models solving only a quarter of typical engineering challenges.
The research team created a test called SWE-Lancer, drawing from 1,488 actual software fixes made to Expensify's codebase, representing $1 million worth of freelance engineering work. When faced with these everyday programming tasks, the best AI model â" Claude 3.5 Sonnet -- managed to complete just 26.2% of hands-on coding tasks and 44.9% of technical management decisions.
Though the AI systems proved adept at quickly finding relevant code sections, they stumbled when it came to understanding how different parts of software interact. The models often suggested surface-level fixes without grasping the deeper implications of their changes.
The research, to be sure, used a set of complex methodologies to test the AI coding abilities. Instead of relying on simplified programming puzzles, OpenAI's benchmark uses complete software engineering tasks that range from quick $50 bug fixes to complex $32,000 feature implementations. Each solution was verified through rigorous end-to-end testing that simulated real user interactions, the researchers said.
Highlighting the final night of competition in Richmond was a pair of Virginia commits. On the women’s side,Anna Moesch earned the win in the 50 free swimming a 25.15. That was a season best as she swam a 25.16 at the US Open in December. Her best time is a 24.87 from 2023 Summer Y-Nats.
David King swam to another best time at the meet, this time in the 200 back. He won the event in a 2:01.11 surpassing his old best of a 2:01.16 that he swam at the end of April. He already was under the Olympic Trials cut of a 2:01.69. He has now almost dropped a second in the event already this season.
Lleyton Arnold swam to improve his seed at Trials as he swam a 2:14.47 in the men’s 200 breast. That was faster than his old best time of a 2:15.76 that he swam at the end of last summer. The Trials cut is a 2:15.99. Arnold heads to Florida State this fall.
The men’s 50 freestyle came down to the final touch as Nien Levy touched in a 22.77 just ahead of Aiden Bond who swam a 22.81. Levy was just under the Trials cut that stands at a 22.79 sneaking under the cut for the first time as his old best time was a 22.81.
The final win on the men’s side came from Nathan Szobota who swam a 8:18.16 in the 800 free, a 16 second add. Evan Mackesy was 2nd in a 8:23.48.
Earning a win on the women’s side was Sara Czirjak who swam a 2:32.63 in the 200 breast. She already has been under the Trials cut that stands at a 2:31.69 as her best time is a 2:30.86 from the end of April.
Andrea Dworak won the 200 back in a 2:16.93 to finish just ahead of 15 year old Alyssa Sagle who swam a 2:17.26. Both were about three seconds off of the Trials cut.
Leading the way in the 1500 free was Claire Stuhlmacher who won in a 17:19.38. 15 year old Ava Jochims dropped over four seconds to finish 2nd in a 17:25.74.
One day, well into Willie Nelson’s long and illustrious career, a TV presenter asked the musician why he wasn’t writing all the great classic songs like he used to.
“When I wrote those great songs I was cold and hungry,” said Willie. “I’m not cold and hungry anymore.”
Let’s call this “The Willie Nelson Problem.” How does a person or an organization keep its creative vitality once they have already become successful, already become comfortable?
This, we think, is one of the cardinal problems of any type of successful business that has seen a lot of growth. How do we keep the hunger? The sexy?
David Ogilvy had a two word answer to this question:
“Divine Discontentment.”
A primal, massively incurable dissatisfaction with the way things are, that permeates one’s entire being.
He called Divine Discontentment “The antidote to smugness.”
If you’ve always been a self-starter, you know exactly what we’re talking about. Not only are you driven, you’ve always had something driving you. Maybe it’s because you grew up poor and didn’t like it. Maybe because you weren’t popular in school. Maybe because your family told you you’d always be a bum. Maybe because you always had a gift and nobody ever saw it. Or maybe because you and your parents arrived in this country when you were three with nothing more than a few hundred dollars and two carry-ons.
Or maybe because you were damn curious and always looking for ways to ward off boredom.
To commemorate Mr Ogilvy, some time after his death, the firm he founded, Oglivy self-published a little private volume for its employees, called “The Eternal Pursuit of Unhappiness,” based on some of the cultural norms they acquired via David’s guidance. Farnam Street has a great 3-minute summary of the book, which lists the EIGHT virtues they believe a company needs to keep its internal/eternal flame alight.
Courage
Idealism
Curiosity
Playfulness
Candor
Intuition
Free-Spiritedness
Persistence
CS Lewis once wrote that of the Seven Deadly Sins, the worst one was Pride. Why Pride? Because Pride is the one that justifies the other six, the one that is their prime enabler.
Similarly, of Ogilvy’s eight virtues, courage is the most important, because without courage, the remaining seven aren’t going to happen. Like it said in the Ogilvy book, people with serious talent are super rare- people with courage, even more so.
While courage can come from all levels, it must be modeled at the top. Leadership by example, and not by decree, has always been the keystone of any culture, especially in the military. Therefore, rebuilding a culture starts as an act of bravery.
Yes, it might be scary, but it sure beats being cold and hungry.
The boys from Middleton High School outside of Madison grabbed their second consecutive team title, and third overall, last weekend in Waukesha, scoring 273 points.
This year’s meet was the 100th boys’ high school swimming & diving championship in Wisconsin.
While those are a few less points than they scored last season (288), their 95-point margin over runners-up Madison Memorial is bigger than last season (77).
Team Scores – Top 10
Middleton – 273
Madison Memorial – 178
Arrowhead – 177
Brookfield Central/East Co-op – 138
DC Everest – 136.5
Sun Prairie East – 132
Hudson – 126
Franklin – 120
Madison West – 119
West Bend West/East Co-Op – 99
Wisconsin high school swimming has two divisions. Division 1 is the state’s bigger high schools, plus some smaller schools that have combined into large co-ops.
Middleton roared to a win thanks to one individual event title and two relays, including a statement-making Meet Record in the 200 medley relay to open the festivities.
The relay of Max Carter (back – 23.35), Sam Wolf (breast – 24.16), Caden van Buren (fly – 21.53) and Vitense Cowan (free – 21.57) combined for a 1:30.61. That won the race by a two second margin and broke Madison West’s State Record that was set in 2019.
They had the fastest split on each leg of that relay aside from the anchor leg, where DC Everest senior and South Dakota commit David Mayer split 20.01.
The first three legs of that winning Middleton relay were all sophomores, with only the anchor Cowan graduating, so they should be back for more next year. That includes Sam Wolf, who also won an individual event title for Middleton.
First he won the 200 IM in 1:47.97, more than three seconds ahead of Franklin’s Jack Paull (1:51.32). There was only one senior in the top 12 of that race, again indicative of the generally young nature of this year’s meet. Wolf was 3rd last year in 1:48.22 at a veteran-heavy meet that had six seniors in the top eight.
Wolf was 2nd in the 100 free in 44.80 behind Mayer (44.29). That swim for Mayer is faster than any USD swimmers have been this season, showing the big impact he’ll have when he arrives on campus in the fall.
He also won the individual 50 free in 20.34, a flat-start best for him.
While Wolf had the fastest breaststroke split on the 200 medley relay, he didn’t swim the 100 breaststroke individually. There he would have run into Madison West’s Abram Mueller, the defending champion and State Record holder.
Mueller won that event again this year in 53.77, about half-a-second off his winning time of 53.20 from last season.
He also won the 100 fly in 48.59 to defend that title as well. This swim was a new lifetime best, shaving .12 seconds off the time he did at last year’s State meet.
Mueller, a senior, is committed to swim at Stanford next season.
Wolf finished his meet by swimming the anchor leg of Middleton’s winning 400 free relay. Again with three sophomores, the relay included Jackson Esteves (47.21), Ben Cutler Heiderscheit (46.71), Max Carter (45.93), and Wolf (44.65).
Wolf was one of two 44-second splits on that closing relay along with Hudson senior Andrew Hanson, who led them off in 44.66.
Individually, Hanson won the 200 free in 1:36.93, which crushed his best time by 1.6 seconds.
In spite of his progress there, Hanson was upset later in the meet in the 500 free by Muskego junior Carter Jewell. Hanson entered the meet as the defending champion and Meet Record holder, but finished 2nd behind Jewell in 4:28.54 – about four seconds off his time from last year.
Jewell, meanwhile, won in 4:25.75, which smashed his personal best coming into the meet by six seconds (he was 4:31.88 a week earlier at Sectionals). In total, Jewell dropped 19 seconds in the 20223-2024 season, putting the uncommitted junior on to a lot of recruiting radars.
While Middleton ran away with the state title, for the second-straight season, Arrowhead and Madison Memorial battled down to the wire for the runner-up spot. Last year, Arrowhead took 2nd place, but this year Madison Memorial took those honors – by 1 point.
Even though Arrowhead was 2nd in the closing 400 free relay, one spot ahead of Madison West, but they needed to be two spots better to overtake for 2nd place.
Other Event Winners
The Co-op from Brookfield Central and East High Schools won the 200 free relay in 1:25.32 – another winning relay with only one senior. That senior was the leadoff Tom Bergin (21.19). He was followed by Caleb Smith (21.59), Luke Thomas (22.11), and freshman anchor Logan Loppnow. Loppnow may be in his first season of high school swimming, but he didn’t show any nerves in that crucial anchor slot – he split 20.43 to run-down and overcome an eight-tenths deficit to keep Middleton from a relay sweep. Bergin and Loppnow attend Central, while Smith and Thomas attend East.
Mayer was again the best split of the field in the 200 free relay, anchoring DC Everest in 20.09 for 3rd place.
Franklin senior Mason Bruhn won the boys’ 100 back in 49.08, using a huge back-half swim to overcome Middleton sophomore Carter (49.25). That’s a new lifetime best for the Denver commit Bruhn, who was 49.45 to finish in 3rd place last season.
junior Brady Huettl from the Waukesha South/Catholic Memorial co-op won 1-meter diving after finishing 3rd last season. He cracked 500 points, finishing with a score of 501.90.
Middle East expert Bilal Saab told Newsweek that Joe Biden was left to choose "the least bad option" in deciding how to react to recent Houthi attacks.
New submitter axlash writes: It has been reported on X that Niklaus Wirth, inventor and co-inventor of several languages including Pascal, Euler and Oberon, died on Jan 1, 2024. He was aged 89. "We lost a titan of programming languages, programming methodology, software engineering and hardware design," writes software engineer Bertrand Meyer
in a post on X. "Niklaus Wirth passed away on the first of January. We mourn a pioneer, colleague, mentor and friend."
Niklaus Wirth, born on February 15, 1934, in Switzerland, is a renowned computer scientist known for his significant contributions to the field of computer science and software engineering. He is best known for developing several programming languages, including ALGOL W, Pascal, and Modula-2, which have had a profound impact on the design and development of modern computer software. Wirth's work emphasized simplicity, clarity, and efficiency in programming languages, which greatly influenced subsequent language design and the development of structured programming techniques. His legacy also includes the development of the Oberon programming language and the design of the Oberon operating system. Wirth's dedication to elegant and efficient software design continues to inspire computer scientists and software engineers worldwide, making him a highly respected figure in the history of computing.
You can learn more about Wirth via A.M. Turing Award, Britannica, and the Computer History Museum.
The U.S. Food and Drug Administration (FDA) disagrees with the World Health Organization's recent assessment that aspartame possibly causes cancer in humans. "Aspartame is one of the most studied food additives in the human food supply. FDA scientists do not have safety concerns when aspartame is used under the approved conditions," an agency spokesperson said. CNBC reports: The International Agency for Research on Cancer, a WHO body, found a possible link between aspartame and a type of liver cancer called hepatocellular carcinoma after reviewing three large human studies in the U.S. and Europe. Dr. Mary Schubauer-Berigan, a senior official at IARC, emphasized that the WHO classification of aspartame as a possible carcinogen is based on limited evidence. Schubauer-Berigan acknowledged during a news conference with journalists Wednesday that the studies could contain flaws that skewed the results. She said the classification should be viewed as a call to conduct more research into whether aspartame can cause cancer in humans. "This shouldn't really be taken as a direct statement that indicates that there is a known cancer hazard from consuming aspartame," Schubauer-Berigan said.
The FDA spokesperson said the classification of aspartame as "possibly carcinogenic to humans" does not mean the sugar substitute is actually linked to cancer. Health Canada and the European Food Safety Authority have also concluded that aspartame is safe at the current permitted levels, the spokesperson said. A separate body of international scientists called the Joint Expert Committee on Food Additives said Thursday that the evidence of an association between aspartame and cancer in humans is not convincing. JECFA is an international group made up of scientists from the WHO and the U.N. Food and Agriculture Organization. JECFA makes recommendations about how much of a product people can safely consume. The organization maintained its recommendation that it is safe for a person to consume 40 milligrams of aspartame per kilogram of body weight daily during their lifetime. An adult who weighs 70 kilograms, or 154 pounds, would have to drink more than nine to 14 cans of aspartame-containing soda daily to exceed the limit and potentially face health risks.
The U.S. Health and Human Services Department told the WHO in an August 2022 letter that JECFA is better suited to provide public health recommendations about the safety of aspartame in food. This is because JECFA reviews all available data, both public and private proprietary information, whereas the IARC only looks at public data. "Thus, an IARC review of aspartame, by comparison, would be incomplete and its conclusion could be confusing to consumers," Mara Burr, who heads the HHS office of multilateral relations, wrote in the letter. The FDA has a slightly higher recommendation than JECFA and says it is safe for a person to consume 50 milligrams of aspartame per kilogram of body weight daily during their lifetime. A person who weighs 132 pounds would have to consume 75 packets of aspartame per day to reach this limit.
You don’t become a billionaire without being clever with money, but Doug Burgum’s latest scheme is a head-scratcher: The North Dakota governor is offering $20 gift cards to people who donate one greenback dollar to his presidential campaign. His fellow candidate Vivek Ramaswamy is offering a 10 percent commission to anyone who brings in donations. A super PAC supporting Miami Mayor Francis Suarez is offering one donor a year of free college tuition.
Critics have long pointed out how big money distorts politics, but these fundraising stunts demonstrate how now even small money has come to warp campaigns. These GOP candidates are trying to reach a Republican National Committee threshold of 40,000 individual donors, including 200 each in 20 states or territories, to qualify for primary debates. The RNC has set a high bar to make sure that candidates have real support—and perhaps to downplay the influence of major donors. But the effect may be the opposite: enticing candidates to try novel tactics to create the illusion of real support, because they know that getting on the debate stage is essential for remaining viable and attracting those big donors. As the old saying goes, it takes money to make money.
These tactics are the product of a new focus in politics on small-dollar donors, which means that they have not been legally tested. “The whole time I was at the FEC I never saw anything like this,” Ann Ravel, a former chair of the Federal Election Commission, told me. Although she called the moves “quite unseemly,” she said she doubted that they clearly broke existing statutes or regulations. But other experts believe that Burgum’s scheme may fall afoul of the law, even if the others are allowed.
An early forerunner of these plans came in 2020, when the Democratic presidential candidate Andrew Yang announced a plan to give away $120,000—$10,000 a month for one year to 10 families, drawn randomly from people who signed up on his website. The idea was to advertise Yang’s proposal for universal basic income for all Americans. (Ravel, for her part, wrote an op-ed at the time arguing that plan was likely illegal.) This year’s schemes eschew such a connection to policy and, unlike Yang’s, require that citizens contribute money, not just enter their email.
Under Ramaswamy’s plan, anyone can gather donations and take a small cut. In a video, he called it a way to “democratize” the old system and break up the “oligopoly” of traditional fundraising. A better way to describe it would be as Uber, but for campaign finance. Candidates have long paid professional fundraisers to bring in cash. They have also relied on “bundlers,” who persuade friends and acquaintances to give, even though each donor is limited to $3,300 in individual federal donations this cycle. Ramaswamy is basically combining those and applying them to small donations.
Rather than let anyone take a small cut, SOS America PAC is effectively holding a raffle for one big winner. They’re asking people to give to Suarez’s campaign, and they’ll give away a tuition gift to one contributor. You’ll have to shop around if you win, though: With a $15,000 limit, a Floridian could cover a full year at the University of Florida, Florida State, or Florida International (Suarez’s alma mater), but an out-of-state student would be able to cover only part of the cost. (That sum would barely reach a quarter of the cost at the University of Miami.)
Burgum’s approach is the most novel. Campaigns commonly offer swag for donations—write a check, get a bumper sticker—but handing out gift cards goes a step further. It sounds like it ought to somehow be illegal. But is it? Candidates aren’t permitted to buy votes, but giving a dollar doesn’t obligate anyone to actually vote for Burgum. They could just be involved in a neat arbitrage and vote for someone else. Federal law also bans “straw donations”: You can’t give money to your friend or spouse or employee and ask them to donate it. The ban serves two purposes. First, it prevents individuals from circumventing the individual contribution limit; second, it ensures compliance with donor-disclosure laws.
But Burgum isn’t subject to contribution limits, because candidates are allowed to give unlimited amounts to their own campaigns, and the scheme obviously doesn’t violate disclosure requirements. “The statutes and regulations are not clear enough to indicate that there’s any real legal problem with this activity,” Ravel told me.
Other experts are not so sure. “I do not fault anyone for thinking creatively about lawful ways to raise money,” Paul S. Ryan, a longtime campaign-finance lawyer, told me. Yet he believes that giving the gift cards and accepting the donations violate the plain language of the straw-donor law: “No person shall make a contribution in the name of another person or knowingly permit his name to be used to effect such a contribution, and no person shall knowingly accept a contribution made by one person in the name of another person.” He worries that if Burgum is allowed to use the scheme, it will set a “horrible precedent,” where a candidate might use the same method to solicit larger donations, or enlist a wealthy friend.
The RNC is arguably also a victim of the arrangement. “Burgum is trying to defraud the RNC,” Ryan said. “These aren’t unique individual donors; these are fake straw donors. If I were at the RNC I would not accept these donors, because they’re illegitimate.”
A Burgum spokesperson did not immediately respond to a request for comment, nor did an RNC spokesperson. The FEC doesn’t comment on specific cases, and in any event the body’s structure and partisan makeup mean that it has deadlocked on most important issues over the past few years. Ravel was skeptical that the body would enforce any violations even if they exist in these cases.
The argument for grassroots donors is not as simple as it might appear on the surface. As David Byler of The Washington Post has written, small-dollar contributions have driven polarization. But if the goal of donor requirements is to increase the power of small donors, a $20-for-$1 exchange actually does the opposite, increasing the power of one massively wealthy candidate to game the system.
My love for the web has ebbed and flowed in the years since, but
mainly it’s persisted — so much so that as of today, I’ve been
writing kottke.org for 25 years. A little context for just how
long that is: kottke.org is older than Google. 25 years is more
than half of my life , spanning four decades (the 90s, 00s, 10s,
and 20s) and around 40,000 posts — almost cartoonishly long for a
medium optimized for impermanence. What follows is my (relatively
brief) attempt to explain where kottke.org came from and why it’s
still going.
A thought that occurred to me when Jason was on my podcast this month (you should listen! — it’s one of my favorite episodes ever): at 25, Kottke.org is over one-quarter as old as The New Yorker magazine, which was founded in 1925. I’ve grown up thinking The New Yorker had been around “forever”. That makes Kottke.org ... one-quarter of “forever” old? My mind boggles.
The Verge writes:
OpenAI CEO Sam Altman has addressed rumors regarding GPT-4 — the company's as yet unreleased language model and latest in the GPT-series that forms the foundation of AI chatbot ChatGPT — saying that "people are begging to be disappointed and they will be." During an interview with StrictlyVC, Altman was asked if GPT-4 will come out in the first quarter or half of the year, as many expect. He responded by offering no certain timeframe. "It'll come out at some point, when we are confident we can do it safely and responsibly," he said....
When asked about one viral (and factually incorrect) chart that purportedly compares the number of parameters in GPT-3 (175 billion) to GPT-4 (100 trillion), Altman called it "complete bullshit."
"The GPT-4 rumor mill is a ridiculous thing. I don't know where it all comes from," said the OpenAI CEO. "People are begging to be disappointed and they will be. The hype is just like... We don't have an actual AGI and that's sort of what's expected of us."
Asked about how far we are from developing AGI, Altman replied "The closer we get, the harder time I have answering. Because I think it's going to be much blurrier and much more of a gradual transition than people think."
And Altman also addressed predictions that ChatGPT will kill Google. "I think whenever someone talks about a technology being the end of some other giant company, it's usually wrong. I think people forget they get to make a countermove here, and they're like pretty smart, pretty competent. I do think there's a change for search that will probably come at some point — but not as dramatically as people think in the short term."
In talking about an upcoming game (more on that in a bit), Nintendo’s Shigeru Miyamoto and Takashi Tezuka discuss the process they used in designing the levels for the original Super Mario Bros. Much of the design work happened on graph paper.1
Back in the day, we had to create everything by hand. To design courses, we would actually draw them one at a time on to these sheets of graph paper. We’d then hand our drawings to the programmers, who would code them into a build.
Here’s the full video discussion:
Now, about that game… Super Mario Maker is an upcoming title for Wii U that lets you create your own Super Mario Bros levels with elements from a bunch of different Mario games. So cool…I might actually have to get a Wii U for this.
This is pretty much the same process I used when designing levels for Lode Runner back in the day.↩
What if more developers developed a sense for the “maximum” number
of things a product should do — and stopped there?
What if more software firms decided, “Hey! We’ve reached the
absolute perfect set of features. We’re done. This product is
awesome. No need to keep on shoving in stuff nobody wants.”
Sure, this would have risks. Standing still risks becoming
obsolete, as other competitors swoop in.
But it can also just mean you have confidence in your
amazing design.
Indeed, some of my favorite pieces of software feel very much like
the “maximum viable product”. They seem like highly mature apps
that realize they don’t need to significantly evolve new gills or
appendages. For twelve years, for example, I’ve used Scrivener for
writing my articles and books. “Word processing” is a
super-competitive area, but Scrivener hasn’t had any feature creep
I can detect. It stuck to its guns. I’d say the same thing about
Logic Pro: I’ve used it for twelve years now for music production,
and while it’s added new instruments and effects, it has done so
gently — it hasn’t larded its UI with endless features. And it’s
facing tons of competition, too, from Pro Tools and Ableton Live
and others.
I think this is common for a lot of apps that have proven to have staying power. It’s why they have staying power. One way to think of it is that software should be designed a little more like hardware. A 2022 MacBook doesn’t have any more buttons or ports than one from 20 years ago. (In fact, MacBooks have fewer ports.) It’s mostly software where there’s a temptation to keep expanding in scope endlessly.
Space rocks that fell to Earth within the last century contain the five bases that store information in DNA and RNA, scientists report April 26 in Nature Communications. Science News reports: These "nucleobases" -- adenine, guanine, cytosine, thymine and uracil -- combine with sugars and phosphates to make up the genetic code of all life on Earth. Whether these basic ingredients for life first came from space or instead formed in a warm soup of earthly chemistry is still not known. But the discovery adds to evidence that suggests life's precursors originally came from space, the researchers say. Scientists have detected bits of adenine, guanine and other organic compounds in meteorites since the 1960s. Researchers have also seen hints of uracil, but cytosine and thymine remained elusive, until now.
How a society that is so good at creating knowledge can be so bad at applying it.
Why don’t Americans trust the experts?
One answer is that experts get a lot of things wrong and often they pay a big price for those mistakes. From the forever wars to the 2008 financial crisis to the botched pandemic response, it seems Americans are constantly careening from one avoidable catastrophe to another.
Another answer is something like Martin Gurri’s thesis in his 2014 book The Revolt of the Public: The digital revolution has transformed the information space in ways that have empowered individuals and undermined the dominant institutions in society — government, media, the academy — and the elites who run them.
Whatever the causes, America has an expert problem and it’s making it harder and harder to solve its societal troubles. Michael Lewis, bestselling author of books like Liar’s Poker and Moneyball, is taking all of this on in the new season of his podcast Against the Rules.
Lewis is as good a storyteller as we have, and he takes a close look at what’s happened to our trust in experts and expertise. The problem, he says, isn’t that we lack experts; in fact, we have lots of experts and some of them have likely saved your life before. The issue is that we don’t value expertise and are therefore really bad at recognizing it when we see it.
I reached out to Lewis for a recent episode of Vox Conversations to talk about how we got here, why it’s an existential problem,and at what point skepticism of authority becomes pathological.
Below is an excerpt, edited for length and clarity. As always, there’s much more in the full podcast, so listen and follow Vox Conversations on Apple Podcasts, Google Podcasts, Spotify, Stitcher, or wherever you listen to podcasts.
Sean Illing
We live in a society that is very good at creating knowledge and very bad at applying it. Why is that, Michael?
Michael Lewis
It’s a really great question. I think part of it is that we have the luxury of being that way. There are various safety nets for our idiocy.
But why are we so bad at this? If you listen to all the [podcast] episodes this season, although I never quite say it, you might say [the answer could be], “Well, in a lot of spheres of life, there has risen new expertise. It’s very complicated and it’s not easy to understand the expertise, especially for people who are not particularly numerate or statistical or scientific.”
I’ll give you an example: meteorologists. So you’re in Gulfport, Mississippi, right now. There’s a famous old weatherman in Alabama named James Spann. He says he’s the second most famous man in Alabama, behind Nick Saban, the football coach. He’s a wonderful weatherman who’s devoted his life to stopping people from getting killed by the weather in Alabama, which happens often.
And he said 50 years ago when he started, he knew basically nothing. When he got up to predict the weather he’d stick his head out the window and say, “It’s pretty sunny today.” He would give you a very crude understanding of what might happen in the next couple of days. Ten days out, he was useless. It was no better than guessing. He certainly had no idea when and where tornadoes were going to touch down. Not very good at predicting movements of hurricanes. But, he said, he stood up on air and his job was to seem as confident as possible about what he was saying.
Now, flash forward 50 years; he knows a great deal. His 10-day forecast is pretty good. His three-day forecast is three times more accurate than it was 50 years ago. He can tell you with some precision where you’re at real risk of being killed by a tornado and when and how to save yourself from it. Still, his life is a constant stream of grievances from the people who feel he’s misinformed them. If he says there’s a 20 percent chance of rain and it rains, everybody thinks he’s an idiot. They don’t understand the probabilistic nature of forecasting.
You find this in a whole bunch of areas where the expertise is complicated and the understanding is probabilistic. Think about medicine. Doctors knew this before Covid, but now it’s so obvious. Doctors will tell you they know so much more than they did 30 years ago as a profession. They are much more useful to patients. But every day more and more people are walking into the office after reading something on WedMD and they think they know what they’re talking about.
It’s not a full answer, but there’s something about our information environment and our ability to understand people who are making probabilistic judgments that make it difficult to evaluate expertise.
Sean Illing
I’ve said a bunch over the years that I think we’re a society that knows the price of everything and the value of nothing, and maybe that’s part of the reason why we don’t recognize expertise when we see it.
Michael Lewis
I think that’s right. Let me tell you a story that dramatizes what you just said. I wrote a book about the federal government called The Fifth Risk, where I just wandered into the federal government and looked at it as a risk-management enterprise and kept meeting these unbelievably impressive, passionate, mission-driven experts who were just trying to save us from ourselves.
After I finished that book, we had our government shutdown in December 2018 and it lasted for over 30 days. [Hundreds of thousands of federal workers were furloughed] as inessential workers and sent home without pay. And I thought of those people I’d met. I asked for a list of people who’d been sent home from an organization in Washington that was monitoring the situation, called the Partnership for Public Service. It was not a random sample of federal employees. It was federal employees who’d been nominated for one of the awards they gave out every year by someone who thought they’d done something good.
So I took this list and picked someone at random. It was a guy whose name was on the top of the list: Arthur A. Allen. He won the alphabet contest. So I call him up and asked him if I could come visit him and just see what he’s doing. He had nothing else to do. He was sitting at home with nothing to do.
This is a guy who spent his whole career as the lone oceanographer in the Coast Guard search-and-rescue division, where he’d started in the late ’70s. There was a particular problem he wasworking on by himself, and the problem was costing a lot of American lives. It was people being lost at sea. The Coast Guard didn’t know how they drifted in the ocean. And Americans have this unbelievable talent for getting lost at sea, which is a whole other thing. On average, every day, the Coast Guard is saving 10 people who are lost in the sea and losing three. So you’re talking about thousands of people who are getting in this situation every year.
The problem is that if you fall off a boat into the ocean, you’re going to drift differently than if you are in a life raft, or if you’re on top of an overturned sailboat, or if you have a life vest on — you get the point. So if the Coast Guard knows where and when you started, as they often do, they should be able to predict where you are in the ocean four hours later, knowing the currents and the wind and your drift. But they didn’t know the drift, until Arthur A. Allen figured it all out. He spent years of his own free time tossing objects into the Long Island Sound, where he lives, measuring the specific drift of like 80 different categories of objects.
That all sounds boring and tedious, I know. But he reduced the drift to mathematical equations and embedded them in the search-and-rescue software program, and instantly they were able to find people they never would’ve found before. Thousands of Americans are alive because of Arthur A. Allen. And thousands of people are alive around the world because of the work he did here. No one knows who he is. No one pays any attention to him. They furloughed him as if he’s useless.
The punchline to all of this, to your point about the way we treat these experts who save our tails over and over again, is thatwhen I went to go see Arthur to talk to him about what he had done with his life, I spent three days with him, interviewing his family, going to see his old office, going to the Long Island Sound to see where he dropped his objects, asking him every which way the story of his career.
After the three days, I’m going back to the airport to head home and he calls me and says, with real wonder in his voice, “Hey, you’re a published author.” And I said, “Yeah, yeah, I’m a published author.” He says, “You’re like a real deal. You’re a real writer.” And I said, “Yeah.” He said, “Are you going to be writing about me?” And I said, “Yeah, that’s why I spent three days learning how objects drift. Yes. I’m going to be writing about you.” He goes, “Wow. I didn’t expect to get any attention for this.” And I said, “Well, what did you think I was doing for those three days?” He said, “I just thought you were really interested in how objects drift.”
This is the mental world of the government expert. They’re so used to nobody caring about what they do, even when what they do is mission-critical, that they can’t imagine us even taking an interest in them. We so don’t value them that they don’t value themselves.
Sean Illing
You make an offhand comment in one of the early shows about the arbitrariness of social status and how that has a way of obscuring someone’s real value, and I can’t help but think of it now.
Michael Lewis
I really think that we exaggerate status differences and create inequality at our peril. These people we end up shoving into lower-status roles actually know things.
We have these big complicated organizations and agencies and corporations, and when some crisis pops up it usually has a very specific solution, and it’s really unlikely that that specific solution is going to be in the heads of the people who are at the top of the organization. It’s going to be something nitty-gritty and the person is going to be six levels down in the organization. And if you’ve created these barriers between the levels, so that someone who’s six levels down will never be heard by someone at the top, you’re essentially saying, we’re never going to surface the expertise that we need to deal with the problem.
Something like this happened in the pandemic. We had this apparatus for dealing with communicable disease. It was called local public health. That’s who did it, local public health officers. Their status was so low. They were so socially powerless. They still haven’t stepped front and center stage and taken over the thing.
But if I were a television producer booking guests who can explain to America what’s going on with Covid, that’s who I’d book. But you don’t see them because they’re invisible because they’re low-status. You see some fancy-pants person who worked in the White House who doesn’t actually know anything. And this is a broader problem that has been exacerbated by the structure of our society, by these widening chasms, between level one and level two, and level two and level three, and level three and level four, and so on.
Sean Illing
A certain amount of skepticism of expertise and authority is healthy, but at what point do you think that skepticism becomes pathological?
Michael Lewis
That’s an unanswerable question, but I’ll give it a whirl.
It becomes pathological when yourunwillingness to take in what the putativeauthority or expert is saying kills you. It’s pathological when you turn up in the emergency room as a 45-year-old healthy police officer with Covid, as someone in one of our stories does, and he’s circling the drain and refuses to be intubated because, in his view, hospitals are trying to kill people in the ICU — that’s pathological.
It’s pathological when you are running a big Wall Street firm and you’re unable to distinguish between the trader, who’s making a lot of money in your firm, making really dumb bets on the subprime mortgage market, and the person who has actually got a bead on how the subprime mortgage market is working and can explain it to you, but you don’t want to hear it — and so your firm blows up.
You can get away with ignoring a lot of expertise in your life as you move through the world. And I agree that you never want to lose your ability to question the things you’re being told, but it’s also not true that everybody has a right to an opinion about everything. I don’t have a right to an opinion about climate change. Neither does Donald Trump. There are people who study this stuff, their whole lives are devoted to trying to understand it. They are state of the art. It is a scientific consensus. My opinion shouldn’t exist, but people think they have a right to an opinion about it.
Sean Illing
Do you think, on some level, that the world has become so big and so complex that it’s too much for people to make sense of, and the temptation to retreat into conspiracy theory or tribalism is just too irresistible?
Michael Lewis
To default into a narrative that’s fueled by anecdote that happens to come from the small circle of people in your world — I’ve seen this. I’ve been amazed with people I admire and who I think are intelligent who will sit down with me and tell me they’re not getting vaccinated because the vaccine is making people sick.
And they’re not wrong in one way. They know somebody who got sick, but that’s the thing that they pay attention to as opposed to the 1 billion studies that show that you were just so much better off being vaccinated. It’s like you walked into the casino for the first time in your life, looked at all the games, and you saw someone pull a slot machine and they hit the jackpot, and you decide, “Oh, well, the slot machines are the smart game to play here.”
It’s people organizing a complicated world with stories that are basically not true stories. They’re not representative stories. They just happen to be the stories they hear. And if you made me God and said, “Michael, how do you fix this problem?” — if I could do anything, I’d probably start with making everyone take a basic course in statistics. Everybody would have to learn a little bit about data and probabilities, just so they understand the notion of a small sample size, especially a sample size of one.
Sean Illing
Do you have any sense at all of what it would take to rebuild trust in our society?
Michael Lewis
I tend to think of this stuff on such a personal, micro level and not as a broader social thing. Again, if you’re handing me God-like powers, one of the things I’d do is, maybe not create some kind of mandatory national service, but at least strongly incentivize people, when they’re 18 or 19, to spend a year or a year and a half working in some government service where they’re all mixed up with other kinds of people.
Part of the problem is we’re not mixed up enough. It’s much easier to think of “us” and “them” if you’re in Berkeley, California, and you’ve never met anybody from Alabama, or if you’re in Alabama and you’ve never met anybody from Berkeley, California. Or if you are poor and you’ve never met a rich person, or if you’re rich and you’ve never had to do anything with a poor person.
It’s amazing how helpful it is when people have personal experience doing something together, trying to achieve something together with people entirely different from themselves. Then we have a living sense that we’re not all that different. There’s no us and them. We don’t belong in these tribes. It’s not the natural order of things. So mixing up the society more in various ways is one answer I would give.
My own personal answer is what I do with my time. I’ve been trying to write about this in ways that invite people who might be deeply skeptical that anything in the government is good for them to see this in a different light. Like, this thing exists to keep you safe. Think about it that way. This is the thing that storytellers can do to help.
Despite considering myself a cryptographer, I have not found
myself particularly drawn to “crypto.” I don’t think I’ve ever
actually said the words “get off my lawn,” but I’m much more
likely to click on Pepperidge Farm Remembers flavored memes
about how “crypto” used to mean “cryptography” than I am the
latest NFT drop.
Also — cards on the table here — I don’t share the same
generational excitement for moving all aspects of life into an
instrumented economy.
Even strictly on the technological level, though, I haven’t yet
managed to become a believer. So given all of the recent attention
into what is now being called web3, I decided to explore some of
what has been happening in that space more thoroughly to see what
I may be missing.
What does this guy know about cryptography, though?
This is your last chance. After this, there is no turning back. You take the blue pill - the story ends, you stay angry, manipulated and being sold to the system of algorithms that feed on your every move. You take the red pill - you read great content, unfiltered by algorithms and connect to the things you care about.
Will you be THE ONE?
During the month of December, any user that plugs into their Old Reader account will be eligible to win a $25 gift card to AMC Theatres to see The Matrix Resurrections. “Whoa!” We’ll be doing 10 random drawings throughout the month. Winners will be notified through the email address associated with their account.
Researchers have found another 17 malicious packages in an open source repository, as the use of such repositories to spread malware continues to flourish.
This time, the malicious code was found in NPM, where 11 million developers trade more than 1 million packages among each other. Many of the 17 malicious packages appear to have been spread by different threat actors who used varying techniques and amounts of effort to trick developers into downloading malicious wares instead of the benign ones intended.
This latest discovery continues a trend first spotted a few years ago, in which miscreants sneak information stealers, keyloggers, or other types of malware into packages available in NPM, RubyGems, PyPi, or another repository. In many cases, the malicious package has a name that’s a single letter different than a legitimate package. Often, the malicious package includes the same code and functionality as the package being impersonated and adds concealed code that carries out additional nefarious actions.
With nearly instant reaction times, superhuman button tapping frequency, and an inability to fatigue, an AI called StackRabbit can play Tetris better than any human player. But how much better? Well, it can play all the way to the end of the game, which…did you know Tetris ended? I didn’t. But before that happens, it plays flawlessly through hundreds of levels while the game itself is throwing up weirdo color schemes and scores from random places in its memory — the game’s creators didn’t imagine anyone or anything would get anywhere close to these levels. Also, I got surprisingly anxious watching this — it was just so fast with so much constant peril! (via waxy)
1Password is launching a new feature to let users create unique email aliases for logins, much like Apple's iCloud Plus Hide My Email function. From a report: 1Password is partnering with Fastmail to bring its masked email feature to the password manager, giving all users the option of hiding their email addresses from apps and services. "Your email address is your online identity," explains Bron Gondwana, CEO of Fastmail. "If your credentials are compromised in a data breach, having a randomly generated email address adds a second line of defense because it can't be associated with your primary email address, and therefore, your identity." This new masked email feature will be ideal for registering accounts for temporary purposes, like a free Wi-Fi network. But they can also be used to hide your personal email address from any app or service as the aliases don't expire unless a 1Password user manually deletes them.
Ever played Charades and acted out the perfect mime to ‘James Bond’, only to be met with blank stares? That moment when you wonder “Are my friends fools!?”, that’s the curse of knowledge.
Your friends witness your mime without any additional information and are confused. But you act it out already knowing it’s James Bond. In your head you see a debonair Daniel Craig, tuxedo on and martini in hand. It’s so obvious, you think, why don’t they get it? Shocking, positively shocking.
The curse of knowledge. Once you know something, it’s hard to remember what it was like to not know it.
This leads to a huge amount of communication frustration. You’ve encountered the curse of knowledge if you’ve ever directed someone to the official documentation to solve their problem, only to have them return scratching their head. They are lacking the extra context that you have built up, possibly over years, that allows you to see the answer in the documentation.
The curse of knowledge is pervasive. Once you see it, you will notice it all over:
Jargon, idioms and colloquialisms: Telling a colleague with English as their second language that you think their idea is “the bee’s knees!
UX: Jump into your new car and try to adjust the air conditioning. There’s always a learning curve while you adapt to the new UI. A month later, shake your head when a friend joins you in the passenger seat and can’t, for the life of them, work out how the air conditioning works.
Pairing: Working through a codebase without verbalising your thoughts and actions leaves out a tonne of context and will leave your pairee confused.
Code: Encountering working code for which you have no idea what it does. The author certainly knew at the time, but their implementation belies understanding.
Note taking: Looking back at a note you wrote from last week and having no clue what you meant. You lack the context you held in your mind at the time the note was jotted down.
A special mention for corporate websites that use language that seems so vague as to be meaningless. The C-level team have, over decades, built up a notion of what this extremely vague phrasing means, but to outsiders it is completely useless.
When we spot a pattern, we reach for abstractions as a way to generalise. When everyone understands these abstractions it allows us to take shortcuts and communicate faster! But if someone doesn’t understand the abstraction they may feel lost.
Perhaps this whole discussion about abstractions is confusing to you? But what if I give a couple of examples?
We don’t talk about "bipedal mammals that aren’t apes”, we say “humans”.
Teams may talk about ‘Boring solutions’ but this 2-word phrase actually encapsulates an entire ethos about choosing technology that is mature, stable, and not the 'new hot thing’.
Abstractions are great, but giving clear, concise examples helps to get everyone on the same page (to be ‘on the same page’ means to have the same amount of knowledge. Aaaah, when does it end!?).
If you’re writing a How To guide, ask a few users to go through it. Watch them, take note of where they struggle, improve the guide, test again.
A lot of our best practices around user testing are born from our learnings around avoiding the curse of knowledge. When watching a user test it’s easy to fall into the habit of: “Why don’t they know where to click? It’s so obvious. The problem is definitely the user and not my UI.” But this completely avoids the value of the user test. The user is confused; we have an opportunity to make things better!
How it feels to watch a user test your product for the first time.
Code review is another great example. It can be humbling to push up a PR you’re really proud of only to have a colleague comment saying, “I don’t get this”. Humbling, but rewarding, as it gives you the opportunity to produce even clearer code!
A lot of this can be encapsulated in the idiom ‘know your audience’. Don’t assume the other person knows what you’re talking about. Consider what they know and what they might not know. When in doubt, over-communicate and over-explain.
When you explain something to someone and they look confused, it’s easy to become frustrated and think, “Why don’t they get this?”. Instead, dig into why they don’t get it and think about how you can explain it to avoid the confusion. This all comes down to ensuring the right level of context. Too much and you’ll end up explaining the English language, too little and you’ll have confused faces.
There’s a famous line in The Blues Brothers movie when they ask the bartender at the Texas honky-tonk what kind of music they play: “Why we play both kinds; Country AND Western!”
The dangers are easy enough to guess: If you just hire the people you like the most, you’ll end up like the Facebook algorithm, just churning up the same ol’, turning everything into a hermetically sealed echo chamber. Country AND Western, indeed.
Another idea that we come across a lot is because everyone is working longer hours and work has an ever more important grip on our lives, we’re more and more looking for traits in our colleagues that have more to do with our socializing needs (“Would you want to have a beer with this person”), rather than just getting the actual job done. When you share a foxhole with somebody, you don’t care if you want to share a beer with them, you just care how many enemy soldiers they kill. But the blurring of the lines between one’s work and social life confuses the issue.
A final point: we have learned firsthand that the word, “Culture”, is a loaded one. Like the word, “Love”, it embarrasses the grownups. This is because, like the word, “Love”, “Culture” for some is a platitude, they don’t understand that when designed, it is a management system and that it has the potential to connect the deepest parts of ourselves, to our work, which is a beautiful and highly effective thing.
Fantastic ad from 2016 I somehow hadn’t seen until this week. Hilarious, and the humor plays directly into the ad’s effectiveness. Trust me, just watch.
