Web Excursions 2021-05-24

The Mysterious Origins of the Cerne Abbas Giant | The New Yorker

The Cerne Giant is so imposing that he is best viewed from the opposite crest of the valley, or from the air. He is a hundred and eighty feet tall, about as high as a twenty-story apartment building. Held aloft in his right hand is a large, knobby club; his left arm stretches across the slope. Drawn in an outline formed by trenches packed with chalk, he has primitive but expressive facial features, with a line for a mouth and circles for eyes. His raised eyebrows were perhaps intended to indicate ferocity, but they might equally be taken for a look of confusion. His torso is well defined, with lines for ribs and circles for nipples; a line across his waist has been understood to represent a belt. Most well defined of all is his penis, which is erect, and measures twenty-six feet in length. Were the giant not protectively fenced off, a visitor could comfortably lie down within the member and take in the idyllic vista beyond.

Hill figures, or geoglyphs, are scattered across southern England, where chalk downs offer ready-made canvases to landscape artists. Some geoglyphs are relatively recent, such as the Osmington White Horse, a representation of King George III on horseback, which was etched into a coastal hillside about ten miles south of the Cerne Giant in 1808, to celebrate the monarch’s patronage of the seaside town of Weymouth. (Local lore has it that the image—which shows the king riding out of town, rather than into it—so offended him that he never returned.)

The Cerne Giant has also been subjected to broad speculation about his age. “It is supposed to be above a thousand years standing,” an anonymous correspondent to the Gentleman’s Magazine wrote in 1764. The text was accompanied by an illustration—the earliest published drawing of the giant, including measurements—which indicates that in the mid-eighteenth century the giant had the additional physical feature of a ring-shaped belly button. It was only when this was—perhaps accidentally—merged with the erect penis directly below it, in the early twentieth century, that the giant acquired the prominent apparatus for which he is known today. “We need to make due allowance for scale,” Rodney Castleden, one scholar of the giant, has written, calculating that the penis as it currently stands is equivalent to nine inches for an adult male of average height—“a prodigious though not unknown length.” The giant’s unmodified member would, at human scale, measure “a perfectly normal” six inches.

Local folklore has long held that infertility might be cured by sitting on—or, for good measure, copulating upon—the giant’s penis. In the nineteen-eighties, the sixth Marquess of Bath, the late Henry Frederick Thynne, told a reporter that when he and his second wife, the former Virginia Tennant, were having trouble conceiving a child, they paid the giant a visit. “We were very much in the dark about what he could do,” Lord Bath recalled. “I explained the problem and sat on him.” A daughter was born about ten months later. She was christened Silvy Cerne Thynne, and the name of G. Cerne was given as godfather.

By the twentieth century, scholars were venturing more grounded theories to account for the giant’s existence. In the nineteen-twenties, Sir Flinders Petrie, an archeologist, argued that the figure’s proximity to nearby earthworks suggested that it was from the Bronze Age, which extended approximately from 2300 to 800 B.C. Stuart Piggott, another archeologist, linked the name Helis with that of an obscure pagan figure, Helith, who, according to a thirteenth-century chronicler, Walter of Coventry, was once worshipped in the Cerne area. (Few contemporary writers have championed this notion.) In the nineteen-seventies, a geophysical survey of the hillside led to speculation that a lion skin had once dangled from the giant’s left arm, which would explain the figure’s somewhat ungainly pose, and might buttress the Herculean identification. Two decades later, Castleden, the historian, carried out further geophysical investigations, which convinced him that it was a cloak, rather than a lion skin, that once swung beneath the left arm, “as if the Giant is running or because he is waving his arm like a matador.”

An equally rich counter-narrative contends that the giant is younger than the Royal Oak pub, which is thought to have been built in the sixteenth century, with stones repurposed from the abbey after it was demolished during the reign of Henry VIII. The fact that a powerful and wealthy monastery once lay at the foot of the hill is often marshalled as evidence against the idea that the giant dates back that far.

Cromwell was sometimes depicted as Hercules. A statue at Highnam Court, a stately home in Gloucestershire, represents the long-haired Lord Protector with a club in hand, naked but for a tastefully positioned loincloth.

Last summer, Brian Edwards, a visiting research fellow at the University of the West of England, Bristol, proposed an alternative seventeenth-century origin story. In an article in Current Archaeology, Edwards argued that the giant was indeed a Hercules figure, and pointed out that the date of the giant’s first recorded renovation, in 1694, coincided with an annual celebration of King William III’s birthday and also with the anniversary of his invasion of England, in 1688, when he was the Prince of Orange. Edwards said that, of all British leaders, William III was the one most often linked with Hercules.

These stowaway snails—which measure only a few millimetres in diameter across their shells, and are typically found in even smaller fragments—are hard to detect, but their presence in a sample indicates that it dates from the medieval period or after. By last summer, Allen had some preliminary data suggesting that soil deposits contemporary with the giant’s creation contained these late-arriving snails.

He grew up at Minterne House, a seventeenth-century mansion two miles north of the giant, and remembers running around the giant’s trenches as a small child. (Lord Digby’s aunt Pamela Harriman, the late Washington hostess and U.S. Ambassador to France, also grew up at Minterne House, as the daughter of the eleventh Lord Digby. According to an obituary, at the age of twelve she rode her horse up to the giant and jumped over his penis, exclaiming, “God, it’s big!”) The current Lord Digby had no opinion on the question of the giant’s age, but he welcomed the National Trust’s investigation. “The more information the better,” he said.

According to an anonymous note that the perpetrator left at the village shop, the intention was “to elevate the giant into a human rather than a binary gendered ‘him.’ ” Irvine told me, firmly, “I took exception to this. It’s an erect penis, and an erect penis is an erect penis.” Several weeks after the incident, on the night before May Day, he and Moore, along with the village electrician and the village plumber, ascended the hill after the pubs closed, carrying battery-run L.E.D. lights, which they set up to illuminate the giant’s penis and eyes, in an effort to restore his compromised dignity.

In April, a little more than twelve months after the National Trust’s excavation of the giant, Phillip Toms, the University of Gloucestershire scientist, finished his analysis, and the results were not what anybody had expected: the figure was neither ancient nor modern in origin but, rather, was created in the murky centuries in between. The sample taken from the deepest layer of the giant dated from between 700 and 1100 A.D., most likely near the midpoint of that range, around the tenth century.

The giant’s view was lovely enough to make any onlooker’s spirits surge. In its mysterious obscurity, the scene was even more beautiful than it would have been if the skies were clear.

Can Horse Racing Survive? | The New Yorker

One thing you could safely say about the horses was that they were thirsty. They had all been injected that morning with Lasix, a diuretic, noted on the racing form with a boldface “L.” The given reason for Lasix is to prevent pulmonary bleeding, which hard running causes in many horses. The bleeding can be dangerous, and can certainly be unsightly, leaving horse and jockey painted with blood—not a good look these days. But only a small minority of thoroughbreds are serious bleeders, and for decades nearly every thoroughbred in the U.S. has received race-day Lasix. The drug’s diuretic function causes horses to unload epic amounts of urine—twenty or thirty pounds’ worth. The advantage of running light is obvious, as is the reason that critics consider Lasix a performance-enhancing med. Race-day Lasix is banned in Europe, Asia, and Australia.

The activists outside, suggesting that horses don’t like to race, were half right. Running fast comes naturally to thoroughbreds, but racers need to be trained to outrun opponents. Most, it is thought, need “encouragement”—whipping—to continue going hard when they’re tired. Racehorses, especially those running on oval tracks, give their lower legs a terrible pounding, straining ligaments, tendons, joints. Mongolian Groom’s lower hind legs were wrapped in blue bandages, which is not uncommon; horses tend to kick themselves. He wore a heavy blue hood, to keep him concentrated on what’s in front of him, and a shadow roll across his nose. Horses can startle at shadows on the ground, and the roll reduces the number they see.

What went wrong at Santa Anita? The abolitionists liked to say it was just business as usual—horse torture and murder. The apologists said it was business as usual, too—racehorses have always died, even before bleeding-heart outsiders started paying attention. But it wasn’t business as usual. Horses were dying every single week. They were dying during workouts, during races, on turf and on dirt. Colts, fillies, geldings. Obscure claimers, first-time runners, a famous stakes winner during a workout. The deaths started to make the Los Angeles Times, and social media picked up the scent. More protesters appeared at the track. People for the Ethical Treatment of Animals demanded that Newsom shut down Santa Anita. “Something is drastically wrong,” Art Sherman, a trainer in his eighties, told the Times. “I’ve been around a long time and have never seen this.”

The fragility of horses is ubiquitous, not confined to the racetrack. “Something spooks them and they run, almost blindly,” Alexander says. “They can break a leg, get hung up on a fence. Their feet are delicate and problematic. Their digestive system tends to back up. A wad of hay gets stuck in their intestine. They can twist a bowel by rolling in the grass when they’re happy. You need to spot that and address it right away. They’re not rugged, like a cow.”

very registered thoroughbred in the world is descended from one of three stallions: the Byerly Turk, the Darley Arabian, and the Godolphin Arabian. These “foundation sires” came to England from the Middle East around the turn of the eighteenth century, and their offspring turned out to have an unprecedented combination of speed, agility, and endurance. Thoroughbred racing was born. Of course, people have been racing horses since shortly after they were domesticated, which is thought to have occurred about six thousand years ago. How shortly? My guess is a week. They started cheating, by my guess, a week after that.

The Romans, according to the veterinarian and scholar A. J. Higgins, used a mixture called hydromel to increase their horses’ endurance. The punishment for cheating in races was reportedly crucifixion. A British prohibition on “exciting substances and methods” is said to have been introduced in 1666. A stable lad named Daniel Dawson, accused of poisoning a racehorse, was hanged on Newmarket Heath in 1812. Once thoroughbred racing crossed the Atlantic, the United States gained a reputation for the innovative use of performance aids: cocaine, heroin, strychnine, caffeine.

Trainers fixate on track conditions, a complex interplay of surface, weather, and horse anatomy. Overly soft tracks cause damage to soft tissues. Overly hard ones cause microfractures in the many bones below the hock, which sometimes heal and sometimes, as with Mongolian Groom, burst into injuries that a horse can’t survive. Alexander had been thinking about stabling his horses at Los Alamitos, a minor track near Long Beach.

Rain alone, a sloppy track, is not necessarily dangerous for racing. But this was more rain than Santa Anita almost ever sees. Alexander said, “If we get half an inch, we can deal with it. We’d see rain coming, seal the track. Half a day later, unseal, harrow, and we’re off.” Sealing a track means compacting its upper layer with rollers or with heavy plates called floats, pulled by tractors.

Used to be one big guy you were riding for. Now it’s syndicates.” Syndicates are partnerships that allow investors to own a piece of a racehorse, often divvying up the shares among hundreds of people. They were unheard of a few decades ago, but now they seem to be everywhere, and their prevalence has strengthened the industry’s bottom line. Racehorse ownership has been somewhat democratized. (There are “microshares” that go for a hundred dollars a year.) The stereotypical impatient owner these days is not some toffee-nosed plutocrat but a clueless hedge funder demanding a Kentucky Derby winner, of which he might own half a hoof.

Perhaps more important to racing’s bottom line, however, have been the extraordinary investments in breeding farms and racing stables by Saudi royals, and, especially, by the ruler of Dubai, Sheikh Mohammed bin Rashid Al Maktoum. These investments have been in Ireland, England, Europe, Australia, and Japan in addition to the United States, and have lately extended to hosting races that offer the largest purses in the world—the Saudi Cup pays twenty million dollars.

TVG began to provide a newcomer’s glossary of racing terms—“tout,” “weanling,” “sloppy.” First-time gamblers were offered three hundred dollars for a risk-free bet.

As sports nearly everywhere disappeared, people were betting on anything that moved. Soccer in Belarus, table tennis in Ukraine, the weather at O’Hare. Organized crime staged “ghost games” in Ukraine—a soccer tournament that never actually happened, brought to you by either the Turkish mob or one from Belarus.

American horse racing was bolstered by the big bettors, known as “the whales,” who came back to the game in the summer. The whales are not obese billionaires sprawled on yachts, as I originally thought, but serried ranks of high-octane computers, operated by individuals who know nothing about horses but everything about betting.

Improving Firefox stability on Linux

  • Every time Firefox crashes, the user can send us a crash report

  • This report contains, among other things, a minidump: a small snapshot of the process memory at the time it crashed. This includes the contents of the processor’s registers as well as data from the stacks of every thread

  • The minidump format was originally designed at Microsoft and Windows has a native way of writing out minidumps.

    • On Linux, we use Breakpad for this task.

  • When it comes to Linux things work differently than on other platforms: most of our users do not install our builds, they install the Firefox version that comes packaged for their favourite distribution.

    • This posed a significant problem when dealing with stability issues on Linux: for the majority of our crash reports, we couldn’t produce high-quality stack traces because we didn’t have the required symbol information

    • We had to write distro-specific scripts that would go through the list of packages in their repositories and find the associated debug information

  • Firefox depends on a number of third-party packages (such as GTK, Mesa, FFmpeg, SQLite, etc.). We wouldn’t get good stack traces if a crash occurred in one of these packages instead of Firefox itself because we didn’t have symbols for them either.

  • Benefits

    • ability to inspect Linux crashes is that it greatly sped up our response time to Linux-specific issues, and often allowed us to identify problems in the Nightly and Beta versions of Firefox before they reached users on the release channel.

    • quickly identify issues in bleeding-edge components such as WebRender, WebGPU, Wayland and VA-API video acceleration

    • identify regressions and issues in our dependencies

Approaches to Undo and Redo

  • What is undo and redo

    • When an “undo” command is executed, the application state is rolled back to the previous step, as if the last action performed had not taken place.

    • When one or more undo commands have been executed, “redo” actions will be available.

    • These can roll the application’s state forward again, bringing it to the moment where the next action in the history list had been performed.

    • Independent of the step we moved to by issuing undo and redo commands, any new action performed will set a new history path in the future and clear the redo actions history, while the history path to the past (the available undo actions) is kept.

  • Two of the main approaches usually considered are the memento and command patterns.

    • The memento pattern may appear to be the simplest and most straightforward way of implementing state history: It involves keeping a copy of the state before any action is performed

    • the command pattern involves executing the necessary commands on the state so as to make it look like it was before the undone action was performed.

  • [The momento approach keeps only] object references to previous states,

    • so as long as we handle state properties as immutable by creating a new object reference if a property is modified, keeping all those copies shouldn’t end up being very expensive.

    • [But if] the object representing the application state contains much more data than before.

    • And keeping copies of the same primitives and object references over and over may start to feel like an unnecessary waste of resources.

    • We may want to handle history a bit differently.

  • [As such, it may be required that] each state in our history stack is a function of the current state:

    • We’ve switched to the command approach to optimize our implementation,

    • and each next state can be obtained by applying a diff to the current state when executing undo or redo.

  • to implement the diff technique, each undo and redo step must include information external to the state data model;

    • we need to know if the diff to be applied is adding, modifying, or removing data, and this extra logic needs to reside somewhere for each step.

  • Problem in a collaboration context

    • Will our history stack be global and keep all the undo and redo actions for all actions performed by the stakeholders? Or will our history stack be local and only keep track of the actions performed by the current user?

    • The additional complication of keeping history in sync compels us to lean to the local history approach, especially because interfering with other clients’ actions through undo and redo history — which is more widely implemented as a local helper — could result in annoying other stakeholders.

  • [There could be] possible conflicts, even if our history stack is local and not shared

  • An approach to solving conflicts like this one — and the one we’ll choose for our example — is to mark each action with a timestamp so that earlier actions override later ones

Inflation Adjusted Hacker News

  • The number of points a story has accumulated on Hacker News is a good indicator for how relevant and overall how good the story is.

  • the average front page story has a lot more points today than it would have had a few years back.

    • very likely a consequence of the growth of the HN community.

    • More people who vote equals more points on average

    • harder to browse through older stories and gauge how good a story is purely based on the point count.

  • Based on past data I created a HN inflation index.

    • The index is calculated on the basis of very simple calculations.

    • It's basically the average point count of the top 100 stories of each day.

    • I used a moving average of one year to account for seasonal changes in HN-behaviour.

  • I made a browser extension which automatically shows the inflation adjusted point score of a Hacker News story next to the actual number of points.

Show HN: Inflation-adjusted Hacker News | Hacker News

  • ineedasername: Pretty soon 100 karma won't even buy you a gallon of milk, much less the small 2-bedroom colonial home it used to back in '08 or '09. At least it's still harder currency than FB likes or Twitter retweets.