Throughout the 20th century, NASA (and its predecessor, NACA) made extensive use of wind tunnels to test and refine designs for airplanes, spacecraft, and many other vehicles and structures. Dozens of specialized tunnels were constructed over the years at Langley Research Center in Virginia and Ames Research Center in California, to test the effects of high windspeed, turbulence, icing, ionization, and much more. Some of these facilities were gigantic—the largest, still in operation, is the 80-foot by 120-foot tunnel at NASA's Ames Research Center. In the 1990s, a surplus of government wind tunnels and advances in computer simulations led to a consolidation, and a number of older facilities were demolished. Gathered here, a collection of images of NASA’s amazing wind tunnels from the past century.
Horseshoe crabs are sometimes called “living fossils” because they have been around in some form for more than 450 million years. In this time, the Earth has gone through multiple major ice ages, a Great Dying, the formation and subsequent breaking up of Pangaea, and an asteroid impact that killed the dinosaurs and most of life on Earth yet again. In other words, horseshoe crabs have truly seen some shit.
Yet, I would conjecture, some of their strangest experiences must have come in just the past few decades, as one of the soft-bodied mammals that came after dinosaurs began using their hands to scoop horseshoe crabs out of the ocean en masse. Contemporary humans do not deliberately kill the horseshoe crabs—as did previous centuries of farmers catching them for fertilizer or fishermen using them as bait. Instead, they scrub the crabs clean of barnacles, fold their hinged carapaces, and stick stainless steel needles into a soft, weak spot, in order to draw blood. Horseshoe crab blood runs blue and opaque, like antifreeze mixed with milk.
And for what exactly do humans need the blood of a living fossil? A sort of witchcraft, you might say, for it literally keeps people alive. Horseshoe-crab blood is exquisitely sensitive to toxins from bacteria. It is used to test for contamination during the manufacture of anything that might go inside the human body: every shot, every IV drip, and every implanted medical device.
So reliant is the modern biomedical industry on this blood that the disappearance of horseshoe crabs would instantly cripple it. And in recent years, horseshoe crabs, particularly in Asia, have come under a number of threats: habitat loss as seawalls replace the beaches where they spawn, pollution, overfishing for use as food and bait. Horseshoe crabs bled for the biomedical use in the United States are returned to the ocean, but an estimated 50,000 also die in the process every year.
There is another way though—a way for modern medicine to make use of modern technology rather than the blood of an ancient animal. A synthetic substitute for horseshoe-crab blood has been available for 15 years. This is a story about how scientists quietly managed to outdo millions of years of evolution, and why it has taken the rest of the world so long to catch up.
Jeak Ling Ding says she was “always a lab rat”—the kind of biologist who wore white coats rather than the kind who waded into mud. Yet, in the mid-1980s, she found herself squelching through mud in search of horseshoe crabs. The estuary where they lived, she recalls in understated fashion, was “not very sweet smelling at all.”
Ding, along with her husband and research partner Bow Ho, had come to horseshoe crabs circuitously, and their ultimate goal was to make the animals no longer necessary in biomedical research. At the time, she was a molecular biologist at the National University of Singapore, and a hospital’s in-vitro-fertilization department had come to Ding and Ho with a problem: Their embryos would not survive long enough—could it be because of bacterial contamination?
A standard test at the time—and now—is LAL, which stands for limulus amebocyte lysate. Limulus refers to Limulus polyphemus, the species of horseshoe crab native to the Atlantic coast of North America. Amebocyte refers to cells in the crab’s blood. And lysate is the material freed from the cells once they have been “lysed” or broken. This is the stuff exquisitely sensitive to bacterial toxins.
The first person to figure this out about LAL was Frederik Bang. Thirty years before Ding—and 9,000 miles away on Cape Cod—he too was collecting horseshoe crabs on the shore. (For reasons not entirely understood, horseshoe crabs are only found around the eastern coasts of North America and Asia.) Bang, a pathologist, was interested in the creature’s primitive immune system. He settled on a protocol of injecting bacteria from seawater directly into horseshoe crabs, which cause their blood to clump into “stringy masses.”
Bang suspected this clotting had a purpose. It immobilized the bacteria, sealing off the rest of the horseshoe crab’s body from an invading pathogen. Intriguingly, their blood turned to gel even if he boiled the bacteria injection for five or 10 minutes first. This should have killed the bacteria and sterilized the injected solution. Bang realized the blood was sensitive not just to live bacteria but to bacterial toxins that persist even after sterilization.
The human immune system may be much more sophisticated than a horseshoe crab’s, but it too reacts to these toxins. Doctors first realized this in the late 19th century, where patients given sterile shots nevertheless came down with “injection fever” or “saline fever.” In the worst cases, the toxins can cause septic shock and even death.
At the time Bang was doing this research in the 1950s, the standard way to test for bacterial toxins was to inject a sample into rabbits. It required someone to come check the rabbits’ temperatures every 30 minutes for three hours for signs of fever, which would suggest bacterial contamination.
Under the microscope, the rabbit’s blood cells also has a tendency to clump toxin, a similarity Bang noted in his 1956 paper on horseshoe-crab blood. Over the next decade and a half, he and a young pathologist named Jack Levin devised a standardized way to extract LAL. It was not until 1977, however, that the Food and Drug Administration allowed pharmaceutical companies to replace their large colonies of rabbits with LAL kits. Now you simply added LAL to the tested material and flipped the vial over to see if it turned solid—much faster and more convenient. The LAL test still required the use of animals, but the grisly process of sticking needles into animals became hidden and outsourced to a different part of the supply chain.
By the time Ding was looking for horseshoe crabs in Singapore, LAL had become a multimillion-dollar industry. One quart of horseshoe-crab blood is reportedly worth as much as $15,000. And the LAL kits she needed to test contamination of IVF embryos were far too expensive. One kit, she recalls, cost $1,000 for her in Singapore.
Which is why she considered making her own lysate. But the horseshoe-crab species she was studying in Singapore, Carcinoscorpius rotundicauda, is much smaller than Atlantic horseshoe crabs, and they couldn’t be bled much without dying. So Ding set out to make an alternative to LAL that eventually wouldn’t require horseshoe crabs at all.
What it would require was manipulating DNA. Her idea was to splice the horseshoe-crab gene responsible for LAL’s toxin-hunting ability into cells that grow easily in a lab, like yeast. Biotechnology as a field was already moving in the direction of recombinant DNA, which entails taking DNA from one species and putting it another. A few years earlier in 1982, Eli Lilly began selling human insulin grown in vats of bacteria.
Ding had a good starting point for her LAL alternative. By then, scientists had identified factor C, the specific molecule in LAL that detects bacterial toxins. So she started hunting for the gene that makes factor C. Her research team took cells from horseshoe crabs that they collected and bled them minimally. (They also tried, but failed, to grow horseshoe crabs in a lab and breed them through IVF.)
The horseshoe crab’s sensitivity to bacterial toxins unfortunately also made it a pain to study. The toxins, it turns out, are everywhere—in water, in test tubes, in petri dishes. “You have to bake all bakeable glassware at 200 to 220 degrees for several hours.” says Ding. They also had to buy special water that had been treated to be bacterial toxin free. If you weren’t careful, your tube of solution could easily turn to gel.
When Ding and Ho finally identified the gene for factor C, they spliced it into yeast. That failed because while the yeast made factor C, it did not secrete the molecule. “The yeast was very difficult to break open. It was very impure and messy,” she says. They tried another type of yeast and mammalian cells—those failed too. In the late 1990s, Ding and Ho attended a course in the United States and learned about baculovirus vector systems. Here, a virus is used to insert the factor C into insect gut cells, turning them into little factories for the molecule. Insects and horseshoes have a shared evolutionary lineage: They’re both arthropods. And these cells worked marvelously.
Finally, a decade and a half after she began, Ding had an alternative to LAL that worked without harming any more horseshoe crabs. She cooped herself up in the library to study patents and drafted the application herself. Then she sent it off and waited for the world to change.
The world did not change, at least not for the horseshoe crabs. It took three years for the first recombinant factor C test kit based on Ding’s patent to come out in 2003, but even then pharmaceutical companies showed little interest.
The companies had a number of reasons. There was only one supplier of the kit, a company that today is part of the Switzerland-based chemicals company Lonza. Pharmaceutical companies were wary relying on a single source for such an important part of their manufacturing. What if something happened to Lonza? Or a natural disaster hit its production plant? Companies that bleed crabs also stand to lose a lot of money if factor C becomes adopted widely. Of the six companies with crab bleeding facilities in the United States, two declined interviews, one did not respond to an interview request, and two have virtually no public presence. The sixth is Lonza, which currently sells both LAL and the recombinant factor.
Lonza, for its part, blamed the slow uptake on regulations. In the United States, the FDA tells companies carrying out bacterial-toxin tests to follow the United States Pharmacopeia, a handbook that lays out drug standards. In a 2012 guidance, the FDA said companies could use recombinant factor C, which does not appear in the Pharmacopeia, if they carried out their own validation tests. “The risk is, of course, the FDA may not accept your validation and you can’t bring your product to market,” says Lonza’s spokesperson Katrin Hoeck. “Pharmaceutical companies are risk averse.” It took the industry decades to move from rabbits to LAL, too.
The realities of business came as a real disappointment to Ding. “We were just so keen as researchers, so happy it is working,” she says. “And we thought the recombinant factor C will be adopted around the world, and the horseshoe crab would be saved.”
Recently, however, a few things have changed the recent risk-reward calculus for pharmaceutical companies. For one, Lonza is no longer the sole supplier. In 2013, Hyglos became the second company to make recombinant factor C. Kevin Williams, a senior scientist at Hyglos, says he sees as a long overdue modernization: Pharmaceutical companies stopped relying on pigs and started making insulin in yeast and bacterial cells decades ago. Why can’t the same technology be applied to the very test used to check that insulin is safe for injection?
On the regulatory side, the European Pharmacopoeia added recombinant factor C as an accepted bacterial-toxin test in 2016, paving the way for change in the United States. A number of pharmaceutical companies, most notably Eli Lilly, have compared the effectiveness of recombinant factor C and LAL.
Jay Bolden, an expert in bacterial toxin detection at Eli Lilly, recalls Lonza coming in their labs with the recombinant factor C kit over a decade ago. He was intrigued at the time but not yet willing to take the plunge. The turning point came in 2013, when Eli Lilly began planning an insulin-manufacturing facility in China, where the native horseshoe-crab species has been declining. “You would hear things about someday the horseshoe crab might get restricted,” says Bolden. In contrast, the supply chain for recombinant factor C looked more secure with both Hyglos and Lonza as suppliers. LAL and factor C are also comparable in cost.
Bolden says Eli Lilly decided to “draw a line in the sand”: All new products after a certain point would be tested with recombinant factor C. The company recently submitted to the FDA its first application for a drug—galcanezumab to prevent migraines—where the final drug will be quality tested with factor C. It has also looked into using recombinant factor C during the manufacturing process to test water and equipment, which currently accounts for the vast majority of LAL use. Bolden says Eli Lilly has been lobbying the U.S. Pharmacopeia to include recombinant factor C.
On Thursday, Bolden will be speaking in Cape May, New Jersey, at an event organized by Revive and Restore, a nonprofit best known for its work on bringing extinct species back to life. “Our mission is to use biotech for conservation,” says Ryan Phelan, co-founder and executive director of Revive and Restore. Phelan first met Ding when she traveled to Singapore for a synthetic-biology conference in 2017, and she realized her research on recombinant factor C sat perfectly in the intersection of conservation and biotechnology.
Revive and Restore and its conservation partners—New Jersey Audubon, American Littoral Society, and Delaware River Keeper Network—chose the Cape May location because horseshoe crabs come here every spring to spawn. You can no longer catch horseshoe crabs here due to their importance to a threatened migratory bird species called the red knot. These birds show up here in the spring, too. Their migration is timed so that birds flying from South America to the Arctic can gorge themselves on the caviar-like horseshoe-crab eggs. The beaches turn black with crabs, their shells clickety-clacking as females scramble to lay their eggs and males to fertilize them. The red knots scramble to eat. They nearly double in weight for their journey to the Arctic.
It is an ancient synchrony between species, one that began long before humans began harvesting horseshoe crabs for blood and will hopefully last long after.
On Tuesday, Google showed off Duplex, a new service the company is testing that allows Google Assistant to call establishments on a user’s behalf to make a dinner reservation or schedule a haircut. The voice synthesis in these calls is jaw-dropping:
With a few millisecond mistakes, Duplex sounds like a human, complete with mmms and uhhs and cheery colloquialisms. The ability of the AI to respond to real, messy language and unexpected sequences is also incredible. Generating conversational human sentences in real time can probably be considered a “solved problem,” as people around here like to say.
The audio suggests that computer voices just skipped right past 2001: A Space Odyssey, and it turns out robots won’t sound like an overlord, but a, uhh, Millennial.
The way Google presented the technology encouraged people to think about themselves as powerful users, casting magic bots out across the world to do our bidding. But it’s the other side of the interaction that deserves attention.
What new skills will service workers develop to identify and respond to voice bots calling on behalf of a “client” (as Duplex put it on one call)? Will the bots get equal service? How will these systems be used to generate better spam? Will the true privilege of high-wage work be the satisfaction of working with other (real) people, while low-wage workers increasingly interact with bots and screens?
This moment we’re in right now—where humans and bots find themselves in an unprecedented admixture—is one more step in the automation of different kinds of human labor. In the quiet, white-collar automation that swept the world in the last quarter of the 20th century, the messiness of human processes required many intermediate steps in the transition from paper and human to computer and computer. Much of what service work used to be was automated over the last few decades. Now, computers make the decisions, and the main role of the human is to deliver this information after pressing some keys on a computer.
Think of a car-rental counter. You’ve booked the reservation online, outfitting the car and contract exactly as you wanted to. That system has told the various workers what they need to get ready and issued the contract and rung up the total. The human involved has only one real job: to run the upsell script before you get in your car and go.
Of course, humans work around the design of the system. They tell jokes, shade competitors, hand out upgrades to people they like, dispense advice about the city, press their lips tightly when a customer says something stupid. But the system merely wants them to run the upsell script. That’s the job. They do the same thing over and over, under time pressure, acting with the grim knowledge that quantitative metrics will be the primary means by which their performance will be judged.
The playwright and author Barbara Garson captured many of these dynamics in a chapter on airline-reservation clerks in her 1988 anthropological study, The Electronic Sweatshop. Garson, encountering this as a new phenomenon, is aghast:
In a feat of standardization even more phenomenal than McDonald’s fry-vat computer, the airlines have found ways to break down human conversation into predictable modules that can be handled almost as routinely as a bolt or a burger.
She finds workers pushing to up their sales bookings and reduce their AHU (After Hang-Up) time stats. They are people who understand that the money they make is a direct result of their ability to hit the quantitative targets by which they are judged, no matter the human experience of the person on the other side of the telephone line. The skilled, long-term reservation agents who knew “all the company’s routes, fares, and policies” and could use that knowledge to help customers by understanding their needs were becoming obsolete. They didn’t know how to optimize themselves for this new, robotic world.
In the end, almost everyone lost their jobs anyway, as most travel booking moved to websites where the work was shifted onto customers who—now that the airline-reservation agents acted like robots—would much rather interact with a web form than have a phone call with a know-nothing trying desperately to sell them something in the shortest possible time.
As with airline reservations, so it goes with many other similar services. Who likes to call a business? Not most young people. We’d much rather hit a button on an app or—soon, I guess—dispatch a bot to talk to the harried, underpaid employee who has been tasked with answering the phone that day.
Automation begets automation. In that sense, Google Duplex feels not like something new and amazing (though it is also that), but something old and stultifying. For decades now, we’ve been forcing human service workers to act like robots. This makes many service interactions unpleasant enough that people want to avoid them, so now, Google will provide everyone with a robot that can act like a human. Finally, technological capitalism has generated the correct match for the robotic service worker: a robot service worker.
It’s almost hilarious, a George Saunders short story. It’s almost tragic, a Samuel Beckett play.
Maybe none of it matters. Who needs small talk? All we’re doing is exchanging information, flipping a bit on a calendar from zero to one. So why not let an AI do that errand? There are bigger thoughts to think, children to spend quality time with, exercise to do, community groups to volunteer for, code to write.
But maybe small talk has a purpose. The urban theorist (and hyper-observer of the city) Jane Jacobs thought that all these dumb interactions were the social fabric. She wrote in The Death and Life of Great American Cities:
The sum of such casual, public contact at a local level—most of it fortuitous, most of it associated with errands, all of it metered by the person concerned and not thrust upon him by anyone—is a feeling for the public identity of people, a web of public respect and trust, and a resource in time of personal or neighborhood need. The absence of this trust is a disaster to a city street.
After Jacobs published her book in 1961, most American cities suburbanized, cutting people off from this kind of casual contact with the people around them. But at the same time, they connected through other means, the telephone primary among them.
Already, the push-button service framework, Uber for everything, has eroded this last bastion of local chitchat. Google Duplex will simply extend that trend even to those businesses who have not given in to the computerization of their reservation systems. And I don’t except myself from this trend: I hate making these little phone calls as much as anyone else.
But if Jacobs is right that simply making conversation with our fellow human beings—“most of it associated with errands”—is what generates trust in the world around us, then what happens when no one is ever quite sure if Alexis or Duplexis is calling?
This article might be trying too hard on the, "Woe to humanity and the lack of interaction" aspect. But wow! I'm super impressed by the ability of Duplex to handle non-scripted interactions with failure modes.
The second interaction - it doesn't start out as me, but it definitely transitions into me into the final, "oooh, I gotcha thanks", which I suspect the AI literally just copied from a conversation I had over google voice.
I'm just disappointed in the whole notion; it feels like giving up. voice has to be the least efficient, least accurate way of a data interaction. the future I want lets me precisely book and confirm appointments via a simple, standard protocol adopted by all. we can at least take consolation in the fact that most of the stuff demonstrated at I/O turns out to be vapour.
I really don't want to have to set up an account and log in to any random place that I'm making an appointment at. If you don't have at least that level of "security" then the potential to get bots spamming appointment times becomes untenable. At least with voice, the handshake takes too long to reasonably DOS via creating fake appointments.
The next phase of this project is to implement it for the user. Surely the appointment automation will use more modern techniques before it falls back on voice. With the voice fallback, though, goog can ensure virtually complete coverage of the service.
I guess I'd rather have everyone have online booking systems too, but since they don't this seems like an effective workaround since it lets me, as the user, asynchronously and non-vocally solve a problem that would otherwise be synchronous and vocal. Basically this gets closer to what we want without forcing universal adoption of a new technology on the part of service providers. Also... kinda think the actual application of the tech is essentially meaningless. It's a tech demo designed to show off real time TTS and STT tech that looks hella advanced, at least when applied to a narrow and definable problem space.
heh, topically, I already saw one security researcher post something of a prediction that this will evolve into a fresh hell of DDOS technology for meatspace and potentially accelerate the end of voice lines if countermeasures against spam and fraud aren't created. my cyberpunk popcorn is ready!
Here’s one to add to the “weird mechanosynthesis” pile. According to this paper, you can do hydrogenation reactions in a stainless-steel ball mill, without any sort of noble-metal catalyst. The hydrogen is produced when you add some n-alkane or diethyl ether to the mix (these actually get converted to gaseous methane and hydrogen under the milling conditions). Some pretty severe reductions are realized, for example, taking biphenyl all the way down to 1-cyclohexylcyclohexane. Olefins go to alkanes under these conditions, of course, and ketones and alcohols tend to just get erased: it’s a bulldozer.
And although it’s surprising, this reaction isn’t quite as much voodoo as it might appear. Well, not more than any other hydrogenation, anyway. The ball mill is, of course, a sealed vessel, and the authors show that the lower-boiling additives seem to be more effective, presumably because they increase the internal pressure. And it’s not like there’s no catalyst present, because you have the stainless steel. If you use zirconium oxide balls instead of the steel alloy ones, for example, you get no hydrogenation at all. Adding bits of different metals back in showed that the chromium is what’s producing the hydrogen from the alkanes or ether, while the nickel is acting as the hydrogenation catalyst itself. The stainless-steel alloy itself (304 steel) was the most effective combination, though.
The balls do get eroded a bit under the conditions, so what you’ve got is finely divided metals, under some heat from the mechanical energy of the milling, and under hydrogen pressure from the decomposition of the co-reactants. No wonder things get reduced! But it’s not a reaction that you would have predicted up front. And even if you’d stipulated that hydrogenation would start under these conditions, I don’t think you would have guessed how powerful it would be. The last time I hydrogenated down a whole aryl system (it’s been a while), I was running at something like 1600 psi (110 bar) at 120C with a rhodium catalyst, and the stuff came out looking – and smelling – like old lawnmower oil. This reaction probably doesn’t get above 60 degrees, by contrast. Admittedly, there’s no word in the paper about the appearance or aroma of the crude products, but I’m willing to bet that that part remains the same. . .