Original image

Automatic for the People: Remembering the Automat Restaurants

Original image

Long before fast food chains invaded the city, New York’s favorite place to grab a quick lunch and a cup of Joe was the Automat.

If you were a working stiff in New York any time during the mid-twentieth century, there’s a good chance that your daily lunch breaks were spent at one of the fifty Automat restaurants around the city. At the height of their popularity, they served around 350,000 customers a day. With their vast walls of chrome-and-glass food-dispensing machines offering everything from Salisbury steak to cream spinach to apple pie, the Automats were famous for simple, hearty fare at a low price.

America’s prototype restaurant chain was the brainchild of Joe Horn and Frank Hardart. At the turn of the 20th century, the business partners envisioned a dining experience that would embrace the industrial revolution’s idea of conveyor-belt efficiency and uniformity. It would feature self-service vending machines instead of waitresses, and have a gleaming modern décor that was more factory than dining room. Inspired by a German eatery called the Quisiana Automat, Horn & Hardart launched their first Automat in Philadelphia in 1902. Ten years later, they opened another in Times Square, and it was there that the Automat really took off.

New Yorkers loved the speed and low prices. And boy, were they low. Until 1952, almost all the food cost a nickel. A hamburger or bacon and eggs were two nickels. Originally, the Automat’s machines accepted only nickels. Cashiers – or “nickel throwers” – changed paper money and larger coins from a booth in the center of the restaurant. Diners dropped the nickels in the slots, turned a knob, then lifted a hinged window and removed the entrée, sandwich, or dessert of their choice. Aside from the novelty, one of the practical advantages of the Automat was that you could see the food before buying it.

Though the restaurants appeared to be automated, behind the scenes there was a busy staff constantly refilling the compartments. And the food, of course, was prepared by human hands, all of it from a central commissary in midtown. Other Automat staples included frankfurters, beef stew, macaroni and cheese, baked beans in a pot, mashed turnips, donuts and huckleberry pie.

Hot Coffee

For all the good food, the Automat’s real secret weapon was its coffee. Horn & Hardart popularized fresh drip-brewed coffee in New York. Prior to the Automat, coffee was often harsh and bitter, boiled and clarified with eggshells. The Automat’s smooth aromatic brew flowed regally from ornate brass spigots in the shape of dolphin heads. In their heyday, Automats sold over 90 million cups of their fresh-brewed coffee each year. And they were committed to keeping it fresh. When an Automat employee brewed coffee, they filled out a time card. After twenty minutes, they discarded whatever coffee was left and made a fresh pot. If there was any doubt about Horn & Hardart’s commitment to java, the Automat even adopted Irving Berlin’s “Let’s Have Another Cup Of Coffee” as their unofficial theme song.

In fact, Berlin was a regular at the Automat. So were Wall Street bankers, cops, sewer workers, office clerks, secretaries, actors, musicians and pretty much everyone else in the city. Uptown, downtown, rich, poor – everyone loved the Automat. That was another wonderful thing about it - it was a truly democratic dining experience.

By the 1960s, suburban sprawl, changing tastes and newer fast-food restaurant chains were taking their toll on the Automat’s business. By the 1970s, Horn & Hardart were replacing Automats with Burger King franchises. The last of the Horn & Hardart establishments closed in April 1991. In 2006, a restaurant called Bamn tried to revive the Automat concept in New York’s East Village, but its success was short-lived. A section of Horn & Hardart’s flagship 1902 Automat in Philadelphia is preserved in the Smithsonian National Museum.

And finally, a few words of love from famous Automat diners:

“I have always thought that the Automat in New York has the best scrambled eggs in the world.” —Gregory Peck
“The Automat was the Maxim’s of the disenfranchised.” —Neil Simon
“I had the same lunch there every day: three vegetables, a roll, and cocoa. All for twenty-five cents.” —Jerome Robbins
“I lived at the Automat. They had the greatest chocolate milk. When I moved to Philadelphia, I apportioned less than two dollars a day to eat on, and the Automat was the only place I could do it.” —Dick Clark
“I went to the Automat all the time. I grew up going to the Automat. The food was delicious. And it was wonderful.” —Woody Allen

This post originally appeared in 2011.

Original image
iStock // Ekaterina Minaeva
Man Buys Two Metric Tons of LEGO Bricks; Sorts Them Via Machine Learning
May 21, 2017
Original image
iStock // Ekaterina Minaeva

Jacques Mattheij made a small, but awesome, mistake. He went on eBay one evening and bid on a bunch of bulk LEGO brick auctions, then went to sleep. Upon waking, he discovered that he was the high bidder on many, and was now the proud owner of two tons of LEGO bricks. (This is about 4400 pounds.) He wrote, "[L]esson 1: if you win almost all bids you are bidding too high."

Mattheij had noticed that bulk, unsorted bricks sell for something like €10/kilogram, whereas sets are roughly €40/kg and rare parts go for up to €100/kg. Much of the value of the bricks is in their sorting. If he could reduce the entropy of these bins of unsorted bricks, he could make a tidy profit. While many people do this work by hand, the problem is enormous—just the kind of challenge for a computer. Mattheij writes:

There are 38000+ shapes and there are 100+ possible shades of color (you can roughly tell how old someone is by asking them what lego colors they remember from their youth).

In the following months, Mattheij built a proof-of-concept sorting system using, of course, LEGO. He broke the problem down into a series of sub-problems (including "feeding LEGO reliably from a hopper is surprisingly hard," one of those facts of nature that will stymie even the best system design). After tinkering with the prototype at length, he expanded the system to a surprisingly complex system of conveyer belts (powered by a home treadmill), various pieces of cabinetry, and "copious quantities of crazy glue."

Here's a video showing the current system running at low speed:

The key part of the system was running the bricks past a camera paired with a computer running a neural net-based image classifier. That allows the computer (when sufficiently trained on brick images) to recognize bricks and thus categorize them by color, shape, or other parameters. Remember that as bricks pass by, they can be in any orientation, can be dirty, can even be stuck to other pieces. So having a flexible software system is key to recognizing—in a fraction of a second—what a given brick is, in order to sort it out. When a match is found, a jet of compressed air pops the piece off the conveyer belt and into a waiting bin.

After much experimentation, Mattheij rewrote the software (several times in fact) to accomplish a variety of basic tasks. At its core, the system takes images from a webcam and feeds them to a neural network to do the classification. Of course, the neural net needs to be "trained" by showing it lots of images, and telling it what those images represent. Mattheij's breakthrough was allowing the machine to effectively train itself, with guidance: Running pieces through allows the system to take its own photos, make a guess, and build on that guess. As long as Mattheij corrects the incorrect guesses, he ends up with a decent (and self-reinforcing) corpus of training data. As the machine continues running, it can rack up more training, allowing it to recognize a broad variety of pieces on the fly.

Here's another video, focusing on how the pieces move on conveyer belts (running at slow speed so puny humans can follow). You can also see the air jets in action:

In an email interview, Mattheij told Mental Floss that the system currently sorts LEGO bricks into more than 50 categories. It can also be run in a color-sorting mode to bin the parts across 12 color groups. (Thus at present you'd likely do a two-pass sort on the bricks: once for shape, then a separate pass for color.) He continues to refine the system, with a focus on making its recognition abilities faster. At some point down the line, he plans to make the software portion open source. You're on your own as far as building conveyer belts, bins, and so forth.

Check out Mattheij's writeup in two parts for more information. It starts with an overview of the story, followed up with a deep dive on the software. He's also tweeting about the project (among other things). And if you look around a bit, you'll find bulk LEGO brick auctions online—it's definitely a thing!

Original image
Name the Author Based on the Character
May 23, 2017
Original image