Original image

Learn the Art of Language Creation From the Guy Who Created the Game of Thrones Languages

Original image

When David Peterson was in middle school, he had no particular interest in language and never dreamed he would one day create languages for wildly successful TV shows—but when he watched Return of the Jedi he noticed that something was not quite right about the scene where Princess Leia, disguised as a bounty hunter, speaks in an unknown language to Jabba the Hutt. She basically repeats what sounds like yaté and yotó a few times, and that somehow, according to the subtitles, stands for both “I have come for the bounty on this Wookie,” “50,000, no less,” and a few other things. That didn't seem like something a language should do. Peterson didn’t know it at the time, but this sense of unease about what he was seeing was an early glimmer of the particular artistic sensibility of the language inventor, the ability to distinguish an intelligent, well-crafted creation, from a lazy jumble of nonsense syllables.

He eventually developed an intense love of language, studying several of them and creating even more. His professional language creations (heard in Game of Thrones, Defiance, and Thor: The Dark World) as well as the personal projects he has been working on since 2000 are of the intelligent and well-crafted type. They have complicated, learnable grammars, extensive vocabularies, and features consistent with fully-imagined cultural practices. In an era where we can watch, re-watch, and pick apart on the internet to our heart’s content, fans demand no less. Yaté yotó just doesn’t cut it anymore.

But what makes a good constructed language (or conlang, for those in the know)? And for those who want to try their hand at language creation for their fantasy novel, secret club, thought experiment, or plain personal enjoyment, where is a good place to begin? Since the early 90s, conlangers have been sharing their ideas and strategies and evaluating each other’s work on listservs and forums and sometimes even at in-person conferences. A sort of technique and artistic standard has emerged, but it can be difficult for a newbie to figure out what it is. Peterson asks,

Where is the collected wisdom of the early conlang community? Why is it not written down somewhere that if you’re creating a naturalistic ergative language, it will most likely be split ergative, and that those splits will happen in one of a small number of likely places in the grammar? This is something that every conlanger knows or eventually learns, but the information is only passed via word of mouth—it’s like we’re living in the 1300s, but we also have the internet and indoor plumbing!

If you are a budding language inventor now thinking “yikes! What’s an ergative language?” Peterson has written the book for you. Full of examples from both natural and constructed languages, The Art of Language Invention will take new conlangers through “the nuts and bolts of language creation so they can focus on the more important question: What do I want to say with this new language that I can’t say in my native language —or any other language that currently exists?”

Even if you have no plans to build your own language, the book is a lively introduction to the important concepts of linguistics, from consonants and vowels, to stress and tone, to verb agreement and case, to how grammar evolves over time. There’s also a section on writing systems with some beautiful examples of invented scripts. And if you’re a fan of HBO’s Game of Thrones or Syfy’s Defiance the “case studies” of the languages Peterson created for those shows will give you a deeper appreciation for just how far we’ve come from the days of yaté yotó.

Read some fun facts about Dothraki or Valyrian, and if you want to know more, check out The Art of Language Invention.

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
Stephen Missal
New Evidence Emerges in Norway’s Most Famous Unsolved Murder Case
May 22, 2017
Original image
A 2016 sketch by a forensic artist of the Isdal Woman
Stephen Missal

For almost 50 years, Norwegian investigators have been baffled by the case of the “Isdal Woman,” whose burned corpse was found in a valley outside the city of Bergen in 1970. Most of her face and hair had been burned off and the labels in her clothes had been removed. The police investigation eventually led to a pair of suitcases stuffed with wigs and the discovery that the woman had stayed at numerous hotels around Norway under different aliases. Still, the police eventually ruled it a suicide.

Almost five decades later, the Norwegian public broadcaster NRK has launched a new investigation into the case, working with police to help track down her identity. And it is already yielding results. The BBC reports that forensic analysis of the woman’s teeth show that she was from a region along the French-German border.

In 1970, hikers discovered the Isdal Woman’s body, burned and lying on a remote slope surrounded by an umbrella, melted plastic bottles, what may have been a passport cover, and more. Her clothes and possessions were scraped clean of any kind of identifying marks or labels. Later, the police found that she left two suitcases at the Bergen train station, containing sunglasses with her fingerprints on the lenses, a hairbrush, a prescription bottle of eczema cream, several wigs, and glasses with clear lenses. Again, all labels and other identifying marks had been removed, even from the prescription cream. A notepad found inside was filled with handwritten letters that looked like a code. A shopping bag led police to a shoe store, where, finally, an employee remembered selling rubber boots just like the ones found on the woman’s body.

Eventually, the police discovered that she had stayed in different hotels all over the country under different names, which would have required passports under several different aliases. This strongly suggests that she was a spy. Though she was both burned alive and had a stomach full of undigested sleeping pills, the police eventually ruled the death a suicide, unable to track down any evidence that they could tie to her murder.

But some of the forensic data that can help solve her case still exists. The Isdal Woman’s jaw was preserved in a forensic archive, allowing researchers from the University of Canberra in Australia to use isotopic analysis to figure out where she came from, based on the chemical traces left on her teeth while she was growing up. It’s the first time this technique has been used in a Norwegian criminal investigation.

The isotopic analysis was so effective that the researchers can tell that she probably grew up in eastern or central Europe, then moved west toward France during her adolescence, possibly just before or during World War II. Previous studies of her handwriting have indicated that she learned to write in France or in another French-speaking country.

Narrowing down the woman’s origins to such a specific region could help find someone who knew her, or reports of missing women who matched her description. The case is still a long way from solved, but the search is now much narrower than it had been in the mystery's long history.

[h/t BBC]