This Is as Close as Disney's Going to Get to Confirming the Pixar Connection Theory


One of the most beloved theories of the internet over the past few years has been the “Pixar Theory”—the wild declaration that posits that every part of the Pixar animation oeuvre is part of the same universe. Disney has released a new video gathering every connection between Pixar films to date, and it’s the closest…

Read more…

Source: io9

We Finally Know What’s Causing Namibia’s Mysterious Fairy Circles

Namibia’s Fairy Circles. (Image: Jen Guyton)

For decades, scientists have struggled to understand the strange circles of barren land that litter the Namib Desert. Called “Fairy Circles,” their formation has been attributed to everything from supernatural forces to poison gas and subterranean insects. Now, scientists may have finally solved this enduring mystery.

A new study published in Nature has found that Namibian Fairy Circles are the product of not one, but two ecological forces. By combining field observations and computer models, researchers from Princeton University, the University of Strathclyde in Glasgow, and several other institutions, have shown that the interaction between both termites and self-organizing plants (i.e. plants that either clump themselves together or disperse according to the availability of resources) is sufficient to explain this enigmatic landscape feature. Importantly, this new theory could be used to solve similar mysteries around the world.

Image: Jen Guyton

Found at the intersection of grasslands and the desert, Namibia’s Fairy Circles are hexagon-shaped patches of bare soil that are surrounded by rings of tall grass. These features, which number in the millions, measure anywhere from six to 115 feet (two to 35 meters) wide, and extend for hundreds of square miles across southern Africa’s Namib Desert.

Over the years, a number of explanations have been posited to explain the strange features. The circles have been attributed to self-organizing plants, grazing ants, subterranean termites, poisonous shrubs, and even toxic gases, such as carbon monoxide, that leak from the ground. Locals say they’re the work of supernatural forces.

Of all the theories, however, root-eating termites and self-organizing plants have emerged as the leading candidates.

According to the termite hypothesis, an industrious species of termite called Psammotermes allocerus is engineering Fairy Circles by killing the plants above them, creating bare patches that concentrate moisture. In their effort to create water traps, the termites are reshaping the landscape around them.

The self-organizing plant hypothesis stems from theories of plant competition, and the observation that grasses growing in areas with limited access to water exhibit distinctive growth patterns. This part of Namibia is the transition point between desert and grassland ecosystems, and there isn’t enough water to sustain continuous vegetation cover. The plants are thus forced to compete for water, and organize themselves accordingly, often forming regular patterns that can stretch for miles.

Image: Jen Guyton

Neither of these two explanations can sufficiently explain the Fairy Circles on its own. The work of termites can’t explain the tall grasses at the perimeter of the circles, and the behavior of plants can’t fully explain why Fairy Circles come into existence and then die, or why similar features aren’t found in similar environments elsewhere in the world. These inconsistencies have resulted in heated debates among researchers over the years, with each camp eager to point out the deficiencies in their rival’s explanations.

But in an interesting twist, it appears that when these two theories are combined, a satisfactory explanation emerges.

“Only by considering the interaction between both termites and vegetation self-organization can we obtain such a comprehensive description of all the main properties reported for Fairy Circles,” said Juan A. Bonachela, an assistant professor from the University of Strathclyde and a co-author of the new study. “We offer a plausible explanation for how these and other regular vegetation patterns emerge, which [includes] what were thought to be mutually exclusive explanations for the system.” The key, said Bonachela, was in understanding the nature of the interactions.

Graduate student Jen Guyton collects data on the “fairy circles,” which exhibit strikingly regular spatial patterns that have mystified many observers. (Image: Tyler Coverdale)

When sand termites create the bare patches, this facilitates the accumulation of water underneath the patch in the soil, which is essential for termite survival. The vegetation ring that forms the outer boundary of the Fairy Circle results from plants taking advantage of that increased moisture, spreading their plastic-like roots underneath the bare patch. The often symmetrical, patterned ground that separates the Fairy Circles is caused by the intense competition for resources; a kind of matrix is formed, where individual colonies of plants and termites come together in an overarching mesh-like pattern, creating evenly-spaced oases.  

By using computer simulations, the researchers were able to reproduce the patterns seen in nature.

“Once the model was validated, our simulations allowed us to study the response of the ecosystem to different scenarios,” Bonachela told Gizmodo. “For example, they allowed us to compare how that response would change if termites were not present, or if climatologic conditions changed drastically.”

Simulations showed that the presence of termites made the ecosystem more robust—in the sense that vegetation remained in the system surrounding the Fairy Rings during times of low rainfall—and they allowed the ecosystem to recover more quickly from environmental stressors, such as long periods of drought.

Excitingly, this research can now be applied elsewhere. Similar landscape features include North America’s Mima Mounds, Brazil’s Murundus, South Africa’s Heuweltjies, and Australia’s own Fairy Circles. It’s not immediately obvious if the new model applies to every situation, but scientists should take note. This is a wonderful example of independent researchers doing good work and coming up with good theories for an unexplained phenomenon—only to find that the integration of competing theories is what was needed to solve an overarching mystery.


from Gizmodo

The PO-32 Tonic is a complete drum synth in your pocket for $89


Teenage Engineering have been charming us for a couple of years now with handheld, pocket calculator, Nintendo Game&Watch-style synth and drum machines. And you might think they’d be out of weird ideas. You’d be wrong.

The PO-32 looks to be both the most surprising, and most serious entry yet. It has an entire drum synth in there. And it’s not just any drum synth – it’s Magnus Lidström’s Microtonic, more or less squeezed into $89 hardware.

Now, at this stage, anyone who’s ever used Sonic Charge’s desktop drum percussion synth pattern sequencer plug-in is going to be a little confused. Microtonic, aka µTONIC, has elaborate on-screen controls for tweaking synth parameters, which you can access via a computer GUI with faders and switches and knobs, all of them labeled.



The PO-32 is a business card-sized circuit board with some tiny buttons on it and some pictures of people out drinking and a mouse apparently making a phone call and … spiders. A number of spiders.

Fortunately, the Teenage Engineers have provided the ability design sounds in the computer plug-in, then load that sound into the standalone hardware.

I’ll be honest: this whole thing was so far-fetched that I had to confirm it with them. But because the hardware has a compatible engine to the plug-in, it’s real. You can make sounds on your computer and load them on the hardware, or move them from PO-32 to PO-32. Jesper Kouthoofd from TE says this is the next-generation Pocket Operator platform, and that the functionality will be used on future tools, too.


Parameters and patterns move between software and hardware and hardware and hardware.

There will be a new, updated Microtonic plug-in to go with it. (Note: there’s only confirmation that you can load tweaked sounds/patterns from the plug-in onto the hardware — not the other way around. I hope they do find a way to go from hardware to plug-in, though, as that would be really useful with patterns.)

Don’t own the plug-in yet? Teenage Engineering are offering a bundle of the plug-in and hardware together for $139, as a limited edition.

Here’s Magnus – who also worked with TE on the CWO effect for their OP-1 – showing how it all works:

Teenage Engineering have also let CDM on another little touch they’ve given this instrument. There’s a “write-protect” tab, inspired by cassette tapes. Jesper explains, “The idea is that you can fill a machine with your personal patterns and sounds and keep them in that state forever. Perhaps give to a friend or sell on eBay? You can still perform live punch-in effects for live performance, but never destroy the original data. sort of a mix tape concept…”


Frankly, it looks like a serious little tool. I think they’re going to be nearly impossible to buy, they’ll be so popular when they ship in worldwide (estimated for April).

This and a Nintendo Switch and basically you’re going to be happy all summer.

Full features:

mic for transferring sounds
16 sounds
16 punch-in effects
parameter locks
built-in speaker
3.5 mm audio I/O
jam sync
LCD display
folding stand
watch + alarm clock
battery powered (2 x AAA)
1 month battery life
pattern chaining -up to 64 patterns
compatible with microtonic

You read that right: serious drum machines users, you get parameter locks. People who oversleep and have fond memories of Casio and Nintendo, you get an alarm clock.

As per usual, if you don’t like the bare board look – or want this to be more road-worthy – there’s an accessory case, which looks like this:



Store (doesn’t look like it’s ready yet)

The post The PO-32 Tonic is a complete drum synth in your pocket for $89 appeared first on CDM Create Digital Music.

from Create Digital Music

Deepgram open sources Kur to make DIY deep learning less painful


Deepgram, a YC backed startup using machine learning to analyze audio data for businesses, is open sourcing an internal deep learning tool called Kur. The release should further help those interested in the space get their ideas off the ground more easily. The startup is also including 10 hours of transcribed audio, spliced into 10 second increments, to expedite the training process.

Similar to Keras, Kur further abstracts the process of building and training deep learning models. By making deep learning easier, Kur is also making image recognition and speech analysis more accessible.

Scott Stephenson, CEO of Deepgram, explained to me that when the company was first getting off the ground, the team used LibriSpeech, an online dataset of audiobooks in the public domain split up and labeled for training early machine learning models.

Deepgram isn’t reinventing the wheel with its release. Coupled with data dumps and open source projects from startups, universities and big tech companies alike, frameworks like Tensorflow, Caffe and Torch have become quite useable. The ImageNet database has worked wonders for image recognition, and many developers use VoxForge for speech, but more open source data is never a bad thing.

“You can start with classifying images and end up with self driving cars,” added Stephenson. “The point is giving someone that first little piece and then people can change the model and make it do something different.”

Getting Kur into the hands of developers will also help Deepgram with recruiting talent. The strategy has proved itself quite useful for large tech companies looking to recruit technical machine learning and data science engineers.

Via, developers will soon be able to share models, data sets and weights to spur more innovation in the space. Deepgram eventually wants to release weights for the data-set being released today so DIY-ers can avoid processor intensive training altogether. Even with a relatively modest 10 hours of audio, models still take about a day to train on a GPU and considerably longer with an off-the-shelf computer.

If you end up exhausting the Deepgram data set, you can also easily expand it with your own data. All you have to do is create WAV files with embedded transcriptions in 10 second increments. You can feed data-hungry deep learning models with more resources in the public domain to improve accuracy.

from TechCrunch