Hackaday Links: July 22, 2018

KiCad Version 5 has been released! Footprints are going to be installed locally, and the Github plugin for library management is no longer the default. You now have the ability to import Eagle projects directly, Eeschema has a better configuration dialog, better wire dragging, and Pcbnew now has complex pad shapes. The changelog also says they’ve gone from pronouncing it as ‘Kai-CAD’ to ‘Qai-CAD’.

Kids can’t use computers because of those darn smartphones. Finally, the world is ending not because of Millennials, but because of whatever generation we’re calling 12-year-olds. (I’m partial to Generation Next, but that’s only because my mind is polluted with Pepsi commercials from the mid-90s.)

Need a NAS? The Helios4 is built around the Marvell Armada 388 SoC and has four SATA ports, making it a great way to connect a bunch of hard drives to a network. This is the second run from the team behind the Helios, and now they’re looking to take it into production.

A while ago, [Dan Macnish] built Draw This, a camera that takes an image, sends it through artificial intelligence, and outputs a cartoon on a receipt printer. It’s a camera that prints pictures of cartoons. Of course, some people would want to play with this tech without having to build a camera from scratch, so [Eric Lu] built Cartoonify, a web-based service that turns pictures into cartoons.

Grafitti is fun to spell and fun to do, and for all the proto-Banskys out there, it’s all about stencils. [Jeremy Cook] did a quick experiment with a 3D-printed spray paint stencil. It works surprisingly well, and this is due to leveraging the bridging capability of his printer. He’s putting supports for loose parts of the stencil above where they would normally be. The test sprays came out great, and this is a viable technique if you’re looking for a high-quality spray paint stencil relatively easily.

Custom Media Player Helps Hacker’s Autistic Son

Getting to play with technology is often the only justification a hacker needs to work on a build. But when your build helps someone, especially your own special-needs kid, hacking becomes a lot more that playing. That’s what’s behind this media player customized for the builder’s autistic son.

People generally know that the symptoms of autism cover a broad range of behaviors and characteristics that center around socialization and communication. But a big component of autism spectrum disorders is that kids often show very restricted interests. While [Alain Mauer] doesn’t go into his son [Scott]’s symptoms, our guess is that this media player is a way to engage his interests. The build came about when [Alain] was unable to find a commercially available media player that was simple enough for his son to operate and sturdy enough to put up with some abuse. A Raspberry Pi came to the rescue, along with the help of some custom piezo control buttons, a colorful case, and Shin Chan. The interface allows [Scott] to scroll through a menu of cartoons and get a preview before the big show. [Scott] is all smiles in the video below, and we’ll bet [Alain] is too.

Pi-based media player builds are a dime a dozen on Hackaday, but one that helps kids with autism is pretty special. The fact that we’ve only featured a few projects aimed at autistics, like this 2015 Hackaday Prize entry, is surprising. Maybe you can come up with something like [Alain]’s build for the 2016 Hackaday Prize.

Continue reading “Custom Media Player Helps Hacker’s Autistic Son”

The Machine that Japed: Microsoft’s Humor-Emulating AI

Ten years ago, highbrow culture magazine The New Yorker started a contest. Each week, a cartoon with no caption is published in the back of the magazine. Readers are encouraged to submit an apt and hilarious caption that captures the magazine’s infamous wit. Editors select the top three entries to vie for reader votes and the prestige of having captioned a New Yorker cartoon.

The magazine receives about 5,000 submissions each week, which are scrutinized by cartoon editor [Bob Mankoff] and a parade of assistants that burn out after a year or two. But soon, [Mankoff]’s assistants may have their own assistant thanks to Microsoft researcher [Dafna Shahaf].

[Dafna Shahaf] heard [Mankoff] give a speech about the New Yorker cartoon archive a year or so ago, and it got her thinking about the possibilities of the vast collection with regard to artificial intelligence. The intricate nuances of humor and wordplay have long presented a special challenge to creators. [Shahaf] wondered, could computers begin to learn what makes a caption funny, given a big enough canon?

[Shahaf] threw ninety years worth of wry, one-panel humor at the system. Given this knowledge base, she trained it to choose funny captions for cartoons based on the jokes of similar cartoons. But in order to help [Mankoff] and his assistants choose among the entries, the AI must be able to rank the comedic value of jokes. And since computer vision software is made to decipher photos and not drawings, [Shahaf] and her team faced another task: assigning keywords to each cartoon. The team described each one in terms of its contextual anchors and subsequently its situational anomalies. For example, in the image above, the context keywords could be car dealership, car, customer, and salesman. Anomalies might include claws, fangs, and zoomorphic automobile.

The result is about the best that could be hoped for, if one was being realistic. All of the cartoon editors’ chosen winners showed up among the AI’s top 55.8%, which means the AI could ultimately help [Mankoff and Co.] weed out just under half of the truly bad entries. While [Mankoff] sees the study’s results as a positive thing, he’ll continue to hire assistants for the foreseeable future.

Humor-enabled AI may still be in its infancy, but the implications of the advancement are already great. To give personal assistants like Siri and Cortana a funny bone is to make them that much more human. But is that necessarily a good thing?

[via /.]