szczys writes: Helium is an important commodity for health care (MRI machines), scientific research, and for the delight of toddlers everywhere. This last use has been called out as a waste, with figures like "$100 per balloon" decried over the actual price of under fifty cents per balloon. But alarms about world supply and inflated cost are overblown. Helium is a non-renewable resource that is found as a byproduct of natural gas production, in rich deposits like the one recently discovered in Tanzania, and other natural sources. Since the US removed price controls on Helium in 1996, supply and demand have kicked in, normalizing use and production to fluctuate with market price.
szczys writes: It's surprising to bump into a NASA engineer, even moreso if they were one of the original software engineers for the Lunar Module used during the Apollo missions. That's what happened recently at a party held at the MIT Faculty club. Don Eyles even brought along an original hard copy of the Lunar Module source code of which he wrote about 2000 lines. He carries the green and white striped stack of paper around in its own suitcase and is happy to share some of his experiences, like working through error codes he wrote which Neil Armstrong faced during the final descent to the moon's surface.
szczys writes: Humanity has already created killer robots — we just haven't yet decided to let them make the choice to kill by themselves As it stands, autonomous weapons systems lack the ability to accurately distinguish an enemy combatant from a friendly and this is currently the distinction that makes killer robots a violation of international law. Soon, advances in Artificial Intelligence will break down this barrier. The Inhumane Weapons Convention has been meeting for three years in Geneva. They just released a preliminary report on Lethal Autonomous Weapons Systems. Progress is very slow; the working group hasn't yet officially agreed on the definition of what constitutes one of these autonomous systems. Meanwhile, military R&D marches on in a race to see if these robots will be banned by convention before seeing a tour of duty.
szczys writes: Fasteners are all around us and most of the time people are making bad choices on the tool used to turn them, the fastener itself, or both. Flat head, Phillips, Robertson, Pozidriv, or other, there is a tool for each screw interface. Spend a bit of time learning about the technology of screws and screwdrivers, apply that next time you reach for a tool, and you will no longer be on the receiving end of broken bits, ruined screw heads, and dented drywall.
szczys writes: How many ways can a crowd-funding campaign fail? When they make headlines it's almost always the biggest failures, frauds, or the highest dollar amount ever raised. What is more useful is a model that can be applied for success; can you hit your goal and deliver on time with the promised features?
Pocket NC is the story of how crowd funding is supposed to work. The company is founded around a desktop 5-axis CNC mill which saw a very popular Kickstarter campaign. But before they launched they had actually fabricated and assembled a number of units, accounting for all material costs and even timing the assembly process down to the second. Everything was in place beforehand to ensure they could deliver on their promises. This is the model that startups seeking crowdfunding need to follow. It's exactly what a traditional investor would look for before pulling the trigger and the benchmark that crowdfunding investors need to demand.
szczys writes: The rise of crowd funding has opened a vector for new hardware companies that simply didn't exist before and with that comes one failed campaign after another. Having been around for some time now, this ground is not unexplored, we just don't necessarily hear the stories of every failure. Kickstarter failures are indeed stereotypical and anyone looking to launch their own crowd funding campaign should learn from the common types of failure, be it lack of interest, failing to raise the goal, underestimating production time and cost, botching quality control, defrauding the customer (or appearing to), and not being prepared to actually succeed.
szczys writes: Peachy Printer made it big on its Kickstarter, raising over half a million dollars on the idea of floating resin on top of saltwater to replace traditional mechanical elements of a 3D printer. The company has now collapsed due too embezzlement of those funds. The original investor stole around $350k in backer's money and funneled it into a new home. This was discovered about 18 months ago and only becoming public now as the company is unable to meet their already delayed delivery dates.
szczys writes: MakerBot may not be dead but it's alive in name only. The company was the first great hope in commercial 3D Printers — you could buy a MakerBot or you could build your own. But one stumble after another led to a transition away from Open Source and a precipitous fall from grace which has ended this week with the announcement that all of their manufacturing staff has been laid off and production will be shifted to China. Brian Benchoff has followed the story since day one and covers the life and times of the first 3D Printer company in this MakerBot obituary.
szczys writes: The USB to mains adapter is an important part of electronics. It's the last line of defense between the device and line voltage. But it also defends the user against shock and guards against accidental fire. Bob recently took on the challenge of finding safe and reliable USB power adapters to ship with a product his company has developed.
USB power supplies are super cheap and omnipresent. They are the Tribble of my household. But they're not all created equal, and some of them may even be dangerous. I had to source USB power supplies for a product, and it wasn't easy. But the upside is that I got to tear them all apart and check out their designs.
szczys writes: The real challenge with alternative energy sources is beating the phenomenal energy densities of fossil fuels. Current battery technology just can't beat gasoline, diesel, and the like but metal powder does have high energy density and can be seen as rechargeable. Recent research points to burning metal particles to power engines. The particles themselves can be regenerated by capturing the waste and performing a conversion powered by renewable energy sources. But this brings up its own set of problems. The waste materials will quickly clog traditional internal combustion engines the same problem faced by coal dust engine experiments from last century. If metal particle engines do prove themselves, you will be driving around an external combustion car: think steam engine.
szczys writes: The general view of hardware startups has become skewed away from what it means to actually start a company. Now when someone from a startup introduces their product, they're apt to immediately face the questions: When is your Kickstarter? and What accelerators have you applied for? But startups old and new have far more options than this. Bob Baddeley is an engineer who has navigated the startup game from both sides of the fence and recently looked at many of the reasons you should choose a different path with your fledgling hardware company.
szczys writes: Self-driving cars are now joined by self-driving farm implements. A farmer in Manitoba, Canada added auto-piloting gear often found in quadcopters to one of his tractors and used it during the season's harvest. The tractor is not doing the grain harvesting, but ferrying loads between the harvester (where he's behind the wheel) and the grain truck which will ultimately take it to market. The key here is that the vehicle is staying in the field rather than taking to the open roads.
szczys writes: Wireless communications is a fascinating race to build bigger networks with faster coverage. The cellular communications tower is where the rubber hits the road, and there is a surprising variety of tower types, antennas, and methods of getting the data from them back to the switch. This field guide to communications towers covers all of that technology. Guyed towers, lattice towers, and monopole towers are the most common way to get the cellular antennas into the air. First-come-first-served decides whose network gets priority elevation, and a combination of microwave, fiber, and copper provide connectivity. But there's a lot more going on in these systems that isn't as obvious.
szczys writes: Charles Lohr is known for implementing incredible protocols on diminutive microcontrollers. Now's he's managed to get 10Base-T Ethernet working on an ESP8266. The modules are well-known for providing dirt-cheap WiFi access, and the chip is quite powerful beyond that single purpose. In this case he's utilized the I2S hardware peripheral along with DMA to achieve the high speeds. This isn't just a party trick, the radio on the ESP8266 is quite versatile, including promiscuous and mesh modes. Adding the support hardware (jack and ISL3177 line driver) only calls for a few more dollars and you end up with a configurable WiFi tool that can be plugged into a network cable.
The iPhone encryption gatekeeper is the screen lock, which uses hardware to cripple the device if it detects a brute-force attack. Circumvention of this is not a huge threat to everyone who uses the same model of phone because an attacker would need access to the hardware — most likely for an extended period of time and in a destructive way. However, if the vulnerability was software-based it has far reaching implications because physical access to the device may not be necessary in that case. The government (and the Israel-based company that did the work for them) have a moral obligation to disclose such a vulnerability.