Year: 2022

Machine Learning Gives Cats One More Way To Control Their Humans

August 26, 2022 by 9 Comments

For those who choose to let their cats live a more or less free-range life, there are usually two choices. One, you can adopt the role of servant and run for the door whenever the cat wants to get back inside from their latest bird-murdering jaunt. Or two, install a cat door and let them come and go as they please, sometimes with a “present” for you in their mouth. Heads you win, tails you lose.

There’s another way, though: just let the cat ask to be let back in. That’s the approach that [Tennis Smith] took with this machine-learning kitty doorbell. It’s based on a Raspberry Pi 4, which lives inside the house, and a USB microphone that’s outside the front door. The Pi uses Tensorflow Lite to classify the sounds it picks up outside, and when one of those sounds fits the model of a cat’s meow, a message is dispatched to AWS Lambda. From there a text message is sent to alert [Tennis] that the cat is ready to come back in.

There’s a ton of useful information included in the repo for this project, including step-by-step instructions for getting Amazon Web Services working on the Pi. If you’re a dog person, fear not: changing from meows to barks is as simple as tweaking a single line of code. And if you’d rather not be at the beck and call of a cat but still want to avoid the evidence of a prey event on your carpet, machine learning can help with that too.

[via Tom’s Hardware]

This Week In Security: In Mudge We Trust, Don’t Trust That App Browser, And Firefox At Pwn2Own

August 26, 2022 by 19 Comments

There’s yet another brouhaha forming over Twitter, but this time around it’s a security researcher making noise instead of an eccentric billionaire. [Peiter Zatko] worked as Twitter’s security chief for just over a year, from November 2020 through January 2022. You may know Zatko better as [Mudge], a renowned security researcher, who literally wrote the book on buffer overflows. He was a member at L0pht Heavy Industries, worked at DARPA and Google, and was brought on at Twitter in response to the July 2020 hack that saw many brand accounts running Bitcoin scans.

Mudge was terminated at Twitter January 2022, and it seems he immediately started putting together a whistleblower complaint. You can access his complaint packet on archive.org, with whistleblower_disclosure.pdf (PDF, and mirror) being the primary document. There are some interesting tidbits in here, like the real answer to how many spam bots are on Twitter: “We don’t really know.” The very public claim that “…<5% of reported mDAU for the quarter are spam accounts” is a bit of a handwave, as the monetizable Daily Active Users count is essentially defined as active accounts that are not bots. Perhaps Mr. Musk has a more legitimate complaint than was previously thought.
Continue reading “This Week In Security: In Mudge We Trust, Don’t Trust That App Browser, And Firefox At Pwn2Own” →

You Paid For This Paper. Now You Can Read It Without Paying Again

August 26, 2022 by 44 Comments

There is probably very little among the topics covered here at Hackaday that doesn’t have its roots somewhere in scientific research. Semiconductor devices for example didn’t simply pop into being in Bell Labs or Texas Instruments, the scientists and engineers who created them did so standing on the shoulders of legions of earlier researchers who discovered the precursor steps that made them possible. As many readers will know, scientific research for its own sake is expensive, so much so that much of it is funded by governments, from your taxes. The research papers with the findings are then hidden from public view behind paywalls by the publishers who distribute them, an injustice which should soon be over for Americans, thanks to a White House memorandum paving the way for federally funded research to be freely available to the public at no cost by no later than 2025.

The academic publishing business originates in the days when paper was king, and it has several tiers. Officially an academic journal is usually the product of a professional body in its field, but it is normal for the publishing itself to be contracted out to a specialist academic publishing company. They accept submissions of papers, edit them, and arrange peer reviewers, before publishing the journals. Originally this was a paper process, but while journals are still printed it’s the Internet through which they are now read. The publishers pay nothing to the researcher for their paper and often only a nominal sum to the reviewers for their input, but charge a hefty subscription for access to the content. As you might imagine it’s an extremely lucrative business, so as this Hackaday scribe saw when she worked in that industry, the publishers and the learned bodies are in no hurry to kill their golden goose.

This move to open access may make few immediate waves outside the world of scientific publishing, but it affirms the principle that taxpayers should be able to see the fruits of their spending. As such it will be of benefit to less-well-off researchers and institutions worldwide. Rest in peace Aaron Swartz, if only you could have seen this day!

White House pic: Matt H. Wade, CC BY-SA 3.0.

Simple Breadboard SDR For Shortwave

August 26, 2022 by 20 Comments

One of the best ways to learn about radios is to build your own, even in the age of cheap SDR dongles. [Aniss Oulhaci] demonstrates this with a simple HF SDR receiver built on a breadboard.

The receiver takes the form of a simplified Tayloe detector. An RF preamp circuit amplifies the signal from a shortwave antenna and feeds it into a 74HC4066D analog switch, which acts as a switching mixer. It mixes the input signal with the local oscillator’s I and Q signals to produce the intermediate frequency signals. The local oscillator consists of a SI5351 clock generator with a 74HC74D flip-flop to generate the I and Q pair. The signals pass through a low pass filter stage and get amplified by an LM358 op amp, resulting in the IQ signal pair being fed to a computer’s stereo sound card.

An Arduino is used to control the SI5351 clock generator, which in turn is controlled by the same program created for the SDR Shield. With the audio signal fed to HDSDR, [Aniss] was able to pick up a shortwave radio broadcaster.

While this is by no means a high-performance receiver, building an SDR on a breadboard is still a great weekend project, with plenty of potential for further experimentation.

Continue reading “Simple Breadboard SDR For Shortwave” →

Buy The Right To Build A Nakagin Tower Anywhere

August 25, 2022 by 41 Comments

We’re guessing that among Hackaday’s readership are plenty of futurists, and while the past might be the wrong direction in which to look when considering futurism, we wouldn’t blame any of them for hankering for the days when futurism was mainstream.

Perhaps one of the most globally iconic buildings of that era could have been found in Tokyo, in the form of the Nakagin Capsule Tower, Kisho Kurokawa’s 1972 Metabolist apartment block. This pioneering structure, in which individual apartments were conceived as plug-in units that could be moved or changed at will, never achieved its potential and was dismantled, looking more post-apocalyptic than futuristic in early 2022, but it could live on in both digital form and reconstructed elsewhere as the rights to its design are being auctioned.

Unfortunately there appears to be some NFT mumbo-jumbo associated with the sale, but what’s up for auction is a complete CAD model along with the rights to build either real or virtual copies of the building. It’s unlikely that any Hackaday readers will pony up for their own Metabolist skyscraper, but the interest lies not only in the love of a future that never quite happened, but in the engineering behind the structure. Where this is being written as in many other places there is simultaneously a chronic housing shortage and a housing system wedded to the outdated building techniques of a previous century, so the thought of updated equivalents of the Nakagin Tower offering the chance of modular interchangeable housing in an era perhaps more suited to it than the 1970s is an intriguing one. Now that we’re living in the future, perhaps it’s time to give futurism another chance.

Regular readers will have spotted this isn’t the first time we’ve brought you a taste of futuristic living.

Header: Svetlov Artem, CC0.

Objective Hotend Performance Measurement Is Hard

August 25, 2022 by 9 Comments

Evaluating the performance of 3D printers and component upgrades is a more difficult than it may seem at first glance, and subjective observations can lead to incorrect conclusions. To objectively determine the maximum flow rates of different FDM 3D printer hotends, [MirageC] is developing a robust testing standard backed by more than just visual observations.

Defining the max flow rate threshold is not straightforward. A common method is to run a test print while slightly increasing the flow rate with each layer, and visually making a judgment on the last acceptable layer. It would be easy to miss errors, or unconsciously be inconsistent with observations over time. [MirageC] wanted to back up observations with measurements. To do this, he is measuring the true feed rate of the filament with an encoder wheel, and the backpressure of the filament on the extruder using a load cell. A Bowden tube helps to isolate the extruder from the vibration of the moving printhead.

After much testing, [MirageC] determined that the numerical threshold would be a specific deviation percentage between the desired and actual flow rate. At temperatures above 230°C, [MirageC] found that the last visually acceptable layer was consistently around 5.75% flow rate deviation for one specific PLA filament. It does not mean that 5.75% will be the magic number for all filaments and nozzle size, but it does provide a measurable parameter to back up visual observations.

In a world of questionable product reviews this dedication to objectivity is a breath of fresh air. If you are looking to upgrade your 3D printer’s hotend [MirageC]’s tests would be a good source of information.

Continue reading “Objective Hotend Performance Measurement Is Hard” →

A tennis racket and a tennis ball with a spinning motor inside

A Self-Spinning Tennis Ball To Surprise Your Opponent

August 25, 2022 by 6 Comments

In many ball sports like golf, football and tennis, controlling the ball’s spin is an important skill. Expert players can make golf balls curve around obstacles, launch footballs towards goal posts from impossible angles, or confuse their opponents by making a tennis ball bounce in a completely unexpected direction.

[Luis Marx], by his own admission, is not an expert tennis player at all, so when he found himself humiliated on the court by his roommate he set about finding a different way to win. In other words, to cheat. The basic idea was to make a tennis ball that would start spinning at the push of a button, rather than by skillful wielding of a racket: a spinning ball that flies through the air will follow a curved trajectory, so if you can make a ball spin at will, you can change its direction in mid-air.

Making a ball spin by itself is not as hard as it may sound. All you need is an electric motor that’s small enough to fit inside, along with a power source and some way to turn it on. When the motor inside the ball starts to spin, Newton’s third law ensures that the outside will spin in the opposite direction. [Luis] found a suitable DC motor and mounted it on a small custom-designed PCB along with an ESP8266 controller and powered it with a tiny lithium battery. A pushbutton mounted on his tennis racket operates the wireless interface to turn the motor on and off.

Although getting this setup to work wasn’t as easy as [Luis] had hoped, turning it into a ball that’s good enough to play tennis with was not straightforward either. [Luis] decided to 3D-print the outer shell using flexible filament in order to create something that would have the same amount of bounce as an ordinary rubber tennis ball. It took several rounds of trial and error with various types of filament to end up with something that worked, but the final result, as you can see in the video (in German, embedded below), was quite impressive.

Tests on the tennis court showed that [Luis] could now easily beat his roommate, although this was mostly due to the erratic bouncing caused by the ball’s spin rather than any aerodynamic effects. Still, the magic tennis ball achieved its objective and even survived several games without breaking. If you’re looking for a more brute-force approach to cheating at tennis, this 180 mph tennis ball trebuchet might come in handy.

Continue reading “A Self-Spinning Tennis Ball To Surprise Your Opponent” →