In a ‘Dangerous and Sinister Step,’ London Police Start Using Live Face Recognition Tech

https://gizmodo.com/in-a-dangerous-and-sinister-step-london-police-start-u-1841201259

The dystopian nightmare begins. Today, London’s Metropolitan Police Service announced it will begin deploying Live Facial Recognition (LFR) tech across the capital in the hopes of locating and arresting wanted peoples.

“We are using a tried-and-tested technology, and have taken a considered and transparent approach in order to arrive at this point,” Assistant Commissioner Nick Ephgrave said in a statement. “Similar technology is already widely used across the UK, in the private sector. Ours has been trialed by our technology teams for use in an operational policing environment.”

The way the system is supposed to work, according to the Metropolitan Police, is the LFR cameras will first be installed in areas where ‘intelligence’ suggests the agency is most likely to locate ‘serious offenders.’ Each deployment will supposedly have a ‘bespoke’ watch list comprising images of wanted suspects for serious and violent offenses. The London police also note the cameras will focus on small, targeted areas to scan folks passing by. According to BBC News, previous trials had taken place in areas such as Stratford’s Westfield shopping mall and the West End area of London. It seems likely the agency is also anticipating some unease, as the cameras will be ‘clearly signposted’ and officers are slated to hand out informational leaflets.

The agency’s statement also emphasizes that the facial recognition tech is not meant to replace policing—just ‘prompt’ officers by suggesting a person in the area may be a fishy individual…based solely on their face. “It is always the decision of an officer whether or not to engage with someone,” the statement reads. On Twitter, the agency also noted in a short video that images that don’t trigger alerts will be immediately deleted.

As with any police-related, Minority Report-esque tech, accuracy is a major concern. While the Metropolitan Police Service claims that 70 percent of suspects were successfully identified and that only one in 1,000 people created a fake alert, not everyone agrees the LFR tech is rock-solid. An independent review from July 2019 found that in six of the trial deployments, only eight of 42 matches were correct for an abysmal 19 percent accuracy. Other problems found by the review included inaccurate watch list information (e.g., people were stopped for cases that had already been resolved), and the criteria for people being included on the watchlist weren’t clearly defined.

Privacy groups aren’t particularly happy with the development. Big Brother Watch, a privacy campaign group that’s been particularly vocal against facial recognition tech, took to Twitter, telling the Metropolitan Police Service they’d “see them in court.”

“This decision represents an enormous expansion of the surveillance state and a serious threat to civil liberties in the UK,” said Silkie Carlo, Big Brother Watch’s director, in a statement. “This is a breath-taking assault on our rights and we will challenge it, including by urgently considering next steps in our ongoing legal claim against the Met and the Home Secretary.”

Meanwhile, another privacy group Liberty, has also voiced resistance to the measure. “Rejected by democracies. Embraced by oppressive regimes. Rolling out facial recognition surveillance tech is a dangerous and sinister step in giving the State unprecedented power to track and monitor any one of us. No thanks,” the group tweeted.

The London police’s decision comes at an interesting time. Just last week, the European Union began mulling a three-to-five-year ban on facial recognition tech in public areas. It’s unclear whether that ban will ever come to pass—the news came via a leaked version of an early draft of a European Commission white paper. It’s also not clear if it will ultimately matter, given the looming specter of Brexit. That said, Liberty has already garnered over 22,000 signatures for a petition demanding Britain’s Home Secretary ban the use of facial recognition tech in public places.

via Gizmodo https://gizmodo.com

January 24, 2020 at 11:21AM

States are suing the US government over 3D-printed gun blueprints

https://www.technologyreview.com/f/615091/states-are-suing-the-us-government-over-3d-printed-gun-blueprints/

The news: A group of 20 states, led by Washington, are suing the federal government over rules that could allow the sharing of 3D-printed gun blueprints online. The lawsuit was filed in a US District Court in Seattle yesterday, with California and New York among the states signing up.

The details: The federal government looking to change the law to allow 3D-printed gun blueprints to be shared on the internet. Yesterday, the Trump administration published rules which would transfer oversight for the export of certain firearms from the State Department to the Commerce Department, claiming it would promote exports and reduce the regulatory burden on gun makers. Critics fear it will lead to an escalation of gun violence. The lawsuit states that deregulation will “make it far easier for individuals ineligible to possess firearms under state or federal law to obtain a deadly weapon without undergoing a background check,” according to the AP.

The background: This fight has been rumbling for a long time. Back in 2013, a pro-gun activist group called Defense Distributed posted blueprints for a 3D-printed gun online. They were ordered to take them down by the Obama administration, but fought back, claiming in a lawsuit against the government that it had a First Amendment right to share the files. It lost. But then in June 2018, the Trump administration granted the company permission to post the blueprints online. Last year a federal judge struck this attempt down. Now the Trump administration is trying yet again to allow the sharing of 3D-printed gun blueprints online.

Sign up here to our daily newsletter The Download to get your dose of the latest must-read news from the world of emerging tech. 

via Technology Review Feed – Tech Review Top Stories https://ift.tt/1XdUwhl

January 24, 2020 at 06:04AM

Adam Savage Puts Boston Dynamics Spot Quadruped Robot Through Outdoor Agility Course

https://geekologie.com/2020/01/adam-savage-puts-boston-dynamics-spot-qu.php


This is a video of Adam Savage putting one of Boston Dynamics Spot quadruped robots through its paces in an outdoor agility course as an introductory video for a full year of TESTED conducting builds and projects with the robot (you can subscribe to those HERE). So that’s something to look forward to. I mean maybe not as forward to as your birthday or mom getting out of jail, but it’s something. So, uh, are we allowed to make suggestions? "Knock it on its back like a turtle!" "See if it can swim!" "The lava test!" Oh you *wiping tears from eyes* you really are my everything.
Keep going for the video while I speculate why Adam won’t put the Nintendo Switch down.

Thanks to Jeffrey S and hairless, who agree call us when they’re big enough to ride like He-Man’s Battle Cat so we officially know it’s time to enter the bunker and seal the door.

via Geekologie – Gadgets, Gizmos, and Awesome https://geekologie.com/

January 24, 2020 at 07:11AM

Is Premium Gasoline Really Better for Your Car? [Science Video]

https://www.geeksaresexy.net/2020/01/24/premium-gasoline-really-better-car-science-video/

Is Premium Gasoline Really Better for Your Car? [Science Video]

Are you slowly killing your car by using below average gas? In this video from the SciShow channel, host Olivia talks about octane ratings and how your vehicle is designed to handle them.

[SciShow]

via [Geeks Are Sexy] Technology News https://ift.tt/23BIq6h

January 24, 2020 at 08:12AM

Pharmaceutical Executive John Kapoor Sentenced To 66 Months In Prison In Opioid Trial

https://www.npr.org/2020/01/23/798973304/pharmaceutical-executive-john-kapoor-sentenced-to-66-months-in-prison-in-opioid?utm_medium=RSS&utm_campaign=news

Insys Therapeutics founder John Kapoor was convicted in a bribery and kickback scheme that prosecutors said helped fuel the opioid crisis.

His sentencing is the culmination of a months-long criminal trial that resulted in the first successful prosecution of pharmaceutical executives tied to the opioid epidemic.

(Image credit: Charles Krupa/AP)

via NPR Topics: News https://ift.tt/2m0CM10

January 24, 2020 at 04:24AM

The math of brewing a better espresso

https://arstechnica.com/?p=1646613

A new mathematical model sheds light on the optimal brewing process for espresso.
Enlarge /

A new mathematical model sheds light on the optimal brewing process for espresso.

Five Senses Coffee (Australia)

Skilled baristas know that achieving the perfect complex flavor profile for a delectable shot of espresso is as much art as science. Get it wrong, and the resulting espresso can taste too bitter or sourly acidic rather than being a perfect mix of each. Now an international team of scientists has devised a mathematical model for brewing the perfect cup, over and over, while minimizing waste, outlined in a new paper in the journal Matter.

“A good espresso beverage can be made in a multitude of ways,” said co-author Christopher Hendon, a computational chemist at the University of Oregon. “The point of this paper was to give people a map for making an espresso beverage that they like and then be able to make it 100 times in a row.”

There’s actually an official industry standard for brewing espresso, courtesy of the Specialty Coffee Association, which sets out strict guidelines for its final volume (25-35mL, or roughly one ounce) and preparation. The water must be heated to 92-95° C (197-203° F) and forced (at a specific pressure) through a bed of 7-9 grams (about a quarter of an ounce) of finely ground coffee over the course of 20-30 seconds. But most coffee shops don’t follow this closely, typically using more coffee, while the brewing machines allow baristas to configure water pressure, temperature, and other key variables to their liking. The result of all those variations in technique is a great deal of variability in quality and taste.

“Most people in the coffee industry are using fine-grind settings and lots of coffee beans to get a mix of bitterness and sour acidity that is unpredictable and irreproducible,” said Hendon, a computational chemist at the University of Oregon. “It sounds counterintuitive, but experiments and modeling suggest that efficient, reproducible shots can be accessed by simply using less coffee and grinding it more coarsely.”

The flavors in espresso derive from roughly 2,000 different compounds that are extracted from the coffee grounds during brewing. So Hendon and his colleagues focused on building a mathematical model for a more easily measurable property known as the extraction yield (EY): the fraction of coffee that dissolves into the final beverage. That, in turn, depends on controlling water flow and pressure as the liquid percolates through the coffee grounds. Modeling the actual grounds—a form of granular media—proved much too daunting. “You would need more computing power than Google has to accurately solve the physics and transport equations of brewing on a geometry as intricate as a coffee bed,” said co-author Jamie M. Foster, a mathematician at the University of Portsmouth in the UK.

Schematic illustrating two strategies to improve espresso reproducibility.
Enlarge /

Schematic illustrating two strategies to improve espresso reproducibility.

Cameron et al./Matter

Instead, Hendon, Foster, and their colleagues based their model on how lithium ions propagate through a battery’s electrodes, which they liken to how caffeine molecules dissolve from coffee grounds. A bunch of simulations and several thousand experimental shots of espresso later (courtesy of Frisky Goat Espresso in Brisbane, Australia), the authors arrived at some surprising findings.

For instance, conventional wisdom holds that a fine grind is best, since more surface area of the resulting tamped-down coffee bed is exposed to the hot water, thus boosting the extraction yield. But this new model, and the group’s experiments, revealed that if coffee is ground too finely, it can clog the coffee bed, thereby reducing extraction yield. It’s also a big factor in the variability in taste. The researchers concluded that there are better methods for maximizing extraction yield, such as using fewer beans and coarser grinds with a bit less water. And the Specialty Coffee Association might be interested to hear that brew time is largely irrelevant.

“Though there are clear strategies, there is no obvious optimal espresso point.”

Coffee is a multi-billion dollar global industry. In 2015 alone, according to the authors, just the US market accounted for some 1.5 million jobs and generated $225.2 billion in revenue. But climate change (along with changing customer tastes) is threatening coffee producers, sparking interest in finding ways to maintain quality while cutting costs and reducing waste. This new model should lend insight into precisely how one might accomplish that, although there is still some wiggle room to account for subjective personal preferences in the flavor profile.

“Though there are clear strategies to reduce waste and improve reproducibility, there is no obvious optimal espresso point,” said Hendon. “There is a tremendous dependency on the preferences of the person producing the coffee; we are elucidating the variables that they need to consider if they want to better navigate the parameter space of brewing espresso.”

Based on a year-long trial waste reduction protocol set up at a local specialty coffee shop in Eugene, Oregon, the authors estimate that a small cafe could save several thousand dollars per year by reducing the mass of coffee used, while the industry as a whole could conceivably save as much as $1 billion per year.

“The real impact of this paper is that the most reproducible thing you can do is use less coffee,” said Hendon. “If you use 15 grams instead of 20 grams of coffee and grind your beans coarser, you end up with a shot that runs really fast but tastes great. Instead of taking 25 seconds, it could run in 7 to 14 seconds. But you end up extracting more positive flavors from the beans, so the strength of the cup is not dramatically reduced. Bitter, off-tasting flavors never have a chance to make their way into the cup.”

DOI: Matter, 2020. 10.1016/j.matt.2019.12.019  (About DOIs).

via Ars Technica https://arstechnica.com

January 22, 2020 at 10:05AM

Firefox’s New Voice Controls Make It As Good As a Smart Speaker

https://lifehacker.com/firefoxs-new-voice-controls-make-it-as-good-as-a-smart-1841066550

I’m used to yelling at my Google Home and Amazon Echo, not so much my web browser. However, Mozilla is now testing voice-control capabilities for its Firefox browser, and you can join me in screaming at your screen—which is actually a little more fun and useful than it might sound at first glance.

To get started, open Firefox and visit Mozilla’s Firefox Voice Campaign page. Click on the “Get started” button, cough up your name and email address, and you’ll get a link to download the Firefox Voice addon:

Once you’ve installed it, you’ll have to approve a few permissions before you can get going. First, you’ll need to grant it access to your microphone—obviously. Mozilla will also ask whether it can collect and analyze voice transcripts of things you’ve said to your browser. It won’t identify who you are, so I’ll leave this one for you to decide. (I don’t like it when companies have recordings of me, but I also doubt I’ll be talking very scandalously to my web browser.)

That’s all the configuring you’ll have to do, though. To start yelling at your browser, you can either click on the microphone icon in the upper-right corner of your screen or tap CTRL + [period]. As for what you can tell Firefox to do, Mozilla has a giant list of commands for you to try, and the websites that support them, including:

  • Searching the web (“Look up recipes for manhattans”)
  • Search a specific website for information (“Search my Gmail for tickets to Hamilton”)
  • Visiting a specific webpage (“Go to Lifehacker”)
  • Asking a general question (“Who is David Murphy?”)
  • Playing music or videos (“Play Green Day on Spotify”)
  • Controlling audio or video playback (play, mute, pause, etc.)
  • Reading websites to you out loud (“Read this page”)
  • Getting the weather report (“What’s the temperature in San Francisco?”)
  • Translating a website to another language (“Translate this webpage to Spanish”)
  • Getting directions (“Find the nearest sushi on maps”)
  • Setting a timer (“Set a timer for five minutes”)
  • Finding a tab among your way-too-many open tabs (“Find calendar tab”)
  • Generic browser controls (“Open a new tab”)
  • Copying and pasting (“Copy full page screenshot”)
  • Creating shortcuts for commands (“open Lifehacker” > “Give that the name hacks”)

The only downside—with an asterisk—is that there isn’t a wake word to simply trigger Firefox’s voice capabilities. From a privacy perspective, that’s great; however, it is slightly inconvenient to have to click an icon or tap a keyboard shortcut all the time.

I also noticed that the CTRL + [period] shortcut actually loaded up my Firefox Containers context menu, not the microphone. However, you can change the microphone trigger to anything you want by clicking on the icon and then clicking on the gear icon to access the add-on’s settings.

(I also recommend turning off the chime that plays whenever you open the mic, lest you go crazy.)

via Lifehacker https://lifehacker.com

January 21, 2020 at 09:05AM