Stories
Slash Boxes
Comments

SoylentNews is people

Log In

Log In

Create Account  |  Retrieve Password


Site News

Join our Folding@Home team:
Main F@H site
Our team page


Funding Goal
For 6-month period:
2017-07-01 to 2017-12-31
Base Goal: $3000.00
Progress So Far:
Approximately: $518.20
17.3%
Stretch Goal: $2000.00
Progress So Far:
Approximately: $0.00
0.0%

Covers transactions:
2017-07-01 00:00:00 ..
2017-09-22 14:37:13 UTC
(SPIDs: [719..741])
Last Update:
2017-09-23 17:50:08 UTC
--martyb


Support us: Subscribe Here and buy SoylentNews Swag


We always have a place for talented people, visit the Get Involved section on the wiki to see how you can make SoylentNews better.

Which search engine do you use the most?

  • Google
  • Bing
  • DuckDuckGo
  • Ixquick
  • Yahoo!
  • AOL
  • Yandex
  • Other - Specify

[ Results | Polls ]
Comments:66 | Votes:376

posted by Fnord666 on Monday September 25, @05:53PM   Printer-friendly
from the surprising dept.

Last week, Minister of the European Parliament, Julia Reda, unearthed a well-hidden 2014 study financed by the European Commission entitled Estimating displacement rates of copyrighted content in the EU [warning: PDF] that studied the effects of copyright infringement on sales. The study cost 360,000 EUR to carry out and although it was ready in 2015, it was only made public last week when Reda was able to get ahold of a copy.

The study's conclusion was that with the exception of recently released blockbusters, there is no evidence to support the idea that online copyright infringement displaces sales. This conclusion is consistent with previous studies, and raises the following question: "Why did the Commission, after having spent a significant amount of money on it, choose not to publish this study for almost two years?"


Original Submission

posted by NCommander on Monday September 25, @04:00PM   Printer-friendly
from the KD2JRT-calling-CQ-on-146-point-52 dept.

As I've said in passing before here on SoylentNews, I'm a ham radio operator, (KD2JRT - Tech). Due to a lack of time and money, I've only been able to afford relatively cheap equipment, primarily two BaoFeng UV-82s, and an external antenna mount for the car. Many of the older ham radio ops decry the Baofengs as cheaply made Chinese junk, but I wanted to see what these radios are actually capable of. Historically, I've had decent success with an external antenna and decent positioning, but I recently conducted an impromptu experiment testing the propagation characteristics with these radios and seeing how well they actually work, and perhaps creating a baseline for more in-depth radio testing in the future.

Over the weekend, I took a day trip to the southern tip of New Jersey, out to Cape May from New York City, and along the way, using a hookup from the radio to my phone, I connected my radio to the national APRS network, and used it to measure distance and propagation effects. Since most of the readership of SoylentNews aren't radio techs, past the fold, I'll talk a bit about Baofeng radios, APRS, digital modes, and my results.

Automatic Packet Reporting System

APRS is one of the more interesting possibilities one can do with a ham radio system. At it's most basic core, APRS is a store-and-forward mesh network that is traditionally used on the 2M VHF band, and interconnected by a series of digipeaters that can use the internet to tie the entire system together. Due it's to architecture, an APRS mesh exists in any location where two radios are broadcasting on the same frequency in the same general area, and APRS itself has a long history of being using in disaster recovery efforts as a low bandwidth/high efficiency reporting system. APRS information is encoded in an AX.25 UI packet (AX.25 is a level-2 protocol used in ham radio).

KD2JRT-7>APDR13,WIDE1-1:=4041.55N/07359.99W$178/039/A=-00045 Out for a drive. Msg for QSO

Broken down, this becomes:

  • Callsign: KD2JRT
  • SSID: 7 (Handheld Radio)
  • Client: APDR13 (APRSdroid)
  • Relay Path: WIDE1-1 (I want to be relayed one hop)
  • Packet Type: = (Positioning data without timestamp)
  • GPS Location: 4041.55N/07359.99W (40 degrees north by 73 degrees west)
  • Course and Speed: 178/039 - Travelling 178 degrees at 39 knots
  • Altitude: A=-00045 I'm 45 feet below sea level
  • Comment: Out for a drive. Msg for QSO

Quite a bit of information in a tiny packet to say the least. Obviously my phone's altitude sensor (which was the source of the GPS information) does have some issues since I'm fairly sure I didn't drive underwater at any point in New Jersey.

It's also possible to use APRS without positional information, and to transmit digital messages from terminal to terminal, which can be relayed over the internet allowing for world-wide sharing of information, as well as post digital bulletins. Gateways exist to allow APRS messages to be sent to email, SMS, and Twitter. Unlike typical repeater systems used for voice, APRS is a simplex protocol; all messages are sent and received on the same frequency and does not guarantee delivery of a given packet. Receipt of a packet is usually in the form of digipeating; that is, a client hears it's own packet back as it passes through relays on the network. In the modern era, APRS packets are also forwarded onto the internet, and are recorded by services such as aprs.fi.

Due to it's nature, APRS is a perfect way to test propagation effects of radio signals from various locations. As I'm driving, the phone's GPS automatically records my position, and it's broadcast over the air to digipeaters without any intervention from myself. Now that we're got the basics of APRS out of the way, let's talk about the radio.

BaoFeng UV-82 and Antenna

The UV-82 is a simple two band handheld radio, which out of the box is capable of transmitting FM signals on the 2m VHF band, and the 70cm UHF band at up to 5 watts of power. Despite the reputation, Baofeng radios have become popular with new ham radio operators for their cheapness. For this test, I was using the Baofeng with a Nagoya UT-72 antenna magnetically mounted to the roof of the car (which acted as a ground plane) as seen in this photo.

Antenna on RAV4 roof

The feed line for the antenna is run in through the tailgate, and then back to the front where the Baofeng sits in a cupholder. A data cable connects the phone (a Samsung Galaxy S6 Active) running APRSdroid. Out of the box, the Baofeng doesn't support any digital modes, so a fair question to be asked is how am I doing APRS with it. The answer is that I'm essentially treating the radio as a simple acoustic coupler. APRS/AX.25 packets on VHF are encoded in Bell 202 audio frequency-shift keying at 1200 bits per second. The radio is connected to the phone's mic/speaker jack, APRSdroid listens for the modem tones, and simply modulates it's replies the same way. The radio is set to operate in VOX (essentially speakerphone) mode, and tuned to 144.390 simplex. In the future, I may buy a simple TNC that can automate this process for me.

VHF 2M Signal Propagation (or What To Expect)

Before going into any test, it's a decent idea to outline what to expect. The 2-meter band is what's known as a very high frequency band. As such, radio signals sent from the surface of the planet aren't (normally) reflected by the atmosphere, and continue out into space. As such, successfully two-way communication can only be achieved via line-of-sight communications with a maximum range of approximately 75 miles/120 kilometers under absolutely ideal circumstances assuming a relatively high receiving tower. In practice, the antenna design and power drastically influences the maximum range to a much lower number.

Without going in-depth into radio propagation theory and boring everyone half to death, there are two basic types of antennas: directional and omnidirectional. As the names suggest, a directional antenna focuses RF emissions in a specific direction, allowing you to get more bang for your buck, at the cost of focusing the beam and cutting out everything else. With a directional antenna, it's generally possible to transmit and receive upwards of 50-60 miles line of sight on 2Ms through obstructions, and there are reports of people using yagi directional antennas and handheld radios to successfully communicate with OSCAR satellites and the International Space Station in low earth orbit.

Omnidirectional antennas, like the Nagoya UT-72 instead emit signals in all directions, drastically lowing the signal strength as the power is dissipated out in all directions. Handheld radios (known as HTs) are inherently low power (also known as QRP) radios. In practice, the general rule of thumb is to expect 20-25 miles (30-40 km) at best. Within the heart of New York City, with all it's obstructions and the same antenna, I usually can get a signal to propagate about seven miles from Manhattan to the W2VL repeater from the waterfront. This is compounded by the fact that the UT-72 is a compromise antenna, and uses electrical lengthening techniques to allow it to properly emanate a 2M radio signal. If I had a proper dipole antenna rig mounted to the car, I could expect far better results.

One final consideration for my mobile setup is the fact that APRS (and it's base AX.25 protocol) is a digital mode. Transmission and reception is an all-or-nothing game. Unlike voice FM contacts which can usually be made out despite static (QRN), a digital signal must be heard clearly to be successfully decoded. In other words, I'm essentially only going to be registered on the network if the signal is extremely clear. This is very much a torture test for the little Baofeng and it's whip antenna.

The Raw Data

Before we go deeper into methodology, let's look at the raw data as seen on aprs.fi - track. aprs.fi only retains tracks for seven days so I'll summarize the most interesting data below.

Here's the route I drove, starting on Roosevelt Island in New York City, and ended in Cape May. Red dots represent location reports from my phone.

APRS map

Testing was only conducted North->South, due to forgetting to recharge the handheld's battery, and having it die on me upon reaching Cape May.

I was heard by the following stations.

Callsign Pkts First Heard - EDT Last Heard Longest (tx => rx) Longest At - EDT
N2MH-15 13 2017-09-24 02:45:02 2017-09-24 03:44:08 17.6 miles 283° 2017-09-24 02:45:02
WA2FPB-5 3 2017-09-24 04:24:24 2017-09-24 04:44:39 2.8 miles 151° 2017-09-24 04:24:24
W2AEE 6 2017-09-17 21:32:51 2017-09-24 02:47:45 6.5 miles 357° 2017-09-24 02:47:45
K2GE-13 2 2017-09-24 03:49:16 2017-09-24 03:56:12 5.4 miles 186° 2017-09-24 03:49:16
KC2QVT-15 1 2017-09-24 06:08:33 2017-09-24 06:08:33 35.2 miles 350° 2017-09-24 06:08:33
KB1EJH-15 1 2017-09-24 12:15:47 2017-09-24 12:15:47 23.1 miles 257° 2017-09-24 12:15:47
CLAYTN 1 2017-09-24 05:46:08 2017-09-24 05:46:08 34.8 miles 282° 2017-09-24 05:46:08
K2DLS-15 9 2017-09-24 03:46:39 2017-09-24 04:02:53 11.1 miles 158° 2017-09-24 03:46:39
WX2NJ-14 1 2017-09-24 05:08:57 2017-09-24 05:08:57 2.1 miles 104° 2017-09-24 05:08:57
BARNGT 8 2017-09-24 04:57:43 2017-09-24 05:46:40 20.8 miles 193° 2017-09-24 04:57:43
TOMRVR 6 2017-09-24 04:25:35 2017-09-24 05:10:58 16.4 miles 198° 2017-09-24 04:25:35
MATWAN 5 2017-09-24 03:04:19 2017-09-24 04:18:34 16.7 miles 210° 2017-09-24 03:04:19
K2RHK-10 1 2017-09-24 02:28:49 2017-09-24 02:28:49 1.9 miles 286° 2017-09-24 02:28:49

NOTE 1: W2AEE in Manhattan is showing more packets than it should as I can't filter by day on aprs.fi and I was testing the APRS setup last week. In total, approximately 50 packets were received by the APRS backbone.

NOTE 2: Despite being in a vehicle, the track shows a phone icon as I forgot to change the APRS reporting symbol

Due to complications with APRSdroid, I don't have an accurate count of how many packets were sent by the handset, but I suspect its approximately 50-100 for reasons that become clear below. Furthermore, I don't have an accurate count of which stations I heard due to stupidity. Now that we've got the raw data out of the way, let's draw some conclusions, and figure out how to refine the testing methodology.

Conclusions

All and all, for my first serious field test with the APRS system, this is a fairly healthy result. As I set out simply to play with APRS, and not actually do an experiment when I went out on Sunday, I didn't have in-depth methodology and logging set up. I also learned a few things about how APRSdroid (or more specifically, the SmartBeaconing system) works which drastically reduced the amount of traffic I sent which degraded the results even more. Despite that, I can draw some initial conclusions from this data, and perhaps do further radio propagation tests in the future if the SoylentNews community finds these types of articles interesting.

As expected, within urban areas, my signal propagation was total garbage, only being heard 1-2 miles under best case scenarios by the APRS gateways in Manhattan. Once I got south of the island and running along the waterfront, and on the Queens-Brooklyn Expressway did the signal start being received by the N2HM digipeater in New Jersey ~10-15 miles distant. Signal propagation continued to increase as I crossed the Verrazano Narrows into Staten Island, and then onward into New Jersey. As I stated before, APRS is a digital mode, and being understood is an all-or-nothing thing; if I was using FM voice, it's likely the effective Rx distance would have been further. Supporting that data, in the relatively flat areas of Jersey Coast, I was routinely being heard upwards of 15-20 miles out by the track due to lack of line-of-sight obstructions, and at two points, I was heard over 35 miles (56 km) away by two separate digipeaters!

Unfortunately, due to a lack of IGates south of Atlantic City, I wasn't heard again until I reached Cape May, and picked up by a digipeater on the other side of Delaware Bay

One thing that partially compromises the testing results however is the APRS SmartBeaconing system. As I put in the overview, APRS includes direction and speed information as part of its transmissions. As an optimization of the system, APRS clients can use SmartBeaconing to reduce the amount of traffic they're reporting. In effect, SmartBeaconing uses dead-reckoning to determine if it needs to send a packet. If I send a position report at a location travelling south at 60 MPH, and two minutes later, I'm two miles south of the previous report, SmartBeaconing will not send a new position since I am where I can realistically expected to be. Since I had cruise control enabled, and the Garden State Parkway runs roughly in a straight line North-South past Sandy Hook, SmartBeaconing drastically reduced the amount of data I was sending. Annoying, APRSdroid still shows phantom reports it didn't send on the log of my station history; and several times, I'd see a report show up in the outbound log, but no red light on the radio saying it was sending.

For receiving characteristics, I checked the phone at each rest area I stopped at, and I was seeing a healthy list of stations show up in the tracker. However, because of the store-and-forward nature of APRS, it's difficult to tell at a glance what stations I was hearing directly, and which were being echoed by a digipeater. It's possible to determine this information by decoding the received APRS packet as repeating stations amend their call sign to the WIDE1-1 line as the packet moves through the network. I also found that I frequently didn't hear myself on the network. I believe this is due to the fact the Baofeng has a relatively slow switch from transmit to receive mode. As such, the radio was cutting itself off and not hearing it's own packets come back.

For future experiments, I need to modify APRSdroid to operate as a beacon station and regularly send positional reports on a given interval, as well as log all transmitted and sent APRS packets to files so I can post-process them into a report. I also need to write a method of gaining data from the APRS-IS backbone; many APRS stations are receive only and as such I won't see my own packets if I'm heard by such a station, and possibly use the APRS messaging functionality to confirm two way operation from the network at regular intervals.

Despite the difficulties in testing, I'm pleasantly surprised by the performance of the UV-82, at least when it's using an external antenna. For the most part, I was getting results consistent with QRP operations, and was being heard at distances that my non-expert mind believe are consistent with HT/2M operation. When time permits, I'd be interested in repeating this test in the future, as well testing other types of radio. If any other SoylentNews readers are ham radio operators, and have equipment they'd like to see tested, feel free to get in contact with me via email or IRC. If people find this article interesting, I may run additional ones on radio operations. I'm hoping before the year is out to get everything together to be able to do satellite operations, and make two-way contact with the International Space Station, or one of the two OSCAR satellites currently in operation. If I can get my hands on the equipment, I may even be tempted to try for Earth-Moon-Earth communications.

I'll also be upgrading my license class to General in the next few weeks and pending money, I can also try propagation experiments in HF bands.

Until next time, 73 de KD2JRT

posted by cmn32480 on Monday September 25, @02:47PM   Printer-friendly
from the eye-in-the-sky dept.

NASA's Hubble Space Telescope has found a binary asteroid pair that exhibits characteristics of a comet:

Hubble was used to image the asteroid, designated 300163 (2006 VW139), in September 2016 just before the asteroid made its closest approach to the Sun. Hubble's crisp images revealed that it was actually not one, but two asteroids of almost the same mass and size, orbiting each other at a distance of 60 miles.

Asteroid 300163 (2006 VW139) was discovered by Spacewatch in November 2006 and then the possible cometary activity was seen in November 2011 by Pan-STARRS. Both Spacewatch and Pan-STARRS are asteroid survey projects of NASA's Near Earth Object Observations Program. After the Pan-STARRS observations it was also given a comet designation of 288P. This makes the object the first known binary asteroid that is also classified as a main-belt comet.

The more recent Hubble observations revealed ongoing activity in the binary system. "We detected strong indications for the sublimation of water ice due to the increased solar heating — similar to how the tail of a comet is created," explained team leader Jessica Agarwal of the Max Planck Institute for Solar System Research, Germany.

The combined features of the binary asteroid — wide separation, near-equal component size, high eccentricity orbit, and comet-like activity — also make it unique among the few known binary asteroids that have a wide separation. Understanding its origin and evolution may provide new insights into the early days of the solar system. Main-belt comets may help to answer how water came to a bone-dry Earth billions of years ago.

The team estimates that 2006 VW139/288P has existed as a binary system only for about 5,000 years. The most probable formation scenario is a breakup due to fast rotation. After that, the two fragments may have been moved further apart by the effects of ice sublimation, which would give a tiny push to an asteroid in one direction as water molecules are ejected in the other direction.

Also at HubbleSite, New Atlas, ScienceAlert, and Space.com.

(300163) 2006VW139.

A binary main belt comet (preprint)


Original Submission

posted by martyb on Monday September 25, @12:59PM   Printer-friendly
from the addiction-sucks dept.

CVS is finally trying to do something about the opioid epidemic:

Drug-store chain CVS Health announced Thursday that it will limit opioid prescriptions in an effort to combat the epidemic that accounted for 64,000 overdose deaths last year alone.

Amid pressure on pharmacists, doctors, insurers and drug companies to take action, CVS also said it would boost funding for addiction programs, counseling and safe disposal of opioids.

[...] The company's prescription drug management division, CVS Caremark, which provides medications to nearly 90 million people, said it would use its sweeping influence to limit initial opioid prescriptions to seven-day supplies for new patients facing acute ailments.

It will instruct pharmacists to contact doctors when they encounter prescriptions that appear to offer more medication than would be deemed necessary for a patient's recovery. The doctor would be asked to revise it. Pharmacists already reach out to physicians for other reasons, such as when they prescribe medications that aren't covered by a patient's insurance plan.

The plan also involves capping daily dosages and initially requiring patients to get versions of the medications that dispense pain relief for a short period instead of a longer duration.

[...] "The whole effort here is to try to reduce the number of people who are going to end up with some sort of opioid addiction problem," CVS Chief Medical Officer Troyen Brennan said in an interview.

It appears this initiative is limited to initial filling of prescriptions — there is no mention of changes in the handling of refills.


Original Submission

posted by Fnord666 on Monday September 25, @10:38AM   Printer-friendly
from the can-it-make-jello dept.

University of Manchester researchers have created a molecular machine that can assemble four different chemical products:

David A. Leigh of the University of Manchester and coworkers made molecular-machine-based chemical synthesis a reality four years ago when they developed a large molecule that picks up amino acids and assembles them into tripeptides (Science 2013, DOI: 10.1126/science.1229753). Now, they've taken the concept to another level by creating a programmable molecular machine that creates four different products by adding thiol and alkene substituents asymmetrically to an α,β-unsaturated aldehyde substrate (Nature 2017, DOI: 10.1038/nature23677). The machine makes each of the products with stereoselectivity similar to, but in some cases lower than, that of corresponding catalytic reactions in solution, Leigh says.

To generate the products, Leigh and coworkers attach an α,β-unsaturated aldehyde substrate to an "arm" in the molecular machine. An acyl hydrazone located in the center of the machine changes conformation in response to pH changes, causing the arm to rotate between two fixed orientations. Rotations position the substrate above one or another of two silyl prolinol activation sites in the machine that mediate reactions with opposite chirality—R for one site, S for the other. So arm position controls reaction chirality.

The substrate forms a reactive iminium intermediate with an activation site, and the intermediate then reacts with a thiol for the first addition. Then the substrate, with its added sulfide, forms a reactive enamine with an activation site, and the alkene adds to that. If the arm is rotated between the two activation sites mid-synthesis by adding acid, the steps have opposite chirality, and the diastereomer products have (R,S) or (S,R) configuration. If the arm is stationary between steps, both additions occur with the same chirality, and the product has (R,R) or (S,S) configuration.


Original Submission

posted by martyb on Monday September 25, @08:17AM   Printer-friendly
from the a-nose-for-innovation dept.

Submitted via IRC for Fnord666:

A recent deadly outbreak of Salmonella has so far sickened more than 200 people throughout the eastern and southern United States. The culprit? Madrol papayas coming from three different distribution companies, all originating from four close farms in Mexico.

[...] Mitigating risk of food-borne illnesses can be a costly and time-consuming business for food manufacturers -- but one that is necessary. ...

This machine works by picking up on possible pathogens, running it through the system and then coming out with results, which founder and CEO Pierre Salameh says have so far yielded results with a 94 percent accuracy in the lab.

[...] "We provide an affordable method, but I don't want to save money for factories. I want to double and triple output within the same budget of what they are doing," Salameh tells TechCrunch.

Source: https://techcrunch.com/2017/09/19/olfaguard-is-an-electronic-nose-for-smelling-pathogens-in-food-factories/


Original Submission

posted by martyb on Monday September 25, @05:56AM   Printer-friendly
from the not-as-hidden-as-you-may-have-thought dept.

Submitted via IRC for SoyCow5743

[On September 12], Bleeping Computer came across a new Dark Web portal that allows users to search Tor Onion sites in the same way users utilize Shodan to discover Internet-exposed services [1, 2].

Named Ichidan — the Japanese word for "first stage/step" — the service is located at ichidanv34wrx7m7.onion, and, in the long run, can prove a useful tool for anyone investigating Dark Web services.

"This search engine is gold," said Victor Gevers, after Bleeping Computer asked the researcher for an opinion. "There is so much we didn't know about many .onion addresses. I am just amazed at things I see."

[...] Under normal circumstances, Tor sites should not be exposing such wealth of information, as they could reveal the whereabouts and identity of the website's owner.

While this much data is good news when tracking down cyber-criminals and other crooks, this might not be good news if you're running a secret news portal in a country with an oppressive regime.

Source: https://www.bleepingcomputer.com/news/security/ichidan-is-a-shodan-like-search-engine-for-the-dark-web/


Original Submission

posted by martyb on Monday September 25, @03:35AM   Printer-friendly
from the synchronicity dept.

Astronomers have found a probable dwarf planet beyond Neptune using data from the Pan-STARRS Outer Solar System Survey. The trans-Neptunian object's perihelion is 39.32 AU. Although 2010 JO179 is a relatively bright (and red) object, it was not discovered earlier due to its high 32° orbital inclination and 30° ecliptic latitude.

Astronomers think they might have found a new dwarf planet beyond the orbit of Neptune. It's about a quarter of the size of Pluto and may be one of thousands of objects awaiting discovery at the edge of our Solar System.

Tentatively named 2010 JO179, it was found using the Pan-STARRS Outer Solar System Survey by a team led by Matthew Holman from the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts. A paper describing the findings is available on arXiv.

From measuring the brightness of the object, the team believes it's about 600 to 900 kilometers (370 to 560 miles) across. For scale, Pluto is 2,370 kilometers (1,473 miles) wide.

The potential dwarf planet [has a semi-major axis] of 78.3 AU (astronomical units – 1 AU is the Earth-Sun distance), beyond the Kuiper Belt. There's an extremely small margin of error in this estimation. It is known as a trans-Neptunian object (TNO). For comparison, Neptune is 30.5 AU on average and Pluto is 39.5 AU.

For every 21 orbits Neptune makes around the Sun, 2010 JO179 makes 5.

[Wikipedia helpfully informs us that Neptune "orbits the Sun once every 164.8 years at an average distance of 30.1 astronomical units." So, these planets synch up every 3460 years or so. --Ed.]


Original Submission

posted by martyb on Monday September 25, @01:13AM   Printer-friendly
from the keeping-an-eye-on-things dept.

Facebook has tightened the reins on its ad-targeting capabilities following a ProPublica investigation that found that Facebook's algorithmically generated categories allowed advertisements to be targeted to individuals who used phrases associated with anti-Semitism. Facebook denied that an algorithm was to blame, instead blaming manual entries by Facebook users (such as listing your occupation as "Jew Hater" with education from "Hitler's School of Hard Knocks"):

In a [September 20th post], Facebook's chief operating officer Sheryl Sandberg made her first public statement on a recent ProPublica investigation of ad-targeting to hate groups, calling the issue "a fail on our part." Last week, ProPublica's investigation found that Facebook clients could target ads using keywords like "jew hater" and "Hitler did nothing wrong."

Sandberg claims the ad-targeting was the result of manual entries in the education and employer fields. (In simple terms, someone listed their job as "jew hater.") That explanation contradicts the initial ProPublica article, which claimed the categories were algorithmically generated. "We never intended or anticipated this functionality being used this way – and that is on us," Sandberg wrote. "And we did not find it ourselves – and that is also on us."

Sandberg laid out three changes in how the company targets ads, although each is largely an extension of existing efforts. After restricting self-reported fields for education and profession, Facebook will now restore approximately 5,000 of the most popular responses, all of which have now been reviewed to ensure they don't violate company standards. The company will also devote more resources to ensuring that "content that goes against our community standards cannot be used to target ads," and add more human oversight to its advertising system more broadly.

Facebook now has a vetted list of around 5,000 targeting options (such as "nurse" or "teacher") and will manually approve new ones.

Also at Recode, Slate, and Marketing Land.


Original Submission

posted by martyb on Sunday September 24, @10:49PM   Printer-friendly
from the headless-horse[less]man dept.

http://www.bbc.com/news/world-us-canada-41355592

A local US police department has urged people not to call 911 after reports that a man had been crushed by a garage door.

The supposed victim in Greene County, Tennessee, turned out to be a scarily realistic, early Halloween decoration. The department said it had received calls about a "suspicious person lying in a driveway with bloody handprints on the garage". Officers investigated, only to find it was a puppet stuffed with straw.

Greene County TN Sheriff's Department wrote on Facebook: "ATTENTION EVERYONE!!! For those of you driving on Chuckey Pike in Greene County: THIS IS A HALLOWEEN DECORATION! Do NOT call 911 reporting a dead body. Instead, congratulate the homeowner on a great display."

[Okay, Soylentils, what is your best Halloween prank? --Ed.]


Original Submission

posted by martyb on Sunday September 24, @08:28PM   Printer-friendly
from the better-whack-it-back dept.

Seeker reports on the finding of Daniel Rothman, a geophysicist at the MIT. The layman version in Seeker

...Daniel Rothman, [is] a geophysicist at the Massachusetts Institute of Technology, who built a database of fossil records going back half a billion years. Rothman found the periods in which large percentages of existing species died off coincided with big swings in the carbon isotopes found in those records, suggesting the planet's carbon cycle was out of whack.

[...] human civilization has been pumping more carbon into the environment by burning carbon-rich fossil fuels like coal, oil, and natural gas. On the current trajectory, the oceans are expected to absorb at least another 300 billion tons of carbon by 2100 — an amount that could end up producing long-term changes to the environment, Rothman concluded.

[...] Rothman isn't alone in warning of a potential extinction. Some scientists argue a sixth such event is under way already, with about two species a year disappearing and thousands seeing their populations and ranges shrink.

The more arid study is published in Science Advances and the full text is freely available.

The abstract goes like this (with my emphasis):

The history of the Earth system is a story of change. Some changes are gradual and benign, but others, especially those associated with catastrophic mass extinction, are relatively abrupt and destructive. What sets one group apart from the other? Here, I hypothesize that perturbations of Earth's carbon cycle lead to mass extinction if they exceed either a critical rate at long time scales or a critical size at short time scales. By analyzing 31 carbon isotopic events during the past 542 million years, I identify the critical rate with a limit imposed by mass conservation. Identification of the crossover time scale separating fast from slow events then yields the critical size. The modern critical size for the marine carbon cycle is roughly similar to the mass of carbon that human activities will likely have added to the oceans by the year 2100.

I hope our grandchildren will be able to live, even if I have doubts they'll be able to forgive us.


Original Submission

posted by martyb on Sunday September 24, @06:06PM   Printer-friendly
from the Not-MegaVegan dept.

A study (open, DOI: 10.1038/s41598-017-11538-w) (DX) has found that large plant-eating dinosaurs opportunistically dined on crustaceans that sheltered in rotting logs:

Large plant-eating dinosaurs are usually presumed to have been strictly herbivorous, because their derived teeth and jaws were capable of processing fibrous plant foods. This inferred feeding behavior offers a generalized view of dinosaur food habits, but rare direct fossil evidence of diet provides more nuanced insights into feeding behavior. Here we describe fossilized faeces (coprolites) that demonstrate recurring consumption of crustaceans and rotted wood by large Late Cretaceous dinosaurs. These multi-liter coprolites from the Kaiparowits Formation are primarily composed of comminuted conifer wood tissues that were fungally degraded before ingestion. Thick fragments of laminar crustacean cuticle are scattered within the coprolite contents and suggest that the dinosaurian defecators consumed sizeable crustaceans that sheltered in rotting logs. The diet of decayed wood and crustaceans offered a substantial supply of plant polysaccharides, with added dividends of animal protein and calcium. Nevertheless, it is unlikely that the fossilized fecal residues depict year-round feeding habits. It is more reasonable to infer that these coprolites reflected seasonal dietary shifts—possibly related to the dinosaurs' oviparous breeding activities. This surprising fossil evidence challenges conventional notions of herbivorous dinosaur diets and reveals a degree of dietary flexibility that is consistent with that of extant herbivorous birds.

Also at BBC and NPR.


Original Submission

posted by martyb on Sunday September 24, @03:45PM   Printer-friendly
from the just-need-a-thousand-monkeys dept.

[The] main problem here is that software development is not an individual sport. Assessing technical traits means that we are looking at candidates as individuals. At the same time, we will put them in a team context and the project's success will depend on their teamwork. A person's resume or LinkedIn profile says close to nothing about their team skills.

What's more, we know quite a lot about what makes teams effective. Anita Woolley's research on collective intelligence [DOI: 10.1126/science.1193147] [DX] provides extremely valuable insight on the topic. First of all, how do we define collective intelligence? It's basically the skill of a group to solve complex problems. Well, it sounds like the definition of everyday work for software development teams if you ask me.

Why is collective intelligence so important? Exploiting collective intelligence, as opposed to going with the opinion of the smartest person in a room, is a winning strategy. To put in Anita Woolley's words: "Collective intelligence was much more predictive in terms of succeeding in complex tasks than average individual intelligence or maximal individual intelligence."

The power is in the team.


Original Submission

posted by martyb on Sunday September 24, @01:23PM   Printer-friendly
from the is-your-fridge-"clean"? dept.

Walmart wants to test "in-fridge delivery" for Silicon Valley customers with August Home "smart locks":

Here's how the test will work: I place an order on Walmart.com for several items, even groceries. When my order is ready, a Deliv driver will retrieve my items and bring them to my home. If no one answers the doorbell, he or she will have a one-time passcode that I've pre-authorized which will open my home's smart lock. As the homeowner, I'm in control of the experience the entire time – the moment the Deliv driver rings my doorbell, I receive a smartphone notification that the delivery is occurring and, if I choose, I can watch the delivery take place in real-time. The Deliv associate will drop off my packages in my foyer and then carry my groceries to the kitchen, unload them in my fridge and leave. I'm watching the entire process from start to finish from my home security cameras through the August app. As I watch the Deliv associate exit my front door, I even receive confirmation that my door has automatically been locked.

While some may find the idea creepy, others have downplayed the creepiness factor:

"Five years ago consumers wouldn't have assumed they'd let a stranger drive them from the airport, much less stay in their house," said Forrest Collier, the CEO of eMeals, a platform that offers shopping lists based on recipes and loads the items into online shopping carts at Walmart and Kroger (KR) . "Now both Uber and Airbnb are billion-dollar companies."

For now, the fridge restocking service will only be available to Silicon Valley users of August Home. Customers will get a notification through their August Home app every time a delivery person drops off their food.

[...] Even though this Walmart service sounds "creepy on the front end," said Collier of eMeals, "it's really not as creepy as letting a stranger sleep in your bedroom."

Also at LA Times, Reuters, SiliconBeat, and CNET.


Original Submission

posted by martyb on Sunday September 24, @11:02AM   Printer-friendly
from the Programming-Jobs dept.

Commentary from The Guardian

The rationale for this rapid curricular renovation is economic. Teaching kids how to code will help them land good jobs, the argument goes. In an era of flat and falling incomes, programming provides a new path to the middle class – a skill so widely demanded that anyone who acquires it can command a livable, even lucrative, wage.

This narrative pervades policymaking at every level, from school boards to the government. Yet it rests on a fundamentally flawed premise. Contrary to public perception, the economy doesn't actually need that many more programmers. As a result, teaching millions of kids to code won't make them all middle-class. Rather, it will proletarianize the profession by flooding the market and forcing wages down – and that's precisely the point.


Original Submission