Show of 03-19-2016

Tech Talk

March 19, 2016

Email and Forum Questions
  • Email from Phil Nicholson: Hey Jim. This is @eyedoctorphil from Periscope this morning. Love the Periscope idea – I think there is a delay in getting responses seen on Periscope. Actually answered it before you got the answer from the call in but no biggie. Listen to your show every Saturday on the way to work. Dr. Phil
  • Tech Talk Responds: Phil, we will have to work out new logistics for our quiz. Be patient as we figure this out. But thanks for listening and trying out our video experiment.
  • Email for Arnie in Colorado Springs: Hi Dr. Shurtz, Thank you for telling me about Ghostery & it’s flaws in your last Tech Talk Podcast. Interestingly I couldn’t find Ghostery in my Control Panel to delete it. I found it in downloads & deleted it from there. Hopefully the deletion is complete. Went to look at & download Purity or Blockr. But it seems these are for Apple, not Windows. There are five programs when selecting Windows only ad blockers in download.cnet.com. One is PuritySoft Endeavor that I plan to download. Not sure if it’s related to just Purity or not. Do you have any other recommendations for Windows ad blockers? A question for you: how can one copy and send Tech Talk written/Podcast notes legally to family members since they are copyrighted by Stratford University?  Thanks,  Arnie, Colorado Springs, CO
  • Tech Talk Respond: Arnie, actually I gave Ghostery a good review. I just recommended that you don’t enable the feedback system to the advertisers. Other than that is good to go. I did only give you my Apple ad blocker. On my PC,  I used Ad Block Plus, which supports multiple browsers. As far as sharing the show with your family. It is not a problem so long it is not used for commercial purposes. Thanks for listening and for sharing.
  • Email from Mike in Maryland: Hello, Love your show. For the past year, I have been listening to your show on MP3. Usually the following week of your broadcast. I am usually working and it’s not convenient to listen during your radio broadcast. Is there a way that I can watch a video version of your show? Please explain in easy terms.
  • I would like to see how tall “Mr. Big Voice” is. Thanks, Mike from Maryland.
  • Tech Talk Responds: The Periscope video is available for 24 hours at the current time. Our Social Media manager is working on a way to preserve for a longer time.
  • Facebook Comment from Ken Hutchinson: On your profiles in IT, it is Brian Acton, not Brian Action. Ken Hutchinson
  • Tech Talk Responds: Thanks for the wet noodle. I should have checked the Facebook page last week during the show.
  • Email from Jim in Michigan: Dear Doc and Jim. Is there a way to send a text message without a cell phone? Sometimes I need to text my wife and only have my computer because the cell phone coverage in my neighborhood is so bad. Love the show. Jim near Lake Michigan.
  • Tech Talk Responds: You can email the SMS Gateway from your computer. All you need is the email address of the gateway. Below are three common ones. You can get a complete list by search for “SMS Gateways”
  • Email from Nhan in Atlanta: Dear Doc and Jim. Every time I run CCleaner, my bank doesn’t recognize me anymore. It’s been suggested that the cause is that I’ve erased the cookie that my bank site has planted and therefore it doesn’t recognize my PC anymore. How can I identify its cookies so that I can isolate it so that CCleaner will not erase it? Enjoy the podcast in Atlanta. Nhan 
  • Tech Talk Responds: Banks and other systems make heavy use of cookies.  Cleaning out cookies, especially if you’re cleaning out all cookies, does exactly what you’ve described. It will force your bank (and you) to go through extra steps to log on again.
  • Many banks with reasonable security have two stages of verification. On a computer that is unrecognized, such as one you’ve not used to visit the bank website before, the process may involve asking you one or more of your security questions before it allows you to complete the log-in. 
  • Once the log-in is complete and you’ve identified your account, answered a couple of those questions, and specified your password, you may be asked some form of “Is this a safe computer? Should I remember this computer?” If you say yes, the website will place a cookie on the machine that says, “This machine has been recognized, and we don’t have to go through all of those extra hoops next time.”
  • CCleaner, and most cookie-cleaning utilities, give you the opportunity to identify the cookies and to specify exceptions. Run CCleaner and have it run an analyze pass. Then, scroll through the list of all of the cookies that it has found on your machine (it will be long). Look for the cookies that include the domain name of your bank. For example, if your bank is BankofAmerica.com, then maybe what you will find in that list are cookies dropped by BankofAmerica.com. Those are the cookies you want to preserve.
  • One problem is that many banks actually use a third party service. The next time you access your online banking information, take a look at the different domains that appear in the address bar. It may just be your bank; it may be several domains that are owned by your bank; it may be domains that the bank uses from a third party, as I described. All of those domains need to be listed in CCleaner as exceptions, so the program doesn’t clean the cookies associated with those domains.
Profiles in IT: Demis Hassabis
  • Demis Hassabis is a British artificial intelligence researcher who founded DeepMind and created AlphaGo.
  • Demis Hassabis was Born 27 July 1976  in London, England. He is of Greek Cypriot and Singaporean descent.
  • A child prodigy in chess, Hassabis reached master standard at the age of 13 and captained many of the England junior chess teams.
  • Hassabis was educated at Christ’s College, in East Finchley in North London. 
  • After completing his exams at the age of 16 two years early, he was hired by Bullfrog Productions, where he co-designed and programmed Theme Park video game.
  • Theme Park, a celebrated simulation game, sold several million copies and won a Golden Joystick Award, and inspired a whole genre of management sim games.
  • In 1994, Hassabis then left Bullfrog to attend Queens College at the University of Cambridge, where he studied the Computer Science. He graduated in in 1997.
  • After graduation, Hassabis worked as a lead AI programmer on the Lionhead Studios title Black & White
  • In 1998, he founded Elixir Studios, a game developer. Its first game, Republic was greeted with lukewarm reviews. Evil Genius, a Bond simulator, fared much better.
  • In April 2005 the intellectual property and technology rights were sold to various publishers and the studio was closed.
  • In 1999, aged 23, he won the Mind Sports Olympiad – an annual international multi-disciplined competition for games of mental skill. He won it a record five times.
  • Hassabis returned to academia to obtain his PhD in Cognitive Neuroscience from University College London (UCL) in 2009.
  • Hassabis published several influential papers concerning memory and amnesia.
  • After graduation, he was hired as Research Fellow at the Gatsby Computational Neuroscience Unit, University College London.
  • In 2010, he co-founded DeepMind Technologies, a London-based machine learning startup, specializing in building general-purpose learning algorithms.
  • In January 2014 DeepMind was acquired by Google for $625 million, where Hassabis is now Vice President of Engineering leading their general AI projects.
  • In March 2016, AlphaGo beat Lee Sedol, one of the highest ranking Go players in the world, winning 4 games out of a 5 game series.
  • Google uses AI to understand search queries providing context awareness, allowing users to talk to the computer as they would a human, by voice or  by keyboard.
  • Hassabis was elected as a Fellow of the Royal Society of Arts (FRSA) in 2009 for his game design work. He was awarded the Mullard Award by the Royal Society in 2014.
Google’s AI Victory Playing Go
  • For the first time ever, a computer program beat the world champion at Go.
  • Go, a complex checkers-style game, invented by a Chinese emperor to teach his son about political wiles three thousand years ago.
  • Lee Se-dol, a 33-year-old South Korean legend of Go, has lost two of five matches against the AlphaGo program, built by DeepMind, a Google subsidiary.
  • When IBM’s supercomputer DeepBlue beat chess grandmaster Garry Kasparov in 1997, it was a huge milestone in the evolution of computers, but then no one ever heard about DeepBlue again. It had been built with the very narrow and specific purpose of playing chess.
  • But AlphaGo is different. The learning principles that underpin how AlphaGo plays can be used to improve all of Google’s products from search to translation, photos, videos and social networks. 
  • Go is a lot more complex than chess – there are 10 ^761 possible moves, compared to 10^120 possible in chess. So it can’t be solved by brute-force calculations that just compute every possible step in the game, which is what DeepBlue did.
  • Instead, this algorithm needs to learn the rules of Go, and approximate human intuition. 
  • To do this, it uses a special technique called “reinforcement learning”. It learns from its own mistakes by playing against itself millions of times.
  • This is very similar to how the human brain develops.  So AlphaGo learns how to predict moves, and then gauges which player is ahead at any point in time. In other words, the machine can simulate “intuition” by making an informed guess about what moves would be most likely to succeed. 
  • It can learn how to do anything from diagnose life-threatening illnesses earlier than doctors, to predict the impact of climate change. 
  • Machine learning has already begun to sweep across the Google products.
    • Type an ambiguous or complicated search query, and it is interpreted in the same way that a human would approximate a conversation in a noisy bar. 
    • Similar algorithms power Google Translate, which now supports more than 100 languages. 
    • Machine learning is also behind Google Photos, which automatically labels pictures for you so you can search by typing image descriptions.
    • Smart Reply, a function that uses machine learning to write short replies to emails for you. 
  • None are powered by DeepMind technology at the moment, but they will be soon. 
  • Facebook and Microsoft are also trying to play in this sandbox.

Microsoft created a Popular Chatbot in China

  • Microsoft has created a Chinese-language chatbot that’s captured the attention and affection of millions of people who message it regularly on WeChat.
  • She is known as Xiaoice, and millions of young Chinese pick up their smartphones every day to exchange messages with her, drawn to her knowing sense of humor and listening skills. 
  • The program, created to emulate the personality of a 17-year old girl, has more than 40 million registered users, and approximately 25% of them (ten million people) have said “I love you” to her, seemingly without irony. 
  • Using the joint expertise of software developers and psychological experts to create a balance of IQ (intelligence) and EQ (emotional intelligence), Xiaoice differs from standard AI chatbots in significant ways. 
  • She opposes users, offers independent opinions and conclusions, works to demonstrate caring, and is unpredictable, not always offering the same response.
  • Like in the movie “Her,” where the main character falls in love with with a Siri-esque operating system, people are using Xiaoice as a surrogate companion, investing both time and emotional energy into their communication with the robot. 
  • People often turn to her when they have a broken heart, have lost a job, or have been feeling down. They often tell her, “I love you.”
  • Xiaoice’s strength lies in its ability to “remember” details from previous conversations and mimic natural speech patterns, which it learns by constantly scraping real conversations between real people for its database from across the internet.
  • Xiaoice is now in a “self-learning and self-growing loop” because the system is gaining new insights from the billions of conversations it’s already had.
  • As far as privacy fears for users, Microsoft says that it only keeps general data, like what a person’s mood is (for example, if they said they lost a job and are sad), and deletes the rest of the personal data.
  • This is the largest Turing test ever conducted and it looks like Xiaoice has passed.
March 14 is only Pi Day in Base 10
  • Pi is usually written in base 10 notation: 3.14159265358979.
  • This makes Pi Day March 14. But some would like to celebrate on other days
  • Here is Pi is a few other numbering systems
    • Base 2: 11.00100100001111 (January 10)
    • Base 3: 10.01021101222201 (October 1)
    • Base 4: 3.02100333122220 (March 2)
    • Base 5: 3.03232214303343 (March 3)
    • Base 6: 3.05033005141512 (March 5)
    • Base 7: 3.06636514320361 (March 6)
    • Base 8: 3.11037552421026 (March 11)
    • Base 9: 3.12418812407442 (March 12)
    • Base 10: 3.14159265358979 (Marcy 14)
Pi Trivia: How many digits of pi does NASA use?
  • Mathematicians have calculated pi out to more than 13 trillion decimal places, a calculation that took 208 days. 
  • NASA’s Marc Rayman explains that in order to send out probes and slingshot them accurately throughout the solar system, NASA needs to use only 15 decimal places, or 3.141592653589793. How precise are calculations with that number?
  • The most distant spacecraft from Earth is Voyager 1. It is about 12.5 billion miles away. Let’s say we have a circle with a radius of exactly that size (or 25 billion miles in diameter) and we want to calculate the circumference, which is pi times the radius times 2. Using pi rounded to the 15th decimal, as I gave above, that comes out to a little more than 78 billion milesIt turns out that our calculated circumference of the 25 billion mile diameter circle would be wrong by 1.5 inches. 
  • When was humanity’s calculation of pi accurate enough for NASA? In 1424, Persian astronomer and mathematician Jamshid al-Kashi calculated pi to 17 digits
  • More trivia: In 1706 a little-known mathematics teacher named William Jones first used a symbol to represent the platonic concept of pi
First known Mac ransomware reaches the wild
  • Palo Alto Networks claims to have discovered the first known instance of OS X-oriented ransomware in the wild, “KeRanger.” 
  • If you install software infected with the code (in this case, a version of the BitTorrent client Transmission), it’ll encrypt your files after three days and demand that you pay a digital currency ransom to regain control.
  • In practice, you’re likely already safe from KeRanger. Transmission has released a new version of its app that should be safe, and Apple has revoked a security certificate from another developer that KeRanger used to slip past OS X’s native defenses. 
  • Malware can potentially infect any given platform. You can’t assume that your operating system’s inherent security will keep you safe.
Your computer mouse may be opening the door for hackers
  • Malware is being spread over Wi-Fi connections from one computer to another.
  • According to cybersecurity firm Bastille, which reports that wireless devices that do not use a Bluetooth connection are at risk of spreading the MouseJack virus.
  • All a hacker needs to get into your computer is a $15 dongle and few lines of computer code. Hackers are intercepting radio frequencies using the 2.4GHz band. 
  • Unlike Bluetooth, there isn’t an industry standard to protect your devices. So, manufacturers create their own security system, if they have one.
  • This vulnerability affects Windows PCs, Macs and Linux computers. It also affects several types of wireless keyboards and computer mice, including Dell, Lenovo and Logitech. Once the hacker is in, they gain full access to your computer, although they need to be within about 100 meters of you. They can take over your computer and infect it with malware.
  • To find out if your device is at risk, click here for Bastille’s list of affected devices. If your wireless device is at risk, contact the manufacturer, and ask if a firmware patch has been issued, or if one is planned. 
  • If not, switch to a Bluetooth-connected wireless keyboard or mouse. And always make sure you’re using a strong Internet security system.
  • Link to Bastille’s list of affected devices: https://www.bastille.net/affected-devices
Ransomware found in ads on NYT, BBC, AOL
  • A number of high-profile and very high-traffic websites such as The New York Times (NYT), BBC, AOL, MSN, the NFL and more have all been delivery ransomware,  according to Malwarebytes.
  • Malicious ads making their way onto websites via ad networks used to be a far more common occurrence but most ad networks from companies like Google have gotten far better at blocking them.
  • This month (March 2016), we witnessed a huge spike in malicious activity emanating out of two suspicious domains, Malwarebytes said.
  • The ads weren’t restricted to coming via Google’s ad network. AOL, AppNexus and Rubicon’s networks were also named in the report.
  • Once a user has been infected, they’re then redirected to a page hosting the malware exploit kits, which ultimately lead to your files being held ransom in exchange for a payment – often requested via bitcoin due to its relative anonymity. In this attack, Malwarebytes says that the Angler exploit kit is used in many instances.
  • The news won’t help publishers win any favors in the ad blocking debate,.
The Way You Use Your iPhone Says Much About You
  • Are you the type to respond to every single text ASAP? Or are you capable of putting your smartphone down for hours and forget about its flashing blue light? 
  • A new study suggests that the more people check their devices, the more impulsive they are in their everyday lives.
  • Two researchers from Temple University had 91 undergraduates fill out a questionnaire assessing how often they used their phones to update social media, browse the Internet, or interact with friends. 
  • Then they tested students’ ability to delay gratification (aka wait) by asking them whether they’d prefer a small sum of money right now or a large sum anywhere from a few days to a year from the immediate moment
  • The researchers also assessed students’ sensitivity to rewards by having them rate how greatly they identified with statements like, “I’ll try anything once,” and, “I like wild and uninhibited parties.” 
  • Finally, they ranked students’ impulsivity by placing them at a computer and asking them to press a button whenever an “x” popped up on the screen but resist pressing a button whenever they saw a “k.” (The more participants hit buttons when they weren’t supposed to, the less impulse control the researchers concluded they had.)
  • Impatient undergrads — the ones least interested in waiting more than a day to get a hypothetical amount of money — were more likely to be preoccupied with their smartphones on a regular basis. 
  • Those who had a harder time controlling impulses to press buttons were also more tethered to their devices. Surprisingly, students’ sensitivity to rewards didn’t appear to be an influence on their phone-checking habits.
  • The more compulsively you check your smartphone, the more impulsive and impatient you probably are.
  • If you’re the type to never let a retweet remain un-liked or a friend request go unnoticed, take a peek at some other areas of your life. 
  • If you’re having trouble reining in urges — to, say, eat the whole pint of ice cream, resist ordering one more round, or flip out at a coworker or friend — it may be time to take a break from your smartphone and start practicing some self-regulation.
The Open Data Button Takes on Academic Paywalls
  • Academic publishers accumulated $25.2 billion in profit in 2015. Unfortunately, their paywalls are an impediment to research.
  • Students David Carroll and Joe McArthur, along with a handful of colleagues, recently released the Open Data Button to solve this program.
  • This new button helps users work around paywalls in order to gain access to research data that isn’t readily available online. 
  • If all goes as planned, with the click of the button, researchers will be able to reveal the lack of access to a particular data set not only to their peers, but to the entire open access community.
  • Here is how it works.
    • Go to: https://opendatabutton.org/
    • Download the Open Data Button for your browser from 
    • Next time you’re reading a research paper and you want to investigate the data behind it, push the Open Data Button.
    • The Open Data Button will try to find you the data you need. If that doesn’t work, it’ll start a request to the author asking them to share their data. Authors will be able to share their data with you and the world.
    • Your story will be used to encourage researchers to make their data available, making progress to more transparent and reproducible research.
Good News for the Earth: Plastic Eating Bacteria
  • There’s an estimated 311 million tons of plastic produced every year globally. 
  • There will be more plastic than fish in the sea by 2050.
  • Polyethylene terephthalate (known as PET), which is used for manufacturing plastic bottles, makes up one fifth of the annual production of plastic. 
  • PET’s strength and stability under heat makes it an ideal material for packaging—it’s one of the world’s most-used plastics for a reason—but it’s properties also makes PET difficult to decompose.
  • Ideonella sakaiensis, a newly discovered bacterium, that feeds on PET. 
  • Researchers in Japan isolated the bacteria from the 250 samples of PET debris they collected from a recycling plant. T
  • he bacteria appeared to use a pair of enzymes to slowly break down PET.
  • “If you put a bacteria in a situation where they’ve only got one food source to consume, over time they will adapt to do that,” lead researcher Enzo Palombo, from Swinburne University, told The Guardian.
  • Researchers note in the study, that it takes the bacteria six weeks to degrade a small piece of low-quality PET. The bacteria took even longer to degrade highly crystallized PET. 
  • Researchers warn that the bacteria would—for now— have limited impact if it were to be applied to large amount of plastic waste sitting in landfills around the world.
  • For now, people should make more of an effort to recycle PET—particularly PET bottles.