Study Opportunities' Blog
Seems like that old nursery chant: 'sticks and stones may break my bones, but words can never harm me' is no longer quite as true as we'd like to believe. Last week a judge in Massachusetts found a 17 year old young woman guilty of manslaughter for using text messages and a phone call to encourage and convince her 18 year old boyfriend to commit suicide.
|“This is saying that what she did is killing him, that her words literally killed him, that the murder weapon here was her words,” |
- Matthew Segal, lawyer with the American Civil Liberties Union of Massachusetts
Our learners live in a world where constant communication and the pressures of social media are pervasive. They often post private messages to each other and public messages on social media without giving a moments thought to the impact, implication, consequences or possible collateral damage these messages might cause. As educators, we need to take the time and effort to repeatedly draw to their attention cases that vividly illustrate this - cases that redefine the law and our society. This is one such case. The New York Times has a good article on the topic, but if you have already used up your free access for the month, here's a link at CNN. If nothing else, this could open the door to Manslaughter / Murder becoming, of all things, a cybercrime.
Human error - again
Computers are stupid. they only do what you tell them to do, so you better make sure that you tell them to do the right thing. One young person, first day on the job, just out of University, unguided and working according to a document telling him how to set up his own test development database did what many of our learners do: he copied the code in the document and executed it unmodified. The result: the entire companies database deleted. Gone. Unrecoverable. The CTO promptly fired him and he posted about his experience on Reddit. Many readers have come out in support of him and believe that the CTO should have been the one to lose his job.
A complete nightmare situation. A business almost down the tubes because of human error. I say almost, because surely there were backups and recovery, though tedious and inconvenient, would be possible? Wrong! Seems like there was a problem with the databases and backups were not being restored. A company that provides a document containing a potentially destructive code snippet to a complete noob for unsupervised use is not likely to make sure they follow the best backup procedure in the world. More human error, compounding the original human error...
Quartz Media has a nice article based on the incident.
Robot, Robots, Robots
Harvard students seem to want to prove that robots can be made from anything - they are developing a spider like robot made from drinking straws and powered by air.
And now, a robot that crawls up your butt - to make an unpleasant medical procedure weirder but less unpleasant. Curious? Check it out at Boing Boing.
Finally, the BBC has a short segment from Ashitey Trebi-Ollennu, chief engineer at Nasa’s Jet Propulsion Laboratory, on 5 robots set to change the world.
Motherboard has an article on a click fraud farm busted in Thailand. Fascinating visuals and a glimpse into the world where fake likes, ratings, etc are manufactured on demand.
In Russia though, this type of click fraud is out in the open. Here's an article about a vending machine that sells Instagram likes and followers!
Whilst Google seems intent on creating AI's that defeat humans at complex games demanding strategy and insight, Facebook has built itself an AI that has learnt to lie to get what it wants. Appropriate for a system where most people create fake representations of a perfect life? A fascinating read.
AI, self driving cars and Insurance
Some interesting questions raised in this article from Readwrite.com.
Scary new malware infection technique
Digital trends has an article on how hovering your mouse over a link in a PowerPoint slide can automatically download and install malware (no clicking required).
Ransomware Ponzi scheme
A Ponzi scheme is a pyramid scheme. Popcorn Time is a new type of malware tries to maximise its profits by using the strategy behind a pyramid scheme - when you get infected and your data is encrypted and held ransom, you are given a choice: either pay up or deliberately infect at least 2 others to 'free' your data. What would you do? The New York Times has the details, also at Fortune.com.
That's it for this week - good luck with the last of exams and the reports....
Well, it does if you live in Ethiopia, Algeria, Iraq or Uzbekistan. We know that it is best practice to make sure IT and CAT learners don't have internet access whilst writing their prac exams. In fact, many of us even restrict their general network access to prevent messaging, e-mail and file sharing being used for cheating. After all, leaving the access provides the temptation... BUT, to disconnect a whole country from the internet for the duration of national exams in all subjects seems a little extreme. There's a bit more behind the story than paranoia and caution though - see The Guardian and Newsweek for details.
Ethiopia cut internet access for a week or two at the end of May this year. Iraq cut internet access for periods of hours (at least not days) last year, and again this year (The Verge). Algeria did it last year (Reuters). Uzbekistan did it in 2014 (Ed Tech Times). In all these countries the internet lockdown was presaged by widespread leaks of papers on social media and general cheating and corruption. This is a fascinating topic for a class discussion. Is the action justified? How important is the credibility of a national exam? Why do we have national exams? What is the impact of an internet shutdown on society? Who should be able to make this kind of decision? In all cases the government seems to have acted unilaterally, without warning and often without comment or indication of how long the shutdown would last. Is this acceptable? This whole topic is pure classroom gold!
Human error - always the biggest problem with IT
I held off from talking about the immense British Airways shutdown that stranded thousands of passengers and cancelled hundreds of flights - because of a computer problem. I was curious to see what the cause of the problem would be - early reports made no mention of hacking or malware - so that left bugs or human error as the cause. I was betting on human error, because.... well, because that's what it usually is. Turns out that the real cause is pretty much the same as in this video that Apple opened its 2017 WWD Conference with (to show the importance of apps, of course)...
Your FACE is being watched - in the most unexpected places and ways
This article caught my eye: Pizzeria billboard in Oslo analyzes people! And I got curious. And did some searching and reading. Turns out there's a lot of hidden commercial face-recognition and customer recording going on out there. In Beijing, authorities are using face recognition to limit toilet paper theft in a public park! Last year Russia saw the release of the FindFace Android app that can track down and identify people from a photo (using the Russian social network Vkontakte). Facebook uses similarly powerful technology not only to tag your photos but to learn and improve its accuracy when you tag a photo. Luckily it has not made it possible for users to use this technology to search for and identify people. This didn't stop the FaceZam app hoax panicking people in March this year. Google has also had its share of the facial recognition creepiness - does anyone remember the thankfully short-lived NameTag app for Google Glass way back in 2014? Wearers of the AR glasses didn't even have to obviously take a photo using their phone to be able to scan your face and identify you!
The article that best summarises what's going on and why is at Readwrite.com and is a good value, in depth read.
|“Attention is the new currency” is the headline here. Your attention (and the meta-data associated with it) is being relayed to advertisers without your permission or awareness, and there is no way to opt–out.|
From an article in Medium.com by Youssef Sarhan.
Moore's Law still survives
It is slowing down, but despite predictions of its death throughout my tech and teaching caree, Moore's Law is still hanging in there. The latest Intel technology (Kaby Lake) sees transistors shrink down to as small as 14 nanometers. IBM has developed new technology which use 'silicon nanosheets' to shrink chip technology down to an incredible 5 nanometers! That means the same chip area can hold double the amount of transistors. Wired has the details.
Did you know there is an Overclocking World Cup? This event has been held at Computex for the last four years and involves the arcane arts at getting a standard CPU to run at ungodly speeds (e.g. 7.5 Gigahertz) long enough to be deemed stable and successful. It typically involves the use of extreme cooling using liquid nitrogen. Engadget writer Richard Lai walks you through the process of learning this extreme (and pointless beyond demonstrating that it 'can be done') art.
SA Internet - where we're at and where we're going.
Short and sweet: ADSL and fixed lines are out, Fibre and Mobile are in. For information, Telkom recently announced two interesting pieces of information:
That's it for this week - never fear, the holidays are near!...
Exams are upon you, so you don't have too much time to deal with in-depth news and articles right now. So this week I'll keep it short and sweet.
That's right: Intel just just announced a new i9 series of core processors that maxes out at 18 cores (36 threads) on a single CPU - trumping AMD's earlier announcement of a Ryzen processor with 16 cores. That's if you are willing to pay $1 999 (over R30 000) just for the CPU in your system! As always though, this development is an indication of how CPUs are developing and where they are going. That, and the fact that everybody (including Apple) seems to be trying to catch up with Google and their Tensor chip by creating silicon designed specifically for AI and machine learning.
This is not something that is on the market, more a proof of concept used in a cool way. Digital Trends has an article on a guy delivering the match ball to a soccer cup final in Portugal.
What's jaywalking? Crossing the street anywhere that is not an official pedestrian crossing - or crossing against the light. Paris has a problem with jaywalking - statistics say that over 4000 people a year are hit by cars whilst jaywalking. So, obviously, authorities would rather people did not jaywalk and have come up with an innovative technical way to discourage people from jaywalking. Digital Trends has the details.
An AI chatbot that helps you get a job
CNN Money has and article on a startup trying to develop a chatbot that deals with recruitment and preliminary job interview questions.
VR Training midwives
Virtual reality has been 'on the cusp' of mainstream use since the early 1990's. I remember the first huge clunky VR sets and the blocky graphics that they displayed. Today's VR is much more realistic and immersive but still faces the problems of lag (even milliseconds of lag can cause headaches and sea-sickness) and expensive hardware. That said, I think VR is closer to mainstream use than ever before and is seeing some innovative use cases. An Australian University is using VR in its final exams to test how good a trainee midwife is at delivering a baby. CNet has the details.
Facial Recognition in education
A French business school will use facial recognition software to check how much attention its students are paying in class. Useful? Invasive? Creepy? You decide. read the article at The Verge. This author strongly agrees with the sentiments expressed in the second last paragraph.
I promised this week would be short and sweet - so that's where we stop! Good luck with exams, invigilation and marking!
086 293 2702
012 565 6469
Copyright Study Opportunities 2016 - 2021. All rights reserved.