Welcome Your IET account
Smartphone used at night
Comment

Polish privacy, dumb smart gadgets, reading on Twitter and more: best of the week’s news

Image credit: Savenkomasha | Dreamstime.com

E&T staff pick the news from the past week that caught their eye and reflect on what these latest developments in engineering and technology mean to them. For the full story, just click on the headline.

 

Vitali Vitaliev, features editor

Poland switches to privacy-conscious contact tracing

I’m not at all surprised that the Polish government, no doubt reflecting the interests of the majority of the country’s population, has opted for a privacy-conscious Covid-19 tracing application. As the poet said, only he who has been exposed to the din of a battle can fully appreciate quiet and silence. Only those who know what lack of privacy means - and all the post-communist nations of Europe, alas, possess that gruesome knowledge in excess - are capable of valuing it properly.

For those who were born and have spent all their lives in (relatively) free democratic societies, it is impossible to imagine the lengths totalitarian regimes would travel to try and control not just the words and movements of their oppressed subjects, but also their thoughts. I, for myself, was dead sure well into my teens that “walls have ears,” as my parents had told me repeatedly. Yet no matter how hard I searched the walls of my tiny bedroom-cum-lounge, no matter how thoroughly I groped them, there were no signs of any ears growing through the wallpaper, not even in those multiple spots where it had become unstuck from the wall to reveal bald patches of peeling stucco.

It would have been nice, I thought, to find somewhere on the wall a pair of pink piglet’s ears. For some reason, I was sure that the wall’s ears, if any, should resemble those of a pig; stirring and rotating like two tiny radar dishes, as if conversing with each other.

One of the most striking examples of how we were blatantly robbed of privacy in the USSR of my youth happened in in 1986, after I came back from my first ever trip abroad – to the ‘brotherly’ Czechoslovakia. Shortly after my return, I bumped into a former university mate in the street, a young lady called Valentina. We all knew that – just like a handful of other students – she was recruited by the local KGB department after graduation to work at its ‘letters readers’ section – a unit entrusted with secretly perlustrating Soviet citizens’ private correspondence, particularly missives to and from other countries.

“Why were you so lazy during your recent trip to Czechoslovakia?” Valentina asked me with a smile after a usual exchange of pleasantries and explained: “You only wrote to your wife once and even that letter was short and hardly readable, as if scribbled in haste”.

I kept staring at her silently, not knowing what to say. She probably realised that she had revealed something she was not supposed to (she had always been a bit of a chatterbox) and, having quickly said her goodbyes, was back on her way, whereas I kept standing there motionless for a while, shocked and mortified by what I’d just heard, for it is one thing to be aware of the proverbial all-hearing ‘ears in the wall’, and a totally different matter when you actually find them!

Tim Fryer, technology editor

Are smart appliances doomed to early obsolescence?

It’s an interesting dilemma this. Those of us who are interested in technology are fond of gadgets, but there are plenty of us who are also quite at ease with turning a light on and off at the switch rather than doing the same thing from our smartphones. I’m one of those people who often thinks that the prehistoric ways of doing things manually are often the easiest solution. How many times have I actually wanted the ability to remotely turn on my heating whilst not in the house for example? Answer – never. That might change if I actually had the ability to do so, but I doubt it would significantly.

Then there is the cost of the new appliances: real premium rates for something that I always have the suspicion will become obsolete not because of lack of manufacturer support, as mentioned in this article, but because the next great thing will come along and supercede it. Rather than having an app to keep an eye on your fridge or heating, there will be more convergence in the capabilities of the apps, plus a confusing divergence of the apps available.

Finally, there’s the 'all eggs in one basket' problem. There are about 58 million smartphone users in the UK and about half a million phones are nicked every year. Probably at least as many are lost or broken. It worries me to have so much reliance on a device that I could be deprived of in an instant.

All in all, I’m favour of smart things where the smart makes a difference – what you can view on a TV these days would bewilder someone fresh from the days of three terrestrial channels, for example - but the battle to conquer the tech is worth it for the viewing feast it offers. However, I’m not sure if it’s worth the risk of paying double for a fridge so that you can be told your milk is out of date (I would probably still drink it if it smelt OK, anyway!), knowing the advanced features of that fridge might be but a passing gift?

Dominic Lenton, managing editor

Twitter nudges users to read before they retweet

I’d like to think that most social media users only pass links on others in their network when they’ve actually bothered to read the story or article they’re implicitly recommending as being worthy of attention. The state of many posts suggests, sadly, that not only is this not the case, but that few people even bother to read back the words that they’ve written themselves before hitting the send button. Covfefe, anyone?

Anything which makes the process of communicating even a little more arduous is anathema to interface designers. The tools that have proved most popular and attracted millions of users – even if it’s only until the next big platform comes along and they slowly migrate – are the ones that are simplest to use and require the least effort. Much of the appeal of Twitter, I’d suggest, is that you can flag a useful bit of content up with a huge number of people simply by liking it.

If implemented more widely than the initial test group in the US, I expect this tool will initially meet with moans from the hardcore Twitter users who are responsible for hundreds of notifications every day. Eventually, they’ll get used to making that one more tap to confirm ‘No, I don’t want to read this article right now, but I want to recommend it before I move onto the next thing and forget’ (I assume the actual warning will be more diplomatic than that).

Will it make any dent in the spread of misinformation? Probably not significantly in the short term. What it might achieve, though, is to gradually educate users about the importance of stopping to think for a second about what they’re doing on social media. Anything that has an impact in that respect can’t be a waste of anyone’s time.

Ben Heubl, associate editor

Smart parking software could help ease congestion

It was astonishing how many news organisations covered this research paper, in which Harvard academics tried using parking analysis to verify that China knew of Covid-19 before the outbreak was confirmed, without factchecking work that hadn’t already been peer reviewed.

The researchers interpreted “an upward trend in hospital traffic in late Summer and early Fall 2019,” together with a surge in online search traffic, as evidence that “the virus may have already circulating at the time of the outbreak”.

This work undoubtedly has some shortcomings. It needs urgent and careful peer-reviewal to confirm the conclusion. I’m not here to make the call. The fact that so many news organisations jumped so enthusiastically at the opportunity to report on it shows we have to be more cautious in our effort to decide which science stories we want to tell. Taking non peer-reviewed findings at face value can be dangerous.

If you want to check yourself, I encourage you to do so. Find the location on Google Earth (30°33'22.2"N 114°20'51.9"E) and check one of the Wuhan Hospitals and its parking lot for late 2019 and compare it with June 2019 and images for 2018. To me, it seems just as busy. Unless and until the research is fully made public with all its evidence, how can we confirm the results?

Then there’s the search traffic data, also used as evidence by the researchers. One commentator on Twitter, a PhD statistics student, pointed out that the researchers used “[the] wrong Chinese words to search. Their findings in hospital traffic doesn’t have statistical significance.” My Chinese is rusty and this would all need to be carefully re-checked. The point is, these questions would usually come up in a peer review process.

I’m not here to fact-check this piece of research, which has also prompted concerns about the cherrypicking of data. All I’m saying is that the media has to be extremely careful in reporting on non-peer-reviewed studies that can have implications for important international political relations. Is it worth taking the risk without a proper peer review in place? I remain sceptical.

Jonathan Wilson

Twitter nudges users to read before they retweet

This is one of the primary problems with social media: people devote so little time to actually reading the posts they 'Like' that misinformation can quickly snowball and gain serious traction within the community. A misguided gut reaction to an oblique headline; a glib interpretation of the salient facts absorbed from a cursory scan of the one-line summary of an article; a genuine misunderstanding; or - increasingly the case - not even reading the material at all and just auto-forwarding in a blinkered passion any stories appearing in the timeline.

It has been proven by repeated studies that many people simply don't read the stuff they appear to have an interest in online. A shallow 'Like' or 'heart emoji' click is enough for some people to signify to the world that they've engaged with a topic, even if their understanding of such a complex issue as, say, endemic racism, or Brexit, or coronavirus is woefully inadequate and ill-informed. Social media platforms have a deadly serious role to play in preventing, and not encouraging, this dumbed-down approach to everything.

Facebook's attitude - well, Mark Zuckerberg's attitude - that people should be able to shoot their mouths off in public online and say anything they like, because this supposedly best espouses the principles of free speech, is disingenuous in the extreme. Walk into any pub (remember being able to walk into any pub?) or stand up on the train and start shouting racist abuse, or about how Brexit was a simply fantastic idea and cannot fail to make the UK great again, and see where it gets you. Hospital, possibly, or a police station. The online world shouldn't be considered any different to the real, offline world - especially as the former now regrettably spills into the latter with almost nothing but disastrous consequences.

Do you really 'Like' everything you see and forward from your timeline? Are you confident that you fully understand the issue at hand and could participate intelligently in a formal and courteous debate? Taking the time to educate oneself about the issues of the day; reading more widely about a topic; keeping compassion and empathy for other people in mind, and remembering always that there are at least two sides to every story. If more people would follow these simple principles, online discussion might not always have to devolve to such extreme, divisive positions.

I realise that this all might be too little, too late and there's nothing that the world's blowhards, bigots and bots love more than anonymously ranting at and trolling the more balanced individuals, but perhaps the rest of us could at least try to be more human in our online conversations. After all, almost none of us are actually worse than Hitler.

Ancient Roman city revealed in great detail using ground-penetrating radar

While there's still no comparison to holding an ancient physical object in your hands or carefully 'trowling back' (one of the core onsite skills of the archaeologist) dusty layer by dusty layer to reveal the stonework of a structure buried for millennia, it's clear that technology can increasingly play a major part in revealing the hidden worlds beneath Earth's 21st-century topsoil. Just last month, archaeologists working remotely due to lockdown used their WFH time productively to analyse data from lidar scans of England's South West region, identifying new medieval and Roman sites in the process. No digging necessary - not even any need to don sturdy walking boots and waterproof clothing.

In this latest story from the ancient-and-modern world, a Roman city in Italy has been mapped out in detail using ground-penetrating radar - again, without a single trench being dug on site. As these technologies continue to improve - and new concepts are invented and perfected - I'm sure we'll see many more fascinating subterranean revelations in the years to come.

Dickon Ross

Will smart guns hit their target this time?

Smart guns are on of those technologies that's been a long time coming - or rather, the technology is there, just not the market, yet. Why wouldn't you want a gun that only you can fire? Isn't that safer than a gun that a baddie can wrestle or snatch from and turn on the owner? Indeed, that fear is one of the factors that keeps guns restricted in the UK. In this feature, Ben Heubl looks at the arguments for smart guns, hears the arguments against and asks why they have been so slow to catch on. In America they have become to be seen as a bit of joke, one movie plot involving the good guys hacking into the smart guns of the bad guys to turn them on themselves. The gun lobby there doesn't like them - although I don't quite understand why not. They're still guns, after all.

Eventually, smart guns may join the long list of innovations that the public fought tooth and nail against as curtailments of freedom, but years later we are mystified that they were ever that controversial. Remember the fight against compulsory seatbelts, which have since save many thousands of lives? In the days before the law was passed, passengers regularly went through windscreens as the result of an impact. Remember when drink-driving was the subject of jolly pop songs? "Have a drink, have a drive, go out there and see what you can find," sang Mungo Jerry in 1970. Now, such behaviour is a social taboo.

We're now hearing strong safety arguments for driverless cars, but as one developer put it at a conference on the subject last year: if you're really concerned about saving lives, you save some now by making it impossible to start your car without the seatbelt on - if your car can issue you a warning as they do now, it can stop you from starting the engine. No one is allowed to drive over 70mph in the UK except emergency vehicles. Why not put limiters in cars? Transport for London says 16 people have lost their lives to traffic accidents in London over the lockdown - lower than usual - but most were due to speeding. Boy racers have been clocked driving at speeds of up to 130mph on the empty roads of lockdown London. That could all be stopped. We have the technology, we just need the will.

Hilary Lamb

Amazon slaps one-year moratorium on police use of Rekognition

IBM nixes ‘general purpose’ facial recognition business

Amazon has paused police use of facial recognition for a year, while IBM has jettisoned its “general purpose” facial recognition business while lawmakers put in place ethical restrictions on its use. These moves are a welcome indication that police facial recognition use seems to have become politically toxic.

Where you stand on the civil liberties debate regarding police surveillance is almost irrelevant. Facial recognition in its current form just does not work, performing exceptionally badly when analysing the faces of women and people with dark skin (up to 30 per cent error rates for dark-skinned women) due to underrepresentation in training data sets. You could call this RIRO (racism in, racism out). For example, when the ACLU compared headshots of every member of Congress against a database of mugshots, Amazon’s Rekognition found false matches of at least 80 per cent confidence between 28 lawmakers and the criminals, with ethnic minority lawmakers massively overrepresented in this group. This is indefensible.

Before we celebrate this recent news too enthusiastically, it is worth noting that:

  1. Use of facial recognition in public is likely to be impeded somewhat over the next year or so anyway due to mask wearing.
  2. IBM’s facial recognition business was reportedly struggling with the axe hanging over it for several months anyway.
  3. Amazon and IBM have not shut down the possibility that they could continue to assist with troubling surveillance in other forms, such as ICE use of Rekognition or unspecified use of IBM data analytics tools for law enforcement.
  4. It is likely that we will see companies like Amazon lobbying for watered-down regulation over the next few months.

Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.

Recent articles

Info Message

We use cookies to give you the best online experience. Please let us know if you agree to all of these cookies.


Learn more about IET cookies and how to control them