Editor's pick: Mass fudging of data begins early
I have read several articles recently describing how the scientific profession has become discredited by scientists changing and manipulating their data (see for example 16 April, p 5 and p 39). This practice is even more deep-rooted than you might think.
For my GCSE biology course I must “investigate what factors affect simple animal behaviour”. We chose to put woodlice in a choice chamber and let them run around, to study the effect of different light intensities.
There are 60 students in my year, and we all came to the resounding conclusion that woodlice are completely psychotic. So did we write a thoughtful evaluation? Did we repeat the test? Did we seek secondary data to explain our findings?
Of course not. How could I expand “woodlice are psychotic” into a 5000-word submission with graphs and tables in 4 hours? We all fudged our data.
We created new data sets from scratch. One student created two data sets, and published one on a blog so he could reference it.
We need better education to stamp out such behaviour and to teach school students that this can't be done. What was worse was that it was considered the norm. Got bad data? Make up some more!
More walking does not reduce traffic
Will increased walking or cycling bring down air pollution (14 May, p 6)? Probably not. The immediate effect would be less congestion and cleaner air. So far, fine.
But congestion is a deterrent to driving. When it is reduced, unless there is a new legal or physical impediment, such as road closure or narrowing, or monetary fees, new drivers will appear from the large reservoir of potential drivers. These will enjoy, briefly, pleasant motorised travel. Soon pollution and congestion will again be in balance.
In the 1980s, tolls on some New York river crossings were lowered for “high-occupancy vehicles” – those with at least one passenger besides the driver. A co-worker responded by picking up a passenger or two at a bus stop before the Holland Tunnel. So transit operators lost revenue, along with the bridge and tunnel agency, and pollution from his large Lincoln car remained as it was. Dreamers and reformers should think beyond step one.
First class post
Let's get rid of the profit aspect of energy. It's what keeps renewables from being mass-produced
Jorge Rodriguez arguments about falling renewable electricity prices (21 May, p 19)
Driverless cars and the tragedy of jam
You rightly say that we should be thinking about the implications of driverless cars now (14 May, p 5). I suggest that the most common number of occupants of a driverless car on the road will probably be zero.
A very important aspect of a highly automated and widely distributed technology is that it doesn't matter to users how slowly it works, within reason. It takes 15 minutes for you or me to wash a pile of dishes. It takes a dishwasher an hour, but we don't mind the extra three-quarters of an hour because it isn't our time.
Similarly a sprawling traffic jam of stationary unoccupied driverless electric cars on their way to collect the week's groceries is costing their owners nothing and emitting no pollution.
Driverless cars will create a vast tragedy of the commons, these commons being road space. The only solution I see is ubiquitous road pricing, which – with the encrypted secure logs that the insurance companies will insist that the cars have – should be easy to implement.
I am not sitting comfortably in that
Are you sitting comfortably, asks the caption on an image of a driverless car (14 May, p 22). Not in those awful seats. No lower back support, horribly concave. It might as well have been designed to make back pain worse.
Axiom of evil would be a big deal
Jacob Aron says that if one of the Turing machines simulated by Adam Yedidia and Scott Aaronson halts, it would prove the set of axioms of mathematics called ZFC to be inconsistent, but mathematicians wouldn't be too panicked because “they could simply shift to a slightly stronger set of axioms” (14 May, p 9). It Aron here means “stronger” in the everyday sense of “tough”, not the sense I use at work as a mathematician – “able to prove more things”.
An inconsistency in ZFC would be a big deal for me. It would undermine virtually all work in modern set theory since before Kurt Gödel's 1931 second incompleteness theorem. This proves that a set of axioms sufficient to define arithmetic cannot prove its own consistency: you can add more axioms to prove the consistency of your first set, but you will not be able to prove the consistency of this bigger set without adding more axioms again – and so it goes on.
Empathy need not be painful
Empathy may not require feeling others' pain (14 May, p 32). It could be defined as imaginatively entering into another's experience. That requires only understanding.
This sort of empathy would provide the means for those with different lives and experience to better empathise – and moreover offers a protective model for those professionals in danger of burnout. It would be an important construct for politicians to adopt before developing policy that excludes some groups from access to fully sharing in society.
Empathy need not be painful
You describe a growing worldwide crisis in empathy (14 May, p 5). I draw your attention to papers such as “From Painkiller to Empathy Killer: Acetaminophen (Paracetamol) reduces empathy for pain”, also published in May (Social Cognitive & Affective Neuroscience, ).
Protecting the internet of things
So John Matherley's Shodan web-crawling engine lists everything connected to the internet (14 May, p 40). That's a creepy crawler if ever I heard of one. For me it highlights the weakest link in the thingternet (internet of things): the Wi-Fi router/receiver. This is the first thing that such a crawler will find in your home. Wi-Fi routers do offer security settings, but I doubt many people bother or know how to use them. Shouldn't firms that manufacture this first point of contact for hackers be obliged to ensure that they continually check security, blocking and warning by default?
Is your data safe in Scotland?
In his letter describing NHS Scotland's approach to privacy (9 April), Harry Burns mentions its reliance on Guiding Principles for Data Linkage, published by the Scottish government (). I was concerned to find the document stresses that it does not set out rules.
It sanctions the dissemination of personal data without consent or notification. It makes no strong requirement for recipients of data sets to secure the data or to be independently audited. It states that recipients should not try to de-anonymise data, but does not require policing of this.
Again and again it mentions privacy concerns but states that these should be addressed “where practicable”.
For want of a 'standard'…
Lisa Grossman attributes the demise of the Mars Climate Orbiter to “failure to convert between standard and metric units” (7 May, p 19). Metric units have of course been “standard” for decades in advanced technological countries.
The US retaining the irrational measurement system of its former colonial rulers is one of its more bizarre features. A mix-up over what is standard may be why the spacecraft was lost.
Asynchronous rhythms of life
Catherine de Lange describes work on biological clocks (16 April, p 30). But some of the findings she reports have long been known, as shown in Gay Gaer Luce's well-researched book Body Time (reviewed in 快猫短视频, ). It is amazing how such ideas can come back as mainstream 40 years later.
Please remember early Australians
As so often when discussing ancient humans, your writers ignore evidence from Australia – and the article on graves was no exception (14 May, p 36).
At Lake Mungo in New South Wales a male was carefully buried lying on his side, slightly flexed, with his hands placed neatly over his groin. He was covered with so much red ochre the sands are still stained with it more than 40 years after the excavation.
This burial has been dated to at least 42,000 years ago.
The nearest source of red ochre is over 200 kilometres away, so considerable effort and resources were put into this burial and it was obviously important. This is by far the oldest formal, ritual burial of a modern human.
Nearby were found the remains of Mungo woman, older than 20,000 years, burnt and then broken up. This represents the oldest evidence of cremation in the world.
Victorians did spot melting glaciers
As I was reading a new acquisition, Ice by Mariana Gosnell, I heard 快猫短视频 drop onto the doormat – with your report that Victorians experienced early climate change but missed the signs (30 April, p 15).
Gosnell mentions that in 1871 physicist John Tyndall gave a lecture to young people about mountain glaciers and reported that “for the last fifteen or sixteen years the glaciers of the Alps have been steadily shrinking…”
At the time, this was attributed to the recovery in temperatures from the Little Ice Age. In 1894, an international group of scientists was set up in Zurich to monitor the state of the glaciers.