Only Humans

Harold Jarche has posted a brief review of “Only Humans Need Apply” by  Thomas H Davenport .

In his review Harold has added the main attributes that he sees as being needed to meet the book’s criteria for human adaptation to a world of automation:

  • Step-up: directing the machine-augmented world (creativity)
  • Step-in: using machines to augment work (deep thinking)
  • Step-aside: doing human work that machines are not suited for (empathy)
  • Step narrowly: specializing narrowly in a field too small for augmentation (passion)
  • Step forward: developing new augmentation systems (curiosity)

I challenge any UK-based educator or politician to identify where we are systematatically encouraging those attributes in our young people.


The machines may eat your job, but that might not be a bad thing – are any politicians acknowledging this?


There are a growing number of indicators that the nature of employment will change radically in our lifetimes, but politicians are all ignoring this.


On BBC Breakfast this morning there was a piece about robots, themed on the forthcoming exhibiton at the Science Museum,

In the piece they interviewed  Michael A. Osborne , Associate Professor in Machine Learning, University of Oxford in which he repeated the research estimate that robots would replace 35% of UK employment by 2030, e.g .transport, taxis, processing invoices etc.

That in turn is highlighted in this 2015 BBC story, extrapolated from this 2013 paper by Frey and Osborne which examined the US jobs market and concluded that estimated 47% of total employment was at risk.

Also this week in “Wanted: Factory Workers, Degree Required” the NYT quoted Eric Spiegel, the recently-retired president and chief executive of Siemens USA:

“In our factories, there’s a computer about every 20 or 30 feet. People on the plant floor need to be much more skilled than they were in the past. There are no jobs for high school graduates at Siemens today.”

Although the Frey and Osbourne paper does not put firm time horizons on their predictions, instead saying “We refer to these jobs as at risk – i.e. jobs we expect could be automated relatively soon, perhaps over the next decade or two”, some of the notable occupations with a predicted probability of replacement by machines of > 75% include:

  • painters, construction and maintenance
  • locksmiths
  • electric motor and power tool repairers
  • bartenders
  • archivists
  • barbers
  • fast food cooks
  • property managers and estate agents
  • electronics technicians
  • executive assistants
  • technical writers
  • pharmacy technicians
  • accountants and auditors
  • budget analysts
  • paralegals and legal assistants
  • credit authorisers
  • drivers
  • telemarketers

A counter view

By contrast, Matthew Yglesias puts forward a counter view – that we should be worried if we don’t automate lots of jobs, because of the dramatic negative impact on productivity.

In his view if societies do not embrace increasing automation (which by implication means finding the ways to invest in it) then they are doomed to a low-income, low living standard future, with a fatal combination of low productivity and spiralling healthcare and social care costs.


There is a clear message in this research – the very nature of employment is already changing far beyond the increased casualisation that is highest on most debates. Many forms of employment will disappear within the working lifetime of children now in schools.

At the same time we need to ensure that we can invest in the technology that will bring high productivity comemrce to the country.

Some questions for the politicians:

  • how will the UK attract investment in the technology needed to operate efficiently in the 2030’s?
  • what changes do we need to be planning NOW to the education system to prepare people for that world?
  • how will our society support the people who cannot reach the levels of education needed to get the jobs that will be available?
  • or is the UK doomed to slump to being a low income, low employment sweat shop?


New York Times link via Doug Belshaw

Featured image CC0 Pixabay

Matthew Yglesias link via Ryan Avent via Michael Osborne


An unremarkable coincidence

A few months ago I spent some time exploring my wife’s family tree. One fragment of jigsaw was a census return from 1851 showing one of her 3–greats-grandfathers, then age 16, living with his parents at 13 Cambridge Street in Soho.

1851 census abstract

I looked at the current London street map and found there is no Cambridge Street in Soho any more, and that was the end of my curiosity.

Skip forward to last week, and I finally got around to reading Ghost Map by Steven Johnson.

Johnson takes as his central theme the story of John Snow, and his pioneering work with Henry Whitehead during and after the 1854 cholera outbreak in Soho which proved for the first time that cholera was a water-borne infection.

One of the key tools used by Snow was to plot the course of the outbreak on a map. Looking at the map I realised I had found Cambridge Street – the old name for the upper part of what is now Lexington Street.

The slightly more detailed general Board of Health Map based on Snow’s work also shows the old house numbers – and this shows number 13, just 7 doors down from the infamous pump which spread the disease.

General Board of Health map

The coincidence is that a direct ancestor of my wife lived in a house which, three years after evidence of their residence, was in the centre of a particularly virulent and well-documented outbreak of cholera. Today, in a modern city, that would be worthy of comment: in Victorian London perhaps less so.

We have no way of knowing if they were still in residence three years after the census, and thus survivors of the outbreak, but even if they had moved on, the descriptions of the Broad Street area provide a fascinating insight into the circumstances in which they lived.

More interesting, to me at least, are the aspects of my curiosity which set up this piece of serendipitous discovery – family history, networks of connections, a lifelong fascination with maps and the story they tell, and the ever-fascinating mental game of “What if… ?”

Unintended consequences…

…or, what if the things you believe are fundamental to keeping your society together are in some way linked to the negative effects that you see around you?

That might be the sort of question you ask after reading a study published in the Journal of Religion & Society which suggests that a high level of religious belief may harm a society.

As reported in the Times, the study, Cross-National Correlations of Quantifiable Societal Health with Popular Religiosity and Secularism in the Prosperous Democracies looks at data across the first world Western democracies, and examines both the level of overt belief in God / disbelief in evolution and the occurrence of various societal measures such as homicides, early mortality, STDs, teenage pregancy and abortion. The paper finds strong correlations between the general level of religiosity in a society and high levels of these negative measures.

The author is clear that this is an initial study of the correlation between data sets, and does not hypothesise a causal link, however he does include a call to arms for social sicences to examine these issues more closely.

Ironically, the scientific method, which to-date has been shown to be the most effective way of exploring links between events “out there” and putative causes is, I suspect, likely to be the last thing that members of a highly religious society will turn to.

Print version of paper [PDF]

[ via Voidstar]

Working in the Twenty-First Century

I’ve been reading Working in the Twenty-First Century, which describes itself as an evidenced-based look at the future of work in the UK over the next 20 years by Michael Moynagh and Richard Worsley, published by the Economic and Social Research Council and The Tomorrow Project.

Four main themes emerge from the report:

The British economy will (be forced to) move up the value chain, changing the nature of the jobs that are available. There will be an amplification of the “hourglass effect”, as jobs in the middle disappear, leading to an economy divided between highly-skilled, high-value work and low-paid service workers.

The UK labour market will continue to be tight, squeezing the supply of key skills and leading to attempts to try and boost labour supply. Areas of likely change are an increase in older workers, an increasing reliance on workers from outside the UK and further moves to greater equality of opportunity for women.

The way people will work will be subject to drastic change (for example a huge rise in mobile working, enabled by technology) but there will be less change in the way they are employed – the authors see a majority remaining in full-time employment. The organisations which provide that employment are likely to adopt a range of structures, from the traditional to newer forms such as virtual and networked organisations.

The nature of work and the management-worker relationship will change as new forms of motivation become the norm, at least at the high-value end of the economy. Here the report authors see that companies will be forced to allow greater autonomy and worker empowerment in order to meet the market demand for customised, responsive services. By contrast, companies providing services in the low-wage, low-value, low-skill service areas are unlikely to see the need to take on the cost of more flexible management methods.

There’s a Tomorrow Project event to discuss the book on 6th October – so I should be blogging some more after that.