Published 26/09/2016 | 12:19
Much automation promises more than it delivers, and has done so for hundreds of years. But what seems to be a failing of the machine may just be a sign that more human expertise is needed at the controls.
In the late 18th Century, Wolfgang Von Kempelen toured Europe with a mechanical device resembling a man who could defeat the best chess players of the day.
This automaton that confounded the great and the good was called The Turk.
When it changed hands years later, the new owner learned that The Turk was a trick, with the mechanical arms operated by a chess expert hidden within the device.
When fiction inspires innovation
The most interesting aspect of the story for me is that in 1784 Edmund Cartwright observed The Turk in action in London. He concluded that if a machine could master the mental and physical challenges of playing chess, then a similar mechanisation could serve his own particular area of interest. Within a year he had invented the power loom.
Neal Stephenson’s polemical Innovation Starvation refers to the Hieroglyph Theory: ‘Good Science Fiction supplies a plausible, fully thought-out picture of an alternate reality in which some sort of compelling innovation has taken place’. The Turk seems a perfect example of this, a trick or fiction that challenged the perception of what was possible at the time, triggering a significant break-through in industrial engineering.
Stephenson’s essay goes on to call out the lack of ambition in addressing real problems compared to the “practical techno-optimism of the Golden Age”, and honed in on the lack of appetite for short term risk and failure to deliver long term benefits: “Today’s belief in ineluctable certainty is the true innovation-killer of our age” which creates “a world where big stuff can never get done”.
Person-machine partnerships are still vital - especially for network security
I’m sure most of us have bought an equivalent of The Turk in our time, having been convinced by what we want it to be, and unprepared for what it is. The gap between expectation and experience in areas of automation can be frustrating: the moment you realise that as well as the machine, you will need a team of chess experts to crawl into the cabinet whenever a game is to be played - and there was no incremental headcount for a 24x7 rota of small chess masters in the business case.
The recent US House of Representative report on two hacks in the Office of Personnel Management, (sub-titled ‘How the Government Jeopardized Our National Security for More than a Generation’) makes for sobering reading, and the recommendations on page 20 in particular. One of these states the need to prioritise a move to a Zero Trust Model, centred on the concept that users inside a network are no more trust-worthy than users outside the network. “In order to effectively implement a zero trust model”, the report states, “organizations must implement measures to visualize and log all network traffic, and implement and enforce strong access controls”.
The network is a critical foundation of defence. Security, and in particular the interpretation of the myriad sources of intelligence and the rapid assessment of risk and response, is one of the most significant challenges for our customers. As I noted in a previous blog, it’s an area in need of automation of detection and decision-making.
I’m optimistic about automation of service management and orchestration, and realistic about the expertise and skills that will be required to bring automation to life and keep it effective. We have a 'plausible fully-thought out picture of an alternative', and it took Cartwright less than a year from seeing The Turk apparently play chess to inventing the power loom.
Andy O'Kelly is Chief Architect in eir Business, providing vision and direction on emerging business and technology trends, and promoting eir solutions to key customers. For more insights on automation, innovation and network security, visit the eir Business blog.