Monday 5 April, 2021
And it snowed this morning in Cambridge. Spring - huh!
Blackthorn blossom, seen on our walk in the Fens this afternoon.
Quote of the Day
”I myself have accomplished nothing of excellence except a remarkable, and to my friends, unaccountable expertise in hitting empty ginger ale bottles with small rocks at a distance of thirty paces.”
James Thurber (who a remarkable English teacher at school encouraged me to read)
Musical alternative to the morning’s radio news
Van Morrison | No More Lockdown
Long Read of the Day
What Data Can’t Do
Lovely New Yorker essay by Hannah Fry in which she reviews two books on data-driven decision-making, Counting: How We Use Numbers to Decide What Matters by Deborah Stone, the other, The Data Detective”, by Tim Harford.
Here’s a sample:
The particular mistake that Tony Blair and his policy mavens made is common enough to warrant its own adage: once a useful number becomes a measure of success, it ceases to be a useful number. This is known as Goodhart’s law, and it reminds us that the human world can move once you start to measure it. Deborah Stone writes about Soviet factories and farms that were given production quotas, on which jobs and livelihoods depended. Textile factories were required to produce quantities of fabric that were specified by length, and so looms were adjusted to make long, narrow strips. Uzbek cotton pickers, judged on the weight of their harvest, would soak their cotton in water to make it heavier. Similarly, when America’s first transcontinental railroad was built, in the eighteen-sixties, companies were paid per mile of track. So a section outside Omaha, Nebraska, was laid down in a wide arc, rather than a straight line, adding several unnecessary (yet profitable) miles to the rails. The trouble arises whenever we use numerical proxies for the thing we care about. Stone quotes the environmental economist James Gustave Speth: “We tend to get what we measure, so we should measure what we want.”
The problem isn’t easily resolved, though. The issues around Goodhart’s law have come to haunt artificial-intelligence design: just how do you communicate an objective to your algorithm when the only language you have in common is numbers? The computer scientist Robert Feldt once created an algorithm charged with landing a plane on an aircraft carrier. The objective was to bring a simulated plane to a gentle stop, thus registering as little force as possible on the body of the aircraft. Unfortunately, during the training, the algorithm spotted a loophole. If, instead of bringing the simulated plane down smoothly, it deliberately slammed the aircraft to a halt, the force would overwhelm the system and register as a perfect zero. Feldt realized that, in his virtual trial, the algorithm was repeatedly destroying plane after plane after plane, but earning top marks every time.
Enjoyable and instructive, like the books themselves.
Corruption as a way of life
Catherine Bennett has a sharp Observer column about Boris Johnson’s sleazy, reckless and ethically vacuous behaviour over many decades. It was bad enough when he was just a journalist, but in office it seems to have got markedly worse. And it leads one to wonder if the current Tory government is actually the most corrupt British administration for at least a century.
In 1994, Bennett recalls, Lord Nolan was tasked by the then Tory Prime Minister, John Major, with rescuing politics from Tory sleaze.
“We seek to restore respect for the ethical values inherent in the idea of public service,” Nolan wrote of the resulting Seven Principles: selflessness, integrity, objectivity, accountability, openness, honesty, leadership. Enforcement was another question. “Formal procedures have a role to play,” Nolan said, “but in the end it is individuals’ consciences that matter.” By the time George Osborne and David Cameron hastened to enrich themselves, this idea was already comical. We are now left with, on the one hand, Nolan’s faded sampler; on the other, Johnson’s expensively wallpapered, ever-expanding development of luxury Augean stables. In a nice touch, Bennett takes Nolan’s ‘principles’ and recasts them as ‘Johnson’s Principles’ to match what the government has actually been doing.
1: Greed (Replaces Nolan’s selflessness.) Holders of public office should take decisions solely in their own interest or that of their friends/families.
2. Shamelessness (Replaces integrity.) Holders of public office should accept gifts from generous individuals and organisations likely to expect favours in return.
3. Self-interest(Previously objectivity.) When making appointments, awarding contracts, etc, holders of public office should not allow merit to affect choices made exclusively to benefit themselves, their supporters, family or friends.
4. Unaccountability (Replaces accountability.) Holders of public office must not submit to scrutiny of their actions.
5. Concealment (Formerly openness.) Holders of public office have a duty to be as opaque as possible about their actions.
6. Fabrication (Replaces honesty.) Holders of public office are expected to lie freely about any private interests relating to their public duties.
7. Entitlement (Previously leadership.) Holders of public office should demonstrate by example their support for these principles, which apply to all aspects of self-enrichment.
Substack link ‘augmentation’
An email from a reader the other day launched me down an interesting rabbit hole.
“This may be a quibble”, he wrote, “but… the links on the newsletter usually don’t show where they link to. Substack corrupts links to make them unreadable, so I don’t know where I’m going to end up when I click on them. Since I can’t afford to subscribe to everything, I’m wary of clicking lest I find that, say, I’ve burned one of my few monthly free links to Atlantic or get taunted by a paywall… Substack’s corruption of links isn’t, of course, your doing … but it violates a basic promise of the Web. It would be great if you were to indicate, say, that the podcast to which you linked was from ‘Big Tech’, rather than our having to go there and figure it out.”
Since I generally get to see the daily newsletter version of my blog as a writer rather than as a reader, I investigated. Before going any further, though, it might be helpful to know about the (minimalist) workflow that goes into the publication of the Substack version (i.e. the newsletter).
It’s quite simple: I compose and publish the online version of the blog (https://memex.naughtons.org) in WordPress. Having done that I copy the online version and paste it into a new page on Substack and then schedule it for distribution early the following morning. On the face of it, the Substack version looks identical to its online version. And the links work the same way as they do in its online twin.
But… it turns out that the Substack software adds something to the link that I hadn’t spotted.
Here’s an example. A few days ago I wrote a post based on a New York Times story about a guy in London who was retrofitting electric motors in reconditioned Vespa scooters. The NYT link was:
‘https://www.nytimes.com/2021/04/01/business/electric-vespa-scooter.html’
But when I examine the link as it appears in the Substack edition this is how it appears:
https://email.mg1.substack.com/c/eJw1kEtuhDAMhk8z2RXlBYFFFt30GigJBtJCghJnKLdvmFEly7b80G9_zi AsMV36iBnJ7Ua8DtABzrwBIiRSMqTRT7qXrJWCTJoq7pQlPo9zAtiN3zQ5it28M-hjuGc FG4aBrNoN3HE2Sxja3rbKSWYHp 2TXTqJnSsxvRVMmD8GBhiekKwYgm14Rj_wQnw_-Ve08zyZc6HfIjYt7rXDKWQ1U3u7ObMk-QM41hQ0cJu8-npAP85FdjPWRZsV9I17fm1RSQTtKZdewRs1KKspN1_dyUDPrHRWzYsIKQ5ll8JB0X1i Ti81o3M99AEk6mLKsGEPtHtHG31e5Yhhr3EvweI0QjN1g0pgKEHwjfvEaFwiQKvppNKhZx5TkQ9e3rOdvI jfCelMrOSVVd4p1K-jvuIZ_2T9AsJIN
Gibberish, ne c’est pas?
Despite that, if I click on it I do get the relevant NYT page.
So the question is: what’s the gibberish for? My initial suspicion was that it was some dark tracking wheeze by Substack — which after all is a tech company, not a charity. But two smart friends surmised that the gibberish is actually designed to provide analytics data for me! For example, Substack tells me what percentage of readers have opened a particular link. Note: just a percentage: I don’t get to know who opened which link. As it happens, I don’t pay much attention to those numbers, and never requested them. But I guess that writers who have paying subscribers might be interested in them.
Coincidentally, Alex Hern (who in addition to being the Guardian’s lead tech writer also has a Substack blog) has an interesting post about all this stuff.