About me
Bookshop

Get new posts by email.

About me

Bullshit engine

In the FT Weekend, Tim Harford recently played with the latest version of ChatGPT and concluded that:

large language models can be phenomenal bullshit engines. The difficulty here is that the bullshit is so terribly plausible. We have seen falsehoods before, and errors, and goodness knows we have seen fluent bluffers. But this? This is something new.

It made me reflect on the types of, erm, lexical nonsense I come across in my day-to-day life.

The kind I find most irritating by far is deliberately obfuscatory language. I recently received an email referring to a ‘pathogen agnostic enabling technology’ and wondered why on Earth someone would write that: I didn’t understand what it meant, I’m fairly certain that most other readers didn’t either, and I’ve a strong suspicion that the author would also be unable to concisely define it. They were hiding their lack of understanding—or perhaps their inability to describe something clearly and concisely—behind a jargon phrase.

Yet, from a reader’s perspective, the advantage of this sort of word salad is that it’s clearly visible: it telegraphs a lack of understanding up front.

Perhaps I really ought to be more exercised about plausible rubbish: the sort of stuff whose clarity of expression lulls one into a false sense of assurance, from which the logical leaps are harder to spot. That seems far more insidious.


The image at the top of this post was generated by DALL·E 3.

This post was filed under: Miscellaneous, , .

31 things I learned in May 2020

1: We’re actually decent people in a crisis – and stories claiming otherwise do harm.


2: A monk’s cowl is “meant to be impractical – you can’t run in it for instance. It slows you down and you can’t do much in the way of work as a result of the long sleeves.”


3: “While the rest of us headed into lockdown worrying about whether we had enough toilet roll and ketchup, the super-rich were desperately trying to recruit live-in staff.” I’m not usually partial to cheap reverse snobbery, but that article had some zinging lines in it.


4: Paul Collier’s critique in The TLS of the UK Government response to covid-19 is the best I’ve read to date (though admittedly I’m trying to avoid reading too much on covid-19 outside of work). I don’t agree with the detail of all of his conclusions, but I think he brings important issues to the surface.


5: “There are many modern thinkers who emphasise the individual’s dependency upon society. It is, on the contrary, only the cultivation of interior solitude, among crowded lives, that makes society endurable.” So said John Cowper Powys, apparently. I tend to agree.


6: “In Europe, bunks on a night train have traditionally been set at ninety degrees to the direction of travel, like the teeth of a comb. In America, the custom was to place them lengthways, so that your body, when horizontal, slotted into the train like a bullet in the breach of a rifle.” I could have lived my entire life without this delightful bit of trivia ever coming to my attention.


7: “The hope is that almost all of us will download the app, that we will be diligent about using it if we develop symptoms, that the detection of identifiers will be reliable, that the subsequent risk calculation will be more or less accurate, and that we will, by and large, self-isolate if the app tells us to. Crucially, the strategy also requires easy access to tests so people can be rapidly alerted if a contact who had symptoms turns out not to have had the disease.” I’m a covid-19 app sceptic: I don’t think the uptake will be anywhere near 80% of smartphones (as is hoped) and nor do I think that there will be comparable compliance with isolation advice given by app and that given in a human conversation. Twelve months from now, when this post is published and the app has proven to be a rip-roaring success, you can comment and tell me what a fool I am for posting such silly predictions.


8: Moving a Bank Holiday to a Friday makes it more difficult to know what day it is. Lockdown and the consequent intense but irregular working pattern already made it hard enough for me.


9: The details in The Economist‘s cover images sometimes pass me by.


10: “Stay alert will mean stay alert by staying home as much as possible, but stay alert when you do go out by maintaining social distancing, washing your hands, respecting others in the workplace and the other settings that you’ll go to.”


11: Gillian Tett’s observation that “Americans are wearily used to the idea that 40,000 die each year from guns, and many accept this as the price of freedom” helped me see grim fatalism as one response to the lifting of the covid lockdown: the polar opposite of the safety first, fear-driven response that many pundits predict will dominate.


12: “Britain is so preoccupied by the virus that it is devoting far too little attention to its Brexit negotiations, increasing the chances that an on-time Brexit will also be a bitter Brexit.” I’m fairly confident that, despite current bluster, the Government will end up asking for an extension of the transition period. (This post is rapidly turning into “31 predictions from May 2020” rather than 31 lessons…)


13: Will Self’s article on the mechanics of freelance journalism, published in the reputedly low-paying TLS, opened my eyes to the basic realities of that profession.


14: My local petrol station is now charging less than £1/litre.


15: “Senior Conservatives have called for all MPs to be allowed to return to the House of Commons as they become concerned Boris Johnson is struggling in the deserted chamber in his encounters with new Labour leader Keir Starmer.” Bless.


16: Uncertainty about the safety and effectiveness of contact tracing apps is growing. The Economist has a published a leader on the topic: “They are an attractive idea. Yet contact-tracing apps are also an untested medical invention that will be introduced without the sort of safeguards that new drugs are subjected to. Inaccurate information can mislead health officials and citizens in ways that can be as harmful as any failed drug. Governments should proceed with care.”


17: “The most important breakthroughs in medical interventions – antibiotics, insulin, the polio vaccine – were developed in social and financial contexts that were completely unlike the context of pharmaceutical profit today. Those breakthroughs were indeed radically effective, unlike most of the blockbusters today.” This is obvious when you think about it, but I’ve never really thought about it before.


18: Multi-person iron lungs existed.


19: Chloe Wilson, who I’ve never come across before, seems to be quite a writer.


20: Cereal taught me the Korean idiom “when tigers used to smoke,” meaning a very long time ago. And also the lovely saying “deep sincerity can make grass grow on stone.”


21: Vitamin String Quartet covered the whole of Daft Punk’s Random Access Memories album and somehow this news has passed me by for the best part of four years, even though I like Vitamin String Quartet and love Random Access Memories.


22: “A local leader characterises PHE’s response to the crisis as ‘carry on covid.'” It seems that even The Economist has now concluded that Public Health England is “unlikely to survive the crisis.”


23: This video introduced me to several new terms unique to the world of antiquarian book repair (though Slightly Foxed taught me the meaning of ‘slightly foxed’ some years ago!)


24: Itsu’s katsu rice noodles are lovely, even if they are basically a posh pot noodle.


25: Going for a drive to test one’s eyesight is, according to the government, an acceptable reason for deviating from “stay at home” advice.


26: How different artists approached drawing the SARS-CoV-2 virus.


27: Dr Bonnie Henry has had some shoes made in her honour. And they sold out quickly.


28: A month ago, I don’t think I could have confidently defined ‘pangram’. Now, I’m coming across them everywhere: there’s been a running feature in The Times diary column, they feature in Exciting Times by Naoise Dolan which I’m currently reading, and The Browser recently recommended an article about them. My current favourite is ‘amazingly few discotheques provide jukeboxes’.


29: The Twentieth Century Society made me aware that tax incentives promote new construction over refurbishment, which is part of the reason why perfectly sound buildings are often demolished rather than repurposed.


30: It’s been lovely to have a day off and go for a walk with Wendy. COVID-19 work has run us both ragged recently. I’ve also had my first takeaway coffee in several months.


31: According to anonymous sources talking to The Sunday Times, “Boris has always been clear that he doesn’t ever say sorry,” “these stories about Boris being fed up with the job are all true” and “the chances of Boris leading us into the next election have fallen massively.”

This post was filed under: Posts delayed by 12 months, Things I've learned, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , .

Review: The Undercover Economist by Tim Harford

It takes a special kind of effort to be so behind the times that I’m reviewing The Undercover Economist more than three years after it was published, despite having it on my “must read” list for most of that time. As a fan of Tim Harford’s contributions to the Financial Times and his presentation of More or Less, I had great expectations for this book – and I wasn’t disappointed.

The Undercover Economist seeks to explain the economic theories underpinning everyday life, from buying a coffee to shopping in a supermarket. I was passingly familiar with most of the theories being discussed, as I would imagine that most people would be, but Harford does a good job of fleshing out the details, equipping the reader with economic vocabulary, and showing how the theories work in everyday situations.

Harford is clearly an excellent writer, and the book zips along at a fair lick for the most part. There are, however, some parts that drag a little. Occasionally, Harford goes to some lengths to explain the same theories repeatedly in different situations in a way that becomes a little repetitive and unnecessary in parts. I sometimes felt like the speed of the repeated explanations was a little slow, as though I was some way ahead of the book. This was a bit frustrating. But it’s not a major issue. I guess what I’m saying is that a well-abridged version of the book would be no bad thing.

There are also bits of the book where I feel like Harford strays a little too far into relating a particular point of view, rather than setting out the wider picture as I would expect from a book like this. I am actually interested in Harford’s opinions on globalisation, for example, but don’t think that this is necessarily the place to communicate his point of view (almost) exclusively. I felt that a little more balance would have been welcome in this introductory book.

But these are niggles. Economics is often portrayed as dry and theoretical, and The Undercover Economist shows it to be anything but. It sheds light on the economic decisions we all make everyday, and the way that we are economically manipulated by corporations and governments – for better or worse. It is absorbing and enjoyable – not words that necessarily spring to mind for books about this particular subject matter.

So whilst there were some rough edges, I enjoyed The Undercover Economist. And perhaps the best I can say is that the sequel is on my “must read” list… and I think I’ll get to it some time before 2017!

 The Undercover Economist

The Undercover Economist is available now from amazon.co.uk in paperback and on Kindle.

This post was filed under: Book Reviews, .

Weekend read: Wonga is the symptom, not the problem

This weekend, I recommend spending some time reading this article, which Tim Harford originally wrote for the Financial Times. Almost all of Tim’s FT articles are brilliant, but this one particularly stuck in my mind for the clarity of its argument on a complex subject.

I was particularly struck by this passage:

A payday loan can do real good, as a cash injection that helps avoid far more serious financial consequences, such as the loss of a job because the car broke down or penalty charges for failing to pay a bill on time. A randomised trial conducted in South Africa showed that this was not just a theoretical possibility. The experiment randomly approved or rejected applications for loans at an annual percentage rate of 200 per cent. Those who received one ended up better off than those rejected.

This is an aspect of this particular debate that’s all-too-often overlooked.

This post was filed under: Weekend Reads, , , .




The content of this site is copyright protected by a Creative Commons License, with some rights reserved. All trademarks, images and logos remain the property of their respective owners. The accuracy of information on this site is in no way guaranteed. Opinions expressed are solely those of the author. No responsibility can be accepted for any loss or damage caused by reliance on the information provided by this site. Information about cookies and the handling of emails submitted for the 'new posts by email' service can be found in the privacy policy. This site uses affiliate links: if you buy something via a link on this site, I might get a small percentage in commission. Here's hoping.