Tim Harford

Two recommendations: Tim Harford - http://timharford.com/ Tyler Cowen - http://marginalrevolution.com/

Tim Harford mentions in his podcast "Cautionary Tales – A Fascination with Failure / Death on the Dance Floor" that checklists might have prevented the 1981 skywalk collapse in Kansas City Hyatt Regency Hotel (killing 114 people). It's really a good podcast (pretty much all of his podcasts are) https://timharford.com/2023/07/cautionary-tales-a-fascination-with-fai…

There is a very good set of podcasts[0] by Cautionary Tales about the German V2 rockets in WW2 which tried to use this. It was so hard the Allies allowed them to continue wasting money on it during the war. Well worth a listen to if you're into that. [0] https://timharford.com/2023/07/cautionary-tales-the-v2-trilogy/

Not quite, I did a search inspired by Tim Harford’s Cautionary Tales podcast, Laser vs Parchment: https://timharford.com/2023/11/cautionary-tales-laser-versus-parchment… Which contrasted this with the Windrush scandal where the Home Office destroyed the legal documents of people welcomed from the Caribbean in the Sixties, and in their absence started to deport them. Update: that is an excellent video, setting the context of the problem and walking through the fix. I doubt I would have found it.

https://timharford.com/2022/11/cautionary-tales-the-inventor-who-almos… This is a podcast worth listening to, on the inventor of CFCs and the global response to the first discovery of the hole in the ozone layer.

On this topic here is an excellent examination of it https://timharford.com/2020/06/cautionary-tales-the-spreadsheet-of-lif…

I thought his might be another case [1] where there was an imperial to metric conversion problem that led to insufficient fuel on board and a semi miraculous glide down landing where thankfully everyone walked away. The story is far more nuanced on examination and for those interested Tim Harford [2] has an excellent Cautionary Tales podcast episode outlining the details within a wider context. [1] https://www.nytimes.com/1983/07/30/us/jet-s-fuel-ran-out-after-metric-… [2] https://timharford.com/2022/09/cautionary-tales-when-the-plane-ran-out…

> SpaceShipTwo is an inherently risky experimental test craft—with death always waiting on each side of a narrow envelope—and it always will be. Management is willfully ignorant of this. This comment reminded me of another story from a previous aerospace era (blimp cruise liners aka airships), the summary of which is: > A British Lord wanted to build the best airship in the world – and so he had two rival design teams battle it out to win the juicy government contract. Competition is supposed to bring the best out of people, but run in the wrong way it can cause people (and the things they make) to fall apart in the most horrifying ways. https://timharford.com/2019/11/cautionary-tales-ep-4-the-deadly-airshi…

This was adapted from Tim Harharford's podcast Cautionary Tales eps Wrong Tools Cost Lives https://timharford.com/2021/05/cautionary-tales-wrong-tools-cost-lives… If you have not listened to the series I highly recommend it. His episode "LaLa Land: Galileo’s Warning" is by far one of my favorite pieces of media. In brief it is why redundant tightly coupled fail safes will often lead to cascading failures. https://timharford.com/2019/11/cautionary-tales-ep-3-lala-land-galileo…

This is an excellent podcast around the Dunning-Kruger effect, how we can all be afflicted by it, and what it actually means. Setup around a (true) airline hijacking. https://timharford.com/2021/03/cautionary-tales-the-dunning-kruger-hij…

Tim Harford has an excellent article on why the polls were wrong: https://timharford.com/2020/11/why-the-polls-got-it-wrong/ The big quote is this: "The pre-election numbers suggest that in a typical poll with 500 positive responses, 255 went for Mr Biden and 243 for Mr Trump. But typical response rates are 5 in 100 — often lower, says Andrew Gelman, a statistician and prominent election modeller. He says that only around 1 in 100 people respond to opinion polls. So now picture 10,000 people, 255 who called their vote for Mr Biden, 243 for Mr Trump and 9,500 who never responded. How confident are we feeling now? " Harford is the host of the excellent More or Less podcast that looks at the validity of statistics in the news.

Thanks :-) Cautionary Tales Ep 3 – LaLa Land: Galileo’s Warning [1] has the following in the shownotes: > Charles Perrow’s Normal Accidents introduces the idea of complex, tightly-coupled systems and has good accounts both of the Three Mile Island and the Fermi reactor accidents. Just after we’d recorded the episode, I heard the sad news that Charles had died on November 12th. He’ll be missed. [1] https://timharford.com/2019/11/cautionary-tales-ep-3-lala-land-galileo…

I also recommend the author's other podcast "Cautionary Tales". It's very well produced, and fascinating. http://timharford.com/articles/cautionarytales/

Christensen's Harvard colleague Rebecca Henderson has done some research how the organizational structure of big companies gets in the way and prevents them from re-organizing around innovations. Tim Harford has an interesting article on the subject: http://timharford.com/2018/10/why-big-companies-squander-brilliant-ide…

Tim Harford’s “Adapt: Why Success Always Starts with Failure”¹, an enthralling collection of tales from really diverse projects. I’ll freely admit to only picking it up because I recognised the name from the More or Less² podcast, but it was well worth the read. 1. http://timharford.com/books/adapt/ 2. https://www.bbc.co.uk/programmes/b006qshd

Before I read Tim Harford's excellent "50 things..." book [1] I had no idea how parchment was made. What struck me most was that to make books available to the general public, the invention of the printing press may have been less important than the (re)invention of paper: for a bible on parchment one needed 500 sheep! [1]: http://timharford.com/books/50things/

> software quality doesn't appear because you have good developers. It's the end result of a process, and that process informs both your software development practices, but also your testing. Your management. Even your sales and servicing. If you only take one thing away from this article, it should be this one! The Therac-25 incident is a horrifying and important part of software history, it's really easy to think type-systems, unit-testing and defensive-coding can solve all software problems. They definitely can help a lot, but the real failure in the story of the Therac-25 from my understanding, is that it took far too long for incidents to be reported, investigated and fixed. There was a great Cautionary Tales podcast about the device recently[0], one thing mentioned was that, even aside from the catasrophic accidents, Therac-25 machines were routinely seen by users to show unexplained errors, but these issues never made it to the desk of someone who might fix it. [0] https://timharford.com/2025/07/cautionary-tales-captain-kirk-forgot-to…

Not at all, even if there was snark intended I think its a fair question. I resisted ranting about it (I hope you'll indulge me now), but in general I don't much like the economics commentariat. As best I can tell, "respectability" seems to mean something like "modal economic views + left of center politics." There are of course some exceptions, e.g. occasionally the economist or WSJ, but it leads to some notable blind spots. For example, Switzerland experienced mild deflation in a good economy 2012-2017, but it wasn't until 2023 or so that articles started coming out that questioned where the 2% inflation target came from [0] and whether deflation could ever be a good thing [1]. This was despite respectable research from at least 2004 that explored the possibility of "good" deflation [2]. Tim, like many in the commentariat, was writing about inflation in 2019 [3]. The title of this essay is "Why inflation is good for us," and discusses the merits of inflation without any real discussion of cases where deflation could be good. At the end, he even raises the idea of a 4% rate target, which is a tad funny knowing the political mood on inflation 2021-24. My point, really, is that I suspect I know most of his opinions without reading much of his work. He's hardly unique in that regard, but it bothers me. When looking for interesting opinions I'd probably look elsewhere. [0] https://www.cfr.org/blog/history-and-future-federal-reserves-2-percent… [1]…

There have been a spree of recent experiments with LLMs solving logic puzzles (specifically Cheryl's Birthday). I wanted to replicate and repeat the tests from [0] with more LLMs. For reference, that article tested whether the trained models handled obfuscation of the text so that verbatim discussions of the solution were less likely to appear in the training corpus. Then I wanted to move further and test whether LLMs were prone to distraction with extraneous and irrelevant data. In a world where RAG may pull in "compromised" data, I wanted to see if LLMs could ignore cruft or if it would alter their answer. TL;DR - it altered the answers. o1 dropped as I was making graphs etc so, I included the results from testing it as an additional section. It was still distractable but was more capable in the obfuscated case. Forgive the bait headline, I'm still trying to find the best balance of information and marketing for posts like this. Suggestions welcome on that front. [0] https://timharford.com/2024/08/ai-has-all-the-answers-even-the-wrong-o…

Funny enough ... I just listened to this 2021 podcast episode https://timharford.com/2021/05/cautionary-tales-fritterin-away-genius/ that goes into some of what you're talking about (of course, it's different - but still amusing to me that the general topic is very similar) Being a "chronic project starter" is not inherently bad It may be "bad" or or may not be "bad" I, likewise, tend to get bored with the middle parts between "idea" and "finalization" I'm really good at architectural-type stuff - big picture, planning, design, pricing, etc And I'm really good at detail stuff (get that bar chart 5 pixels further left, and 10% shorter) The middle (ie really boring parts) ... not so much - I can (and do) do them: but they are distinctly not my favorite :) The only way that has worked for me to move from "project starter" to "releasing projects" is to have an outside deadline imposed - whether it's needing a new chicken coop before the chicks are big enough to move from the mini coop, or something else: I personally need a hard cutoff from the clock to help me get there Otherwise ... it just sits on the pile of "I'll get a round tuit"

A chief tool used by industry to deceive was to cast doubt on independent science and scientists. A popular book on misuse of statistical methods, one that I'd read and liked myself, How to Lie With Statistics, turns out to have been part of that effort, as detailed by Tim Harford in his own book, How To Make The World Add Up https://timharford.com/books/worldaddup/ The ultimate cautionary tale here is Darrell Huff’s 1954 classic, How to Lie With Statistics. Huff’s book is clever, insightful, and impish, and it may be the best-selling book about statistics ever written. It is also, from cover to cover, a warning that statistics are all about misinformation, and that one should no more believe in them than in stage magic. Huff ended up testifying at a Senate hearing that the evidence linking smoking and cancer was as spurious as the evidence linking storks and babies. His unpublished sequel, How to Lie With Smoking Statistics, was paid for by a tobacco-lobby group. https://www.theatlantic.com/ideas/archive/2021/03/the-conspiracy-theor… People are very well aware that the war over what is truth and real continues to be fought. And that it is not merely corporations engaged in it, as control over truth and perception is at the heart of power. "All warfare is based on deception." https://suntzusaid.com/book/1/18/ As a hint, if one side has a long record of lies, and another a long record of honest relations and owning their…

Software engineers who have had to deal with spreadsheets being used inappropriately. Spreadsheets are an extremely powerful tool, but they have limits. Their accessibility is part of what makes them so ubiquitous, but it also means people get comfortable and start using them beyond their limits. Spreadsheets (Excel ones in particular) are great and presenting tabular data and they're mediocre at most other things including crunching numbers, taking notes, storing computational data, providing user interfaces, and much more. It's often okay to use a sub-optimal tool, but you have to draw the line somewhere. A recent HN post[0] highlighted an example where Excel was being used to keep track of large-scale contact tracing data for covid in the UK. Excel is a mediocre database. It has hard limits on cell sizes and row/column counts among other things. They ran into one of those limits and lost track of 16K positive cases because of it. From my own experience, I've had to deal with repositories containing tens of thousands of Excel spreadsheets. They were used to capture verification data. Excel files are large and difficult to parse with scripts, which was bad enough. But the worst part was that Excel doesn't really have a syntax or schema, so users editing the spreadsheets would frequently create changes to the table layouts which would have to be accounted for as edge cases in scripts. I'd be lucky to even recover data from half the files using automation. I even…

I find your certainty about something you appear to know nothing about mind boggling. Every crash is a deficiency of sorts, so let’s just tell Airbus to build planes that don’t crash! Easy! Let’s just have peace instead of war! Obvious! It is not sufficient to point out “obvious deficiencies”, one must also point out possible improvements. As other posters (with more patience than I can muster) have pointed out, it is not obvious how to improve this. Putting a lock in place to prevent movement of the lever adds another level of complexity that can itself fail. This is well known and has been extensively studied (a classic paper is No good deed goes unpunished: Case studies of incidents and potential incidents caused by protective systems [0]). Listen, for example, to this episode of “Undercover Economist” Tim Hartford’s Cautionary Tales [1] (though the example given might itself be problematic [2]). Airplanes are extraordinarily safe already (look at the amazingly low accident rate, given those fallible, sometimes even negligent and incompetent people that fly them). Contrary to your facile assertion “There is lots of room for improvement”, there are very few obvious improvements. Change something and you might prevent one accident but enable another one. As an example: terrorists enter the cockpit. Easy, obvious: lock the door! Ok. Suicidal pilot locks out the other pilot and flies into a mountain. Easy, obvious: unlock the door. GOTO 10. It’s not obvious. These are…

Replace calculator with the modern equivalent: Excel. It does make mistakes and is not reliable[0]. The user still needs to have a "feel" for the data. (to be pedantic "Excel" doesn't make mistakes, people trusting its defaults do) [0] https://timharford.com/2021/05/cautionary-tales-wrong-tools-cost-lives…

Good 3 part podcast on Houdini: https://timharford.com/2025/03/cautionary-tales-houdini-magician-impos…

I caught a podcast the other day, telling the story of a plane crash that hinged on the ambiguity of what the word "clear" meant. One of the fixes was to alter the language, so that they'd only use the word "clear" when things actually were clear. They wouldn't say things like "proceed when it's clear", just in case they mis-heard. https://timharford.com/2025/01/cautionary-tales-frozen-in-a-burning-74… I'm probably not describing it very accurately, but that's the gist. And it sounds as if a similar error occurred here. I'm a little surprised that such ambiguities would still exist after all this time.

The other really impressive part is that they also treated the workers really well during the construction. Cautionary Tales did an episode on this: https://timharford.com/2024/09/cautionary-tales-steel-and-kindness-the… They paid good wages but also kept a high bar on quality and performance. The workers in turn were incentivized to do a good job and gave feedback like ideas on how to improve efficiency, e.g. build an internal railroad to carry bricks up as they got higher and higher in height.

> who patented a product called Rose’s Lime Juice. This worked as well as lemons, and the Royal Navy mandated that the product be issued to all ships Limes did not work to prevent scurvy. Steam ships just happen to be fast enough that nobody noticed. This is why everyone in Scott’s polar expedition got scurvy - limes didnt work, but they thought they would. https://timharford.com/2022/08/cautionary-tales-south-pole-race-3/

This Cautionary Tales podcast episode talks about him [0]. Both were invented in the service of DuPont. CFC at least helped making refrigeration safer and helped getting it to the masses. Leaded gasoline on the other hand was always known as a bad idea. In both cases though, DuPont fought back hard to prevent bans of these solutions. They rather kill the planet than their profits. Suitably enough he killed himself with another one of his inventions that used pulleys to overcome mobility issues due to his Polio infection and helped him get out of bed. I guess health and safety was really not his thing. [0] https://timharford.com/2022/11/cautionary-tales-the-inventor-who-almos…

Put into much better words than I managed: https://timharford.com/2022/10/what-we-keep-getting-wrong-about-inflat…