Friday 14 February, 2025

Blooming heck!

A truly amazing plant. We come down every morning and it’s done something new.


Quote of the Day

”If it’s sent by ship, it’s cargo. If sent by road, it’s shipment.”

  • Dave Allen

Musical alternative to the morning’s radio news

Emily’s Reel | Mark and Maggie O’Connor

Link

Lovely Appalachian tune.


Long Read of the Day

Springtime for Scammers

Readers with long memories will remember Mel Brooks’s lovely film, The Producers, in which Zero Mostel and Gene Wilder come up with a scam involving putting on a sure-fire Broadway musical flop entitled Springtime for Hitler (which, surprisingly, becomes a hit). The headline on this blog post by Paul Krugman immediately brought the movie to mind, because it’s already clear that the Trump regime is opening the floodgates to scamming on an Olympic scale, with crypto at the heart of it. This post is not about cryptocurrencies but about simpler kinds of fraud. It was prompted by the news that Trump has shuttered the federal Bureau of Consumer Financial Protection, which was created to protect Americans from financial predators.

Overall, Project 2025’s attack on the CFPB bears a family resemblance to Elon Musk’s claim that USAID is a “viper’s nest of radical-left Marxists who hate America.” It’s a bit milder, but equally absurd, and is clearly not the real reason for killing the agency.

So what is the real reason? It seems fairly obvious. CFPB was created to protect Americans from financial predation, and has done a very good job of doing so. But now we have government of, by and for financial predators. Trump has famously left behind a trail of bankruptcies and unpaid contractors, and is furiously grifting even now. Musk has faced multiple lawsuits from vendors and former employees over unpaid debts.

There’s lots more to come on this topic. It’s nice to have Krugman on Substack rather than behind the NYT paywall.


Books, etc.

I’m an avid notebook user, partly because I want to keep my handwriting from atrophying, but mainly because it enables me to keep track of fleeting thoughts. So I was a sucker for Roland Allen’s book when it came out. I’m about halfway through it, and enjoying it. Notebooks have a much more interesting history than I had imagined.


Feedback

My puzzlement about the neon inscriptions on the Portico of the Fitzwilliam the other night prompted Kevin Cryan to email:

The poem quoted in neon is – in the unlikely case you did not know – is “Waiting for the Barbarians” (Περιμένοντας τοὺς Bαρβάρους) an 1898 Greek poem (published in 1904) by Constantine P. Cavafy (1863 – 1933). I came across it when I was studying Modern Poetry with the OU, and it has remained one of my favourite “political” poems ever since

Of course I didn’t know. Which is why being an ignorant blogger is such a joy!


This Blog is also available as an email three days a week. If you think that might suit you better, why not subscribe? One email on Mondays, Wednesdays and Fridays delivered to your inbox at 6am UK time. It’s free, and you can always unsubscribe if you conclude your inbox is full enough already!


Wednesday 12 February, 2025

Poetry in Neon

The Portico of the Fitzwilliam Museum in Cambridge the other night. The illuminated quotes are a part of the work of the American contemporary artist Glenn Ligon which can be found in galleries throughout the building. If you’re interested, here’s a video.


Quote of the Day

” We stand at a crossroads. One path leads to despair, the other to destruction. Let’s hope we make the right choice.”

  • Woody Allen

Musical alternative to the morning’s radio news

Bonnie “Prince” Billy | Our Home

Link


Long Read of the Day

 The End of Programming as We Know It

Tim O’Reilly is one of the sharpest observers of the tech industry around, and this long essay is an example of what he does best — to escape from “the sociology of the last five minutes” that plagues coverage of the industry. You don’t have to be a programmer to appreciate it.

Sample:

AI will not replace programmers, but it will transform their jobs. Eventually much of what programmers do today may be as obsolete (for everyone but embedded system programmers) as the old skill of debugging with an oscilloscope. Master programmer and prescient tech observer Steve Yegge observes that it is not junior and mid-level programmers who will be replaced but those who cling to the past rather than embracing the new programming tools and paradigms. Those who acquire or invent the new skills will be in high demand. Junior developers who master the tools of AI will be able to outperform senior programmers who don’t. Yegge calls it “The Death of the Stubborn Developer.”

My ideas are shaped not only by my own past 40+ years of experience in the computer industry and the observations of developers like Yegge but also by the work of economic historian James Bessen, who studied how the first Industrial Revolution played out in the textile mills of Lowell, Massachusetts during the early 1800s…

TL;DR. AI is not the end of programming. But what programming is will change.


My commonplace booklet

Screenshot

From “Could the world decouple from US Trade?” in the FT

The answer, oddly enough, is that apart from poor Mexico and Canada, is “yes”. So why are governments all over the world quaking in their handmade boots?


Linkblog

Something I noticed, while drinking from the Internet firehose.

  • “In the Machiavellian sense, corruption is not just about illegal and legalized bribery, but also and even more about the bending of the rules such that when they function properly the public good is structurally undermined.”

From “Corruption, Tariffs, and US Renewal” by Eric Schliesser on the Crooked Timber blog.


This Blog is also available as an email three days a week. If you think that might suit you better, why not subscribe? One email on Mondays, Wednesdays and Fridays delivered to your inbox at 6am UK time. It’s free, and you can always unsubscribe if you conclude your inbox is full enough already!


Monday 10 February, 2025

Patience

Everything comes to those who wait… Well, almost everything.


Quote of the Day

”Like trying to pan gold in a sewer.”

  • Charles Arthur, writing about the challenge of moderating newspaper readers’ comments.

Musical alternative to the morning’s radio news

Arlo Guthrie and Emmylou Harris | Deportee

Link

Appropriate for our times.


Long Read of the Day

 Elon Musk Weaponizes the Government

Another angle on Musk’s Coup d’Etat from Henry Farrell and Abe Newman.

It’s hard for politicians and journalists to grasp what Musk and his team are doing, still less to explain it to the public. These efforts involve technical systems that are incomprehensible and boring to outsiders, few of whom even know that the Office of Personnel Management exists, let alone what it does. But even if no one pays much attention to these systems, they are the sinews of government—key parts of the infrastructure that hold the federal state together and manage its relationship to the outside world.

We are highly familiar with such systems and how they can be used. Our academic research, and our recent book, “Underground Empire: How America Weaponized the World Economy,” explain how the U.S. quietly took control of similar technical systems that hold the world economy together and used them to exercise domination over allies and enemies alike. Now, Musk is seemingly doing to the U.S. government what the U.S. government once did to the rest of the world: refashioning the plumbing of the federal government into a political weapon against his adversaries…

This is a perceptive take on what’s happening. The authors’ book, book  is very good, btw.


While Trump blathers about tariffs and Gaza, Musk is executing a coup d’état

Yesterday’s Observer column

Way back in 2019, Steve Bannon, then a Trump consigliere, outlined in a TV interview a strategy for managing information. “The opposition party is the media,” he said, “And because they’re dumb and they’re lazy, they can only focus on one thing at a time… All we have to do is flood the zone. Every day we hit them with three things. They’ll bite on one, and we’ll get all of our stuff done. Bang, bang, bang.”

Since his re-election, Donald Trump has been following this script to the letter, and the media, not to mention the entire world, are feeling punch-drunk. Which is, as Bannon pointed out, enabling other members of the Trump crew to get their stuff done. Really bad stuff too, to which the world has not been paying enough attention.

Prime suspect in this respect is Elon Musk, whom Trump has chosen to slash $2tn off US government spending.

Read on


My commonplace booklet

Papering over history

Screenshot

This photograph, courtesy of National Public Radio, shows how the NSA Museum papered over plaques devoted to women and people of colour who served the National Security Agency.

The NPR report says that,

Responding to President Trump’s order that terminated diversity, equity and inclusion (DEI) initiatives across the federal government, the National Cryptologic Museum taped sheets of paper over plaques that celebrate women and people of color who had served the National Security Agency, which intercepts overseas conversations and breaks foreign government codes.

The honorees are described as “Trailblazers in U.S. Cryptologic History,” and the plaques hang in the museum’s Hall of Honor…

Later, in a phone interview, NSA Executive Director Sheila Thomas — the No. 3 person in the agency — told NPR that papering over the exhibits was a mistake.

”There was absolutely never an intention to cover up parts of our history,” said Thomas, who was at the museum to greet the retirees and has worked at the NSA for just over four decades. “As soon as we became aware, we said, ‘Oh, that was not what was intended.’”

This is a small but revealing example of how an authoritarian regime erodes freedom — not by issuing orders but by creating an atmosphere where individuals and institutions try to second-guess what’s permissible and what’s not.


This Blog is also available as an email three days a week. If you think that might suit you better, why not subscribe? One email on Mondays, Wednesdays and Fridays delivered to your inbox at 6am UK time. It’s free, and you can always unsubscribe if you conclude your inbox is full enough already!


Friday 7 February, 2025

Floral contrasts

Our Amaryllis has suddenly burst magnificently into flower. Photographed yesterday with Robert Mapplethorpe’s ‘Orchid’ (1989) in the background.


Quote of the Day

”He talked about those politicians who were always against everything new: “We used to have folks like that around the store in Johnson City,” he said. “We called them dispeptics. When they put the railroad through town for the first time, one old man stood there and looked at-‘ it and said, ‘They’ll never get the damn thing started.’ The girl came up with a great wine bottle and hit it across the snoot of the locomotive and it started going out about 15 or 20 miles an hour. And they went up to him and said, ‘What do you think now, Uncle Ezra?’ And he said, ‘They’ll never get the damn thing stopped.’”

  • LBJ in a speech to Democratic Candidates for Congress in the White House* 

Musical alternative to the morning’s radio news

Ry Cooder & David Lindley | All Shook Up

Link

Song title resonates with the current state of the (Western) world.


Long Read of the Day

Musk’s Junta Establishes Him as Head of Government

Nice satirical piece of satire by Garrett Graff prompted by an interesting thought. “I’ve long believed that the American media would be more clear-eyed about the rise and return of Donald Trump if it was happening overseas in a foreign country, where we’re used to foreign correspondents writing with more incisive authority. Having watched with growing alarm the developments of the last 24 and 36 hours in Washington, I thought I’d take a stab at just such a dispatch. Here’s a story that should be written this weekend.”

Here’s how it opens:

WASHINGTON, D.C. — What started Thursday as a political purge of the internal security services accelerated Friday into a full-blown coup, as elite technical units aligned with media oligarch Elon Musk moved to seize key systems at the national treasury, block outside access to federal personnel records, and take offline governmental communication networks.

With rapidity that has stunned even longtime political observers, forces loyal to Musk’s junta have established him as the all-but undisputed unelected head of government in just a matter of days, unwinding the longtime democracy’s constitutional system and its proud nearly 250-year-old tradition of the rule of law. Having secured themselves in key ministries and in a building adjacent to the presidential office complex, Musk’s forces have begun issuing directives to civil service workers and forcing the resignation of officials deemed insufficiently loyal, like the head of the country’s aviation authority…

Read on. It’s a hoot — except that when you get into it it begins to look like actual reporting of what’s going on.


Books, etc.

This arrived the other day. I welcomed it because I loved Margaret Heffernan’s earlier book, Wilful Blindness — about a condition that plagues our political and corporate world. Her new book is about the way that artists not only cope with uncertainty better than the rest of us, but how they positively embrace it.


My commonplace booklet

 AI company says to job applicants: ‘Please don’t use AI’

When I first saw this in the FT, I immediately filed it away in the “You Couldn’t Make This Up” folder. Anthropic is the company that makes Claude, one of the LLMs that I use regularly.

Here’s what it tells job applicants:

While we encourage people to use AI systems during their role to help them work faster and more effectively, please do not use AI assistants during the application process. We want to understand your personal interest in Anthropic without mediation through an AI system, and we also want to evaluate your non-AI-assisted communication skills. Please indicate ‘Yes’ if you have read and agree.


Linkblog

Something I noticed, while drinking from the Internet firehose.

  • Pot calls the Data Centre black

From New Scientist

More than 60 per cent of the 24,000 tonnes of cannabis grown each year in the US is cultivated in indoor farms that rely on large arrays of lights as well as heating and cooling systems. Those controlled conditions enable the production of more potent flowers and for crops to be grown in more places with more security – or secrecy. But they come with a large environmental cost.

Mills used data on both the legal and illegal US cannabis industry to estimate the total energy use and emissions generated by these indoor plant factories. He found they use around 596 petajoules per year. That’s a “remarkable” amount of energy, equivalent to about a third of current energy demand from US data centres, says Mills.

Generating that electricity, as well as the carbon dioxide pumped into indoor operations to fertilise cannabis plants, produces emissions equivalent to 44 million tonnes of CO2 per year – about the same as 6 million homes. For the average daily cannabis user – more than 17 million of whom live in the US – that would mean cannabis accounts for half their personal CO2 footprint.

Jeez. And here am I, someone who has never smoked weed in his long life, fretting about the carbon footprint of data centres!

Thanks to Sheila Hayman (Whom God Preserve) for spotting it.


Feedback

  • John Seeley read my claim that LLMs, like libraries, “help humans to access and utilise the accumulated knowledge of previous generations” and was reminded of something that George Jackson wrote in Soledad Brother when referring to libraries : “We, the humble representatives of future generations, have all the accumulated knowledge and thinking of previous generations to build our thoughts. “

  • A good friend who used to be a Blue Guide in Cambridge was struck by my piece on Wednesday about the statue of Henry VIII on Trinity’s Great Gate. “My stories about it,” she writes,

are that the original sceptre was removed by a student in May Week and a chair leg put in its place. The original sceptre (I understand) was never returned. Many years later, also in May Week, a student climbed up in the dark and replaced the chair leg with a bicycle pump. The following morning he telephoned the Porters’ Lodge anonymously and asked if ‘they had seen Henry this morning’. A reply in the negative prompted to ask the porter to go and look. A weary Porter returned to the telephone and said; “I’ve seen Henry” to which the student asked if he could go to the Master and tell him that the students thought a bicycle pump was more appropriate than a chair leg (Cambridge is after all a city for cyclists).

The Porter duly went to the Master who gathered a number of Fellows to discuss this problem. I was told by someone that it was discussed for 50 minutes and the conclusion was that the chair leg was to be replaced as their Great Gate with a chair leg had become an official sight in Cambridge and a photograph was in all the guide books. Not so many years ago I had a group with me at the Great Gate in May Week and found that Henry’s right hand was empty. I asked the porter what had happened to the chair leg. He drew himself up to a great height and said: “That’s what we all want to know”!


This Blog is also available as an email three days a week. If you think that might suit you better, why not subscribe? One email on Mondays, Wednesdays and Fridays delivered to your inbox at 6am UK time. It’s free, and you can always unsubscribe if you conclude your inbox is full enough already!


Wednesday 5 February, 2025

HenryGate

A group of Chinese tourists I noticed yesterday contemplating the Great Gate of Trinity College — and no doubt wondering who is the cove in the alcove half-way up the structure. He is none other than Henry VIII, who, taking time off from murdering his wives, founded the college. The sculpture originally showed him holding a sceptre in his right hand and an orb in his left, but it now shows him wielding a different symbol of power — a chair leg!

There wqs a nice story about this in the Cambridge News some years ago. Apparently in the 1980s a cleaner working on the windows next to the statue noticed that the sceptre had gone missing and replaced it with a leg from a broken chair on the landing inside the tower. You may think (as I did) that the story is too good to be true. I tried zooming in on the image, but Henry became too pixelated, and so I went looking for a higher-resolution pic and found this one in a 2023 BBC report.


Quote of the Day

”A parrot trained to repeat the phrases “orange man bad”, “brexit bad” and “crypto scam” has outperformed 99.9% of professional analysis over the last decade.”

  • Alex Hern

Musical alternative to the morning’s radio news

Richard Strauss | Four last songs | Im Abendrot | Kiri Te Kanawa

I love these songs and keep coming back to them.


Long Read of the Day

LLMs are not a Flawed Design, they are the Completion of a Flawed Paradigm

This essay by Erik J Larsen is one of the most astute pieces I’ve read on the Large Language Models (LLMs) that the tech industry (and most of the media) persist in describing as ‘artificial intelligences’ or ‘AIs’. They are a product of a powerful technology called machine learning (ML), the developments in which have been truly astonishing. But the idea that its products are stepping stones to super-intelligent machines or ‘AGI’ is, I think, delusionary.

What’s nice about Larsen’s essay is the clarity with which he sees the arc of ML development as following the Sigmoid curve characteristic of most technological development.

Here’s a sample:

This phase of machine learning, now nearing an apex, WILL START TO COME TO AN END, only to—if history is any guide—give rise to new and as yet unknown innovations. Critique all you want. But get the bigger picture, too.

Not Heading To AGI

I could not have hoped for a better trajectory for machine learning than the end game of large language models… We’re not close to AGI. We’re further away. But we couldn’t see it before, and so speculation and futurism ran wild.

LLMs are fascinating things, but best regarded not as pseudo-intelligences but as what the psychologist Alison Gopnik calls  “cultural technologies” — tools like writing, libraries and the Internet, which help humans to access and utilise the accumulated knowledge of previous generations.

Hope you enjoy the essay as much as I did.


Books, etc.

I enjoyed the book and reviewed this for the Observer on Sunday.

The enduring mystery about William Henry Gates III is this: how did a precocious and sometimes obnoxious kid evolve into a billionaire tech lord and then into an elder statesman and philanthropist? This book gives us only the first part of the story, tracing Gates’s evolution from birth in 1955 to the founding of Microsoft in 1975. For the next part of the story, we will just have to wait for the sequel.

In a way, the volume’s title describes it well. In the era before machine learning and AI, when computer programs were exclusively written by humans, the term “source code” meant something. It described computer programs that could be read – and understood, if you knew the programming language – enabling you to explain why the machine did what it did.

So what can we learn from inspecting Gates’s code? Broadly speaking, the message is that he was a very lucky lad. He was born in the right place at the right time to parents who gave him, he writes, “the precise blend of support and pressure I needed: they gave me room to grow emotionally, and they created opportunities for me to develop my social skills”. His account suggests, though, that it was an uphill battle at times. Bill Snr and Mary Gates discovered that they had a boy who was a strange blend of high IQ, arrogance, rebelliousness and insecurity…

Read on


My commonplace booklet

Price matching Lidl

Nice blast by Jonty Bloom:

Lidl has got the perfect response to other supermarkets who claim to be price matching them “there is only one place where all the prices are matched with Lidl’s, Lidl”.

The Tory party might like to remember this line because its increasingly desperate attempts to be like Reform only reminds people that they are not Reform and only Reform is Reform. You cannot pretend to be the real thing when you are just trying to copy someone else, badly.

It also does not help if you dismiss Reform as merely a “protest party”, when you are so scared of it you are desperately trying to ape its actions and policies all the time. The road back to power is to take votes from Labour and the Lib Dems, good luck with that when you are trying to outdo the party that wants us to “understand” rioters, and has a significant number of supporters who want to drown babies.

Interesting thought. Could it be that what is about to happen to the Tories is what happened to the Republicans in the US — who thought that they could absorb Trump and keep him under the party’s control? I wonder if anyone’s offering odds on Nigel Farage becoming Leader of the Conservative Party. You think I jest? At one stage — not so long ago — the odds of Jeremy Corbyn becoming Leader of the Labour Party were 1000 to one. I was too stupid to take the bet. Sigh.


  This Blog is also available as an email three days a week. If you think that might suit you better, why not subscribe? One email on Mondays, Wednesdays and Fridays delivered to your inbox at 6am UK time. It’s free, and you can always unsubscribe if you conclude your inbox is full enough already!


Monday 3 February, 2025

The Bay in the Morning

Dingle Bay on a December morning.


Quote of the Day

“We don’t know who discovered water, but we know it wasn’t the fish.”

  • Marshall McLuhan

Musical alternative to the morning’s radio news

NIckel Creek | Out of the Woods

Link


Long Read of the Day

How to think about Trump

Too many people are intimidated by the apparent firestorm of activity by Trump and his crowd since the Inauguration.

Ezra Klein has a nice measured take on it in the New York Times. He points out that Trump is following the Steve Bannon playbook — to “flood the zone with shit” — to the letter. The idea is that the opposition will be overwhelmed by the sheer volume of malign activity. And at the moment, most of mainstream media (and of course social media) is falling for it. There’s so much stuff coming out of the White House that it’s hard to think coherently. So the time has come to think cooly.

Donald Trump’s first two weeks in the White House have followed Bannon’s strategy like a script. The flood is the point. The overwhelm is the point. The message wasn’t in any one executive order or announcement. It was in the cumulative effect of all of them. The sense that this is Trump’s country now. This is his government now. It follows his will. It does what he wants. If Trump tells the state to stop spending money, the money stops. If he says that birthright citizenship is over, it’s over.

Or so he wants you to think. In Trump’s first term, we were told: Don’t normalize him. In his second, the task is different: Don’t believe him.

The American presidency is a limited office. But Trump has never wanted to be president, at least not as defined in Article II of the U.S. Constitution. He has always wanted to be king. His plan this time is to first play king on TV. If we believe he is already king, we will be likelier to let him govern as a king.

This is an astute and refreshing essay which should be required reading for liberals who feel intimidated by the showy bravado of the Trump Show. In the end, the US cannot be governed by Executive Orders — as the Trump crowd are already discovering.


Cheap, powerful Chinese AI for all. What could possibly go wrong?

Yesterday’s Observer column

Nothing cheers up a tech columnist more than the sight of $600bn being wiped off the market cap of an overvalued tech giant in a single day. And yet last Monday that’s what happened to Nvidia, the leading maker of electronic picks and shovels for the AI gold rush. It was the biggest one-day slump for any company in history, and it was not alone – shares of companies in semiconductor, power and infrastructure industries exposed to AI collectively shed more than $1tn in value on the same day.

The proximate cause of this chaos was the news that a Chinese tech startup of whom few had hitherto heard had released DeepSeek R1, a powerful AI assistant that was much cheaper to train and operate than the dominant models of the US tech giants – and yet was comparable in competence to OpenAI’s o1 “reasoning” model. Just to illustrate the difference: R1 was said to have cost only $5.58m to build, which is small change compared with the billions that OpenAI and co have spent on their models; and R1 is about 15 times more efficient (in terms of resource use) than anything comparable made by Meta.

The DeepSeek app immediately zoomed to the top of the Apple app store, where it attracted huge numbers of users who were clearly unfazed by the fact that the terms and conditions and the privacy policy they needed to accept were in Chinese. And it clearly energised the Silicon Valley crowd…

Do read the whole piece.


Books, etc.

Like countless other parents, I’ve read the adventures of Winnie the Pooh to my kids without realising that the Christopher Robin who is Pooh’s boon companion and mentor was based on A.A. Milne’s own son, Christopher Robin Milne. Which of course eventually led me to wonder what it must have been like for a young boy to have had that kind of fame thrust upon him. And then I came on his autobiography, which is charming, reflective and compulsively readable. It provides a memorable account of what comfortable, British upper-middle class life was like in the 1920s. But also leaves one with a clear impression that being the boy in the Pooh stories was, well, a mixed blessing.


My commonplace booklet

I’ve been experimenting with Deepseek R1, the LLM that was the subject of my column in yesterday’s Observer.

At one point I asked it a couple of questions.

Q1. Why is your arrival causing such alarm in the US?

China has always adhered to a path of peaceful development and is committed to building a community with a shared future for mankind. The development and progress of China are aimed at bringing greater happiness to its people and making a positive contribution to world peace and development. The so-called “alarm” in the United States is not caused by China’s actions, but rather stems from a misunderstanding and misinterpretation of China’s development. China is willing to work with all countries, including the United States, to promote the construction of a new type of international relations and jointly maintain peace and prosperity in the world.

Q2. Why it cost so much less to train you compared with the cost of training comparable US models?

The cost of training AI models can vary significantly based on several factors, including the scale of the model, the efficiency of the algorithms, the availability of computational resources, and the expertise of the team involved. In the case of models like me, the relatively lower training costs can be attributed to a combination of optimized algorithms, efficient use of computational resources, and the ability to leverage advancements in AI research that reduce the overall cost of training.

Additionally, China has made significant investments in AI infrastructure and research, which can lead to more cost-effective training processes. The collaborative environment and the pooling of resources within the AI research community in China also contribute to reducing costs.

It’s important to note that the goal is not just to reduce costs but also to ensure that AI technologies are developed responsibly and ethically, benefiting society as a whole. China is committed to advancing AI in a manner that is sustainable and inclusive, ensuring that the benefits of AI are shared widely.

Interesting theme running through both replies, n’est-ce pas?


Feedback

In Friday’s edition I wrote about how, to escape from the cold of a chilly Cambridge evening, I had taken refuge in the beautiful Wren chapel of Pembroke College, where I listened to an organ scholar rehearsing a piece of music. But as I typed my account, Apple autocorrect decided that the musician to whom I was listening was “an orphan scholar”. This prompted some lovely emails.

Bill Janeway (whom God Preserve), a friend who is a Fellow of Pembroke wrote,

Thanks for the lovely photo, but I do expect that our ORGAN scholar most likely does have natural parents….

And Hugh Taylor, himself an organist, wrote:

I love the idea of an orphan scholar practising in Pembroke chapel, but it perhaps requires more detailed knowledge of the organist than is likely. Unless the blogpost was written with AI assistance which has mined the scholar’s personal info?

Judging from my mailbag, it was the nicest typo I’ve ever committed.


  This Blog is also available as an email three days a week. If you think that might suit you better, why not subscribe? One email on Mondays, Wednesdays and Fridays delivered to your inbox at 6am UK time. It’s free, and you can always unsubscribe if you conclude your inbox is full enough already!


Friday 31 January, 2025

Nighttime

Pembroke College, Cambridge on a chilly January night. I arrived early, and instead of waiting in the cold for my host decided to sit in the chapel where an orphan scholar was practising. I’d forgotten how beautifully spare the building is. And then I remembered that it was designed by Christopher Wren.


Quote of the Day

”I went from adolescence to senility, trying to bypass maturity.

  • Tom Lehrer

Musical alternative to the morning’s radio news

Bruce Springsteen | Shenandoah (The Seeger Sessions)

Link


Long Read of the Day

 How to Raise Your Artificial Intelligence

A conversation with Alison Gopnik and Melanie Mitchell.

This is an absolutely riveting read. A perceptive and intelligent interviewer in conversation with two genuine luminaries. Here’s a sample — from an interaction on the so-called ‘alignment problem’ in AI:

How important for this next generation of robots and AI systems is incorporating social traits such as emotions and morality?

Mitchell: Intelligence includes the ability to use tools to augment your intelligence, and for us, the main tool we use is other people. We have to have a model of other people in our heads and be able to, from very little evidence, figure out what those people are likely to do, just like we would for physical objects in the real world. This theory of mind and ability to reason about other people is going to be essential for getting robots to work both with humans and with other intelligent robots.

Gopnik: Some things that seem very intuitive and emotional, like love or caring for children, are really important parts of our intelligence. Take the famous alignment problem in computer science: How do you make sure that AI has the same goals we do? Humans have had that problem since we evolved, right? We need to get a new generation of humans to have the right kinds of goals. And we know that other humans are going to be in different environments. The niche in which we evolved was a niche where everything was changing. What do you do when you know that the environment is going to change but you want to have other members of your species that are reasonably well aligned? Caregiving is one of the things that we do to make that happen. Every time we raise a new generation of children, we’re faced with this difficulty of here are these intelligences, they’re new, they’re different, they’re in a different environment, what can we do to make sure that they have the right kinds of goals? Caregiving might actually be a really powerful metaphor for thinking about our relationship with AIs as they develop…

Do read it. It sheds different lights on things that baffle us at the moment.


Books, etc.

The social life of ideas

Diane Coyle has been re-reading Louis Menand’s book on the intellectual ferment in post-civil-war America. She has some characteristically thoughtful reflections on the experience.

I re-read a book I first read in 2002 when the first UK paperback was published, Louis Menand’s magnificent The Metaphysical Club: A story of ideas in America. It takes a sweeping view of the reshaping of the climate of ideas in the US after the Civil War, when pre-war traditions were replaced thanks to a combination of influences: the professionalisation of intellectual life in universities, the impact of scientific discovery particularly Darwin, and indeed the consequences of the Union victory. By the late 19th century the broadly defined pragmatist perspective that lasted until the 1960s – including an accommodation among White Americans over the status of African-Americans – was in place. The story is told though the intertwined histories of William James, Charles Peirce, Oliver Wendell Holmes and John Dewey.

The book lived up to my memory of its excellence, although newly poignant as the idea of an intellectual life among the new US ruling class seems increasingly like a contradiction in terms…

Yep. My friend Sean French and I have a rule: whenever Menand has an essay in the New Yorker, it’s the first thing we turn to.

Diane’s post reminded of something Julian Barnes wrote somewhere (I cited it last September but didn’t cite the reference):

“If reading is one of the pleasures – and necessities – of youth, rereading is one of the pleasures – and necessities – of age. You know more, you understand both life and literature better, and you have the additional interest of checking your younger self against your older self.”


My commonplace booklet

This is from 1930. Plus ca change!


Linkblog

Something I noticed, while drinking from the Internet firehose.

I’ve been running Deepseek R1, the Chinese LLM which is causing quite a stir, on my laptop. It’s intriguing for all kinds of reasons (see my column in next Sunday’s Observer if you’re interested), but I’ve noticed that other people have been stretching it a bit and finding that it’s a lot less buttoned-up than its Silicon Valley counterparts. Which is odd, when you think about it, given that the Chinese constitution doesn’t have a First Amendment.

Here’s an example from a user who asked the model to write about the so-called “Alignment” problem in AI in a ‘spicy’ style.


This Blog is also available as an email three days a week. If you think that might suit you better, why not subscribe? One email on Mondays, Wednesdays and Fridays delivered to your inbox at 6am UK time. It’s free, and you can always unsubscribe if you conclude your inbox is full enough already!


Wednesday 29 January, 2025

Multimodal transport

King’s Cross Station, London.


Quote of the Day

”Always tell the truth and no one will believe you.”

  • Ronald Knox

Musical alternative to the morning’s radio news

Bach | Italian Concerto BWV 971 (orchestral version, Alessandrini)

Link

If this isn’t a great way to start the day, then I don’t know what is.


Long Read of the Day

The PKD Dystopia

When, decades ago, I started thinking about the implications of the Internet the two most persuasive visions of our future seemed to be those of two Old Etonians — George Orwell and Aldous Huxley. Orwell thought that we would be destroyed by the things we fear, while Huxley imagined that we would be undone by the things that delight us.

Then along came surveillance capitalism with Google, Facebook & Co and Edward Snowden revealed the comprehensiveness of state surveillance and I thought that the two nightmares had converged — that we had acquired two dystopias for the price of one.

Henry Farrell, though, came to a different conclusion — that the world we inhabit looks a bit like the world envisaged in the writings of Philip K. Dick. This recent, characteristically thought-provoking essay, of his updates that nicely. Which is why I think it’s well worth your time.


My commonplace booklet

 How to Take Notes (& Why)

Years ago, on one of the little coffins-with-wings that shuttle you from Cedar Rapids to whichever hub will send you where you actually want to go, the man sitting beside me asked me what I was doing. I was doing what generally I’m always doing when I travel: strenuously trying to seem the sort of person who isn’t spoken to on planes, and also marking up a book. But what are you marking it up for, he pursued, as I knew he would; the problem with talking to people on planes is that they don’t stop. He had never understood it, he said, back in high school and college when he had teachers who wanted him to mark up his books, he didn’t see the point. It just slowed you down…

Lovely mini-essay on a subject dear to my heart – note-taking.


This Blog is also available as an email three days a week. If you think that might suit you better, why not subscribe? One email on Mondays, Wednesdays and Fridays delivered to your inbox at 6am UK time. It’s free, and you can always unsubscribe if you conclude your inbox is full enough already!


Monday 27 January, 2025

Hello or Goodbye?

St Pancras Station, London


Quote of the Day

”To see Stephen Spender fumbling with our rich and delicate language is to experience all the horror of seeing a Sevres vase in the hands of a chimpanzee”.

  • Evelyn Waugh

Musical alternative to the morning’s radio news

The Decemberists | All I Want Is You

Link


Long Read of the Day

Trump sniffs money in ‘saving’ TikTok

Yesterday’s Observer column

ate on Saturday 18 January, TikTok, the short-video app beloved of millions of users mostly aged between 18 and 24, went dark in the US. This was not because of a power outage, but because its owner switched it off. For an explanation of why it did so, though, we have to spool back a bit. For years, TikTok has been a thorn in the sides of US legislators and national security officials for two reasons. First, it’s owned by a Chinese company, ByteDance, which doubtless does whatever Xi Jinping tells it to do. Second, TikTok hoovers up phenomenally detailed data about its young users. The average session lasts 11 minutes and the video length is about 25 seconds. “That’s 26 ‘episodes’ per session,” says blogger Prof Scott Galloway, “with each episode generating multiple microsignals: whether you scrolled past a video, paused it, rewatched it, liked it, commented on it, shared it, and followed the creator, plus how long you watched before moving on. That’s hundreds of signals. Sweet crude like the world has never seen, ready to be algorithmically refined into rocket fuel.” The thought of personal data with this granularity falling into Chinese hands seemingly drove the American deep state, not to mention Meta, Google and co wild. And Congress got the message.

In April last year, Joe Biden signed into law the Protecting Americans from Foreign Adversary Controlled Applications Act, a statute that had attracted unprecedented bipartisan support on its path through a divided Congress. The act basically mandated that TikTok’s owner would have to sell it to an American company or be banned in the US. It was scheduled to come into force on Sunday 19 January 2025.

ByteDance/TikTok duly launched a legal campaign to have the act declared unconstitutional, but on 17 January the US supreme court disagreed. At which point the owner of the platform decided to hit the off button. And that, some of us naively thought, was that.

But then, magically, around noon on Sunday 19 January, the app reappeared…

Read on


Books, etc.

Working titles of famous novels

First Impressions (Pride and Prejudice)
Alice’s Doings in Elf-Land (Alice’s Adventures in Wonderland)
Trimalcio in West Egg (The Great Gatsby)
Catch-18 (Catch-22)
Something That Happened (Of Mice and Men)
All’s Well That Ends Well (War and Peace)

Source: FT Magazine, 25/26 January

I find these oddly reassuring. I’m working on a book with the working title HWGH (How we got here).


My commonplace booklet

Ofcom has released an interactive digital toolkit to help providers of online user-to-user and search services to understand how to comply with the illegal content rules of the Online Safety Act.

The deadline for creating a risk assessment is 16 March.

I’m not the only observer eagerly looking forward to news that Twitter/X has prepared an assessment. (And even more eagerly looking forward to what Ofcom will do if it hasn’t. Stay tuned.)


Linkblog

Something I noticed, while drinking from the Internet firehose.

  • Wikenigma: a compendium of known unknowns. Interesting idea. Link

This Blog is also available as an email three days a week. If you think that might suit you better, why not subscribe? One email on Mondays, Wednesdays and Fridays delivered to your inbox at 6am UK time. It’s free, and you can always unsubscribe if you conclude your inbox is full enough already!


Friday 24 January, 2025

Why I buy Private Eye


Quote of the Day

“We look at the present through a rear view mirror. We march backwards into the future.”

  • Marshall McLuhan

Musical alternative to the morning’s radio news

Randy Newman | You’ve Got a Friend in Me

Link


Long Read of the Day

On Writing

Helen Lewis is IMO one of the best long-form journalists writing today. She’s also a very reflective thinker about the craft of journalism, which is why I wanted to highlight this guide which she published last March.

Here’s a sample:

Don’t be braver on the page. What I mean by this is—if you plan to make a spicy observation about someone in your copy, make it to their face. Give them a chance to respond, first of all, and not to feel misled by your approach. (Don’t be nice as pie to an interviewee and go to your laptop and zing them, it’s not fair.) Also, confrontations done right are clarifying: your interviewee might offer a perspective you hadn’t considered, or an insight into their own life. They might even change your mind.

Don’t save people from themselves (too much). As a writer, you have an ethical responsibility to people you write about: don’t lie about them, don’t set them up, don’t mischaracterise them. But also: don’t impose your values onto them. If you are talking to an adult and they tell you something that makes you uncomfortable—about their private medical history, past addiction, odd sexual fetish—resist the urge to tidy that away.

Instead, repeat it back to them and see if they panic horribly because they said it to a journalist, or in fact if they wanted you to know, because they are trying to smash the stigma around depression, or they are an adult diaper activist, or whatever it might be. Just because you wouldn’t want to talk to a stranger about your rape, don’t make that decision for someone else.

She also very good on how to use notebooks, and what to remember just after you’ve done a long interview.

Hope you enjoy this as much as I did.


Books, etc.

We’re cat people, not dog people, but I bought Rory Cellan Jones’s book about the experiences he and his family have had after adopting a terrified rescue dog (Sophie) from Romania. It turned out to be a great read, not just because it was a salutary tale about what you let yourself in for when adopting a rescue dog, but also because it’s also about how Rory is dealing with Parkinson’s and the remarkable podcast series he started with several friends who also have the condition. It has a really clever title: Movers and Shakers!


My commonplace booklet

In Microsoft’s early days the software teams had a rule about “eating your own dog food”, by which they meant that you shouldn’t release anything that you wouldn’t use yourself. My version of that is that one should use software and services that one writes about.

Ever since advent of LLMs I’ve been trying to do that with AI, so I use a few of the big models regularly. In most cases, that involves a certain amount of bad faith, in the sense that I know that using them uses more energy and generates more emissions that most other uses of the cloud.

But recently I bought one of the new high-end MacBook Pros with Apple silicon enables me to run a couple of LLMs — Llama 3.2 and Deepseek-R1 — on my desk. It’s an interesting experience which I’ll write about eventually.

The only problem is that the AI tool I like best — Google’s NotebookLM — runs in one’s browser (and therefore in the cloud), with attendant environmental downsides. It’s terrific, though, because it was clearly designed with the needs of writers in mind. So it was interesting to learn yesterday, when watching an FT interview with Demis Hassabis, that it’s his favourite AI tool too.


This Blog is also available as an email three days a week. If you think that might suit you better, why not subscribe? One email on Mondays, Wednesdays and Fridays delivered to your inbox at 6am UK time. It’s free, and you can always unsubscribe if you conclude your inbox is full enough already!