Rajesh Jain

Life Notes #15: Memories – 2

A Schoolteacher and a Professor: In a week recently, I met a schoolteacher and a professor from IIT. Both had left a mark on me with their teaching, and it is just one of life’s delightful coincidences that I got to meet both separately. I have many happy memories of St. Xaiver’s school – I wrote about them on my previous blog in 2010. [Parts 1, 2, 3, 4.] As Bhavana, Abhishek and I spent a wonderful afternoon with my schoolteacher, memories of school came flooding by. I was back in her classroom, sitting on a bench, paying attention, listening, and learning. Those were the formative years of our lives. A week later, I was at an IIT Alumni event, and I met a professor whom I saw after almost 35 years. And for a moment, I was a student again. His teaching was something I loved – because it was so different and alive. Even though we only got a few minutes together, it was a meeting to remember for a lifetime. As I was walking out, I told him, “Thank you very much. You made me what we I am today.” And the professor’s words are ones I will not forget, “Rajesh, and so did you. A professor is also made by students.”

Mumbai Rains: Come June, and so do the Mumbai rains. Just yesterday, we were driving on the Sea Link, and the intense downpour reminded me of how magical Mumbai’s rains are. When it pours, it really comes down. And all one can do is watch. This is the way it has been every rainy season. There was a time I meticulously tracked the daily quantity of rain – Times of India had a table showing how much rain fell over the past 24 hours in different cities of Maharashtra. I was always envious of Matheran and Mahabaleshwar because they were always ahead of Mumbai! Every year, there would at least be one occasion when I would get thoroughly drenched despite having an umbrella. As one grows older, the rains become background noise – and some of nature’s wonders are lost on us.

Homes: During May, Bhavana, Abhishek and I were searching for a new home. We have lived in our current home for 20 years. And perhaps it is time to explore something different a few years down the line. My first home that I remember was at Taikalwadi in Mahim. Then, we moved to Nepean Sea Road, and that was home for over 30 years, until we moved to our current home at Kemp’s Corner. We may not realise it, but our homes and neighbourhoods mould us. They share our happy moments and sad ones, we have our favourite spots and the ones where we sit when we are down and feeling let down by the world. For me, it was the pandemic which created a much greater attachment to the home. The table where I sit to write this was bought during the early days of the pandemic as I needed to create makeshift working space. All my writing is done here. The view of the Arabian Sea expanse, the garden behind, the new buildings coming up nearby. And memories of me growing up, me raising Abhishek. Our homes are us.

Thinks 1282

Tyler Cowen on “Best Things First”: “The author is Bjorn Lomborg, and the subtitle is “The 12 most efficient solutions for the world’s poorest and our global SDG promises”. Here is what Lomborg presents as the twelve best global investments, in no particular order: Tuberculosis, Maternal and newborn health, Malaria, Nutrition, Chronic diseases, Childhood immunization, Education, Agricultural R&D, e-procurement, Land tenure security, Trade, Skilled migration.”

David Friedman: “One of the things you learn in economics, and life, is the principle of revealed preference: Believe what people do, not what they say.” More: “One effect of being brought up as I have described is to make you less willing to be persuaded by bad arguments, even when offered by high status people. Another is to make you more willing to be persuaded by good arguments, even ones nobody else is making.”

Indrajit Gupta: “The physical store has a new role: Not just showcase what is available, perhaps help the customer to try it out — but also offer the added benefit of closing the sale, in case the customer is open to having it delivered from their nearest store or warehouse. It isn’t without reason that blended commerce is emerging as the holy grail of retail. Especially since covering a country as large and diverse as India has its challenges. The opportunity in Tier-II towns and beyond is starting to open up. Major retailers are scrambling to open physical stores, but that invariably takes time. Instead, a thoughtfully crafted combination of e-commerce and physical stores could help bridge the gap.”

Mint: “One of the most critical yet often overlooked parts of the reform process is determining what should be the ideal ‘unit of change’ of the proposed transformation. What do we mean by this? A ‘unit of change’ is a component of a system that’s the focal point for transformation efforts. It could be a geography, a function, a process, an institution, a policy or any other distinctly identifiable element that, when modified, is capable of catalysing broad systemic improvements. By way of example, consider reform efforts in education. One approach could be to target the ‘school’ as the ‘unit of change.’ When this is the case, the reform agenda could include developing new curricula, different pedagogy or making changes in the administrative processes of the school. On the other hand, if the ‘unit of change’ is the child, reform efforts would focus on improving learning and development outcomes. Identifying the correct unit of change is crucial for the success of a given reform, as it sets the stage for the depth, scale and scope of transformation.”

Economist: “Might it be better to design specialist AI chips from scratch? That is what many companies, small and large, are now doing in a bid to topple Nvidia. Dedicated AI chips promise to make building and running AI models faster, cheaper or both. Any firm that can mount a credible threat to the reigning champion will have no shortage of customers, who dislike its lofty prices and limited supplies.”

Life Notes #14: Memories – 1

As one grows older, reminiscing about the past tends to increase! In this sub-series, I will write occasionally about memories that come up.

Bangkok: In late April, I went to Bangkok for a Netcore international sales meet. As we drove from the airport to our hotel, many memories came flooding by. Bangkok was a destination Bhavana and I used to visit a lot in the late 1990s and early 2000s (before Abhishek was born). After Abhishek was born, and our international travels (during summer vacations) ended up being mostly to New York or Singapore because there was a lot more for all of us to do. We had made a trip to Bangkok and Phuket a few years ago (during which I bought my BB), but that was now the exception than the norm. Bangkok, just a 5-hour flight away, was a different world from India. We used to stay at the Riverside Marriott. Bhavana liked the shopping experiences, and for me it was a good break from work. Taking the boat ride across the river was such a calming experience. I also liked the Kinokuniya bookstore which had a good collection of English books. Bangkok, even with its traffic, reminded me of what Indian cities should have become. We would visit the various markets, especially the night market. The city was (and is) a shopping hub. Even today, that allure stays. At 28 million, Thailand gets almost four times the tourists India gets annually.

Old Mumbai-Pune Highway: In early May, Abhishek and I were going to Talegaon to see a factory of a company a friend had invested in. (Abhisek is keen to do Industrial Engineering and wanted to see some manufacturing happen.) Because of some traffic on the Expressway and the fact that we needed to go to Talegaon, we took the old Mumbai-Pune highway. I had not travelled on that road for over 20 years – since the Expressway started in 2002. We passed Khopoli and I was reminded of the obligatory stops at Ramakant just before the ghats. The old highway, as it made its way slowly up, always offered spectacular views, especially during the rains. We passed a small temple where I remembered throwing coin for good luck, as we motored along first in our Ambassador car, then a Fiat, then a Maruti. The speedy expressway and its ‘professionalism’ has perhaps taken away some of the charm of the ride through the ghats – one almost doesn’t notice them because the speed barely reduces. For a brief time that day, I was a teenager in the 1980s looking out of the car window and enjoying the beautiful views of the world below.

Factories: Talking of factories, I was also reminded of my childhood summer and Diwali vacations spent at Abu Road, against the whirring of marble cutting and polishing machines. My father had set up a marble and granite unit in Rajasthan at the foothills of Mount Abu. Since he needed to visit, we (my mother, sister, and sometimes, a few cousins) would go to Abu. Even as we visited a few temples, there were visits to marble mines. And then stay at the factory! During the summer heat, we slept in charpoys in the open which meant the noise from the machines was even more. The Talegaon factory visit also reminded me that there is a world beyond software and finance, a world where real stuff is made.

Thinks 1281

WSJ: “Politicians and the press mislead voters and readers when they claim that tax cuts for the rich don’t benefit other economic classes. We all gain from new, improved products made possible by innovative startups funded by the wealthy. Excessive taxation, doubtless a feature of a “middle-out” plan, could deplete the funds that entrepreneurs use to start and sustain useful ventures. Americans shouldn’t worry so much about wealth distribution. Instead, we should be grateful for how the wealthy enable entrepreneurial ideas to come to life, allowing everyone to prosper.”

Noah Smith: “[Here is] a theory of how totalitarianism might naturally triumph. The basic idea is that when information is costly, liberal democracy wins because it gathers more and better information than closed societies, but when information is cheap, negative-sum information tournaments sap an increasingly large portion of a liberal society’s resources. Remember that I don’t believe this theory; I’m merely trying to formulate it.”

Ryan Bourne: “Price controls are making a comeback across the United States. Federal, state, and local governments increasingly control rents, minimum wages, interest rates on short-term loans, healthcare prices and premiums, credit card late fees and even food delivery service charges. The sharpest inflation burst since the early 1980s has also seen Democrats demand a federal anti-price gouging law, anti-”junk fees” rules, and even efforts to prevent companies engaging in “shrinkflation” or algorithmic dynamic pricing. I call all this The War on Prices, the title of a new Cato Institute book. And when it comes to this war, economists are typically pacifists. The long sweep of history, from ancient Egypt to modern America, shows us that price controls can’t quell inflation, because they don’t fundamentally change the money supply or aggregate production. What controls on market prices guarantee is inefficiency. They squelch the delicate coordination mechanism that prices and their movements provide to encourage economical action. Price ceilings, the most common incarnation, are thus a tried-and-tested recipe for shortages, declines in product quality, and black markets.”

Anirban Mahapatra: “A study published in Frontiers in Psychology recently found that students who took handwritten notes had higher levels of brain activity across the regions responsible for movement, vision, sensory processing, and memory. In contrast, typing led to minimal activity in these areas. Researchers Audrey van der Meer and Ruud van der Weel suggest handwriting forces students to process information more deeply. When you’re writing by hand, you can’t copy everything down verbatim, so you have to think as you write, unlike typing, where the temptation to transcribe lectures word-for-word is high. In addition, handwriting reinforces memory and learning pathways. The action of forming letters creates a feedback loop with our visual and sensory systems, embedding information more deeply in our brains. This process is like drawing or building something. In all these cases, the process helps strengthen the concept and makes it stick in our memory.”

VentureBeat: “Researchers from the University of Chicago have demonstrated that large language models (LLMs) can conduct financial statement analysis with accuracy rivaling and even surpassing that of professional analysts. The findings, published in a working paper titled “Financial Statement Analysis with Large Language Models,” could have major implications for the future of financial analysis and decision-making. The researchers tested the performance of GPT-4, a state-of-the-art LLM developed by OpenAI, on the task of analyzing corporate financial statements to predict future earnings growth. Remarkably, even when provided only with standardized, anonymized balance sheets, and income statements devoid of any textual context, GPT-4 was able to outperform human analysts. “We find that the prediction accuracy of the LLM is on par with the performance of a narrowly trained state-of-the-art ML model,” the authors write. “LLM prediction does not stem from its training memory. Instead, we find that the LLM generates useful narrative insights about a company’s future performance.””

Life Notes #13: Gym

A few months ago, I started gym training. My sister, a doctor, has been telling me for many years that I need to do more than just a 30-minute walk five times a week. Her point was that the body starts losing muscle mass after the age of 40, and unless I do strength training, things could become harder as I grew older.

Here is the medical explanation from Henry Ford Health: “As we age it’s normal to experience some reduction in muscle mass, strength and function, a condition known as sarcopenia. These changes begin as early as your 30s and continue at a rate of 3% to 5% per decade. The good news is that strength training can help you maintain and rebuild muscle at any age. Research shows that older adults see even greater improvements in their muscle strength versus younger adults. “So it’s never too late to start,” says Pamela Webert, MS, ACSM-CEP, an exercise physiologist at Henry Ford Health. “Everyone should be doing strength training as part of their exercise program.””

We have a gym in our building. I never once entered it during the past two decades. Just seeing all the equipment through the glass door scared the daylights out of me! Walking was so much easier.

Last October, after another push by my sister, I decided to change. Under the guidance of a trainer, I now workout for about 35-40 minutes three times a week. The first few sessions were tough given that my body had seen no real exercise through life (except for a few years of yoga a long time ago). Having a trainer come is good because there is no “I will do it tomorrow” excuse. Now, even though I don’t fully enjoy it (!), it has become part of my weekly routine. As one grows older, the realisation starts dawning that good health must become a priority.

I asked ChatGPT to list the physical health benefits of gym training:

Improved Muscle Strength and Tone: Regular weight training helps build and maintain muscle mass, which naturally declines with age. This can enhance your strength, endurance, and overall physical performance.
Increased Bone Density: Weight-bearing exercises, such as weight lifting and resistance training, can help increase bone density, reducing the risk of osteoporosis and fractures.
Enhanced Cardiovascular Health: Incorporating cardio exercises, like treadmill walking or cycling, can improve heart health, lower blood pressure, and reduce the risk of heart disease.
Better Joint Health and Flexibility: Strengthening the muscles around your joints can improve stability and reduce the risk of injuries. Stretching and flexibility exercises can help maintain joint flexibility and reduce stiffness.
Weight Management: Regular exercise helps burn calories and can aid in maintaining a healthy weight or losing excess weight, which is crucial for overall health.

So, one piece of advice: don’t wait until it’s too late. Begin training now (especially if you were over 50 years of age).

Thinks 1280

WSJ: “The tech labor market is in an unbalanced state. There is demand for a specific type of tier-one AI talent—namely those who have the technical knowledge or experience working with large language models, or LLMs, that fuel chatbots with the ability to generate content. There are companies seeking candidates with those skills, but not enough workers who are qualified to do them. Then there is everyone else. Thousands of people have been laid off in the past few years, and many of those who remain employed are dealing with new management styles, reorganizations and microcuts, as more resources get shifted into AI. Those workers are now taking courses in AI, adding buzzwords to their résumés and competing in an increasingly crowded field.”

NYTimes: “In the fall of 2020, Amazon looked unstoppable. The pandemic lockdowns had supercharged its e-commerce business, and executives were busy hiring an army of new workers and expanding the number of Amazon fulfillment warehouses across the US. But the tech giant would soon face the biggest threat to its online shopping empire in recent memory, and it would come from an unlikely place: China. Over the last few years, two Chinese-owned e-commerce platforms—Shein and Temu—have quietly become some of the most popular shopping websites in the US. Shein specializes in low-end women’s fashion, while Temu focuses on home decor and household items—the same type of affordable, made-in-China products that many American consumers have come to associate with Amazon.”

Mint: “Systems that rely on more than one chatbot or use multiple Large Language Models (LLMs) are called multi-agent systems (MAS). A chatbot on the website of a bank, auto, insurance, edtech or any other company has improved over the years, thanks to advances in AI and natural language processing. But it still falls short of answering questions that might require more ‘human-like’ capabilities or pulling data from different domains. The solution lies in using multiple bots or more than one AI (artificial intelligence) model—each with a different, specialized capability, and complementing each other to deliver a better response…Just as teams of humans are better in tackling complex problems, more than one chatbot will improve customer interactions. So, a Hindi language chatbot could combine with a math teaching bot and deliver math lessons in Hindi. Each is a separate LLM (which helps chatbots understand human input and offer answers) and is better at executing specialist tasks.”

Fei-Fei Li and John Etchemendy: “An LLM is a mathematical model coded on silicon chips. It is not an embodied being like humans. It does not have a “life” that needs to eat, drink, reproduce, experience emotion, get sick, and eventually die. It is important to understand the profound difference between how humans generate sequences of words and how an LLM generates those same sequences. When I say “I am hungry,” I am reporting on my sensed physiological states. When an LLM generates the sequence “I am hungry,” it is simply generating the most probable completion of the sequence of words in its current prompt. It is doing exactly the same thing as when, with a different prompt, it generates “I am not hungry,” or with yet another prompt, “The moon is made of green cheese.” None of these are reports of its (nonexistent) physiological states. They are simply probabilistic completions. We have not achieved sentient AI, and larger language models won’t get us there.”

NYTimes: “China’s industrial dominance is underpinned by decades of experience using the power of a one-party state to pull all the levers of government and banking, while encouraging frenetic competition among private companies. China’s unrivaled production of solar panels and electric vehicles is built on an earlier cultivation of the chemical, steel, battery and electronics industries, as well as large investments in rail lines, ports and highways. From 2017 to 2019, it spent an extraordinary 1.7 percent of its gross domestic product on industrial support, more than twice the percentage of any other country, according to an analysis from the Center for Strategic and International Studies.”

Life Notes #12: IndiaVotes and Elections

IndiaVotes is a website I had created in 2011 during the time I was working for the 2014 election campaign. I had found a gap in the easy availability of election data. While the raw data was available from the Election Commission, it was in PDFs and almost impossible to search. So, a group of us had extracted the data into a database and designed a nice navigation and search interface. I am glad to say that even after 13 years, it remains the best elections data website in India. A Google search for “india elections data” shows IndiaVotes as the #1 search result.

During the recent elections, traffic spiked by a magnitude. I was very surprised. Post 2014, while the site had been updated with the results of all the elections, the interface had not changed, and the site had become slow. I guess its simplicity and authenticity has helped make it popular and the go-to destination for elections data.

It is perhaps time for an overhaul of the site. It needs a Gen AI chat interface – “show me all the constituencies that were won by a margin of under 5%” is much easier than doing the same query on the results pages. It also needs better visualisation of results – we had tried maps once but somewhere it has got lost. Another idea is a “swing calculator” – so enthusiasts can see what would happen if voting patterns changed. I would also like to bring in Form 20 data so it becomes a permanent repository of granular elections data. The one innovation we had done during the recent Lok Sabha elections was a prediction game called YouPredict which attracted a few hundred players. I am hoping that over the next few months, we can create an improved IndiaVotes site – the best needs to become even better!

A few friends had called me for my take on the recently concluded Lok Sabha elections. My take: after two wave elections in 2014 and 2019, India has gone back to seat-by-seat, state-by-state, caste-by-caste voting, as was the norm between 1989 and 2014. This is what I had written about in my prescient 2011 essay about how the BJP could win 275 seats in the 2014. For any party to win a majority on its own in India, it needs to create a wave and have a very high hit rate in contested seats. This did not happen in 2024.

PS: Here is the IndiaVotes page summarising the 2024 results. An interesting page to see is the alliances summary – focus on the contested voteshare.

Thinks 1279

WSJ: “Researchers looked at more than 800,000 online reviews for beauty and fashion items listed on Amazon’s website. They found that online reviews written in the present tense were more likely to be “liked”—that is, receive a thumbs-up emoji from readers—than reviews written in the two other tenses. “A 50-word review only using verbs in the present tense is about 5.5% more likely to be ‘liked’ than a 50-word review using no present-tense verbs at all,” says David Fang, a doctoral student at Stanford University and one of the paper’s authors.” Also: “We found that consumers give, on average, a higher overall rating when asked to rate the overall experience and several attributes of the experience compared with when they are asked to rate the overall experience alone. In other words, when consumers can speak to what specifically they didn’t like about an experience, they weight those negative aspects less heavily in their overall ranking.”

Ruchir Sharma: “In his farewell address, Ronald Reagan described America as the “shining city on a hill”, open to “anyone with the will and heart to get here”. I was one of those inspired to try, and today the dynamic mix of academics and entrepreneurs who energise the world’s technology leader still strikes me as a marvel. Of the top 100 US companies, 10 now have chief executives who were born in my home country, India, a breakthrough that could have happened only in a capitalist meritocracy. Nonetheless, I worry about where the US is leading the world now. Faith in American capitalism, which was built on limited government that leaves room for individual freedom and initiative, has plummeted. Most Americans don’t expect to be “better off in five years” — a record low since the Edelman Trust Barometer first asked this question more than two decades ago. Four in five doubt that life will be better for their children’s generation than it has been for theirs, also a new low…America has become unhealthily dependent on loose money and big government.”

Wired: “[Microsoft’s] Bubeck says these results seem to indicate that making future AI systems smarter will require more than just scaling them up to still greater sizes. And it also seems likely that scaled-down models like Phi-3 will be an important feature of the future of computing. Running AI models “locally” on a smartphone, laptop, or PC reduces the latency or outages that can occur when queries have to be fed into the cloud. It guarantees that your data stays on your device and could unlock entirely new use cases for AI not possible under the cloud-centric model, such as AI apps deeply integrated into a device’s operating system.”

Ted Sarandos: “I think that A.I. is a natural kind of advancement of things that are happening in the creative space today, anyway. Volume stages did not displace on-location shooting. Writers, directors, editors will use A.I. as a tool to do their jobs better and to do things more efficiently and more effectively. And in the best case, to put things onscreen that would be impossible to do. Think about this gigantic leap from hand-drawn animation to computer-generated animation, and look how many more people animation employs today than it used to…I have more faith in humans than that. I really do. I don’t believe that an A.I. program is going to write a better screenplay than a great writer, or is going to replace a great performance, or that we won’t be able to tell the difference. A.I. is not going to take your job. The person who uses A.I. well might take your job.”

Economist: “A resumé is not a list of every job you ever had. It is not your autobiography. It is, like that hair-care advert, a marketing tool. Your audience is made up of recruiters and hiring managers. Like cocktail-party guests, they do not take a long time to decide if they want to keep talking. According to one study, such professionals spend an average of 7.4 seconds skimming a job application.”

Life Notes #11: Blogging and AIs

A couple of years ago, I had written about my blogging system. “At any time, I have a list of topics that I can write on. And that is good enough to get started. Once I start writing, the ideas flow. Some of the writing is original, while at other times I will aggregate what others have written in a single place. This reading and collating also helps me learn new topics. At times, I find myself going back to my earlier writings to refresh my own thinking.”

I have now settled into a routine. Most of my blogging happens in the early mornings on weekends. Sitting at my home desktop, I write for about 2-3 hours in a flow. (On weekdays, it’s either gym or walk so I don’t get that kind of contiguous time.) I will typically have thought through the essay theme, the sub-topics, and the opening sentences by the previous night. This way, I can begin writing without wasting precious time thinking about what to write.

Much of my recent writing has been around new ideas in marketing. Since I write for myself, there is no pressure to be perfect. Each essay moves my thinking forward. Even as I have a broad structure in mind, as I write the story takes its own course. This is what I like the most: the process of writing itself begets more writing. For example, during the process of thinking about how AI will impact marketing, many sub-themes came up: large customer models, co-marketer, and digital twins. I then also explore synthetic data and vector search, both of which were new to me. In email too, a similar process has followed. Writing about AMP led me to Email 2.0, 3.0 and eventually Action Ads and Epps (Email Apps). What I like most is naming the ideas. Velvet Rope Marketing, Generative Journeys, Co-Marketer, Inbox Commerce, Segment Twins, Singular Twins, Profipoly, Action Ads, Epps – these are all words and phrases I have come up with.

In the past year or so, one big change in my writing has been the use of AIs. I use a combination of ChatGPT, Claude, and Gemini to help with several tasks: exploring ideas, summarising my past writings, expanding and improving on briefs, helping write stories from my ideas, and preparing briefs to encourage others to read what I have written, critiquing what I have written, and adding new ideas to what I have written. Working with AIs has been the biggest change in my blogging process, and one that has helped make the process faster and better. It is almost as if I am working with a co-blogger: one who understands my thinking and works interactively with me. More than my specific voice (which is of course always there), what I want is to ensure a proper and full exploration of the ideas I am writing about. This is where the AIs excel.

The key to getting the most from the AIs is of course prompting. This is where my past writings and briefs come in. It is not just about “Write an essay about email apps” but more like “Here is all that I have written so far. Improve and expand on it.” To borrow a phrase from Kazuo Ishiguro’s “Klarna and the Sun”, the AIs have become my “Artificial Friend”. I have used the AIs to improve my titles and help me with naming (“Give me a better phrase for Singleton Twin starting with the letter “S” – and it came up with Singular Twin). I talk to it as if it’s a person working alongside me.

For me, blogging has become a way of life, an integral part of me. It frees my imagination, fuels ideas, and births innovations. The words I wrote four years ago still ring true: “I think every entrepreneur should write a blog. Not just tweets or pithy LinkedIn posts. But write about one’s ideas and aspirations in real-time. Give people a glimpse into the world that you see. Because that is what you are really doing – creating a future ahead of others. And blogging is a great way to accelerate that future.”

Thinks 1278

WSJ: “Getting angry doesn’t just hurt our mental health, it’s also damaging to our hearts, brains and gastrointestinal systems, according to doctors and recent research. Of course, it’s a normal emotion that everyone feels—few of us stay serene when a driver cuts us off or a boss makes us stay late. But getting mad too often or for too long can cause problems. There are ways to keep your anger from doing too much damage. Techniques like meditation can help, as can learning to express your anger in healthier ways.”

Economist: “The conventional language of career success moves in only one direction: up. You scale the career ladder or climb the greasy pole. If you do well, you have a rapid ascent. And if you really succeed, you reach the top. No one ever rings home to share the news that they have reached a plateau. But there is another type of career trajectory. Sideways moves, to jobs that don’t involve a promotion or even necessarily a pay rise, can be a boon to employees and organisations alike. A study carried out by Donald Sull of the Massachusetts Institute of Technology and his co-authors in 2021 found that the availability of lateral career opportunities has a marked impact on employee retention. Their research found that chances to move sideways were two and a half times more important than pay as a predictor of workers’ willingness to stay at a firm. Another paper, by Xin Jin of the University of South Florida and Michael Waldman of Cornell University, concluded that lateral moves did not just benefit organisations: employees who experienced them were more likely to be promoted and to enjoy higher wage growth later in their careers than employees who did not. You can move up by first moving sideways.” [via Mint]

Arnold Kling: “Suppose that within the next three years, I am using an AI-powered app to write this substack. I can tell the app to summarize substacks and other sources that I follow. I can tell the app to grab passages to quote. I can tell the app the main point I wish to make in an essay and have it compose a draft in my style. Once an app can do those things, I no longer need a personal computer. I just need a headset that can handle AI, using the cloud if necessary. I could keep up this substack while on a walk or on a bike ride. In the future, either AI realizes its promise or it doesn’t. If it does, then we will be accessing computer resources by using our voices, possibly with gestures. If AI doesn’t pan out, we will still need personal computers and Microsoft software, but an AI PC won’t be much of an improvement over the laptop I am using now. I see no future scenario in which masses of people will be using an AI PC. That is what I mean when I say that an AI PC is an oxymoron.”

WSJ on a great vacation: “You don’t do plan as much, you do less of what you like, you take a break from enjoying yourself, and you make yourself do things that are actually uncomfortable. If that sounds like the recipe for a lousy trip, that’s understandable. But science suggests otherwise. So my advice is to forget about what feels like the smart thing to do and embrace the opposite. The result may very well become the stories you cherish and share long after your return.”