There is a new (ish? to me?) malaria vaccine being used in Africa. It is effective but requires four doses over a period of months to remain effective long term, which presents some challenges. Researchers are working on a one-dose version.
A site called Contrary Research has a deep dive on prediction markets. They are similar to sports betting and options trading mathematically, but different in terms of legal and regulatory treatment (which is constantly litigated and evolving). There are concerns about insider trading. Despite all this, I like them just because they give us some more data points to look at. On political events, sometimes they echo polls and sometimes they tell a different story. On geopolitical events, they provide a somewhat objective, independent data point that might help us spot bullshit in the narratives we are being offered by the media.
Anyway, Polymarket and Kalshi are the big ones. There are others, and there is a blurring of the lines between sports books and prediction markets, which both starting to overlap each other. I was curious if there was an aggregation site for multiple prediction markets, and a popular one is Election Betting Odds. They are pretty true to their name and don’t cover everything on the prediction markets. At the moment, they show J.D. Vance and Marco Rubio leading Gavin Newsom for U.S. President in 2028. They show Democrats with an 80% chance of taking the House in November but the Republicans with a 56% change of continuing their majority in the Senate. Now, prediction markets are not always even consistent internally. If you go to Polymarket itself, they match that 56% chance of Republicans controlling the Senate, but another contract shows a 44% chance of a Democratic sweep of both levels of Congress (and a 41% chance of a Democratic House, Republican Senate. I’m having trouble wrapping my head around the probabilities here – are these things in fact inconsistent?)
Here is CBS News telling me “Senior defense officials looking at Cuba military options“. Well, today Polymarket thinks there is a 14% chance that the U.S. will invade Cuba in 2026 and a 31% chance of U.S. military action against Cuba by December 31, 2026. Well, I guess military action could fall short of an invasion so these two markets are not necessarily inconsistent.
Survey after survey shows a shift in favor of Americans saying they want to live in walkable communities.
Nearly two-thirds (64%) of respondents said they would pay more to live within walking distance of parks, shops, and restaurants, according to the latest Community and Transportation Preference Survey from the National Association of Realtors®—shared first with Realtor.com®.
But intelligent people who probably did just fine in geometry in high school do not connect what they learned to the two-dimensional reality of their neighborhoods.
But enthusiasm drops when walkability becomes a question of what gets built nearby. While 63% say small-lot single-family homes would be a good idea in their community, support falls to 51% for townhomes, duplexes, and other attached housing, 44% for rental apartments, and 40% for condos.
When you set up the geometry that includes those large lots and lots of car storage storage space both at peoples’ homes and nearby shopping centers, the car infrastructure ends up taking up a lot more space than the actual houses, yards, and stores. Things get more spread out and quickly become unwalkable.
There are some potential solutions I can think of. Mixed use (commercial and residential together) is the obvious one. But people have at least an impression that this means more crime and noise. Setting up cities in a more linear way is a possibility – a long, narrow commercial main street with a long, narrow strip of housing behind it including single family homes and lots if that is what people want. Safe, protected infrastructure for bikes, scooters, golf carts, and whatever light low-momentum vehicles we come up with next – these let you cover those distances that take too long or are too boring to walk. Maybe they need to be covered in some climates. But most importantly, there has to be a safe, protected middle ground for light, practical transportation modes that fall in between walking and heavy highway vehicles. And the resistance to this basic concept is just enormous, for some reason.
Traditional public transportation definitely should play a role, particularly along that linear main street. But when things are more decentralized and randomly laid out, vehicular transportation may need to be more on-demand, whether public or private, human driven or autonomous, and I think we should just experiment with all combinations of these until we figure out what works for a particular community. If we can decouple the need to get around in vehicles from the need to store all those vehicles in a stationary manner steps away from our homes and businesses, that can free up a lot of space.
I can picture all this in my head just fine, and then the painful process of change on the ground is measured in decades if not generations, so now at age 50 I wonder if I will see a freaking safe protected bike lane in my neighborhood during my natural lifetime (and my neighborhood is one of the most walkable neighborhoods in one of the most walkable cities in the USA!).
Here’s another article (by an American who toured Chinese factories) on how great Chinese electric vehicles are, and how far behind American industry seems to be. So are we going to just keep protecting and bailing out our failing auto industry, while using propaganda to try to keep the public from realizing what we are missing out on? There are, apparently, some possibly legitimate concerns about imported vehicles as a way to hijack the internet somehow (?).
Given the affordability and quality of Chinese EVs like those from Xiaomi, these cars could go over well in the U.S. Perhaps too well, undercutting Western automakers who fear China’s advantages will put them out of business.
But the U.S. government is concerned about the perceived security risks of internet-connected vehicles — that they could be used as surveillance tools given all the data they collect or even be remotely controlled by the Chinese government through back door software access. There were similar Chinese fears about Teslas until it reached an agreement in 2024 to use Baidu’s mapping platform.
The Biden administration’s “connected vehicle rule”, issued in January 2025, effectively banned Chinese vehicles from American roads, with prohibitions on software starting in 2027 and hardware kicking in 2030.
As a start, maybe we could provide public benefits to our citizens to take that burden off our corporations trying to compete with foreign counterparts that provide those modern benefits? Beyond that, I wonder how we can be so far behind on industrial robots when we are supposedly so high tech?
The obvious superiority of electric vehicles is a dam that eventually has to break in the minds of the U.S. public.
Homicide clearance rates in Philadelphia are way up in 2026. I learned from this article that clearance rate is basically the number of arrest warrants issued on open homicide cases divided by the number of homicides during a calendar year. Some homicides get solved in later years, so it’s not a perfect measure of cases solved but it’s a reasonable proxy. The rate in 2025 was 71% compared to numbers that often hovered in the 40s or 50s over the past decade or so. This is partly due to homicide rates falling so that a fixed number of police can devote more time to each case, and partly due to surveillance technology. It supports the idea, which I first thought seriously about after reading the book Ghettoside: A True Story of Murder in America, that most communities support police if they feel like they are being policed fairly and if they feel that crimes have a good chance of being solved. And it’s not discussed in this article, but I suspect that when people feel like they can call the authorities in and have a good chance of justice being served, they are less likely to take matters into their own hands. And over time, a culture of escalating revenge and retaliation might slowly change.
Sebastian Mallaby is a Washington Post columnist, among other things. I found this too-long-didn’t listen podcast had some interesting things (I skimmed the transcript – thank you to all podcasters who choose to post a transcript!). Here’s just one quote, but there is a lot in the article.
So the case for believing that getting first to artificial general intelligence or AGI is the key metric is that once you get there, you have such a powerful model that nobody will ever catch you up. And there’s this notion, the singularity idea, that when you get to a position where the model is strong enough to write all of the code for the next model and then that next model will create the one after that, the power of the frontier goes vertical and at that point it’s game over. And this is a view held, I think, particularly strongly at Anthropic.
And that’s an argument. There’s another argument, though, which is, the frictions associated with getting the model into deployment, whether that’s in a military setting, in a government setting, in a corporate setting, are going to be significant. And so just having the best possible frontier model in your server inside a frontier lab isn’t the whole enchilada.
I’ve had this same thought – that some industries (like tech companies with intangible products and very short product cycles) will adopt AI very quickly and it will be very disruptive, while in others (particularly government) friction will be enormous. I suspect traditional professions like engineering and medicine will be somewhere in between.
He also talks about government and international regulation of AI. His point is that we have structures, like the Food and Drug Administration, the Securities and Exchange Commission, and Nonproliferation Treaty to emulate. It’s complex but it has been done and can be done. It requires some moral leadership and political will, which seem to be in short supply in today’s world at the moment.
This article describes what even wealthy Americans would have gone through around the Revolutionary War era.
Lice abounded. Bed bugs interrupted sleep. Scabies, ringworm, rashes from numerous unknown sources and unwashed skin was wrapped in clothing of stiff linen, smelly woolens or coarse calico. The byproduct was irritated, itchy skin with the discomfort of scratches, scabs and the stink that accompanied it…
Doctors, following accepted practice, would likely have bled or purged an ill person for humoral balance. Surgeons washed their bloody hands in contaminated water and dried them on their equally bloody apron or clothes, unaware of germs…
When fluid accumulated from infection, a practitioner might use a small sharp spear nested in a metal tube, called a trocar and cannula. The pair were pushed into the body wherever swelling threatened a patient’s health, or exploration of an inner cavity was warranted. Then the doctor removed the perforating trocar, with its triangular shaped head, and left the cannula in place, as a conduit for fluids going in or coming out…
Ouch – yes, I will take my disinfected drinking water, safe food, (mostly) painless dentristry, and climate controlled house and office, thank you.
This just in – the Earth’s rotation speed is not perfectly constant, and it has recently sped up. This has to do with tiny changes in the relative movements of the Earth, Moon, and Sun. It probably has always been happening and we have just recently developed the technology to measure it more accurately.
…starting in 2016, Earth started to spin unusually fast.
For instance, on July 4 of 2024, Earth set a new record by completing its daily spin 1.66 milliseconds faster than usual. Around one year later on July 10, 2025, it spun 1.36 milliseconds faster than usual, with other shorter-than-usual days occurring on July 9 and July 22. This has led experts to warn about a “negative” leap second—subtracting a second from UTC—to keep UTC aligned with UT1 as early as 2029.
Subtracting a second doesn’t seem like that big a deal. I certainly remember a few seconds being added in my lifetime. But apparently, this has caused serious problems for modern infrastructure and industries such as computers, GPS, and banking. The decision is made by something called the General Conference on Weights and Measures, and the leap seconds were causing enough problems that they decided in 2022 not to add them any more. The solution they are considering is to add (or subtract? I had trouble following the article) a whole leap hour, which would have to be done much less frequently. It makes sense to me – over decades or centuries, I suppose not making the small time adjustments would result in astronomical events (solstices and equinoxes – what are the plurals of these words) and the weather eventually not matching the calendar (although let’s be honest, the weather is f-ed anyway…). That would actually happen pretty fast if we didn’t add the one leap day every four years – in my 50 year lifetime, the calendar would already be off by 10 days or so compared to the Earth’s actual journey around the sun.
Neil deGrasse Tyson has a good podcast that explains (along with some fart jokes) that this actually happened in the centuries after the Roman calendar was invented. At that time it became noticeable, the authority on timekeeping was the Pope, and the Pope decided to skip 10 days in October one year. Problem solved. (I suppose people outside the Catholic world had their own calendars, their own observations/calculations of astronomical events, and their own ways of dealing with these changes.) Even not considering the recent changes scientists are now able to measure, Tyson explains that there would need to be some additional adjustments from time to time.
Now, I say if we start messing with the calendar it is also time to go to 13 months, where each month has exactly 4 weeks, except there is an extra New Year’s Eve. Then every four years there can be two New Years Eveses. Let’s throw ourselves one little bone every four years and make leap day a holiday. Or maybe an election day which is also a holiday. It just makes sense. And to hell with daylight savings time! Let’s make it happen.
Albert Einstein might not have been a guy you would want to sit down and have a beer with. He was happiest thinking alone in a quiet room (or entire house). His first marriage sounds quite rocky, although it lasted 16 years and produced three children. His second marriage was to a cousin – a first cousin, their mothers were sisters and she did not have to change her name because it was already Einstein. They did not have biological children, but she had two children from a previous marriage which he adopted. But of course, he is not known for developing any special or general theories of harmonious human social relations.
I feel like I’ve been at least somewhat of a cheerleader for the AI boom, so let’s talk a little bit about the case for an AI bubble. The Guardian talks about it mostly in stock valuation terms, with a little discussion of corporate debt.
In focus at the moment is the concentration of equity in just seven companies, the Magnificent Seven: Amazon, Alphabet (Google), Nvidia, Meta (Facebook), Microsoft, Apple and Tesla (possibly soon to merge with Elon Musk’s other venture, Space X)…
It is disturbing that the 10 largest companies in the S&P 500 account for about 40% of the index’s total market capitalisation, which is well above the 27% peak reached during the tech bubble of 1999-2000…
Yet the AI bubble has further to run because the top 10 are making huge profits, they have a US president who is prepared to lose wars to keep the financial markets happy, and the world is awash with savings looking for a home.
In my mind, we need to see an acceleration in real GDP growth and productivity growth to go along with this stock market surge, and then we might be able to hope for a soft landing. If I were much smarter I would make some kind of dashboard to look at all this data side by side. There must be smart people doing this, no?