
Tech, cyclist, 'master'-ranked backgammon player, nlhe dilettante, lapsed banker. early @founderscoders. currently CTO at @BookPebbleUK. opinions mine alone
London
Joined December 2009
- Tweets2.7K
- Following828
- Followers333
- Likes3.4K
adam h retweeted
"Our initial evaluation suggested that GPT-6-Astra's score was in the late eighties followed closely by Opus 5.5. But, as we started to investigate the mistakes that the models might have made, the models, especially Astra, insisted that they were right and that it was the expert-supplied answers that were erroneous. When we went back to the experts it turned out, in multiple instances, that that the experts found some subtle errors buried deep inside their calculations. As they revised their answers, Astra's score started to inch up and it finally ended up with 98% (49/50)."
adam h retweeted
YOU LOVE TO SEE IT
Life expectancy for people with cystic fibrosis going up and up and up and uo
That's what REAL science does.
😍
adam h retweeted
Margaret Hamilton, who led the 100-person team that wrote the onboard software for the Apollo missions, has died. nyti.ms/3U9IrTj
adam h retweeted
Scott Aaronson's take on "The Mathocalypse". Always worth a read.
scottaaronson.blog/?p=10169
adam h retweeted
Either you are with us, or you are with the mosquitos.
There is no good reason for mosquitoes to exist in wealthy countries today.
New technologies can get rid of the mosquito species that cause diseases like dengue, yellow fever, and malaria; and others can prevent them from carrying those diseases.
worksinprogress.co/issue/mos…
So why do mosquitoes still exist?
In a new article for Works in Progress, Maya Rosen explains how state incapacity and risk aversion led to over a decade of delays and billions of unnecessary bites in the United States.
Oxitec, for example, engineered an Aedes aegypti mosquito in such a way that its female offspring would die before adulthood. Through sustained releases, they showed they could reduce local mosquito populations by 80% or more, in field trials.
They sought approval from the USDA in 2010, which then passed it to the FDA, which passed it to the EPA. In 2020, the EPA allowed them to conduct limited field trials in the Florida Keys. Two years ago, their experimental permit expired, and they still need another review before it can be used routinely. Their proposed field trial in California was withdrawn after public opposition.
Another company, MosquitoMate, developed a technique involving Wolbachia, which also suppresses local mosquitoes. They then spent over fifteen years working through the system before finally receiving commercial registration nationwide in 2024.
It seems that, sometimes, the barrier to getting rid of diseases isn’t technology at all. It’s policy and political will.
New products shouldn’t get stuck in regulatory review for so long, especially if the delays are about basic questions like which agency has oversight over the product, rather than testing for safety.
Though some of those uncertainties were being addressed, the broader changes haven’t materialized. After Biden’s 2022 executive order, the agencies planned to update how their coordinated framework was implemented. But Trump rescinded the order last year.
Meanwhile, the EPA unit that reviews new mosquito control products is still a small office backlogged in reviewing a wide range of products.
Mosquitoes are a policy choice. We can get rid of them. Will we choose to?
worksinprogress.co/issue/mos…
There is no good reason for mosquitoes to exist in wealthy countries today.
New technologies can get rid of the mosquito species that cause diseases like dengue, yellow fever, and malaria; and others can prevent them from carrying those diseases.
worksinprogress.co/issue/mos…
So why do mosquitoes still exist?
In a new article for Works in Progress, Maya Rosen explains how state incapacity and risk aversion led to over a decade of delays and billions of unnecessary bites in the United States.
Oxitec, for example, engineered an Aedes aegypti mosquito in such a way that its female offspring would die before adulthood. Through sustained releases, they showed they could reduce local mosquito populations by 80% or more, in field trials.
They sought approval from the USDA in 2010, which then passed it to the FDA, which passed it to the EPA. In 2020, the EPA allowed them to conduct limited field trials in the Florida Keys. Two years ago, their experimental permit expired, and they still need another review before it can be used routinely. Their proposed field trial in California was withdrawn after public opposition.
Another company, MosquitoMate, developed a technique involving Wolbachia, which also suppresses local mosquitoes. They then spent over fifteen years working through the system before finally receiving commercial registration nationwide in 2024.
It seems that, sometimes, the barrier to getting rid of diseases isn’t technology at all. It’s policy and political will.
New products shouldn’t get stuck in regulatory review for so long, especially if the delays are about basic questions like which agency has oversight over the product, rather than testing for safety.
Though some of those uncertainties were being addressed, the broader changes haven’t materialized. After Biden’s 2022 executive order, the agencies planned to update how their coordinated framework was implemented. But Trump rescinded the order last year.
Meanwhile, the EPA unit that reviews new mosquito control products is still a small office backlogged in reviewing a wide range of products.
Mosquitoes are a policy choice. We can get rid of them. Will we choose to?
worksinprogress.co/issue/mos…
adam h retweeted
Waymos are really safe (80-95% safer than humans depending on the metric), but they also seem to be getting *safer* over time. This is the worst they'll ever be.
A quick analysis of the police-reported crash rates in three cities from Waymo's published safety data.
adam h retweeted
This is the software trilemma.
You can only pick two.
Possible combinations:
1) High quality + fixed scope = flexible release date
* You only ship it when you're done implementing the defined scope.
* You don't cut corners. Instead, you change the release date.
2) Fixed date + high quality = flexible scope
* You cut scope to meet the date.
* You complete the most important parts first so that you're done early, then you implement what's left until you meet the date.
* You might not be able to ship the entire initial scope by that date.
3) Large scope + fixed date = cut corners
* If you have to deliver a large scope by a fixed date, you'll have to trade-off quality and cut corners so that the same amount of things gets done faster.
* There's no magic to doing more work in less time. You can obviously find clever ways to make certain things easier, but then I'd argue that's actually "high quality" and it takes time to think that way.
* That's why high quality is actually also faster (not infinitely faster, but usually faster).
Choose wisely
adam h retweeted
Awesome news today -- @sunyesf scientists are de-extincting lost American trees!
@USDA just approved a genetically modified version of one of the great American trees, the American Chestnut, that went effectively extinct due to an Asian fungal blight introduced in 1904 that killed 3-4 BILLION trees (some 120 feet tall) by 1950.
By adding in one gene from wheat to the American Chestnut genome, conservation scientists at @sunyesf have shown they can make the American Chestnut resistant to the fungal blight and revive and bring back this American classic.
Let's finish this approval by @EPA and get these planted so our kids can visit these forests!!!
adam h retweeted
Trial lawyers are lobbying against self-driving cars because they're too safe. They need people to be killed and injured so that they have material for lawsuits.
adam h retweeted
Wild animal vaccines are a thing. In 1978, Switzerland airdropped 4,000 vaccine-laden chicken heads from helicopters over the Alps. And it worked: fox rabies was eliminated.
For most of history, our approach to sick wild animals was to treat it as not our problem. Turns out that was a bad idea, because roughly 60% of human infectious diseases come from wild animals.
So scientists got creative. Rabies was spreading through raccoons, foxes and coyotes despite culling hundreds of thousands of them — which, besides being grim, didn't even work. The solution? Hide vaccines inside food and let the animals vaccinate themselves while having no idea they're participating in a public health program.
Switzerland went first. Their method of choice was vaccine-stuffed chicken heads, which were thrown out of helicoters. This eliminated fox rabies from the entire country by 1999.
The US followed with fishmeal blocks dropped near dumpsters to vaccinate raccoons. They now distribute up to 10 million baits a year. This is how Texas wiped out two rabies variants entirely. Vaccinating wild animals is cost effective: every dollar spent saved up to $13 in human treatment costs.
Now we're going further. Australia approved a koala chlamydia vaccine in 2025 (yes, koalas have chlamydia). Researchers are also developing anti-fungal gels for bats.
And we are moving on the "philosophical" dimension too.Disease isn’t bad solely because it threatens a species’s survival but because it causes suffering. Every koala or raccoon that contracts a fatal illness spends days or weeks vomiting, seizing, blind, feverish, or paralyzed. None can hope for treatment. Then, alone and in pain, they die. We can prevent this!
We should be vaccinating wild animals.
60% of all infectious diseases, including Ebola, Lyme disease and rabies, come from animals, mostly wild ones. Vaccinating wild animals would help to control these diseases before they spill over to humans.
worksinprogress.co/issue/why…
Doing it is surprisingly easy and cheap. Switzerland stopped the spread of fox rabies in the 1970s by airdropping chicken heads filled with oral vaccines from helicopters. By 1999, fox rabies had been totally eradicated.
Many American states vaccinate rabies for coyotes by hiding oral vaccines in fishmeal blocks covered in fish oil or vanilla. In Texas, every dollar spent on that program saved between $4 and $13 in human treatment.
But we should also vaccinate wild animals for their own sake! Infectious diseases kill animals in agonizing ways and can drive species to extinction. One fungal disease alone has caused 90 amphibian species to go extinct in the last fifty years.
In 2015, 60 percent of the global saiga antelope population was killed in three weeks by a bacterial outbreak. White-nose syndrome has killed over six million North American bats since 2007.
We could save all these animals from suffering by vaccinating them the way we do pets and humans. It would make the world safer for humans, and *much* better for the animals themselves.
adam h retweeted
the Belgian Prime Minister today at the “One Europe, one market” event: "there are twice as many people working on compliance compared to innovation in Europe."

































