inglor_cz

Once upon a time, there was also the Soviet propaganda effort in the West. Many of the Cold War reactors could be used to produce not just electricity, but fission material for nuclear warheads. Meaning that if you could persuade big countries to give up nuclear energy for good, you'd also reduce your adversary's capability to build and maintain their nuclear arsenal, or preempt its future acquisition in the first place.

It is still quite tribal, with whole mutually hostile clans that stem from a particular professor or workplace. Plus the economy of academy tends to be quite zero-sum.

Also, the US academy specifically cares about your race a lot.

On a similar personal note, I had a half-year health scare that ended just today.

In May, they told me on ultrasound that my dilated aortic root has grown by more than 2 mm in the last year and that if the progression does not stop, an operation will sooner or later be necessary.

I got a referral to magnetic resonance and had to wait for it until October. Was there last week, nervous. Today I got the result: compared to the MR from 2021, there is no progression at all. The same 44 mm as then. The ultrasound was just not as precisely done, or the doctor measured it slightly askew or what.

Ooof.

I would argue than no one is really competent in a language that has been dead and unused for millennia.

The entire cultural context of that society and work is gone and we have but a few shards thereof.

This really reminds you of the Chernobyl station computer printing all the warnings and recommendations to shut the reactor down when they started playing around with it ... and Comrade Dyatlov overriding it.

Dyatlovs seem to be born everywhere.

"bypissing procedures"

I know that this was likely a typo, but it is also a genuinely great word for the sort of attitude related to these situations.

I started thinking about agentic programming from a totally different perspective.

Personal accountability in software development is actually hard, and less useful than it seems. What good is it if you discover a 12 y.o. vulnerability and can blame a specific person? (Like the recent one in PHP?) You may learn the name of that programmer who made the mistake, but it does not really tell you anything about that person, unless their overall record is terrible, and you are not entitled to any damages anyway.

In closed source, this is even worse. So a certain Windows function sucks. Now what. You never learn any other provenience than "Microsoft did it". Tough luck even getting support, and you certainly don't get your license fee back. This sort of accountability is worse than useless.

The only exception I see is knowledge within a specific development team. We usually know what sort of errors our colleagues are prone to, and we may be able to work around their personal weak spots.

Agentic coding done well has a dimension of trust that human coding cannot achieve. I can tell my agents to aim for 100 per cent test coverage without overburdening them or running out of budget and time. I can ask other agents and other models to review code written by the first agents, and then again. Every layer of such review adds some credibility to the result, and its mechanistic character ensures certain level of repeatability. Sure, it can happen that all involved mechanisms share the same weakness and let a particular bug through. But it is similar to several sieves over one another lining up perfectly to let something through, and that is quite rare.

If I had to choose, I would trust code designed, implemented and reviewed by three different SOTA models over code designed, implemented and reviewed by three median programmers today. Three elite programmers would have been a different story, but that would also be expensive like hell, and much like you don't eat most of your dinners in a Michelin restaurant, you have to economize with your code development in a similar way.

While this is indeed not how trees work, it puts the number into a useful perspective.

Finland, comparable in size to Montana, is basically one huge forest (75% tree cover) dotted with lakes, and a lot of those forests are harvested every year. Of the entire EU, this is probably the place where trees are the least rare.

250 thousand trees cut down in England would have been a very different story.

This is quite similar situation indeed. It was once easier to adopt the metric system when the infrastructure was not yet as developed around the previous ones.

Austria-Hungary adopted the metric system in the 1870s. Back then: no cars, no highways, no speeding laws etc. Oxen-drawn wagons in the countryside were not built to a precise spec.

Only the railway, the architects and the artillery had to really switch. Ordinary people would still ask the grocer for half a pound of ham in the 1930s.

USA in the 1970s would have to remeasure all the land communications etc., or end up with a weird hybrid like the UK.

WHat we have in IPv4 and IPv6 is the weird UK-like hybrid...

Now this is an interesting map.

First impression: Poland stronk. Even Czechia strong. Charles IV.'s realm restored. Russia in shambles, not a very efficient conqueror of the address space. And the glory of Rome restored, with pitiful Constantinople looking at their backs from afar.

Second impression: Morocco, Kenya, Egypt and South Africa as the African heavy-weights. Corresponds to what I know.

Third impression: the real bosses are Taiwan and South Korea - just wow.

"You might be not interested in the war, but the war is interested in you."

The point about finding problems in documented processes (Enveda: manufacturing and filings) is unexpected but interesting. One of our civilizational problems is that the sheer volume of documentation has slowly exceeded capabilities of humans to produce it reliably and without flaws.

Same applies to construction projects, BTW.

I found this essay interesting in the "out-of-the-box" sense. In a time when anti-AI sentiment dominates everywhere, its author argues that AI can actually bring people back from the fringes where social media pushed them.

Unfortunately only paid subscribers can comment there, and I am not one. I would definitely ask him about at least one failure mode: if you simply omit some information from the learning context. Chinese models will likely never hear about Tiananmen Square Massacre, and a Chinese user will get "accurate" information in that context only.