kzrdude

Yes: most mainland Norway ice was gone during the holocene climatic optimum (9–6 thousand years ago), and glaciers recovered since then (until approx the 20th century, when deglaciation started accelerating again).

The climate is varying in long cycles, and the deglaciation 20 thousand years ago (last glacial maximum) was driven by a spike in summer temperatures, and this peak has already long passed. Ice ages and interglacial periods have a certain momentum to them, so it takes more than just that to go back to full ice age..

Is it different between major models how they scale with reasoning level? Some models where it hardly pays off? I expect it to be very similar across the board.

And we note that these specializations are part of production Python and will not be removed when the JIT is removed.

That's not how academic literature works. The structure is like this: Statement (Citation). The citation is not an affirmation of truth but a content declaration; this statement is supported by this publication. The receipts are there, so that you can trace the argumentation and make your own conclusions if necessary.

'Knowledge' formed this way is a living, evolving graph, and sometimes links are broken (findings are reevaluated or disproven); having the content declaration so clearly in the open makes it easier to deal with this evolving landscape.

LLMs cook up new sentences based on probability, based on everything in their training and their context. There's just nothing in their model that traces provenance of facts, they physically don't store that information. Once they have written a sentence, they can't tell you from which work they got this information.

There are tools where they try to retroactively add references, the LLM interfaces try to do this, but it's usually inaccurate.

I think of the way a scientific article works. "The sky is blue (Pearson, 1992)." The writer is fully aware what the factual statement is and which publication supports it (not just that, but which kind of publication - primary, secondary source, peer reviewed?). An LLM can try to imitate this, but the actual information to support this does not exist in its training.

There is a reason why LLMs can not be good at this: They don't know what they are quoting or citing. A different kind of model with factoid provenance could do it?

Google has flash-lite, too. I liked OpenAi's scheme better, as long as it is parallel and makes sense.

A dispersant should divide the oil into much smaller particles = larger surface area, which by itself should speed up any natural biodegradation significantly. Of course, once it settles on the sea floor, most of that effect is gone again (it's not dispersed in water anymore). But IF there is something that can consume the oil, dispersing it into microdroplets is not necessarily a bad idea?

That announcement arrived 2 hours ago (on x), so now they have.

This is called a hybrid reasoning because the reasoning can be mapped to text and is therefore introspectable, with tooling. That should address the concern, I think.

Just yesterday there was just one public revision of 4.1, and they listed they were using 'Deepseek v4.1 flash'. I assume they will update their website if they update the model version (and use the version in the name).

Presumably. The current opencode slug for the model is just 'deepseek-flash' suggesting they will just update that one going forward, not keeping versions in parallel.

It's a 51 hour video but I still don't think that's enough time to play 1000 games. How did it work?

I saw a lot of sad horses in Sevilla, with above 36°C that day