tal

Many text editors today just load the whole file into RAM.

been the case for decades

One data point: emacs normally loads the whole file, unless you're using the vlf package or similar.

TECO and ed might not. Dunno.

Another user in the BlueSky thread showed a photo that appears to be a Best Buy case of RAM, showing a 32GB set of two DDR5 DIMMs going for over $400 USD, a 64GB kit for over $900.

If I hit Google Shopping, which indexes a ton of retailer sites, I can find 2x16GB DDR5 DIMMs for far less than that at various retailers that haven't jacked up prices yet.

https://www.google.com/shopping?udm=28

My first hit for "2x16gb 32gb ddr5" sorted by price is this:

https://pcpartshawaii.com/products/kingston-fury-ddr5-32gb-2x16gb-5200mhz-cl40-ram

Kingston Fury DDR5 32GB (2x16GB) 5600MHz CL40 RAM KF556C40BBK2-32

$100.00

They say that they have two in stock.

These guys are next lowest:

https://www.barcodediscount.com/catalog/kingston/part-kcp548us8k2-32.htm

Price: $103.06

I mean, they did make a lot of money, but they also had an extremely high valuation.

https://www.macrotrends.net/stocks/charts/NVDA/nvidia/pe-ratio

NVIDIA PE ratio as of November 20, 2025 is 48.45.

Something like 20 is typical for a mature company. Tech companies have, in the past, often had higher ratios, but that's based on their expectation to grow a lot rapidly, and expecting NVidia to dramatically grow from their current

already very high

valuation is asking a lot.

If NVidia were a small tech company that was doing well and clearly had a lot of market to expand into rapidly, that would be one thing.

I think that in general, the market has been pretty good to NVidia. Their share price is up 31.22% since the start of the year. 1,247% over the past five years.

https://en.wikipedia.org/wiki/Taurus_KEPD_350

Germany apparently has 600 Taurus air-launched cruise missiles.

They apparently have a next-gen longer-range variant coming out in 2029, and are ordering 600 of those.

If I had to make a guess, the second batch

exactly the same size

presumably is to replace the first, which means that they're presumably not gonna need (all?) the first batch in four years.

Ukraine apparently also requested some.

In May 2023, the German Federal Ministry of Defence said that Ukraine had requested the missile during the ongoing Russian invasion of Ukraine.[16] In interviews in June and July 2023, German Chancellor Olaf Scholz and Minister of Defense Boris Pistorius said that Germany would not supply Ukraine with long-range missiles.[17][18][19] In January 2024, the German Bundestag voted against the supply of the Taurus missile to Ukraine.[20] In February 2024, the German Bundestag and Chancellor Olaf Scholz again expressly refused Ukraine's request while agreeing to deliver longer range weapons.[21][22] In May 2025 newly elected chancellor Friedrich Merz made more ambiguous statements regarding Taurus, that their delivery to Ukraine was within the 'realm of possibility' and that the discussion about their delivery to Ukraine would not be public.[23][24]

There's Mono. I don't know what portion of .NET compatibility issues that addresses in 2025.

Meta’s chief AI scientist and Turing Award winner Yann LeCun plans to leave the company to launch his own startup focused on a different type of AI called “world models,” the Financial Times reported.

World models are hypothetical AI systems that some AI engineers expect to develop an internal “understanding” of the physical world by learning from video and spatial data rather than text alone.

Sounds reasonable.

That being said, I am willing to believe that an LLM could be part of an AGI. It might well be an efficient way to incorporate a lot of knowledge about the world. Wikipedia helps provide me with a lot of knowledge, for example, though I don't have a direct brain link to it. It's just that I don't expect an AGI to be an LLM.

EDIT: Also, IIRC from past reading, Meta has separate groups aimed at near-term commercial products (and I can very much believe that there might be plenty of room for LLMs here) and aimed advanced AI. It's not clear to me from the article whether he just wants more focus on advanced AI or whether he disagrees with an LLM focus in their afvanced AI group.

I do think that if you're a company building a lot of parallel compute capacity now, that to make a return on that, you need to take advantage of existing or quite near-future stuff, even if it's not AGI. Doesn't make sense to build a lot of compute capacity, then spend fifteen years banging on research before you have something to utilize that capacity.

https://datacentremagazine.com/news/why-is-meta-investing-600bn-in-ai-data-centres

Meta reveals US$600bn plan to build AI data centres, expand energy projects and fund local programmes through 2028

So Meta probably cannot only be doing AGI work.

https://en.wikipedia.org/wiki/Romani_people_in_Slovenia

According to the 2002 census, there were 3,246 Romani individuals living in Slovenia.[1] They constitute 0.5 percent of the total population.[2] The Slovenia Roma speak Balkan Romani and Italian.[3] The Roma have been living in Slovenia since the 15th century.[4]

So did they just now become a threat, or was it for the past six centuries too?

Took down Framework's website, which I was using.

It's back up now.

Just keep in mind that the long run trend for storage prices is pretty strongly downwards; that's a log-scale graph.

The cases of old ones are on eBay, as I mention in another comment.

And it looks like someone has made ATX mounting kits.

https://thelaserhive.com/product/mac-pro-atx-kit-with-psu-mount/

Good point!