pantsforbirds

The sol max looks like it's absolutely ripped for some reason

The constant goalpost moving is insane. This is objectively an amazing result.

ah, i think i misunderstood your original post. if you mean the wayback-machine/arkive, then I suspect it'd be hard to justify? You are essentially paying a third-party source to validate that diffs didn't go through on the source material.

with llms, at some point it probably becomes easier to use your paid api connection to manage your own cached version yourself?

I mean we already paid for the article from the source itself. I guess I'd expect a better "diff" source from them, but if they dont even update the article itself, i guess i wouldn't expect a paid service to have those updates either?

We used to always "scrape" the wayback machine for any sort of news article we actually paid to consume. I was absolutely shocked by major news sites making very important edits to an article without any sort of editorial notice!

Sadly this sort of thing is probably not really possible anymore, but I can't really blame anyone for making this sort of decision. I can't imagine how much more traffic they get now vs 2021 when we were doing this.

I think it's VERY project specific. If you are looking for anything technical at all, then keyword search almost always does better (in my experience). I'd actually recommend starting with keyword search, and then expanding with embeddings after you have a better idea of what your users are trying to determine.

I'm hoping this means we get an rds + s3 implementation of ducklake from AWS: https://ducklake.select/

I have a Google account that I had tied to a domain I originally purchased from google (when they had the .dev offerings), and now that it's been purchased by square space my ability to use AI through that account is in a bizarre state. Even my free gmail account has more AI offerings, and I'm not allowed to pay for improved AI offerings on my custom domain & google workspace account.

I know this is probably a pretty small edge case, but it is a bit frustrating. Any other provider lets you sign up with an email and give them a payment processor/card, but because google wants me to only use their unified workspace for signing up, I'm completely locked out now.

I appreciate the comparison to Timescale, DuckDB, and ClickHouse. It's interesting to see the differences between a more specialized db, like Quest, and DuckDB/ClickHouse.

Ah yes, I specifically mean you can keep earning points infinitely. After the board is full, it resets the game, and you get points for each piece on the board. I usually stop the simulation after it clears the board 10 times or so.

I'm fairly sure the code snippets aren't equal in the last python example:

```python names = sorted({name.lower() for name in users if name}) ```

vs.

```gossamer let names = users |> iter::filter(|n: String| n.len() > 0) |> iter::map(|n: String| n.to_lower()) |> iter::sort_by_key(|n: String| n.len()) ```

Python is sorting a set (unique values only), but I'm not seeing a unique or set approach for gossamer.

I've never seen the level of anti-scientific, straight misinformation as I have with data centers. I genuinely think it's to the point where the antivaxxers have a stronger scientific backing for their stance.

It did not feel organic at all. It's to the point now where that initial seeding of ideas has gained legitimate traction, but the initial burst of anti-datacenter content was wild to see in real time.

The way the game is seeded, it's 100% deterministic. So if you can scan enough potential moves / second, it's very easy to infinitely play the game at a fast rate.

Shockingly, Claude was able to find the exact random generators that the js library uses, list out all of the differences between the languages that might impact the replays (like rounding being treated differently), and include those differences in the simulation. Claud then back-tested on a huge number of games using the original implementation as a golden set generator.

I started by pulling the game source locally and doing some initial testing with a few solvers. I ended up having Claude make a 1-for-1 perfect sim of the original game using rust with some optimizations. The original engine was able to check around 180 moves/s and the rust port was hitting 2,150 moves/s.

After that, I tested a few solvers. Single Player Monte Carlo Tree Search seemed promising, but didn't do as well as I expected. A simple beam search does appear to go infinitely pretty easily.

I had a lot of fun playing the game. I also made a local solver to see how far you can theoretically get, and it's definitely infinite if you play perfectly. Working on the local sim / solver was almost as fun as playing the game :)

Fun article and worth the read, but sadly none of the LaTeX was rendered for me (assuming it was supposed to).

I would have loved to buy my children some small C&H figurines or, especially, a stuffed tiger, but I do respect Waterson's decision

How many existing data centers do you think you rely on each day? Almonds do not realistically improve my life in nearly the way the internet does, but also who am I to say whether you should be allowed to farm almonds or build a datacenter?

Like producing more tax revenue for a county than the rest of the county combined? That payoff allows for insane levels of infrastructure improvement at literally no cost to locals.

In the case of the Stratos / Box Elder County mega-datacenter, they are planning to use gas turbines to run off of the nearby Ruby Pipeline. It's a pretty tidy solution, if you ask me.

The mega-scale ones almost all have power as part of the ramp-up process. They are also a massive tax benefit.

Looking at the controversial Stratos data center in Box Elder County:

Box Elder County baseline (before Stratos): 2026 General Fund revenue: $36.1M 2026 General Fund tax revenue: $21.1M 2026 Municipal Service Fund: $18.8M

Stratos (Phase 1): Power: 3 GW Revenue expected: $30M/year Revenue as % of existing county tax rev: 142%

Stratos (Full Buildout): Power: ~9 GW Revenue expected: $108M/year Revenue as % of existing county tax rev: 512%

Planned power details: - Box Elder says it will use natural gas from the nearby Ruby Pipeline. - MIDA describes it as dedicated on-site generation designed to limit demand on the existing grid. - The project ramps from ~3 GW in Phase 1 to ~9 GW at full buildout.

I'll acknowledge that the MIDA structure is not a normal county tax bill and comes with explicit energy-rate discounts, but given that the county could hit 500%+ of its previous tax revenue, I think that's a very easy deal for the county to make.

Data sourced from: https://www.midaut.org/stratos

Except datacenters are actually very low environmental impact. As long as they provide their own power, they have MUCH lower impact than most farms would.

It's a neat product, but I am also a bit confused by the choice to use C as the primary language. I would think you'd want something more ergonomic since this is entirely I/O bound.

I'm 100% on board with using C for the rag or sqlite sync solutions, but the agent itself seems like it would be much easier to write in another language and build on those tools. Even something like go would be much easier to adopt.

I don't understand why people on this website hate Grok for being tied to Elon Musk, but also have no problem with models that are directly sponsored by the CCP.

Grok and Gemini are the ones I tend to use for finding news related to breaking events. Both were really nice during the Iran incident when I wanted to find out things as they were being reported.

If your data follows a relatively uniform distribution, you can try out interpolation search: https://en.wikipedia.org/wiki/Interpolation_search

The US doesn't have a desire to annex Canada; that's very silly. And the reason Russia doesn't have the capability is because of Canada's alliance with the US.

That seems entirely plausible. I based my comment on one of Elon's tweets (xeets?) about it: https://x.com/elonmusk/status/2028261823678759335?s=20

Ignoring the ... less substantive portions of your response

> I can both deny you the means to defend yourself, forcing you to rely on me for protection. That's the definition of a protection racket.

The US didn't deny Europe the means to defend itself. Europe chose not to build those means because it was cheaper to rely on the US. These were domestic political choices made by governments whose voters preferred social programs over defense budgets. A protection racket requires coercion; what the EU received is much closer to a subsidy.

> This is revisionist history at best. The US has done their best to undermine and dismantle European social programs.

Can you cite a specific example? The US has broadly pushed for capitalist markets or free trade via policy, but "done their best to undermine and dismantle European social programs" is a very strong claim without evidence. Norway's sovereign wealth fund being "tolerated" because of strategic positioning is, at best, a conspiracy theory. There has been some tension over Norway divesting in American companies for political reasons, but that's hardly the claim you've made.

> America was protecting Israel's trade routes. Let's be clear. European trade routes largely just rerouted around the Cape of Good Hope.

Rerouting around the Cape added weeks of delay and a high monetary cost to European shipping. Just because European ships could reroute doesn't mean the European economy wasn't significantly impacted. Why did the European trade association publicly beg for more governments to join the operation if the Red Sea shipping was only about Israel?

> You can't both have a protection racket and expect military help

You expect America to adopt a one-way obligation where it provides for the defense of its allies, and receives no help in return? Why wouldn't that deal fall apart?

> Yes, literally nobody wanted the US and Israel to launch an unnecessary, unprovoked and ill-planned war on Iran

You can disagree with the decision to strike Iran. But when Iran retaliates by launching missiles and drones into 12 different countries (11 of which had not participated in the initial strikes against Iran in any way), the question of whether allies will support defensive operations is separate from whether they endorsed the initial strikes.

> Not a single defense analyst would even seriously consider such a prospect

No country would seriously consider it a prospect because the entire might of the US Armed Forces would immediately engage anyone who tried. This despite the fact that Canada has anemic defense spending, a large arctic border with Russia, and strategic assets I'm sure Russia would love to have.

> Many people believe that the US annexing Canada is a higher probability than either China or Russia doing so. All three are very low probabilities.

I believe those people are being a bit silly, and their position probably comes from a strong dislike of Trump as a person, and not a genuine belief.

Russia annexed a warm-water port and then shortly after attempted to incorporate Ukraine as part of a plan to remake the USSR. The only thing keeping China from taking Taiwan is the United States.

The US has no desire to annex Canada, and it also has no need to. If Canada proposed statehood or even a territory agreement with the US, I genuinely don't think it would even pass a vote.