Sancus

I really enjoyed The Raven Scholar and of course A Drop of Corruption though it's not quite epic fantasy...

the polish figure skating team in the kiss and cry holding a plush pierogi

IMPORTANT OLYMPIC NEWS:

THE POLISH FIGURE SKATING TEAM HAS A PLUSH PIEROGI

The GOP has offered nothing this century except for catastrophic wars. No healthcare plan; no re-industrialization. Now they're losing their only issue, immigration, as people see the violence and havoc that comes with mass deportations.

(Dude suddenly stops pushing his broom across the floor of the Death Star)

Does this look like a big laser to anyone else

Pork tenderloin has almost identical macros to chicken breast but is a nice change to mix in.

If the US were a geopolitical rival the headlines would be like: US Secret Police Execute Second Dissident In the Streets As Militarized Occupation Enters Sixth Week

The Canada-China relationship has been distant and uncertain for nearly a decade.

We’re changing that, with a new strategic partnership that benefits the people of both our nations.

AOC: I want everybody to understand that the cuts to your health care are what’s paying for this. You get screwed over to pay a bunch of thugs in the street that are shooting mothers in the face.

there needs to be a sex offender registry but for ice guys. you should have to go door to door when you move to a new area and inform your neighbors that you used to be an ice guy. they shouldn't be allowed with 500 feet of a school

If the worst single-family house in your n’hood costs over $1 million, then your n’hood should be zoned for apartment buildings.

(It’s pretty much that simple)

OpenAI, specifically, clearly has an unsustainable plan for future training. I just think their plans are commercially unnecessary. A model with order of magnitude higher training costs won't come close to break even. But that doesn't mean much to the viability of all models.

Now if your business model is only barely profitable at current API costs or even requires cheaper, good luck with that. I think you're fucked.

Basically I'm saying that even once you account for companies going out of business, the ones left standing will still be operating pretty good models. And perhaps training will become research only. Plenty of technologies only exist from public research spend that was never profitable.

Of course. But my point is the models that have already been trained are pretty good and there's no long term commercial need to burn $2T/year training. Those costs *will* drive companies bankrupt. But they'll be bought and some companies will operate and tune models much closer to inference cost.

So even if the cost of an H100 worth of compute never goes down, inference is worth it. This analysis ignores the cost of training, which is the big question mark and enormously hard to calculate. But training costs are already sunk for existing models and we don't need 1000 new models per week.

Now yes, Sonnet 4.5 is more capable. But the output token cost is $15/M. There is just no way the cost of inference is many times that amount. And I think that's a very fair price for Sonnet 4.5 right now as a coding assistant.

It's definitely hard to say, but let's say we are renting H100s at $3/hr, which is about 50% higher than the going price. Using Deepseek R1 f ex, you're going to get maybe 150 output tokens/s per H100, or a bit more. That is ~$6/1M *output* tokens which is actually cheaper than Sonnet 4.5.

I don't think it's necessarily true they'll need to get more expensive. We run open LLMs in the cloud, done the math on the costs. Sure, training costs can expand infinitely but models that are not on the bleeding edge will only get closer as LLMs plateau. Which I personally believe has begun.

One fun thing about the Japanese is we love the FUCK out of Christmas and have only the loosest concept of what is going on

"the prompt" means what exactly? The whole conversation with all context? I don't think much code is generated by one shot prompts anymore, for me the context and conversation is always many times longer than the PR itself. Nobody is gonna read all that.

Every technology problem is three social problems in a trench coat

Epstein: I’m best friends with this pedo running for President, here’s a bunch of stories about how big a pedo he is

NYT: but her emails

If I give the bully my lunch money every day eventually he will die of old age

They're going to change the clocks this weekend and the president is powerless to stop it. Because he's weak and a loser. That's what they're saying sir

I can see this being popular with some devs but does everybody even want to code with voice? If you type actually fast there's not a significant speed gain from it and I would personally not enjoy talking constantly all day.

The official home of the Python Programming Language

TLDR; The PSF has made the decision to put our community and our shared diversity, equity, and inclusion values ahead of seeking $1.5M in new revenue. Please read and share. pyfound.blogspot.com/2025/10/NSF-...
🧵

I tweet: "people who see 50% of their monthly paycheck go to rent don't care if a candidate eats with their hands. they want to hear what your candidate is going to do about housing." John Galt 2025 replies: "Move if the cost of living is too high."

who creates the lively art and music scenes if the rent is too high? who makes cultural life? who serves you food or washes your dishes? who creates the fancy little restaurants that draw you to urban living? affordable housing is at the heart of what makes city life worthwhile.

cry havoc and let slip the frogs of war

Problem is the predatory techniques are always more lucrative and we've chosen not to regulate this stuff. If it was legal to open a casino on every street corner, there would be a lot of those. Mobile games are the unavoidable consequence of anything goes monetization.