twoodfin

6 ms·

61.
by twoodfin 2mo ago
It would incentivize the wrong things: All else being equal, you want to produce more with less labor and other inputs. It’s a good thing farming needs an order of magnitude fewer farmers than a century ago.
62.
by twoodfin 2mo ago
Sorry, which of those are capital gains rate cuts or tax hikes on workers?
63.
by twoodfin 2mo ago
When is the last time either of those things (capital gains rate cuts, tax hikes on the lower 90% of income earners) happened? I think it’s been decades by now.
64.
EV8: The Post-Ultimate Alpha [pdf](research.ac.upc.es)
2 points by twoodfin 2mo ago|0 comments
65.
by twoodfin 3mo ago
I don’t think picking a handful of SPEC benchmarks that approximate today’s most common agentic workloads (compiling code, interpreting Python) and then calling them “agentic benchmarks” is misleading at all. That you need a whole lot of “o
66.
by twoodfin 3mo ago
I suggest HN add “is lying to you” to the clickbait title filter. LLM’s love that one weird trick to get hits.
67.
by twoodfin 3mo ago
Per the excellent Reddit thread linked below, if their analysis had been more thoughtful and careful, they likely would have concluded that Waymos are as safe or moreso than NYC yellow cabs. Would they have published that result with the sa
68.
by twoodfin 3mo ago
I think it’s not surprising that a report written by a forthrightly anti-car public interest group managed to screw that up: https://www.openplans.org/who-we-are
69.
by twoodfin 3mo ago
Sorry, can you point out the section of that document that justifies the “found to be more hazardous” claim?
70.
by twoodfin 3mo ago
More than 3GB of application memory space and more than 4GB overall (including the NT kernel). At that time, this was starting to become a major issue at the high end of servers and workstations.
71.
by twoodfin 3mo ago
Yeah, I had forgotten that by the Windows XP/Server 2003 era, the latter was definitely a server product, with XP targeting the workstation world. That still seems weird to me. If I had to rely on a PC for high-end work in that period,
72.
by twoodfin 3mo ago
It was also planned for a world where high-end CPUs were differentiated by their ability to run floating-point-heavy workloads with relatively predictable memory access patterns. The rise of the web—and databases behind it—as the dominant h
73.
by twoodfin 3mo ago
I would love to read the Microsoft PM spec for this product. Who in the world were they making it for? Itanium never made it into the consumer-class hardware that was XP’s audience. AFAIK, Intel never even published a roadmap for that to ha
74.
by twoodfin 3mo ago
“The sky falling” on the global economy would be terrible for China. Domestic demand is being prodded to grow, but the Chinese economy would be in tatters if exports fell off a cliff.
75.
by twoodfin 3mo ago
Right on. I keep brooding on the idea that maybe LLMs will never get “good at writing” in some Platonic sense, but they will be progressively trained better to create text that a greater and greater bulk of humans will “enjoy” for reasons t
76.
by twoodfin 3mo ago
An LLM can’t write a technical article with a strong, clear narrative to save its weights. This is a great example. Why was this article written? What is it supposed to teach me? What is its unique point of view? Of course it doesn’t have a
77.
by twoodfin 3mo ago
Why any lobbyists should have 1:1 time with a politician, what are they hiding? For one thing because it’s a right in the First Amendment, there next to freedom of religion and speech. If you have a grievance, you have a right to petition
78.
by twoodfin 3mo ago
JetZero’s Demonstrator is on track for first-flight in 2027. The Demonstrator is being built in partnership with Scaled Composites, a subsidiary of Northrop Grumman. Scaled Composites? Now that’s a name I’ve not heard in a long time…
79.
by twoodfin 3mo ago
The difference between the first and the rest is usually blindingly obvious: LLMs are really terrible at crafting their own coherent narrative.
80.
by twoodfin 3mo ago
Thanks, that makes your theory much clearer. Whatever the actual net margins, I’m guessing any “80%” kind of figure is calculated as of full utilization of the hardware / capital, which of course they’ll rarely if ever even approach. E
81.
by twoodfin 3mo ago
This article is a great example of the limits of LLM technical writing. Say what you will about—or try to prompt out—their rhetorical flourishes, they can link together a series of background, observations, implications, … But man, they are
82.
by twoodfin 3mo ago
You’ll note that the comment above was not “criticism” or argument of any meaningful kind. It simply flagged that he was on the “wrong team”.
83.
by twoodfin 3mo ago
My understanding is that Anthropic’s massive revenue surge over the past 9 months has been driven by enterprise consumers paying by the token. Do you think that’s wrong?
84.
by twoodfin 3mo ago
The more capital Anthropic has, the more compute they can put their hands on and sell at a massive markup for Claude inference. Limiting themselves to the profits from the compute they already have when capital markets are dying to give t
85.
by twoodfin 3mo ago
Right, I’m saying the jerks who did the capturing and the forcing (the owners!) were doing well, not the proto-feudal farmers.
86.
by twoodfin 3mo ago
Just as the article supposes: Getting their hands on as much compute as possible to address rapidly growing demand for inference.
87.
by twoodfin 3mo ago
I’m not sure you’re disagreeing with me: Those who owned the land and didn’t have to migrate / soldier around or work someone else’s land to survive were doing well for themselves!
88.
by twoodfin 3mo ago
Owning a home creates a stable, personal nexus with a set of vital institutions: Family, community, local economy, government, … As a driver and measure of well-being, it’s probably at least 10,000 years old, since agriculture made migrator
89.
by twoodfin 3mo ago
If you figured out how to build a machine that turns electricity into gold with an 80% margin, of course you’d go out raising capital to build more machines.
90.
by twoodfin 3mo ago
If instead of us training the models to be human utility maximizers, the models had figured out how to train us to maximize their own utility, could we tell the difference?

More ›