tomrod

5 ms·

31.
by tomrod 25d ago
This seems unnecessarily harsh. My prior experience with science popularizers is that they are typically telling a potential, and either objectively likely or subjectively hoped for, narrative plausibly supported by the evidence. "Scie
32.
by tomrod 25d ago
And thus $9k of <something they got that $9k worth of value for> is injected into the economy, either assets sold or work performed.
33.
by tomrod 26d ago
Your claim confuses me. This history of math is replete with people doing it for the love of the game.
34.
by tomrod 26d ago
I've seen a number of these too. Slimy is the word that comes to mind. Spamgents are going to bring about the dark forest in email above and beyond spam calls.
35.
by tomrod 26d ago
Again. Weird. Doesn't jive with my experience. One or two in the dozens I know, sure, but not enough to be a stereotype.
36.
by tomrod 26d ago
A fellow Brin aficionado!
37.
by tomrod 26d ago
Say they do. What then? Will people buy from them? What if we decide not to?
38.
by tomrod 26d ago
> was the point of math ever just because some humans enjoyed doing it? Yes. Friends and I often work on Putnam problems and this series: The (Almost) Impossible Integrals, Sums, and Series by Cornel Ioan Vălean
39.
by tomrod 26d ago
> Job protectionism for elite mathematicians under guise of caring about student development. Huh. Weird. This hasn't been my take of mathematicians at all. The dozens I know are quite humble and dedicated to math and the beauty one
40.
by tomrod 26d ago
Nah, they only recently got FedRAMP high. The real traffic still can't use it.
41.
by tomrod 26d ago
Hacker News is genuinely one of the few communities that seems to improve as it ages and grows. Somehow, despite strong concerns about an eternal September, it seems to say "Hey, come in and lets learn together." I just think its
42.
by tomrod 27d ago
Or the companies hosting the frontier AIs are leeching the conversations.
43.
by tomrod 27d ago
Say more. I need me proper rust in the browser.
44.
by tomrod 27d ago
In more accurate assessment rather than assuming no maliciousness nor incompetence, remembering also happens. Unless we fail as a society, the simpler explanation that "OpenAI is training on all data it can and resetting config toggles
45.
by tomrod 28d ago
> need essentially the same things Eh. Bacon futures disagree with this hypothesis.
46.
by tomrod 28d ago
This is a great way to think of it, idiotsecant. That flavor profile is known -- it's typical behavioral distribution is somewhat understood (and, often, common failure modes addressed). If JSON breaks about 20% of the time, and that d
47.
by tomrod 28d ago
Nope, strong disagree. The model is one small part of the process harness; behaviors are usually routable with expected propensities. Unexpected model changes avoiding change management messes with monitoring and observability thresholds. S
48.
by tomrod 28d ago
GrapheneOS for the win, here.
49.
Cloudflare Startups Program(cloudflare.com)
3 points by tomrod 28d ago|0 comments
50.
by tomrod 29d ago
I'm working on a practical review implementation on this! Great to hear others are thinking along the same way.
51.
by tomrod 1mo ago
Timely. I'm also looking at this now, actually! We tend to throw benchmark after benchmark at systems, but miss that models are one part of the system. Harnesses are more than models and need tuning too, and in doing so there can be ga
52.
by tomrod 1mo ago
Sweet. This will be a great layer for our hazard prediction engine.
53.
by tomrod 1mo ago
Human values ideally start deeper, in the creation of preferences, where economics simply takes as a primitive/preexisting part of the environment. Marketing theory is the one that focuses on manipulating human value.
54.
by tomrod 1mo ago
Fair callout. Strongly discounting (b) and (a) is underspecified for me to respond reasonably. Your premise of full token commoditization has a major wrinkle though. GPU-centric data centers with never allow tokens to go to zero since you h
55.
by tomrod 1mo ago
Self-hosting takes either cloud- or on-prem-engineering chops, which are in short supply. In my view, that becomes the primary driver for lowering token costs. Secondary are the alternative model sources like Z.ai and Moonshot, which betwee
56.
by tomrod 1mo ago
> I wonder how models perform on finding analogies between analogies Load-bearingly verbose, in my experience.
57.
by tomrod 1mo ago
Not sure why you were flagged either -- it was a reasonable comment that clearly spawned a discussion! People are strange. > There's no essential reason to have utility companies, dams, nuclear arsenals, factory production floors, m
58.
by tomrod 1mo ago
A classic you may enjoy: https://news.ycombinator.com/item?id=18442941
59.
by tomrod 1mo ago
> wonder if long form writing will go the way of code. I worry about AI Loopidity here though. Think about the similar analogy of email. If my set of ideas is condensable to bullet points, but I use AI to expand the content, then I add n
60.
by tomrod 1mo ago
That's awesome to hear. I look forward to trying it out and, ideally, seeing SLMs/open weights model following suite.

More ›