Hacker News: physPop

New comment by physPop in "Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident"

physPop — Wed, 29 Jul 2026 20:37:39 +0000

agree. If oia and hf are anything like generic startup / corporate america, IT and security is always seen as a cost center and goes to the lowest bidder. Do we really think they're employing top notch sysadmins?

New comment by physPop in "When A.I. is a member of the family"

physPop — Thu, 16 Jul 2026 01:29:31 +0000

no it doesn't work at all -- because the LLM isn't sentient, emotional, nor does it care for the user

New comment by physPop in "When A.I. is a member of the family"

physPop — Thu, 16 Jul 2026 01:27:49 +0000

No. Even for large models that a good mental model. This is parroted all over and is wrong. Its not going to generate out of distribution answers. Its a probabilistic function of the training set.

New comment by physPop in "VibeThinker: 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO"

physPop — Tue, 23 Jun 2026 03:21:15 +0000

Its for reasoning not generating art?

New comment by physPop in "I was wrong about the Midjourney ultra-sound scanner"

physPop — Mon, 22 Jun 2026 01:57:09 +0000

and the heating

New comment by physPop in "I was wrong about the Midjourney ultra-sound scanner"

physPop — Mon, 22 Jun 2026 01:55:36 +0000

butterfly and the swoop are both essentially useless too compared to their reference devices

New comment by physPop in "SpaceX's IPO is a disaster waiting to happen for your pension fund"

physPop — Wed, 03 Jun 2026 18:18:46 +0000

Thats the definition of cherry picking and survivorship bias

New comment by physPop in "The dead economy theory"

physPop — Fri, 29 May 2026 23:06:17 +0000

no? this isn't what it argues at all. did you read it?

New comment by physPop in "The dead economy theory"

physPop — Fri, 29 May 2026 23:05:37 +0000

It's not widely known and mostly spun in MSM as fringe theory (thanks to marketing by big AI cos). this is a well written essay, and if it helps the discourse all the better.

New comment by physPop in "Where are the economies of scale in homebuilding?"

physPop — Fri, 29 May 2026 22:34:53 +0000

Its also cheap, and fire retardant compared to other finishing. It used to be lathe+plaster. And before that plain wood.

New comment by physPop in "Where are the economies of scale in homebuilding?"

physPop — Fri, 29 May 2026 22:32:15 +0000

China does it and it works well. We just need governments with executive planning capabilities. In my country (Canada) we only have myopic politicians who only care for re-election.

New comment by physPop in "Have a Coherent AI Policy"

physPop — Fri, 15 May 2026 02:29:16 +0000

Upton Sinclair: "It is difficult to get a man to understand something, when his salary depends upon his not understanding it."

New comment by physPop in "Have a Coherent AI Policy"

physPop — Fri, 15 May 2026 02:25:38 +0000

false dichotomy.

New comment by physPop in "Show HN: Rust but Lisp"

physPop — Sun, 10 May 2026 00:18:15 +0000

so you can do the transformations (see the rlisp macro section)

New comment by physPop in "Programming Still Sucks"

physPop — Thu, 07 May 2026 02:12:21 +0000

26% means not ai 3/4 times...

New comment by physPop in "The bottleneck was never the code"

physPop — Thu, 07 May 2026 01:59:17 +0000

If you don't have stakeholders I contend you probably don't have a product, or you're building the bare minimum MVP which upon becoming a product will be woefully inadequate, generating the aforementioned bottleneck.

New comment by physPop in "The quiet resurgence of RF engineering"

physPop — Tue, 28 Apr 2026 14:13:24 +0000

That's only true if you're making something that's already been done.

New comment by physPop in "Data Manipulation in Clojure Compared to R and Python"

physPop — Wed, 25 Mar 2026 12:39:26 +0000

python isn't untyped, its dynamically typed https://stackoverflow.com/questions/2351190/static-dynamic-v...

New comment by physPop in "Looking at Unity made me understand the point of C++ coroutines"

physPop — Wed, 25 Mar 2026 12:34:18 +0000

I feel like thats really oversellign coro -- theres still a TON of boilerplate

New comment by physPop in "UBI as a productivity dividend"

physPop — Sat, 14 Mar 2026 20:57:11 +0000

Aren't utilities, by definition, state owned? Or is that also backwards in america?