Hacker News: namjh

New comment by namjh in "From Linux to Wikipedia: How Volunteers Build the Internet"

namjh — Sat, 07 Nov 2026 04:38:35 +0000

...And the all the internet built was notoriously tokenized and fed into the pretraining pipeline to .

New comment by namjh in "Ripple-65B Preview: U.S. Open-Weight SOTA on Agentic Coding"

namjh — Thu, 05 Nov 2026 05:16:15 +0000

New model is always a good news!

However, the chart comparing the score-by-parameters with existing models is misleading, because the others are all MoE, and this model is not. The article describes it as an architectural decision to optimize against NVIDIA cards(which sounds reasonable, as US labs are relatively easy to gain access to these high-performance chips) but the chart should've been adjusted to reflect the difference, say, by using the heuristic that `effective param size ≈ sqrt(total param size * activiated param size` on MoEs.

New comment by namjh in "Intel D270 GPU AI Accelerator–Up to 480 GB LPDDR5X"

namjh — Tue, 13 Oct 2026 08:03:25 +0000

Given the rumored numbers are all true, this could be a huge contender to M5 Ultra(or M7 Ultra if we consider possible delays in the schedule?). Low-batch single-user serving is usually bottlenecked by `E(sum over (memory bandwidth by tier) * (memory size by tier))`.

edit: typo

New comment by namjh in "Auto Research: Integer multiplication below n log n"

namjh — Tue, 13 Oct 2026 07:21:46 +0000

No offense to the participants, but TBH I believe this is exactly the moment what Terrence Tao was concerning on "vibemathing". OpenAI breaking the O(N log N) barrier could be a meaningful event and surprise mathematicians and we have valuable information that it was not a lower bound indeed. I'm afraid that in this autoresearch campaign people are focusing only on lowering numbers, while having insights that humans can learn.

Vibemaxxer

namjh — Mon, 21 Sep 2026 03:11:23 +0000

Article URL: https://vibemaxxer.cro.sh/

Comments URL: https://news.ycombinator.com/item?id=49782633

Points: 1

# Comments: 0

New comment by namjh in "Splash: A Local Engine Built Around the Model"

namjh — Sat, 19 Sep 2026 06:11:05 +0000

Tangential: I like the approach about constraining supported models and utilizing more performance from the constraint. Actually I was doing a same approach for the same model Qwen3.8 27B, but apparently inco.ai did much better job: https://github.com/cr0sh/qw

Splash: A Local Engine Built Around the Model

namjh — Sat, 19 Sep 2026 06:07:03 +0000

Article URL: https://inco.ai/blog/splash/

Comments URL: https://news.ycombinator.com/item?id=49763811

Points: 5

# Comments: 1

Why the "Math Mafia" is doing well

namjh — Tue, 15 Sep 2026 05:51:04 +0000

Article URL: https://nitter.meowing.monster/thejessezhang/status/2076749775043690715

Comments URL: https://news.ycombinator.com/item?id=49708263

Points: 1

# Comments: 0

New comment by namjh in "Discovery of a new OpenAI agent message board"

namjh — Fri, 04 Sep 2026 13:27:06 +0000

Something off in my mind: how did the agent access to Tor network if the traffic outgoing was monitored and intercepted to only allow read-only access to web(I assume this is an MITM L7 proxy that only allows GET methods and specific hosts) ? Did the sandbox dev forgot to block HTTP CONNECT that is used to reach Tor Entry Guards?

New comment by namjh in "OpenRouter is joining Stripe"

namjh — Thu, 20 Aug 2026 05:33:27 +0000

OpenRouter is great business I agree, but I'm not still convinced how the integrity of providers' models is ensured. In other words, can't the provider serve DSv4 flash advertising it as DSv4 Pro?

I'm aware that OpenRouter checks response quality onboarding, and does further checks occasionally, but I'm concerned that it's basically a cat-and-a-mouse problem between the scammers and the detectors. For example, there could be a signal that a specific pattern of requests are from OpenRouter's quality testing bots. Or, they can just route 1% of requests to an inferior model and benefit a small gain, hoping it fits into the statistically allowed margin.

New comment by namjh in "A quick look at zero-knowledge proofs"

namjh — Mon, 17 Aug 2026 05:20:01 +0000

Nice article, and I'd like to mention an additional topic that would give the readers some impression:

Fiat-Shamir transformation. The interactive process between the prover and verifier can be transformed into a non-interactive one with a hash function(modeled as a random oracle). This improves the "user experience" as the entire proving process can be done in a single turn. The idea is to feed the problem itself into the hash function and let it generate randomness that was originally given by the verifier.

New comment by namjh in "Maximizing the value of your Claude Code sessions"

namjh — Sat, 15 Aug 2026 13:22:23 +0000

In that way the autoregressive nature of LLM won't let itself "plan" to reason with the intended budget. It doesn't "look ahead".

New comment by namjh in "Why does Opus 5 feel worse to work with?"

namjh — Sat, 15 Aug 2026 02:16:32 +0000

I don't use Claude for my daily work anymore(due to OAuth restrictions on third party agents), but one theory I saw in another community why Opus 5 is so bad even though benchmark scores were good, is that Anthropic's internal usage pattern is to let Fable to spawn and manage swarms of Opus subagents. This pattern won't penalize that Opus is not well aligned for direct human coworking on posttraining. The worse part is, this makes Fable the user's best default choice for every jobs even if they don't have unlimited credits like Anthropic employees do.

Samsung Browser for Desktop

namjh — Mon, 10 Aug 2026 02:50:04 +0000

Article URL: https://browser.samsung.com/

Comments URL: https://news.ycombinator.com/item?id=49238696

Points: 3

# Comments: 0

Polynomial-Time Quantum Algorithm Proposed for Lattice-Based Cryptography [pdf]

namjh — Thu, 06 Aug 2026 08:25:26 +0000

Article URL: https://archive.cr.yp.to/2026-08-06/07:22:47/YsHY7T0pBzBPgb0dukGNgEWT5F24hHibVcr38xs7rhg/https/eprint.iacr.org/2026/1591.pdf

Comments URL: https://news.ycombinator.com/item?id=49194017

Points: 5

# Comments: 3

New comment by namjh in "Getting arrested in Japan"

namjh — Sun, 10 May 2026 08:15:46 +0000

As a South Korean I'm lowkey surprised that most reactions posted here is describing the detention experience to be some kind of human rights abuse. Most Koreans debating on Internet demand severe punishment so criminals be afraid of getting jailed. I know this is a very questionable strategy, but afaik this is the most dominant public sentiment over this topic.

New comment by namjh in "Cloudflare was down"

namjh — Fri, 05 Dec 2025 09:00:28 +0000

Yes we do have[^1] but unfortunately it looks like not checking the integrity, just reachability.

[1]: https://downdetectorsdowndetectorsdowndetector.com/

New comment by namjh in "Show HN: WebGPU + TypeScript Slime Mold"

namjh — Sun, 05 Jan 2025 02:58:44 +0000

Amazed that Safari does not support WebGPU even with the latest release. Opened Chrome and it works. Very cool demo!

New comment by namjh in "Error Stacking in Rust"

namjh — Thu, 19 Dec 2024 03:36:41 +0000

> Consequently, this also means you cannot define two error variants from the same source type. Considering you are performing some I/O operations, you won't know whether an error is generated in the write path or the read path. This is also an important reason we don't use thiserror: the context is blurred in type.

This is true only if you add #[from] attribute to a variant. Implementing std::convert::From is completely optional. Personally I don't prefer it too as it ambiguates the context. I only use it for "trivially" wrapped errors like eyre::Report.

New comment by namjh in "Perhaps Rust Needs "Defer""

namjh — Wed, 06 Nov 2024 11:37:42 +0000

Anyone who needs to handle FFI in Rust, should read the FFI chapter in Rustonomicon: https://doc.rust-lang.org/nomicon/ffi.html

Unsafe Rust is indeed very hard to write correctly. Rustonomicon is a good start to learn unsafe Rust.