Hacker News: 3371

New comment by 3371 in "Micron Taiwan increasing salaries for factory workers by 2.7X"

3371 — Tue, 27 Oct 2026 12:29:22 +0000

Haha, I'm surprised no one brought this up: In Taiwan, there are Micron Taichung and Micron Taoyuan. In the past days, It is a hot topic that the executives of Micron Taichung betrayed the 10 thounsand members and sneakily signed the agreement, including a term to forbid any more strike in the next 3 years.

You may use your translation tools to read the Threads post: https://www.threads.com/share/_pvBLXXpx/

I'm not a Micron employee nor in chip industry.

New comment by 3371 in "Claude Haiku 5.5"

3371 — Wed, 07 Oct 2026 19:10:22 +0000

Seeing people talking about the Agents SDK -> credits change makes me wonder does it impact Zed or likes.

New comment by 3371 in "On social reality in China"

3371 — Sat, 03 Oct 2026 12:26:24 +0000

I'd say it's more about the education. Your education prioritize loyalty to the nation/party more than loyalty to humanity and oneselfs.

New comment by 3371 in "Gemini 4 Argon"

3371 — Wed, 30 Sep 2026 21:18:27 +0000

More importantly, they can stuck at reasoning loop!

New comment by 3371 in "The AI Race Just Got Awkward"

3371 — Wed, 30 Sep 2026 16:58:19 +0000

Maybe look up "China deflation"

Claude Desktop Silently Associates with Common Files

3371 — Thu, 17 Sep 2026 14:38:26 +0000

v1.44121.1 2026-09-02 > Added Claude to the “Open with” menu for common work files, including spreadsheets, PDFs, Word and PowerPoint documents, text files, and images, on macOS and on Windows (Microsoft Store and MSIX installs).

I don't recall I ever approved anything, somehow the Claude app associate itself with some common formats like .json, .csv, .markdown, .rtf, .docx, etc. on my Windows 10 machine, even though I already have Cursor, Notepad, New Notepad App, LibreOffice... that can handle these files and associated.

I am not sure how this happened and how would Windows (10) allowed this to happen. Regardless, I don't appreciate the move trying to become "the default" at all, considering it's a rate-limited, connection-required app that does not really allow the user to directly edit the file.

Comments URL: https://news.ycombinator.com/item?id=49741481

Points: 6

# Comments: 1

New comment by 3371 in "Detecting and countering misuse of AI: September 2026"

3371 — Fri, 11 Sep 2026 12:40:38 +0000

Unfortunately it's much less likely to happen for their managed models because surveillance is built-in in every public-facing service since day-one.

New comment by 3371 in "DeepSeek v4.1 Flash"

3371 — Fri, 11 Sep 2026 05:24:57 +0000

Interestingly the latest report from Anthropic explained that they may be just routing requests to managed models to Anthropic.

New comment by 3371 in "DeepSeek v4.1 Flash"

3371 — Thu, 10 Sep 2026 19:19:43 +0000

This raise a question: why do open source models sometimes identify themself as Anthropic's models. I recall seeing some plausible theories in the past but I can't recall.

New comment by 3371 in "Portal by Spotify cut my Claude Code token usage by 90%"

3371 — Sat, 05 Sep 2026 07:30:58 +0000

codegraph + context mode are all I need.

New comment by 3371 in "Any Human Ever – One life, drawn at random from all who have ever lived"

3371 — Fri, 04 Sep 2026 07:18:47 +0000

"He was born, lost his father at 13, was orphaned at 20, had 4 children, and died at 90 of a parasitic infection, with 3 surviving children, 10 grandchildren, and 15 great-grandchildren."

This 3727 BCE Korean boy died at 90.

New comment by 3371 in "Agent memory as a file format"

3371 — Mon, 31 Aug 2026 19:14:18 +0000

You may be interested in my No3371/projex repo, which is a whole solution based in this idea, I wonder how much do people do differently in this direction

New comment by 3371 in "Highlighting My Code Based on How Much I Care"

3371 — Mon, 31 Aug 2026 14:45:44 +0000

This makes we wonder if anyone have tried to track attention to code with eye tracker

New comment by 3371 in "Claudette: Make Claude stop talking like a BuzzFeed article"

3371 — Sat, 22 Aug 2026 06:11:53 +0000

I don't think it's possible to effectively control LLM's output, unlike Human the difference between "thoughts" and "words" is very weak.

Better just clean stuff up afterward with a dedicated agent.

New comment by 3371 in "Maximizing the value of your Claude Code sessions"

3371 — Sun, 16 Aug 2026 15:43:18 +0000

Yeah, making it a skill is just much more convenient. It can be a very simple skill, a one liner asking the agent to gather the context and write a handoff for another agent. My skill is unnecessarily complex if that's the only requirement.

New comment by 3371 in "Maximizing the value of your Claude Code sessions"

3371 — Sat, 15 Aug 2026 15:57:39 +0000

Yes it's just another (but my own) impl of SDD

New comment by 3371 in "Maximizing the value of your Claude Code sessions"

3371 — Sat, 15 Aug 2026 15:56:12 +0000

It's just good old prompting in skill format. You can check out the skill repo: no3371/projex

New comment by 3371 in "Maximizing the value of your Claude Code sessions"

3371 — Sat, 15 Aug 2026 04:17:50 +0000

> Plus your handoff files becomes a useful piece of project memory that you can reference later.

I've been doing this since I started agentic development, and have a whole framework based on this; Simply put I define workflow s that output templated files for everytype of tasks that happens in development.

It's a powerful pattern I'd recommend everyone.

New comment by 3371 in "Firefox is now the last major browser that still supports uBlock Origin"

3371 — Sat, 15 Aug 2026 04:06:23 +0000

People believe ublock because it's reputable and not affiliated with browsers IMO.

New comment by 3371 in "Docker Sandboxes – Disposable, isolated sandboxes for AI agents"

3371 — Mon, 10 Aug 2026 08:01:57 +0000

I used this for a while then decided to build my own suites that pack individual harness and respective host state (config, plugins, skills, etc.) into an image. Works better and much flexible in my opinion.