frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Ask HN: Thoughts on maintaining anonymity against state-level actors?

3•adamanteye•1h ago
It has been nearly five years since Ruan Xiaohuan (https://en.wikipedia.org/wiki/Ruan_Xiaohuan) was arrested by Chinese authorities in May 2021. For those unfamiliar, he ran the legendary anonymous blog Program-Think (https://program-think.blogspot.com/) for over a decade.

What haunts me is that his identity was compromised despite his elite background. He was the chief engineer for the 2008 Beijing Olympics network security system. His OPSEC was rigorous: he operated on a cash-only basis, avoided all e-commerce, and never discussed his digital life with anyone, including his wife, who only learned of his "second life" after his disappearance.

Despite his expertise, it's still a mystery how his anonymous persona was deanonymized.

As a hacker in China, I’m interested in your thoughts on the feasibility of maintaining a truly anonymous identity long-term. Is it even possible to win a "battle" where you have to be perfect 100% of the time, while the adversary only needs to find one leak?

What are the most likely failure points in a high-level threat model like this?

Comments

johncoltrane•1m ago
> Despite his expertise, it's still a mystery how his anonymous persona was deanonymized.

The only explanation is that his OPSEC wasn't as rigorous as you think it was.

Reading the Wikipedia page, the most obvious gap would be the medical treatment he received in 2017. If he mentioned it publicly, then it was basically game over: finding him would have been routine police work at that point. The process that led to the interview by a German newspaper might have been leaky as well. There are so many opportunities.

As a rule of thumb, you can consider that while your OPSEC might _theoretically_ be the tightest in the world, you will eventually have to deal with other people and orgs at some time, who might not care as much as you do. In which case your OPSEC is really only as strong as _theirs_.

> Is it even possible to win a "battle" where you have to be perfect 100% of the time, while the adversary only needs to find one leak?

It is not. Simply because you can only control so much. A few years ago, there was a story about a mafia boss who successfully escaped justice for 20 years… until a Google Street View passed by while he was shopping groceries in a Spanish village. The guy certainly had the strongest OPSEC his money and relationships could buy, but it eventually amounted to nothing in the face of pure randomness.

All you can do is try to compartmentalize as best as you can for as long as you can, but something will eventually leak.

Also, it will be harder…

- the longer you keep it going,

- if it involves others,

- if you are married, have kids, etc.,

- if you have complicated needs (sex, drugs, health issues, etc.),

- if money is tight,

- if you are not geographically and socially mobile,

- etc.

Scout AI Revolutionizes Security Intelligence with Amazon OpenSearch Service

https://aws.amazon.com/solutions/case-studies/maxsecurity-bigdataboutique/
1•synhershko•1m ago•0 comments

The Befunge Programming Language

https://esolangs.org/wiki/Befunge
1•askl•1m ago•0 comments

Show HN: PhotoCraft – an AI photo editor I built and shipped as my first iOS app

https://apps.apple.com/us/app/photocraft-ai-photo-editor/id6756682393
1•devavinoth12•2m ago•0 comments

Achieving Kafka reliability at scale with the Streaming Platform (2025)

https://www.datadoghq.com/blog/engineering/streaming-platform-kafka-custom-abstractions/
1•teleforce•7m ago•0 comments

Kuo: Apple's AI Deal with Google Is Temporary and Buys It Time

https://www.macrumors.com/2026/01/13/apple-google-ai-deal-is-temporary/
1•mgh2•8m ago•0 comments

Lore, A reasoning engine that stores the "why" behind code changes

1•almonerthis•11m ago•0 comments

UK police chief admits policy relied on CoPilot hallucination

https://www.telegraph.co.uk/news/2026/01/14/maccabi-police-chief-admits-misleading-mps-by-using-ai/
3•nanna•12m ago•1 comments

Jensen Huang Is Begging You to Stop Being So Negative About AI

https://gizmodo.com/jensen-huang-is-begging-you-to-stop-being-so-negative-about-ai-2000709335
2•robin_reala•12m ago•0 comments

London cracked mobile phone coverage on the Underground

https://www.ianvisits.co.uk/articles/how-london-finally-cracked-mobile-phone-coverage-on-the-unde...
1•ganonm•15m ago•0 comments

Wine stable release 11.0.0 is now available for Linux FreeBSD and macOS

https://www.wine-reviews.net/2026/01/wine-stable-release-1100-is-now.html
1•twickline•16m ago•0 comments

Show HN: I got PyTorch models running on WebGPU without ONNX export

https://github.com/jmaczan/torch-webgpu
1•yu3zhou4•21m ago•1 comments

UK government rolls back key part of digital ID plans

https://www.theguardian.com/politics/2026/jan/13/government-rolls-back-digital-identity-card-plans
4•chrisjj•21m ago•0 comments

Premature Optimization in Entertainment Development

https://medium.com/luminasticity/on-premature-optimization-in-entertainment-development-d2f66083cb26
1•bryanrasmussen•28m ago•0 comments

Private Inference

https://confer.to/blog/2026/01/private-inference/
1•_____k•29m ago•0 comments

Rockstar vs. Union: We Went to Court and Saw the Evidence [video]

https://www.youtube.com/watch?v=dnuipPQDd_w
2•aragilar•29m ago•0 comments

A 16-inch Laptop that Expands to 23.8 inches of Screen Space

https://www.youtube.com/watch?v=PTyQnfXHKQs
1•guytv•30m ago•0 comments

I Love You, Redis, but I'm Leaving You for SolidQueue

https://www.simplethread.com/redis-solidqueue/
6•amalinovic•30m ago•2 comments

Police chief admits misleading MPs after AI used in ban justification

https://www.bbc.co.uk/news/live/c394zlr8e12t
3•asplake•31m ago•0 comments

Bulletproof Type Safety in Gleam: From Database to Client

https://blog.andreyfadeev.com/p/bulletproof-type-safety-in-gleam
1•andfadeev•33m ago•0 comments

Show HN: Visibility and Controls for Browser Agents (ContextFort YC S25)

https://contextfort.ai/
1•ashwinr2002•34m ago•1 comments

Signal creator Moxie Marlinspike wants to do for AI what he did for messaging

https://arstechnica.com/security/2026/01/signal-creator-moxie-marlinspike-wants-to-do-for-ai-what...
2•_____k•35m ago•0 comments

LLMs are a 400-year-long confidence trick

https://tomrenner.com/posts/400-year-confidence-trick/
4•Growtika•36m ago•0 comments

Let LLMs Visualize Data for You

https://medium.com/altitudehq/let-llms-visualize-data-for-you-a8a494abbbbf
1•vietthangif•38m ago•0 comments

Klarna launch peer-to-peer payments

https://www.klarna.com/international/press/klarna-expands-digital-bank-offer-with-peer-to-peer-pa...
1•danols•38m ago•0 comments

AI writes code faster. Your job is still to prove it works

https://addyosmani.com/blog/code-review-ai/
2•TheAnkurTyagi•42m ago•0 comments

Show HN: GLM-Image Online – 16B AR+Diffusion model for accurate text

https://glmimage.online/
1•hugh1st•43m ago•0 comments

In Praise of Writing (and the Case Against AI)

https://jaapgrolleman.com/ai-writing/
1•jaapgrolleman•44m ago•1 comments

AI Memorization Research

https://www.theatlantic.com/technology/2026/01/ai-memorization-research/685552/
2•tagyro•45m ago•0 comments

Show HN: AI Contract Reviewer – Flags Risks and Suggests Fixes in Minutes

1•Saurabh_Kumar_•45m ago•0 comments

New tech and tools for retailers to succeed in an agentic shopping era

https://blog.google/products/ads-commerce/agentic-commerce-ai-tools-protocol-retailers-platforms/
2•geoffbp•45m ago•0 comments