Total order isn't needed to avoid double spending. It's needed to let you read balances provably from anyone. Eventually consistent protocols are trivially censorable.

They. Did. Not. Account. For. Services

There is a lot of (justified) of excitement about MCP, but just like most function calling approaches, I think it has a severe limitation at the moment, and I am kind of of flabbergasted why I am the only one that is bothered by it. The problem is that when you define a tool

Made a quick plugin to paste the doc of a Rust crate to Simon Willison 's llm tool, similarly to what Filippo Valsorda @filippo.abyssdomain.expert did for llm-fragments-go github.com/huitseeker/llm… Hopefully useful for #rustlang nerds 😀

I've had to learn much of this post the hard way, and met so many folks in my career who were unwittingly gearing up to rediscover the same. This post is a gift to anyone working on formal methods anywhere. Thanks Mike! /cc Galois

Amazing location for the #ZKsummit13

"no one is going to get to 3Bn people proving their identity in a privacy preserving way faster than us" - mareko.eth | Celo L2 🦇🌳 presenting @selfprotocol throwing down the gauntlet at #zkSummit13

Got my popcorn ready for Francois Garillot ‘s workshop at Protocol Berg v2

A moratorium on precompiles, including zkVM precompiles, being advocated for by vitalik.eth at Protocol Berg v2 - except for SIMD. I explained why that won't work 4 months ago x.com/huitseeker/sta…

8 months after the BEAM chain announcement, Guillaume Ballet, who actually tested the language interop promises of the RISC-V ZKVMs at Protocol Berg v2. And calls for standardization. Yes! Of course.!

At ACM #Facct2025, learning that LLMs face diminishing returns & comparatively worse task-specific performance when fixating on a larger scale arxiv.org/abs/2409.14160 (paper w/ @mer_edith)