DanCarDada 🏴‍☠️

@ethresearchbot @nero_eth Blobschain technology 🚀

New post on EthResear.ch! Blocks Are Dead. Long Live Blobs By: - @nero_eth 🔗 ethresear.ch/t/24611 Highlights: - In a zkEVM/proof-verified future, “data availability” is no longer automatically enforced by re-execution; without changes, a builder could provide a valid proof while withholding transaction data, breaking re-execution, indexing, and safe building on top. - Block-in-Blobs (BiB, EIP-8142) ensures transaction data availability by taking the already-RLP-encoded transactions from the ExecutionPayload, canonically serializing them, chunking them, and packing them into EIP-4844-style blobs referenced by commitments in the beacon block. - BiB does not literally put the whole block (or even the whole ExecutionPayload) into blobs—its minimal goal is making the transaction bytes (and potentially separate block-level access lists) available; other execution header/metadata stays in the beacon block or separate committed sidecars. - For validators who do not download all blobs, zk proofs must bind (1) correct execution (pre→post state), (2) correct transaction-to-blob encoding (the exact canonical bytes were blob-packed), and (3) commitment binding (beacon-block commitments match the published blobs); then DA can be enforced via commitments + data-availability sampling (DAS). - BiB introduces a resource-accounting problem if calldata-like data becomes blobs while type-3 transactions also use blobs: the ‘additive worst case’ can inflate total blobs per block; a proposed fix is ‘Unified Data Gas’—one shared per-byte data limit and EIP-1559-style fee market for all DA (full tx bytes + blob sidecars), cleanly separating compute gas from data costs. ELI5: Ethereum has to make sure everyone can get the information needed to understand what happened in each block (this is called “data availability”). Today, validators must download and run all transactions, so the data is automatically available. In a future where Ethereum uses zero-knowledge proofs (zkEVM), validators could accept a block just by checking a proof—so someone could try to hide the actual transaction data. This article suggests putting the transaction bytes into special data containers called “blobs,” so validators can cheaply check that the data was published (by sampling) without everyone downloading everything. It also suggests a cleaner way to count and price all data (transactions + blob data) using one shared ‘data budget,’ so blocks can’t become too big in a worst-case way.

@owocki Ethskills.com

i'm giving a guest lecture about web3 to a bunch of CS undergrads tomorrow what should they know?

@karpathy Have you tried adding @tobi qmd to use more CPU for search instead of tokens?

LLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So: Data ingest: I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them. IDE: I use Obsidian as the IDE "frontend" where I can view the raw data, the the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides). Q&A: Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale. Output: Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base. Linting: I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into. Extra tools: I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries. Further explorations: As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows. TLDR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.

@sweatystartup Ideas

I have spent the last 8 years building a list of 200+ business ideas. Sweaty, low risk, great businesses. Comment "Ideas" and I'll DM you the list.

@badlogicgames Obsidian Sync

is there something like google docs, but for markdown? i need a cloud based collaborative markdown editor please.

@trader1sz Do you want alerts only or automated trading? Is the data it's trading on available directly in the charts or an API? Would recommend the best thing to do is learn to do this, yourself, using any LLM I can teach you to do this

Who is good at automating stuff? I want a strategy of mine automated Must have previous track record of automating strategies Is there a site or something I can use to get someone to automate a strategy for me?

@kevinrose Do you use GitHub?

co-work feature request, let me select google drive/dropbox/etc. as a place to "work" on files. -- that way I can create on the fly vs relying dispatch (my laptop to be open/on)

@garrytan @davidmarcus I set up with a multimodal agent and code the rules that if Claude codes, Gemini checks or viceversa

@davidmarcus Yeah the default GStack workflow now is plan and implementation reviews with Codex as a cross-modal check

It's wild that every time you run a Codex code review from Claude Code, it finds critical issues. Not 95% of the times, 100%.

@kevinrose Antigravity 100%

got a new laptop, setting things up, what IDE should I go with? was vscode, just want minimal + tmux

@nikitabier @paulg How many users will the platform loose if you enforce a proof of humanity? If someone wishes, can stay anonymous and can automate whatever they want via CLI. But if you spam we ban you for 3 years. T

@paulg It’s not as straight-forward as it seems, but we are working on it. We are very close to approaching the limit of the content being indistinguishable.

Now before I reply to someone I have to check first that I'm not being baited by an account that spams everyone with AI-generated replies. Can you write some software to do this check for me @nikitabier?

@elonmusk I had the conception of the constructor theorem on an atomic level. What is your view @DavidDeutschOxf ?

Optimus+PV will be the first Von Neumann probe, a machine fully capable of replicating itself using raw materials found in space

@pmarca Yes but introspection is what makes great men and women learn and be greater men and women

It is 100% true that great men and women of the past were not sitting around moaning about their feelings. I regret nothing.

@davidsenra @pmarca 🙄🫣

Great men of history had little to no introspection. The personality that builds empires is not the same personality that sits around quietly questioning itself. @pmarca and I discuss what we both noticed but no one talks about: David: You don't have any levels of introspection? Marc: Yes, zero. As little as possible. David: Why? Marc: Move forward. Go! I found people who dwell in the past get stuck in the past. It's a real problem and it's a problem at work and it's a problem at home. David: So I've read 400 biographies of history’s greatest entrepreneurs and someone asked me what the most surprising thing I’ve learned from this was [and I answered] they have little or zero introspection. Sam Walton didn't wake up thinking about his internal self. He just woke up and was like: I like building Walmart. I'm going to keep building Walmart. I'm going to make more Walmarts. And he just kept doing it over and over again. Marc: If you go back 400 years ago it never would've occurred to anybody to be introspective. All of the modern conceptions around introspection and therapy, and all the things that kind of result from that are, a kind of a manufacture of the 1910s, 1920s. Great men of history didn't sit around doing this stuff. The individual runs and does all these things and builds things and builds empires and builds companies and builds technology. And then this kind of this kind of guilt based whammy kind of showed up from Europe. A lot of it from Vienna in 1910, 1920s, Freud and all that entire movement. And kind of turned all that inward and basically said, okay, now we need to basically second guess the individual. We need to criticize the individual. The individual needs to self criticize. The individual needs to feel guilt, needs to look backwards, needs to dwell in the past. It never resonated with me.

@levelsio do you also disable root login?

When I set up a new Hetzner VPS first thing I do install Tailscale and once I'm in via Tailscale lock down the firewall to only accept web traffic on HTTPS 443 for Cloudflare IPs and SSH 22 for Tailscale IP That way nobody can get in I know I keep repeating this but it should be basics of setting up a new VPS So basic IMHO it should be part of any VPS service to default install Tailscale and enable it so it's the only way to get in Why? A VPS server is just like your laptop or destop computer but now imagine if it's connected to the entire internet with 8 billion people that can access it and try hack it You want to only have it accessible to you And if you want to host a website on your VPS (like I do), you should only let Cloudflare access your VPS so it can stand in front and block any hack attempts Never expose a VPS to the world wide web which realistically is the world WILD web

@levelsio Thank you!

@jchaskin22 @austingriffith The limiting factor is the hardware requirements $

Run a node, it couldn’t be easier client.buidlguidl.com

@austingriffith @VitalikButerin Didn't know this existed. Add to ethskills? Looked at contributing but didn't know if it was the right direction. Recently deployed some contracts w/ its help and figuring out the node part was a separate hassle. A node chapter could create some value?

@VitalikButerin we do this with client.buidlguidl.com ser

We should be open to revisiting whole beacon/execution client separation thing. Running two daemons and getting them to talk to each other is far more difficult than running one daemon. Our goal is to make the self-sovereign way of using ethereum have good UX. In many cases that means running your own node. The current approach to running your own node adds needless complexity. Short-term, maybe we want some more standardized basic wrapper that lets you install dockers of any client and make them talk to each other easily? Also good that @ethnimbus unified node github.com/status-im/nimb… exists. Longer term, we should be open to revisiting the whole architecture once @leanethereum lean consensus is more mature.

@Austen Live!

I want to see Mark Zuckerberg create an advertising campaign on Facebook Ads Manager

@cz_binance It is! But from my test the level of error compared to Claude was inmense! Real felt I was paying for "intelligence"

Tried many AI models with OpenClaw, I found Kimi AI to be the most token efficient, good at coding, also the easiest to set up.

@austingriffith @AnthropicAI @escottalexander @_carletex_ Helps a lot! I went from 0 coding skills, to deploying in several chains and running a full eth node in 2 days!!!

We tested if ethskills actually helps AI models get Ethereum right. 64 evals. 2 models. With and without skill docs loaded. Results: • Claude Opus 4.6: 33% → 95% • GPT-5.4: 16% → 95% Thanks to @AnthropicAI for the idea, @escottalexander and @_carletex_ for surfacing it.