Sabitlenmiş Tweet
Some random AI project on Base was using the majority of transactions per block when I was watching the chain yesterday with my homegrown chainwatching tool.
English
Daniel Von Fange
3.5K posts
Daniel Von Fange
@danielvf
Skilled Professional (most days). Defends against the bad guys.
East Coast Katılım Eylül 2006
1.2K Takip Edilen12.2K Takipçiler
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.
English
@functi0nZer0 I just want batch transactions, and someone else able to sign to pay a TX. And that could be in Hegota.
So I'm almost a simple man.
English
@Alessiozazza Top two concerns are that if we move to world where we are executing arbitrary code and loading lots of state from disks, just to know if a transaction should be considered, it gets hard to do thousands of transactions per second.
Secondly, this one:
x.com/danielvf/statu…
Daniel Von Fange@danielvf
TCP/IP is great but it would be a lousy spec for intrabank transfers. No one would know what to say over the channel. The great benefit of a standard is that it solves the O(N²) complexity of intraparty communication, and turns it into each party just coding to a single spec. It would be hell-on-earth for app and wallet developers to be unable to know what kinds of frame transactions would actually get through the mempool. The current plan is not only unknown evolution over time on what is accepted, but also having per provider differences in acceptance criteria. That's really bad combo if I'm trying to design a smart account wallet that I'm going to be putting 400K into auditing and testing. At a minimum there should be at least a spec on what is guaranteed to be universally accepted.
English
When you say “downstream” here, do you mean wallet UX getting weird, infra like bundlers and relays centralizing, or app devs losing control over how users sign and pay fees? Curious which failure mode you’re most worried about and what you’d want to see in the spec to make it feel sane
English
I made an ugly chart.
Ethereum is currently discussing Frame Transactions as a future headliner. I'm concerned about downstream complexity but the spec has gotten much better over the past two weeks.
English
It's clear that EIP-8130 is born from real experience with real pain points. Something I wish all EIP's had. I've only read 8130 two times, and haven't thought deeply on it enough to intelligently comment.
My gut feel is that there's a simplification somewhere that would make it much smaller.
English
TCP/IP is great but it would be a lousy spec for intrabank transfers. No one would know what to say over the channel. The great benefit of a standard is that it solves the O(N²) complexity of intraparty communication, and turns it into each party just coding to a single spec.
It would be hell-on-earth for app and wallet developers to be unable to know what kinds of frame transactions would actually get through the mempool.
The current plan is not only unknown evolution over time on what is accepted, but also having per provider differences in acceptance criteria. That's really bad combo if I'm trying to design a smart account wallet that I'm going to be putting 400K into auditing and testing.
At a minimum there should be at least a spec on what is guaranteed to be universally accepted.
English
@danielvf why not let people do whatever at the protocol layer, but enforce more restrictive rules at the mempool layer?
that pattern has existed in bitcoin for over 10 years ("standard transactions")
mempool layer is much easier to upgrade over time than the protocol
English
@_prestwich eh, this EF Mandate is basically improving core parts of Ethereum. I can't complain about that.
English
the EF Mandate has always been to pursue whatever poorly thought out philosophy VB is huffing this year
English
@raulvk @TobiasKlenze Each minute Ethereum sends out 5 blocks and Monad sends out 150.
So if you convert those bandwidth requirements to to per block numbers, then Monad requires considerably less bandwidth.
(There may be other reasons not to use raptorcast, but bandwidth shouldn't be the problem)
English
this is not using gossip, it's erasure-coded broadcast.
- Monad and Solana have deterministic topologies
- this involves permissioning the validator network plane and publicly listing endpoints
- Ethereum's network is permissionless, we cannot route deterministically
- Monad requires 300Mbps bandwidth, Solana more.
- Here are Ethereum's requirements, keeps the network universally accessible and decentralized
English
A 500-node P2P network decoding a 2 MiB message with ethp2p Reed-Solomon erasure-coded broadcast (0.25x viz speed).
Initial benchmarks: ~1.4s p95 decode latency vs ~7.15s with gossipsub. 5x faster, 40% of the bandwidth.
This is with minimal tuning. Plenty of room to tighten.
Realistic topology modelled after Ethereum mainnet (downscaled), deterministically simulated on Shadow.
The visualizer handles 1.7M trace events without breaking a sweat, built on WebGL2/WebGPU via deck.gl.
Purple = sender, blue = receiving, green = decoded.
English
Cheap as air blockspace.
This week AAVE forks were getting hit on fast chains because the cost to loop a call that makes only the tiniest smidge of money costs so little in gas that it was profitable.
kaden.eth@0xKaden
back when flash loans were first popularized, it opened up a whole new vulnerability class before then, oracle manipulation never seemed to be feasible, but suddenly attackers realized they had access to billions of dollars needless to say, a lot of protocols got hacked i wonder what vulnerability classes still lay dormant, waiting for a novel mechanism to unlock them 🤔
English
@danielvf Now turn it into sound and listen to it while you go to sleep, such that anytime an incorrect note is played you know there is a bug present
English
It just looks like noise, until you see the patterns.
Made a matrix eye view of every blockchain transaction as they happen. I keep this up on second monitor and see all kinds of things go by. Site link below.
English
ZXX
Introducing Edge, a high level, strongly statically typed, multi-paradigm domain specific language for the Ethereum Virtual Machine (EVM).
github.com/refcell/edge-rs
English
Ah, I wasn't speaking of the morality of the attacks, but of their expediency. At first glance attacking these other countries appears to be strongly against the government of Iran's own interests.
As an example, the UK was blocking the United States from attacking Iran from its territories, which made things harder for the US. Even if the UK does never retaliate against Iran, that restriction is almost certainly lifted from the US now.
English
@danielvf Better question: Has any nation in history kidnapped or killed this many head of the states in a few months?
You don't get to skip the provocation and start the clock at the response.
English
My one public question on the current war:
Has any nation in history ever physically attacked so many non-belligerents in one week? This is a new record for adding enemies, right?
English
@dhh It also defacto mandates that all applications on your computer be able to infer your child's birthday, since apps can watch for age zone rollovers.
English
No one has yet figured out what do with the freakish amounts of bandwidth for transactions that new generation blockchains have.
English