Technical AMA - will respond to any technical Q about blockchains, vms, performance, privacy, security, tooling, open source etc.

Hoping to meet programmers or architects with similar interests at the frontier :)

cto & general partner @paradigm. working with /rust on /ethereum. building /reth and /foundry. https://gakonst.com

(casted when boarding plane, just landed, to get to these by the morning!)

Technowatermelon. Elder Millenial. Building Farcaster. 

nf.td/varun

I think somebody will need figure out how to connect account abstraction with stealth addresses plus hiding payment amounts. gas sponsorship will let anyone pay for your tx, batch txs will mean you can spend from multiple authorized stealth addrs solving the fragmentation.

Whats the most likely path to get practical, private transactions in the Ethereum world ?

fractalist. tech lead at cabin.city. founder of LBRY.com, Odysee.com, vibe.camp, @farcon, grin.io

will require a wallet that is built for multiple chains and produces aggregate views across your assets independent of chainid, will require being opinionated, can't support all seamlessly.

Building ponder.sh. EVM, TypeScript, open-source enjoyer.

I'd add a cursor field to log requests / responses so we can have sane pagination, I think. You?

Not obvious also why we went JSON-RPC and not REST but that ship has sailed

If you could make one breaking change to the Ethereum JSON RPC spec what would it be and why?

Researcher at Clave and
Co author of EIP-7212 

don't have one, think address/secret mgmt is already a problem..what's yours?

What’s your favorite solution for cross chain account management on AA?

Working on Terminally Onchain /toc. Join The Terminal here: toc.fun

bet yes on 4844 happening in 2024

blob prices and hard to model 

I'd expect all blobs per block to be almost always full. 

I think it's possible there's high blob prices at times, as blobs will need to cover builder reorg risk due to larger payloads taking more time to propagate and more likely to reorg

Pragmatism (Security) at OP Labs.
Prev: ConsenSys & Coinbase.

Attribs: Friendly, curious, introverted, mid-witted.
Chans: /eth-security /maurelian /stoicism




state becomes larger, means state doesn't fit in memory, this means your iops go up and your performance degrades.

each disk has max iops it can handle, determined by your load

parallelization let's you fire more iops for the same load, so for large state parallelism is even more effective than when it fits in mem

Is state size the real bottleneck to increasing throughput? If so, what parallelization get us?

Eng @ Coinbase, working on Base
https://bayardo.org

all of the above. if a chain committed to a state management policy with rent or expiry it could go vroom vroom on speed. Solana nodes for example all bootstrap from state snapshot, discard tx history older than a few days, and now are building a system for expiring / witnessed revival of state

What's most holding back scaling up transaction throughput on EVM chains?  Slow clients (e.g. geth vs reth)?  Database bottlenecks (e.g. I/O limits of a single commodity NVME)? Lack of parallel execution / tx pipelining?  State growth?  All of the above?

🖖| via sequence. pimlico. poap. polygon. consensys. paxful. flipside crypto. dash. bridgewater associates. darpa. secdef. navy. dia | linktr.ee/alexanderchopan

yes correct, we need to separate key management / rotation from the account, 4337 getting that right

haardik.dev | Founder @learnweb3 | lead instructor /uniswaphooks | building @openfarcaster

I don't have any complaints and never thought the VM was an issue in blockchains....I really like Cairo 1.0 for the verifiability, and I like-not-love Move because of security features for newbs.

What’s the worst thing about EVM in your opinion? What alt-VM (Cairo, Solana, Fuel, etc) feature would you put into the EVM if you could - pick one except parallelization

huge problem don't have a good answer, I'd look at Aztec and Penumbra. 

you could introduce a bloom filter-esque data structure maybe to avoid the linear scan through all data?

UTXO note discovery for privacy protocols have always been a challenge, as the client need to filter through all the published UTXOs to sync its state. What do you think are viable path to solve this challenge, while still maintaining certain degree of trustlessness?

Co-founder & CEO @goldchain a gaming L3 | Space cadet on @skyoneer | MIT | Blockchain dev since 2016 | /lostfounder

Conduit will offer all DAs and largely see it as a competitive market with little look for a differentiation at the moment, but very exciting

Realized I rattled off like 5 Qs 😅 last set I promise: there’s a growing war in DA land but do you think it’s overhyped? With the emergence of hundreds of variants on VM/DA/consensus stacks, wouldn’t that make the DevEx much more confusing than it already is? I guess Conduit could be a solution to that?

What’s your take on shared sequencing? Is this the much desired solution for interop or is the challenge of liquidity fragmentation better solved at the wallet layer?

/ethereum #RunANode | control the memes control the minds paragraph.xyz/@chaskin

like it, think needs to be deployed to a Devnet for people to play with

I'm excited for EIP-4844!

However, I suspect people who are hoping for significant gas price decreases are underestimating induced demand.

https://en.wikipedia.org/wiki/Induced_demand

What are your thoughts on this cast by Dan?

I think we'll settle more rollup txs per block, that's great. 

price per tx should prob go down but it can also be the same as today, or more if there's a lot of demand unlocked and it meets or exceeds the new supply at the old price as Dan says!

What are your thoughts on this cast by Dan?
https://warpcast.com/dwr.eth/0xd368dd63

You can do parevm without consensus changes, so IMO it'll make it's way everywhere in some form. Others may do breaking changes to improve it further.

What’s your take on how/if the EVM could adopt parallel execution or is this something that is better done off L1 and tested on L2s (like we currently see experimentation around)?

Stay Curious // iixii // Lazer Technologies // CoinDesk

need some analytics tool, nansen and the like, don't have a rec but would love one

whats the best way to view or identify an address’s activity across everything across the evm?

Id love to be able to identify what dapps a given address has interacted with and view the address activity more holistically across ETH, L2s and other EVM blockchains

Aspiring mad scientist,
support team @ /rainbow

no way to force a node to forget your transaction without replacing no - the signature cannot be revoked. tx retention time depends by node config, not in consensus :/

Are there any tools to clear a pending transaction from the ethereum mempool without cancelling/replacing the nonce?

Any consensus on when a tx will be dropped?

idk, prob something the rollup operator would choose? for me prob tiny amount?

How much would you pay per month for a service that stores all your relevant blob data for you?