Sabitlenmiş Tweet
In this blog post I introduce a useful concept for reasoning about rollups, sharding, light clients and bridges, called State Commitments Arguments.
pep.wtf/posts/sca-and-…
English
big pep
1.2K posts
big pep
@pepyakin
Protocols, databases, execution (wasm, RISC-V). Ex Irreducible, NOMT, Polkadot.
Berlin Katılım Kasım 2012
606 Takip Edilen1.2K Takipçiler
How tf are you adding 78k more LOC to a newsletter website
Garry Tan@garrytan
Absolutely insane week for agentic engineering 37K LOC per day across 5 projects Still speeding up
English
When I die my crypto is gonna go to the 0 address as a donation to everyone
Ante ⚙️@AnteOrg
Last week we talked about the crypto handoff problem. Every approach either gives up too much control, or leaves too much to chance. What would a real solution look like? 🧵
English
> be random guy on the internet
> makes decisions purely on vibes
> misses a few huge opportunities in crypto
> wonders why his life keeps looking random
> stumbles on a weird article about probability theory
> realizes every decision has expected value
> realizes markets are just Bayesian machines
> realizes most "genius trades" are survivorship bias
> realizes most people size bets completely wrong
> realizes he’s been playing the game with no math at all
> opens Polymarket
> starts thinking in probabilities instead of opinions
> suddenly the world starts looking like a giant EV calculator
turns out most life outcomes are just probability problems people never bothered to model:
> career decisions
> investments
> relationships
> risk
all of it is just EV + Bayes + Kelly
> the crazy part?
none of this math is complicated
> you can literally learn the models in this article
> use AI to help you apply them
> and completely upgrade how you think in a few months
but most people will keep making decisions the same way
> vibes
> emotions
> scroll Twitter
> one lucky success story
and wonder why nothing compounds
darkzodchi@zodchiii
English
@geoffreylitt i vibe coded it.
Not even sure if joking since it kinda lost its original meaning becoming “i worked with the agent” or so.
English
We need a shorthand way of saying:
"An AI did the work, but I vouch for the result"
Saying "I did it" feels slightly sketchy, but saying "Claude did it" feels like avoiding responsibility
English
@Sirupsen @jpountz Reminded me of this HN post of the extreme gains you can have from dumb & serial news.ycombinator.com/item?id=174922…
English
on modern machines, increasingly:
dumb & serial > smart & random
read this excellent article from @jpountz on a massive improvement to text search
turbopuffer@turbopuffer
yesterday: short human queries, scalar CPUs today: long LLM queries, wide SIMD lanes for FTS v2, we use a vectorized MAXSCORE algorithm instead of WAND, because dumb & serial beat smart & random algorithms on modern CPUs tpuf.link/maxscore
English
🚀 After millions of payments on x402, we’re excited to introduce x402 V2.
Listening to community feedback, we’re releasing V2 to evolve internet-native payments with greater flexibility and power.
What’s new 🧵
English
I happened to be involved in building Substrate. It's great, but it still constrains you in a box. I realized that it would've been better if the dev builds their chain from components specializing to their use where it matters and using off the shelf components where it makes sense. Then commonware was announced.
x.com/_patrickogrady…
English
"nethermind and go ethereum are both faster [than reth] in normal tests that can be run independently by anyone. nethermind is usually about 50 percent faster. [...] they still run faster since they have superior architectures and better written codebases."
Also: a chain with only one client always lives under the Sword of Damocles of breaking down when this one client has a bug.
Overall corp chains have nothing to do with permissionless L1s.
- not decentralized
- security grade of companies
- low longterm scalability since not using L2 model
- low network effects
- higher friction when interacting with L1s like ETH vs. L2s. Less friction always wins
That said appreciate them when becoming ETH L2s and for their role with onboarding mainstream.
Fede’s intern 🥊@fede_intern
congrats on the @tempo testnet launch. i want to point out one issue in the readme. it says that tempo is “the most performant evm execution client.” this is not correct when we compare it with real benchmark data. i don't get why @paradigm insist with this meme that reth is fast when it isn't. nethermind and go ethereum are both faster in normal tests that can be run independently by anyone. nethermind is usually about 50 percent faster. this is important because both clients use programming languages with garbage collectors, which should make them slower than rust. but today, they still run faster since they have superior architectures and better written codebases. @ethrex_client, our own execution client, is also a bit faster than reth, but the difference is not very big. in any case, the numbers do not support the claim that tempo is the fastest client. after the fusaka update, reth has also been using very high amounts of memory. on our servers, it stays around 80 gb of ram. in the last few months it also had several OOM crashes because of problems with concurrency and how tokio is used. my criticism has nothing to do with the @ethereum versus tempo debate. it's related with keeping objective engineering measurements in a technical subject. it's not ok to keep repeating things that are not true. one last note, commoware’s codebase is very well written and performs strongly. congrats for choosing it.
English
English
English
When you start thinking about "your workload," rather than "all workloads," you'll find non-trivial optimizations all over the place.
Turns out that constructing a database that only works well for DA is both elegant and shockingly performant.
Bravo @eigen_da 👏
EigenDA@eigen_da
Bringing crypto to the cloud era means GB/s throughput. No general-purpose database could get us there. So we built our own: LittDB.
English
@protolambda Do you see any positive changes? what's your long-term stance?
English
@_prestwich @signetsh in reality this is a baitpost and the thing is an actual pepper spray.
English
@SuhailKakar @mert Here is a plausible argument for why you may be in a for a surprise in the next 2-5 years youtu.be/OkVYJx1iLNs
YouTube
English
@mert true, q breaks ecdsa eventually but you still need quantum machines that are way powerful compared to what exists today
and that’s before bitcoin even thinks about a post-quantum upgrade
chamath throwing a 2 to 5 year timeline is straight fear bait
English
quantum computing will never break bitcoin seed words
12-24 word seed phrases have 2^128 to 2^256 entropy - that's more combinations than atoms in the universe
even a quantum computer testing billions of combinations per second would take billions of years to crack one
English
Irreducible's time has come to an end.
Long story short, @radi_cojbasic and I ultimately came to the realization that we couldn't sustainably build the kind of deep tech business in ZK that motivated us.
English
@jimpo_potamus @radi_cojbasic F
It was an incredible experience for me to work with you and @radi_cojbasic , what a ride!
English
@shitpost9000 @alifarhat79 well, it wouldn’t be as bad, if it wasn’t a russian jet
English
“The US dollar isn’t even backed by anything.”
Me:
English