Will

building & breaking software

[Cloudflare @Cloudflare]

We're opening the waitlist for our Monetization Gateway, which will allow you to charge for any web page, dataset, API, or MCP tool behind Cloudflare. The charges will settle in stablecoins over the x402 open protocol. https://t.co/pvICtEIixj

[Will @will_break_it]

@relai_app Unrealized P&L

[David Sacks @DavidSacks]

True. And guess where those Biden AI staffers went to work as soon as the admin was over? Anthropic.

[Will @will_break_it]

@matt_elevenlabs 11B

[Will @will_break_it]

https://t.co/vDQPKxZW1h

[Will @will_break_it]

We’ve integrated a great Leios tracking site on our website for those curious to follow engineering progress!

#leios #ouroboros

Check it out - feedback is welcome!
https://t.co/3xcmCtyaM0

[Will @will_break_it]

To be fair, every highly concurrent system must deal with similar challenges. Account-based ledgers with high TPS often trade off deterministic characteristics for more throughput, leading to subjectively bad UX from paying for failed transactions that never make it on-chain.

At some level of throughput, any system trades latency for throughput, as latency is required to sequentialize - define an order - for concurrent events/transactions. Some systems go further, trading off decentralization (number and geographic distribution of block producers) or security by replacing probabilistic Nakamoto-style consensus with deterministic BFT chains.

[Will @will_break_it]

@yeaookk Yes that’s exactly the idea. Though, any further increase will come at likely more trade offs.

[Will @will_break_it]

Just published updates to the #Leios webpage:
https://t.co/A432PCKHqq

Check it out and share your questions or feedback on our proposal!

[Will @will_break_it]

Tech-curious? Here's the high-level scoop! 😎🚀

[Will @will_break_it]

On a high level, bc going any higher would force us to deal with transaction conflicts due to higher concurrency. And that would mean we will have to introduce more fundamental ledger changes like sharded or collateral designs which would increase adoption/ downstream cost for existing dApps, wallets etc.

The proposal is only a first of a family of designs which can be further extended but any increase in throughput will come at a trade-off of existing properties which users and developers have been familiar with.

[Will @will_break_it]

... after a lot of back and forth we published our proposal for #Leios. Great work team! 🎉

https://t.co/lCS59sS9xl

[Will @will_break_it]

@pumpTokens The proposed version aims for less complexity. There is no definite date yet but Sebastian and his team are working on having that discussion as soon as possible.
It will also depend on CIP reviewers and potentially requested changes to the design.

[Will @will_break_it]

@HeikkiRuhanen It’s an early estimate. I think by the end of this year it will be clearer after CIP reviews what needs to be done. Then Sebastian might have a public discussion around the release.

[Will @will_break_it]

Designing protocols like #Ouroboros #Leios means navigating huge trade-off spaces across many dimensions.

This illustration only scratches the surface of what the team had to explore to reach the best solution.

[Will @will_break_it]

... after a lot of back and forth we published our proposal for #Leios. Great work team! 🎉

https://t.co/lCS59sS9xl

[Will @will_break_it]

@MladenLm It‘s a proposal at this stage.
So in the next weeks we will see if what reviewers have to say.

[Will @will_break_it]

Some #Leios ideas for partial Endorser Block certification via Merkle Trees

https://t.co/kiXDxbLVJL

#cip #edgecases

[Will @will_break_it]

@cryptstitution Cost estimates:
https://t.co/MYfexspXlk

[Will @will_break_it]

@shaggyrax @cryptstitution @intertreeJK We have weekly simulation results and back of an envelope calculations in our repository which show how higher throughput rates dictate different hardware requirements.

Weekly parameterized sims:
https://t.co/to4pQ2Hs4u

Cost estimates:
https://t.co/MYfexsqvaS