Another day, another block ordering proposal. If you're into these things, check out my alternative proposal on the Ordering FIP. I think I'm aware of all the benefits that come with it, you can help by pointing out flaws and inefficiencies :-)

https://github.com/farcasterxyz/protocol/discussions/193#discussioncomment-10722075

Dad, geek, conversation liquidity provider. vrypan.net (home), /weatherxm (work), @l--o--l (fun). Also https://paragraph.xyz/@purplesubmarine

Note to self: I see that the reordering mechanism needs work. It's currently a patchwork of concepts, sometimes overlapping, and needs some refactoring. What gets invalidated and what not, which messages are considered new and how. Will do it later today.

I think so. However
- Flooding the network with messages has similar effects in any model. For example, snaps will have a size limit, which means the attacker could create a situation that messages are competing for snap space.
- It seems that what you describe has a measurable cost, i.e. security budget. It's easy to increase this cost (for example staking?), in which case that question is how much is it worth to attack Farcaster, and for how long?

I have 100,000 accounts under my control. I publish new messages privately to my block to make it win. I also control 101 miners. I now rotate programmatically between the miners with my bot accounts (so to maximize my block score). Now legitimate casts are stuck waiting?