The proposal aims to defend against stronger attacker models, where attackers actively try to learn which nodes were involved in the stem phase. To address this issue, the Dandelion team has developed a new variant of the Dandelion++ mechanism called Per-Incoming-Edge routing. This variant causes each transaction from the same source to traverse the same pseudorandom path, preventing intersection attacks that occur when a client broadcasts multiple transactions over a short period of time.The Dandelion team has validated the new variant with new analysis and simulation. Additionally, the team plans to re-engineer their reference implementation to use per-incoming edge routing, simplify orphan transactions, and adopt an approach of avoiding the mempool data structure for stem transactions. However, Greg Maxwell has raised concerns about the description of the Mempool Embargo approach used to defend against attackers and suggests that it should be simplified and clarified. He also proposes an alternative construction that allows non-Dandelion capable nodes to take on the role of the last stem hop, improving the anonymity set during the transition phase.Furthermore, the Dandelion proposal includes an experiment that compares the rate at which an attacker can learn the main P2P graph under diffusion compared to using Dandelion. While diffusion leads to a higher learning rate, the use of Dandelion reduces the adversary's learning rate. Lastly, Maxwell asks about the possibility of Dandelion propagation measuring properties of the inter-node connection graph.In summary, the Dandelion project is proposing a modification to Bitcoin's transaction propagation mechanism that aims to enhance privacy. The team has developed a new variant of the Dandelion++ mechanism called Per-Incoming-Edge routing to prevent intersection attacks. However, there are still concerns about the Mempool Embargo approach, and Maxwell proposes an alternative construction. The proposal also includes an experiment to compare the learning rate of attackers under diffusion and Dandelion.