Before Gloas there's a unique post-state per tuple (block_root, slot). After Gloas there are two possible post-states, one if the slot is Full and one if it's Empty.
I am concerned that Gloas will break the dependant root mechanism. We should proof / disproof this concern before a Gloas release.
Below is a long writeup on the topic, please read it if interested https://hackmd.io/@dapplion/gloas_dependant_root
I also ask about this topic in the #eps channel https://discord.com/channels/595666850260713488/874767108809031740/1451721054849339516
Before Gloas there's a unique post-state per tuple (block_root, slot). After Gloas there are two possible post-states, one if the slot is Full and one if it's Empty.
I am concerned that Gloas will break the dependant root mechanism. We should proof / disproof this concern before a Gloas release.
Below is a long writeup on the topic, please read it if interested https://hackmd.io/@dapplion/gloas_dependant_root
I also ask about this topic in the #eps channel https://discord.com/channels/595666850260713488/874767108809031740/1451721054849339516