It seems like the biggest desire to go with the rich transaction is to avoid mempool complexity of introducing batch transactions.
At this point, there couldn't be a consensus on any transaction type that would be happily introduced in Berlin.
Until the next upgrade or until there is agreement on the best mechanism, Lightclient will continue looking into the batch transaction understanding the complexity of mempool.聽
The next meeting to continue this discussion will be scheduled based on the suggestion in the Breakout room channel.
The next meeting will be announced in the Eth R&D discord #breakout-room.
Most helpful comment
Here's some notes from the call: https://hackmd.io/CccouoTaRUu8gCrFPLB1fg