Vitalik Buterin: "Reducing L1 slot time" is worth discussing

Golden Finance reports that Bankless founder David Hoffman posted on X stating that, in a sense, the rollup-centric roadmap has been "completed", with current progress including the launch of 4844, DA, and cheap L2. The primary focus for future optimization is L1.

Vitalik Buterin responded to this by saying: "I don't think it's a matter of priority. The next step for DA is peerdas, and those working on this are almost completely detached from those improving SSF or EVM execution. I do believe that declaring 'the rollup-centric roadmap has been completed' is premature victory. We're only at 1.3 times peak usage away from full blob data (=21600 blobs/day). We need to initiate peerdas and start secure scaling while improving L1."

Hoffman continued: "If we increase blocktimes for L1, wouldn't we also increase the supply of blobspace? Focusing on L1 seems synergistic with focusing on both L1 and L2 simultaneously, whereas focusing on DA only benefits L2."

Vitalik explained: "If we reduce slot time without changing other variables then yes. But both limits are determined by bandwidth constraints so if we're satisfied with today's gas/blob limit then we'd be directly satisfied with the gas/blob limit too. So I usually consider bandwidth and latency as two separate issues. I do think reducing slot time for L1 is worth discussing though it needs to be done cautiously to avoid 'killing' independent stakers or even geographically dispersed stakers."