An Adaptive Block Size for Bitcoin
Our Analysis, Code and a BIP
A couple months ago I wrote about an adaptive approach to the block size limit based on the median size of recent blocks. Since that time, we have done some additional analysis and still believe that this is the best approach to the block size limit. It allows the limit to adjust with scalability advancements and changing user demand while still providing protection against denial of service attacks.
We have published a BIP that outlines the details of the proposal. The BIP specifies a number of parameters (i.e. the look back period for the median calculation) that are reasonable choices among many that would probably be satisfactory. We have also published a number of graphs of historic block sizes and the corresponding block size limit that would have been in place if this approach was used. These visualizations were useful in choosing various parameters. Finally, we have written the code needed to implement this median based adaptive block size limit.
While we would love to see this approach eventually employed in Bitcoin, we are not planning to launch a new Bitcoin project around it, nor are we going to advocate for a fork of the Bitcoin block chain. We also do not want this work to derail any current efforts to address the capacity of the Bitcoin network. Our goal is merely to draw attention to one possible long term solution to the issue and bring closure to our work on it.