Fountain Codes: a way to reduce blockchain storage costs

The discussion around Fountain Codes in Bitcoin, as referenced during the Scaling Bitcoin conference in 2019 and further conversations within the 2140 office and at the Floresta/utreexo retreat, brings forth several critical considerations regarding the implementation of Fountain Codes in blockchain technology. The primary concern articulated is that blockchain storage and serving are not pressing issues at this moment due to nodes' ability to prune when space is restricted. This suggests that the urgency for such a solution may be overstated.

Moreover, the use of Fountain Codes is identified as potentially exacerbating the current bandwidth bottlenecks faced by the blockchain network. The critique points out that any proposal increasing the initial block download (IBD) time is counterproductive. Specifically, Fountain Codes would require downloading approximately 10% extra data for block recovery, which slows down the IBD process. Additionally, the need to source blocks from about 110 different peers due to each peer retaining only about 1% of necessary data complicates the process further. This not only increases memory requirements but also leads to dependency on multiple peers, which could result in slower data transmission rates if many peers are slow.

Another significant limitation highlighted is the loss of flexibility in how blocks are downloaded. With Fountain Codes, blocks must be downloaded in sets or "epochs" rather than individually. This restriction impacts the ability to request single blocks from the network, crucial for certain types of clients like those using BIP157/158 protocols, and hinders IBD optimizations that involve sequentially validating the chain from multiple points in parallel.

Lastly, concerns about peer fingerprinting through their specific choices of droplets in Fountain Code implementations pose additional privacy and security issues. This factor needs careful consideration to prevent unintended identification of peers based on the data they transmit.

In summary, while the academic exploration of Fountain Codes in the context of Bitcoin provides an interesting avenue for discussion, the practical implications and current technological constraints suggest that it may introduce more complexities and inefficiencies than benefits at this stage. Further advancements in network capabilities and a more pressing need for such a solution might change this assessment in the future.