Satoshi's paper refers to UTXO pruning, while Pieter Wuille's "ultraprune" has been part of Bitcoin Core for more than three years. Block pruning, on the other hand, is not mentioned in that paper because nodes store historical blocks merely for administrative purposes. The only reason nodes store historical blocks at all is to serve it to other nodes. Therefore, right now, pruning nodes don't serve block data at all as it would require a change to the P2P protocol. A user can enable block pruning by setting prune= on the command line or in bitcoin.conf, where N is the number of MiB to allot for raw block & undo data. This allows the user to choose how much hard disk space they want to dedicate to block storage. Any value of N is intended to be safe, and there is no security compromise with pruning. Very low values could delete undo data that may be necessary in a reorganization, but this is prohibited by argument checks. There is no "auto" setting for N, but the lowest acceptable value is just as safe as storing everything. It is important to note that while release notes are not meant as a replacement or supplement for documentation, the documentation for -prune is available. It is also crucial to understand that the minimum storage quota of 550 MiB is necessary for pruning nodes even if the block data is not served to other nodes. If a client did transaction verification and transaction relaying and only kept the block(s) being verified on disk, it could result in deleting security-relevant transaction data if N is set too low.