TL;DR

We detect the median fee size of transactions that qualify for the discount, over the last 2 weeks of time.

Then we multiply that median fee by a little more than 3x to create an economic equilibrium where inscriptions become just as expensive as fragmenting NFTs across thousands of transactions.

Spam unilaterally becomes 3x more expensive than all basic transactions, while preserving enough outputs for Lightning anchors (which Luke has blocked). No OP_Codes are penalized, only bucket sizes—it strives to be timeless… future proof.

“1. Collect all qualifying transactions from blocks [H-2016, H-1]

2. Calculate MEDIAN_FEE = median(qualifying_fee_rates)

3. Required fee for non-qualifying tx ≥ MEDIAN_FEE × 3.14

4. Reject blocks violating this rule”