Supporters of BIP-110 view Bitcoin as a public utility whose scarce block space should be reserved primarily for monetary settlement. Inscriptions and other data-heavy applications represent consumption of a limited resource that should be protected for financial transactions, even if doing so requires introducing new consensus rules.
DOG Mode starts from the opposite premise.
Leonidas argued Bitcoin should remain a neutral marketplace for block space, where any valid transaction is equally legitimate provided the sender pays the prevailing fee. From that perspective, there is no objective distinction between a bitcoin payment and an Ordinals inscription.
Rather than seeking permission through a protocol upgrade, the intention for DOG Mode is to remove policy restrictions that its supporters argue Bitcoin itself never required.
The proposal also raises a more subtle question about Bitcoin’s infrastructure.
If enough nodes begin running different policy software, the network’s mempool — the collection of unconfirmed transactions waiting to be mined — could become increasingly fragmented. Consensus would remain intact, but different parts of the network could relay different transactions, affecting fee estimation and how quickly some transactions reach miners.
That fragmentation already exists to a degree, but DOG Mode could widen those differences by encouraging broader acceptance of transactions that many default nodes currently refuse to relay.