Ok, to be fair, that’s a better answer than I usually get.
Still not compelling, though. It only models worst-case scenarios materializing while ignoring second-order effects from aggressively countering spam, and the uphill battle that creates for spammers who are determined to build a marketplace on Bitcoin.
You’re partially correct in assuming fees were one of the tools historically used to combat spam — particularly DDoS-style flooding and steganographic data stuffing.
But fees alone are not equipped to fight the new form of gamified spam that emerged in early 2023. That situation could have been prevented if decisive action had been taken early on the Core side, before an entire economic ecosystem formed around this behavior. The simplest mitigation was fixing the witness discount loophole that allowed spammers to bypass existing filtration policies. Instead, it was reframed as a feature rather than a bug, which effectively incentivized spammers to organize crowdfunding campaigns, build specialized tooling, and even open direct submission channels to miners.
Once that marketplace for gamified spam took root, fee pressure stopped being an effective deterrent. Spammers weren’t paying fees as a cost — they were treating them as speculative inventory, expecting to recover them by selling their “art” to the next buyer. At that point, the lenient posture from Core — which effectively held a monopoly over standard spam filtration policies — signaled that this new bazaar would not face meaningful resistance.
The results are visible three years later. The UTXO set expanded from roughly 4 GB to 12 GB. Spam occupies approximately 36% of blockspace, and around 76 GB of additional blockchain data now exists that likely wouldn’t otherwise. Meanwhile, for the first fourteen years of Bitcoin, mempool policy and fee pressure were sufficient to keep non-financial data at near-zero levels. That wasn’t accidental. Developers and node operators largely shared the same expectations regarding arbitrary data storage, and spammers respected those norms because they knew they would face consistent resistance otherwise (Vitalik’s early experiments are a good historical example).
Today is a different environment. The core issue is that protocol maintainers shifted from treating spam as adversarial behavior to treating it as inevitable (Gregory Maxwell’s posts from 2014-2015 illustrate the earlier philosophy clearly).
Your argument against BIP110 — particularly scenarios one and two — assumes spammers are still deciding whether Bitcoin welcomes them. That decision has already been made. The derogatory label “Knotzis,” often thrown around to dismiss policy-focused Bitcoiners, originated within those spam-aligned communities. BIP110 is one of the first proposals in years that has visibly unsettled them, as evidenced by recent commentary from the primary Ordinals wallet provider. For the first time in a while, their business assumptions are facing uncertainty.
Regarding the claim that spam revenue disadvantages small miners and accelerates pool consolidation — current data does not support this. Spam transactions have contributed less than 0.76% of total miner revenue over the last three years. Mining centralization is far more strongly tied to structural incentives created by the FPPS payout model. If the mining ecosystem genuinely wanted to address centralization, transitioning toward DATUM-style template construction could significantly reduce pool dominance, while simultaneously making spam dynamics even less economically relevant.
Now consider OP_RETURN. The justification for expanding its size limit by roughly 1200x was that it would channel spammers toward less harmful data-embedding methods. If that theory were correct, inscription-based spam should have declined. Instead, both vectors are now heavily utilized, transforming a single problem into two parallel ones. There has been no meaningful accountability, no serious exploration of alternative mitigations, and no acknowledgment that the policy experiment failed.
You argue that fighting spam at the consensus level becomes an endless cat-and-mouse game. Ironically, the same argument was previously used by Core developers against policy-level filtering: that filters were insufficient and only consensus changes could address spam. The goalposts appear to move depending on which layer is being discussed.
Spam mitigation is inherently adversarial, but the asymmetry still favors defenders. The objective is not to eliminate spam perfectly — it is to disrupt the economic viability of spam markets. Once profitability is destabilized, fee pressure and policy friction naturally reinforce the defense, exactly as they did for over a decade prior to 2023.
If BIP110 fails to produce a measurable reduction in spam incentives, alternative proposals like CAT directly target spam marketplaces themselves. In combination, they would likely be devastating to the current spam economy. Realistically, activating both simultaneously may not even be necessary. Once complacency disappears and profitability declines, speculative “Bitcoin-native” token experiments will follow their historical pattern and migrate elsewhere.
I think the problem is black/white thinking on both sides.
There's a difference between spam and legitimate non-monetary use cases.
Ideally, a reasonable balance could be found by earest and pragmatic developers and Bitcoin service providers.
Uncapped OP_RETURN with no filters is too extreme and an obvious hazard.
I support L2s, timestamps, etc. but they don't need such extremist and dangerous default settings.
