The first real soft-fork fight since Taproot

For six years, BIP-119 has been the upgrade everyone argued about and no one could activate. That changed this quarter.

There's now an activation client, with real deployment parameters and a real start date - the first time CTV has had actual code with a calendar attached to it since Jeremy Rubin drafted it back in 2019.

So it's worth understanding what's actually on the table, because this is the closest thing Bitcoin has had to a genuine consensus fight since Taproot activated in 2021. And unlike Taproot, this one is not a foregone conclusion.

What a covenant actually is

Strip away the jargon and a covenant is simple: it's a rule about where a coin can go next, beyond just "whoever holds the key can spend it."

Today, if you control the key, you control the coin - full stop. Your bitcoin can be sent anywhere, by anyone who has access to that key. That's the source of Bitcoin's freedom, and also its sharpest edge: if someone gets your key, there is nothing stopping them emptying your wallet to an address of their choosing. The coins go wherever the thief points them.

A covenant changes that. It lets you attach a condition to your own coins in advance: this UTXO can only ever be spent to these specific destinations. CTV - OP_CHECKTEMPLATEVERIFY - is the minimalist version. It doesn't try to do everything. It just lets you pre-commit a transaction to a fixed template, and checks that whatever spends it matches.

The clearest example is a vault. You set up your wallet so that even if your key is stolen, the coins can only move to a backup address you chose months ago - and only after a delay long enough for you to notice and react. The thief holds the key, but the coins still won't go where they want them to. For anyone who has ever lain awake wondering what happens if their seed phrase leaks, that's a genuine change to the threat you're carrying.

Other use cases: congestion control (pre-sign a batch of payments now, broadcast them later when fees drop), payment pools, and more efficient Lightning and Ark channels. Useful stuff - but it's worth being clear-eyed about who gets the most out of it.

The numbers

Why this one's a fight (and Taproot wasn't)

Taproot was easy to love. Schnorr signatures, better multisig privacy, MAST - it made Bitcoin better for nearly everyone and it had near-universal support. It sailed.

CTV is different, on two fronts.

First, the philosophy. Covenants are genuinely contentious. One camp sees CTV as the smallest, safest possible covenant - it commits to an output set and nothing more. The other camp sees any covenant as the thin end of a wedge: open the door to restricting where coins can go, and you eventually get recursive covenants, and from there, theoretically, coins that carry permanent conditions - the raw material for censorship. Reasonable, technically literate people sit on both sides of that line. This isn't noise versus signal; it's a real disagreement about what Bitcoin should be.

Second, the constituency. Taproot's benefits were universal. CTV's benefits mostly accrue to infrastructure - vaults, congestion control, payment pools. Here's what that means: the people who feel the upgrade first aren't everyday holders buying and self-custodying their stack. They're the wallet developers, the businesses and the builders who construct tools on top of Bitcoin. A vault is something a wallet has to build for you; congestion control is something an exchange uses to batch its payments. The benefits are real, but they reach most users second-hand, through the software they use - not directly in the way Taproot's privacy gains did. That's the real activation problem: not the mechanics, which are a proven Speedy-Trial-style design, but whether something that mostly benefits builders can reach 90% miner support and broad user buy-in.

It's worth remembering how these fights have gone before. SegWit needed the threat of a user-activated soft fork to break a miner standoff in 2017. Taproot went smoothly because it was uncontroversial. CTV sits somewhere we haven't really tested.

Why we're watching this at Stacked

Interestingly, the first beneficiaries of CTV aren't retail holders - they're the people building self-custody infrastructure. Wallets. Exchanges. The layer Stacked operates in.

Vaults are the clearest example. A non-custodial setup where your coins can only move to your own pre-set destination, on a timelock, is exactly the kind of building block that makes self-custody safer for non-technical people - which is the entire problem we work on. If CTV activates, the toolkit for building genuinely hard-to-steal, genuinely sovereign wallets gets meaningfully better.

So we're paying attention. We're not going to tell you covenants are an unalloyed good or a looming threat - the people who say either with total confidence are usually selling something. But the direction is clear: the building blocks for better self-custody are getting closer, and that's worth tracking whether or not this specific activation succeeds.

There's a couple of ways covenants can happen

CTV isn't the only path. OP_CAT now has an official BIP number. LNHANCE is being discussed as a Lightning-focused route to similar capabilities. And BitVM2 - which needs no soft fork at all - is already live in production, powering Citrea's mainnet bridge.

The takeaway: even if CTV stalls, the broader move toward more expressive, safer Bitcoin contracts is happening on multiple fronts. The question isn't really if anymore. It's which, and how carefully.

How Bitcoin actually decides

It's worth stepping back to notice how strange this process is. No one is in charge of this decision. There's no CEO of Bitcoin to sign off on the upgrade, no board to push it through by Q3. A meaningful change can take six years to get to the starting line - and might still not cross it.

That's not an accident, and it's not dysfunction. It's the same machine that runs Bitcoin every ten minutes, just playing out on a longer timescale. Every block, miners and nodes are independently deciding which set of rules to enforce - no one tells them, they simply run the software they choose and reject anything that breaks the rules they agreed to. Consensus isn't announced from the top; it emerges from thousands of parties separately arriving at the same answer. A soft fork is that exact process, stretched out: miners signal, nodes choose whether to upgrade, and the change only sticks if enough of them independently say yes. The 90% signalling threshold is just consensus made visible.

That's why it's slow, and why slow is the point. A network where someone could simply decree a new rule is a network where someone could decree a rule you'd hate - and change the money out from under you. The friction that makes covenants take six years is the same friction that means no one can quietly alter your bitcoin. You don't get one without the other.

Stack accordingly,

Simon

Stacked - Bitcoin-only, non-custodial. New Zealand.

stackedbitcoin.com