Your keys, your coins, both with and without covenants. Your addresses encode your spending conditions—the spending conditions you opted into (single sig, multisig, etc.) None of the covenant proposals alter this.

What covenants allow is construction of an address that pays out to pre-specified list of one or more other addresses. So someone who owes you coin could wrap it in an address that aggregates coins to you and others. That’s the point of it, in fact, to create density so that more than one beneficiary can share a UTXO.

Whether you accept this as payment is up to you.

On the question of whether an adversary could use covenants maliciously. Current multisig would be a better adversarial tool. That is, adversary sends the coin to a 2-of-3 where your address is just one, so you have to get permission from another signer to spend. You would be within your rights to refuse this as “payment” since you’re not in full control of the coin. (I would refuse)

If a State wanted to enact spending controls, inserting themselves as signers in multisig is superior to trying to negotiate covenant addresses that encode spending criteria. With the multisig, they can approve or withhold approval arbitrarily over time. With a covenant, they’d have to decide up front how the coin could be spent and work with you to create that script. An adversarial despot would much prefer the flexibility and ease-of-use of mandatory multisig.

For this reason, on the point of adversarial encumbrance, I do not see any of the proposals to be worse than the status quo.

Coinbase is historically bad at adopting technology that’s good for them. Look how long it took them to batch transactions. They’re only just now starting to implement Lightning. Even after CTV activates, I wouldn’t expect Coinbase to take advantage for years.

We already have been diluted. FTX is a great example. They took Bitcoin deposits, then sold them rather than holding them on the books. People leaving their coins on FTX, believing in the IOUs, caused reduced price for everyone (dilution).

While you and I agree that Bitcoin is Bitcoin (and IOUs are shitcoins), other people believing in the IOUs does and has caused dilutive effects on the value of our coins.

In and of itself, this is somewhat unavoidable. We can’t stop people from buying, holding and believing in IOUs. However, if we fail to densify transactions, we guarantee that people will do this. IOUs are the alternative to scaling, and they are dilutive.

Built to last

It could be prohibitively expensive to land the TX that releases your funds by yourself, yes.

The recipient is the one who decides the unlocking requirements, not the sender.

When you send Bitcoin to someone, they provide an address. That address is a hash of a script they’ve chosen. You don’t get to choose the script parameters.

What the sender could do is make a UTXO that, say, promises you some coin, and also promises coin to a bunch of other folks. To use your coin, you first need a transaction that pays out to you (and everyone) and then a second transaction to actually spend. So, the fees could be potentially high, since it takes two TXs and there may be e a lot of outputs (many bytes) for the first one.

The burning need is to start building consensus on a UASF. This will take time. It is not prudent to wait until there’s a “burning need” to get the ball rolling. That’s a recipe for a rushed rollout.

Plus, I’m looking to reduce the likelihood of more big custodial collapses like Mt. Gox and FTX. If we wait, the next one could be Coinbase, or some as-yet-unknown custodian that’s even bigger.

Getting the UASF ball rolling now is capacity planning for the “suddenly” rush that’s around the corner.

After listening to Shinobi a while back on a podcast, I came to understand that covenants allow denser use of block space (sharing UTXOs) and that without this, people holding keys on chain would necessarily be a minority of users.

In such a world, as more people want to use Bitcoin, they’ll have to fall back on custodial solutions. I heard Francis of Bull Bitcoin talking on a pod about Liquid being a possible solution. I’ve heard Livera talk about fedimints. I’ve seen people using eCash lately. People buying into ETFs. etc. etc.

We must not give up on peer-to-peer cash. If we cede to banks, we’re guaranteed to get diluted by IOUs and risk centralization (as happened with fiat currency).

We can’t stop people from using custodians. But if we don’t continue to densify, we guarantee that they will.

To use a UTXO in a transaction, you have to meet its demands. Often this requires providing signatures (proof of keys) but it can also depend on other things such as the block height (check-lock-time-verify etc.). You, as recipient, encode these rules into the address you provide to the sender. You opt-in to the rules you want to apply.

“Covenants” allow you to opt in to more rules, such as where the funds can be spent to. It’s still opt-in. You choose to encumber your addresses or not. You can choose not to, just as you can choose not to use multisig.

Even if you choose not to use this feature, you benefit when others do so. Covenants allow denser use of block space, which means (all other things equal) lower fees as more can get done in fewer/smaller transactions.

Be the persuasion you want to see in the world. Don’t others to change.

Agreed. Cross-Input Signature Aggregation (CISA) is another mechanism for densifying transaction data.

It’s already the case that you don’t know how much it’ll cost you to transact in the future. That is, how much value (non-fees) you’ll get out of spending UTXOs you hold. In this way, covenant UTXOs guaranteed to you, and non-covenant UTXOs you have are not all that different.

The covenant case does mean it’ll take two TXs to move funds to someone else. This is why further densification is also desirable, such as Cross-Input Signature Aggregation (CISA) as Guy mentioned.

The reason to push for it now is that it takes time to reach consensus and then even more time to develop UX. There’s a delay of potentially years between when a protocol improvement rolls out and when wallets offer the ability to use it.

Data density -> more capacity -> more participation -> less demand for custodians -> less IOU printing -> less dilution (NGU)

I agree that arguments so far have been unpersuasive. The rationale that persuaded me was combatting custody (and thus dilution via inflation and also centralization risk).

Thankfully, governments live in a perpetual state of anarchy with respect to each other.

The benefits of defecting increase the more that others join the cabal. The States that don’t abide will get all the Bitcoiners (and their intelligence).

The positive framing is NGU. Without densifying, people will use custody solutions which will inflate and thus dilute.

“The same” is a quagmire. Two things that are different are never the same, by definition.

Sharing UTXOs is a way to increase data density. Increasing data density lets us scale participation without sacrificing decentralization.

SegWit and Taproot increased data density. Lighting is a form of UTXO sharing. Covenants will continue the process of densification.

We can’t stop people from using custody solutions that inflate supply. But if we don’t continue to densify, we guarantee that people will do so.

No. I’d pursue other avenues. A few options off the top of my head:

- lie about holdings

- send holdings to friends/family in other jurisdictions

- flee

- burn coins

In the fullness of time, the size of the UTXO set is probably unbounded.

This is about getting more people to be able to hold keys and be able to settle to chain in any given fixed time period. We do this by increasing the density of data encoded in transactions.

SegWit and Taproot represented similar density improvements. Increasing data density is how we preserve decentralization while scaling participation.

When we don’t scale participation, we get inflation by custodians.