How Bitcoin Fixes the Internet
From hard money to hard signals
Bitcoin is hard money because it’s hard to inflate. Like gold, it takes real cost to mine Bitcoin.
The internet, however, still functions on easy signals for attention and credibility, and we’re paying for it. Accounts, reviews, posts, and messages are all easy and cheap to create, which means bots can create hundreds of accounts and thousands of spam messages, comments, likes, and fake reviews to sway collective attention.
The web’s signals have become even easier now that AI can produce intelligible text and agents can code legitimate-looking scam websites at very low cost.
As with easy fiat money, the system has to get really bad before we notice that there’s something fundamentally wrong with the game.
An easy signal can be easily automated and fabricated, carrying little cost and no consequence. It is a poor currency for coordinating human attention, as it can be exploited endlessly, but it’s the currency the web runs on. It knows no other.
Bitcoin, as a scarce source of value, can be used to create a scarce source of signal, which I call hard. A hard signal can’t be easily inflated, as it carries real cost and a real history associated with the pubkey that signed it.
When the internet gains an improved currency for attention, it naturally tends toward order, low-time-preference behavior, and civility.
How Chronicle Uses Bitcoin to Create Hard Signals
Hashcash created a precedent for hard signals with proof of work. The idea is that a proof of computational cost attached to, say, an email can prove that the underlying signal is harder than a normal message that can be easily replicated and sent en masse. The hardness of PoW emails could be used to filter out easy email spam.
The solution wasn’t practical for email, since it would be bad UX to require a user to wait a long time for the PoW to complete before they could send the message1. Since then, the email protocol has centralized because only large providers can address spam2.
Bitcoin provides a more flexible solution for proving cost: an irreversible Bitcoin burn attached to a digital subject such as an email address, message, or domain name. The hardness of the signal is then proportional to the amount of Bitcoin sacrificed.
Another dimension of hardness is the public key that signs the signal: pubkeys with a credible history of activity are harder than new pubkeys. The hardness of the signal is therefore a function of both the cost and the credibility of the pubkey that signed it, because both are hard to produce.
Chronicle is an open protocol that provides a real-time publication layer for these signals and defines the format for events.
The signal, or event, is just a six-field JSON array:
[kind, subject, amount, pubkey, ts, sig]Here’s Bob verifying his domain name with 1 million msats:
[”web:verify”, “bob.com”, 1000000, bob_pub, ts, bob_sig]Or backing his Nostr post with 1 thousand msats:
[”nostr:event”, <event id>, 1000, bob_pub, ts, bob_sig]Theoretically, any digital subject with a unique identifier can be backed in this way, from phone numbers to social media posts or comments to content blobs with a unique hash.
To make the network move at real-time speeds, a Chronicle node batches many signals together, burns the total amount of Bitcoin, and includes the Merkle tree root hash of the events in the OP_RETURN field.
Chronicle nodes store the events off-chain in a publicly accessible record, which can be read by any internet client or service.
How Clients Use the Record to Improve the Product
Web browsers, email clients, social media platforms, and open protocols are all struggling under the pressure of spam: increased moderation, operational costs, and degraded UX.
Once a public record of hard signals exists, clients can start indexing signals by the cost and credibility of the actors who signed them.
Email clients can suppress unknown senders with low credibility. Web browsers can warn against dangerous domains, URLs, and content. Social media platforms and forums can filter out low-credibility posts and comments.
Users can avoid most of the noise by setting the cost and credibility thresholds they are comfortable with. A very popular person who might otherwise receive a thousand spam emails per day could set the threshold higher to receive only the most relevant messages.
The same clients can be used to create Chronicle signals. When you’re emailing a stranger you’d like to connect with, your email client can automatically attach a hard signal signed with your pubkey and help you select a cost that would be effective in getting past the recipient’s email filter. If your pubkey lacks a credible history, you might need to attach more cost to make the signal harder.
Now, Chronicle has an interesting feature. If you abuse the credibility behind your pubkey and spam someone, they can publish a counter-signal against your event. They can specify a negative amount, for example, -1000 msats, and reference your signal’s event ID:
[”chronicle:event”, <your event id>, -1000, victim_pub, ts, victim_sig]The significance of this is that opposition to a signal linked to your pubkey creates a blot in your history, which makes your pubkey a little less credible. The effect is that your future attempts at spamming or intruding into people’s inboxes will have more friction.
This example extends to all online activity where we interact with other people: posting, commenting, upvoting or downvoting. Every action can be linked to a hard signal, signed by a public key that is accountable for the consequences.
This is important. Current platforms and clients allow users to block and report spam email addresses, phone numbers, and accounts, but this does very little to deter spam. Spammers can cheaply create more email addresses and phone numbers, and reporting stays platform-local: reporting a number on Telegram doesn’t transfer to WhatsApp.
But if email addresses, phone numbers, and accounts must be backed by pubkeys with credible histories in order to have any sort of effect on other users, then they can’t be easily and cheaply spun up. And if the record of hard signals exists in the public domain, where all platforms and clients can consult it, then the misuse of a phone number in one client can transfer to all clients and online services that use phone numbers for verification and identity.
Chronicle only provides a neutral record for publishing events. It is the clients that shift the incentives of the game when they start to index by hard signals. It is the individual users who choose the thresholds of cost and credibility they would like to engage with online who create the consequences for those who misbehave.
How This Changes the Game
In the game of easy signals, the cost of spam is nearly zero, and the consequences are nonexistent because the identity can be tossed away and cheaply recreated.
We can see how a system of easy signals rewards bad behavior. Since there are no costs or consequences, the expected loss from bad behavior is essentially zero. For a bad actor, this leaves only the upside of spam, scams, and manipulation.
This is the game we currently inhabit. All the “yuck-y” feelings we get from the web and current social media stem from this immoral structure.
A game of hard signals flips that equation.
Information can remain cheap to publish, but the signals that influence other people should be costly and accountable.
If a person doesn’t have a credible history to stake, then, until they build one, they must sacrifice more cost to get other people’s attention. A reputable person can rely more on the credible history of the pubkey they’re putting at risk each time they use it to engage online. If a person abuses the credibility of their pubkey, other users can oppose their actions, diminishing that pubkey’s future reach. These two factors turn the expected loss from bad behavior from zero to positive: the initial cost and the potential consequences of a diminished pubkey.
But the very fact that opposition and consequences are possible creates a natural restraint. People become accountable for what they do online, and accountability is a precondition for civility.
The greater the number of clients and platforms that start using hard signals to improve their products, the more valuable pubkey reputation becomes, and the more one has to lose by behaving badly. At some point, the profitability equation of bad behavior flips from the expected gain being much greater than the expected loss to the expected gain being much less than the expected loss. The game begins to reward good, useful behavior.
This has major implications for the internet. In a hard-signal game:
Spam and scams become rare
Social media becomes civilized
Search results become useful
Reputation becomes portable between platforms and clients
Escrow between reputable actors becomes optional
Captchas and other slow verification methods can be replaced with immediate pubkey verification
Mass manipulation and propaganda become prohibitively expensive
Free speech, not merely free publishing, becomes possible and accountable
We’ve gotten so used to these externalities that we’ve accepted that this is just the way the web is. But these are only symptoms of the easy-signal game. The easy currency of attention and influence has been abused for years and is now becoming too problematic to ignore.
Bitcoin fixed easy money. It can fix the easy currency of attention and influence by backing signals with real cost and accountable pubkeys.
In this way, Bitcoin can progressively reorganize the internet around goodness, truth, and civility.
Thanks to Curtis Heinen for reading the drafts.
—
Chronicle is under active R&D
Chronicle has a working prototype, a whitepaper, a protocol specification, and a registry of events. Hard signals are a new primitive that can reshape the internet and the way we interact online. We’re developing new UI/UX patterns, reputation algorithms, simulations, mathematical models, and small-scale experiments to understand and refine the system before launch.
If you want to get involved in early beta experiments or work on Chronicle, let me know. You can leave a comment or get in touch through our website.
Ben Laurie and Richard Clayton, “Proof-of-Work Proves Not to Work”, 2004.
Jameson Lopp, “The Death of Decentralized Email”, 2022.