This is a segment from The Breakdown newsletter. To read more editions, subscribe
“Nobody goes to the store to buy a Swiss Army knife. It’s something you get for Christmas.”
— Jensen Huang
Great businesses start as scalpels, not Swiss Army knives.
By choosing one thing to do, companies are more likely to get great at that one thing — and customers are more likely to know that’s the thing they do.
Here, for example, is Yahoo’s homepage, circa 1999:
There’s a lot going on there — search, auctions, news, email, instant messaging — and Yahoo wasn’t particularly good at any of it.
Here’s Google’s homepage, circa 1999:
It’s easy to see why that proved to be the winner: the single-purpose website made it clear to users what Google does, and that helped Google get very good at doing it.
The fact that lowercase “google” has since become a verb and Yahoo doesn’t do anything more important than host my fantasy baseball league is evidence that being great at one thing generally beats being average at many things.
Will that apply to blockchains, too?
Bitcoin is a single-purpose chain — all it does is send bitcoin — and its simplicity might be the main reason it’s been such a success.
But Ethereum and Solana are also general-purpose chains and they, too, have had some success.
And neither approach seems to be encroaching on the other: Bitcoin has so far failed at DeFi and Ethereum has so far failed to become money.
So maybe both approaches can peacefully co-exist?
It might be too early to answer, because general-purpose chains will soon have a new, single-minded competitor.
Last week, Stripe and Paradigm formally announced the development of a stablecoin-focused blockchain, Tempo, that feels like an instant favorite to win the emerging crypto payments business.
Tempo will be purpose-built for stablecoins, offering predictable fees (paid not in a native token, but stables), near-instant finality, “opt-in” privacy and compliance functionality, “lanes” exclusively for payments, and high throughput — all things that general-purpose chains struggle with.
Matt Huang, who leads the development of Tempo, suggests that being more narrowly focused will allow the chain to develop faster: “We feel urgency to build for the demand that’s coming and want fewer dependencies, including on the rate of Ethereum L1 progress.”
Calling out Ethereum like that might suggest that Tempo will have ambitions beyond payments.
Perhaps tellingly, Huang says that, while Tempo will start with a permissioned validator set, the chain will be “permissionless” from day 1 and will “decentralize further from there.”
A fully decentralized and permissionless blockchain that excels at payments sounds a lot like what you’d want a general-purpose blockchain to be.
Could Tempo pose a threat to Ethereum and Solana not just in payments, but in everything?
The Google example suggests it’s possible: they beat Yahoo at search, then expanded horizontally into email, cloud computing, smartphones, self-driving cars — becoming even more of a Swiss Army knife than Yahoo was in 1999.
There are several such examples: Microsoft was just BASIC, Amazon was just books, Apple was just PCs, Southwest Airlines was just Texas.
But there are counter-examples, too.
To name just one, single-purpose calculators used to be better for doing quick math than general-purpose computers, but who owns one now?
You’re more likely to have a Swiss Army knife in a drawer somewhere than a Texas Instruments calculator.
So, if general-purpose computers could make calculators unnecessary, maybe general-purpose blockchains can eventually make payments blockchains unnecessary, too.
Max Resnick is confident they will: “Decentralized blockchains,” he predicts, “will be superior to centralized ones in every way” — including speed, scale, reliability, and even regulatory compliance.
If so, Tempo may have fired the starting gun in a race to make general-purpose chains super fast before a super fast, single-purpose chain can be decentralized.
This might be a race with no finish line: it might be impossible to be both decentralized and optimized for payments.
Mert Mumtaz, for example, thinks Tempo won’t even be a blockchain, let alone a general-purpose one: “How can you possibly have something that’s a payments-only chain?”
For Mumtaz, blockchains are decentralized by definition — and a decentralized blockchain is by definition general-purpose.
Tempo says they will progress toward decentralization.
But Mumtaz thinks that will inevitably flood the chain with frivolous things like Fartcoin, clogging things up and degrading the chain’s performance in payments.
There are only two ways to be a payments chain, he says.
One is to make the chain “not Turing complete,” like Bitcoin, “where you can’t do anything other than send money” — and the other is to “permission the chain.”
If so, Ethereum and Solana can rest easy that they’re not going to become Yahoo to Tempo’s Google.
But what if Tempo gets all the users without decentralizing?
