It’s been over 12 years since first development, and 6 years since anyone has dusted off the codebase for this legacy Bitcoin project. But its finally back, thanks to a new Bitcoin hub on the Western Coast of the USA called “localhost research.”
Dubbed the “Fast Internet Bitcoin Relay Engine (FIBRE),” it helps address Bitcoin’s thorny mining centralization tendencies on the protocol level, by offering a method for miners to relay blocks as quickly as possible across the Bitcoin network. The project was first developed in 2014 by OG Bitcoin developer Matt Corallo, before being abandoned in 2020 due to maintenance constraints.
More technically speaking, FIBRE was designed to improve how bitcoin blocks are produced and distributed, plus reduce some toxic behavior incentives for bitcoin miners.
Now it’s back, operated by localhost, and has some modest upgrades.
So what exactly does FIBRE solve?
First, some Bitcoin 101.
In “vanilla” Bitcoin, when a mining node finds a block, it sends that block to the nodes it’s connected to which then sends the block to the nodes they’re connected to, and so-on. Another mining node sees that block and starts mining on top of it.
But that’s often slow. And with blocks worth hundreds of thousands of dollars, there’s a solid recipe for shenanigans: information dislocations, arbitrage, and possibly toxic behavior.
One of these toxic behaviors is “selfish mining:” where a miner uses that time delta to mine on their own blocks. When you have large mining pools, this can create an unfair advantage and more centralization pressure. Selfish mining can lead to a list of issues, including censorship of transactions.
Bitcoin mining is designed to be a fair lottery: If you have X% of the hashrate you should find X% of the blocks. But due to Bitcoins relay policy, mining can start to look more like a race than a lottery. Not only do you have to find the block, you have to tell everyone about it first!
Creating a highly performant network–possibly with FIBRE–collapses many of the disproportionate advantages that currently exist, and likely improves network security.
What’s Different This Time
The last time FIBRE was operating, the world looked a lot different: COVID shutdown travel, mask mandates proliferated, Bitcoin price tumbled before a ascendent rise to all-time-highs. So what’s different with FIBRE this time around? Honestly, not much!
“The delta between OG FIBRE and current FIBRE is not huge,” a localhost representative told Blockspace. “We are working on just bringing it back and then observing where we can improve upon it.”
The changes to FIBRE are largely operational. Where Corallo’s original network was a one-man operation with a captcha-gated onboarding process and minimal monitoring, Localhost is building FIBRE to be run by others. They’ve open sourced a monitoring suite. There’s now a dashboard that offers a real-time view of relay health, tracking block reconstruction times and propagation performance across the network.
Resurrected FIBRE also improves compact block relay. Compact blocks, or basically highly efficient blocks, didn’t work over the previous network as they ran on TCP. FIBRE replaces TCP with UDP. This may sacrifice some reliability in delivering packets, but makes up for it with raw speed. The new FIBRE also introduces forward error correction (FEC) to account for possible issues with UDP.
A decade ago “it wasn’t easy for people to run their own FIBRE networks” Localhost says. The team’s goal for the resurrection of the project is to lower the barrier for anyone to operate their own FIBRE network.
At launch Localhost-run FIBRE network has 6 highly performant nodes in Frankfurt, New York City, Tokyo, Beijing, Seattle, and London.
Network decentralization
FIBRE seems to have a similar goal as the BitAxe and Open Source Mining movement: make mining more secure and less centralized.
Yet, FIBRE and BitAxes don’t single-handedly fix points in mining centralization. Fundamentally capital access is downstream of protocol design–such as block variance–to which there are legitimate constraints.
But like the Bitaxe movement, it doesn’t need to see big numbers to matter. Localhost’s goal is to channel FIBRE’s users towards protocols like Stratum V2 or P2Pool, where client-side template construction makes decentralization operationally meaningful.
Both FIBRE and Bitaxe provide narrative tailwinds for decentralization while providing tools participants can actually use. Neither is a silver bullet, but both lower barriers to entry.
“I don’t think it’s [going to] make a deep impact in centralization pressures from a number standpoint, but I think philosophically it has a big capture and that’s really important,” Localhost says.
Enjoyed this read? Join us at OPNEXT, the Bitcoin scaling conference in NY, NY April 16th. Localhost will be joining us for a presentation.
