brk/README.md at e9680afdff05cffa987819a9478eda74b887801f

Mirror/brk

Fork 0

mirror of https://github.com/bitcoinresearchkit/brk.git synced 2026-04-30 01:20:00 -07:00

Files

k e9680afdff readme: update

2024-09-18 19:31:33 +02:00

8.4 KiB

Raw Blame History

A better, FOSS, Bitcoin-only, self-hostable Glassnode.

Description

While mempool.space gives a very detailed view of the network where you can follow the journey of any address, this tool is the exact opposite and very complimentary by giving you a much more global/macro view of the flow and various dynamics of the network via thousands of charts.

To promote even more transparency and trust in the network, this project is committed to making on-chain data accessible and verifiable by all, no matter your intentions or financial situation. That is why, the whole project is completely free, from code to services, including a real-time API with thousands and thousands of routes which can be used at will.

Having anyone be able to easily do a health-check of the network is incredibly important and should be wanted by every single bitcoiner.

Donations

This project was started as an answer to the outrageous pricing from Glassnode (and their third tier starting at $833.33/month !).

But it is a lot of work and has been worked on full-time since November of 2023 and has also been operational since then without any ads.

At the time of writing (2024-09-12), this project has made around 2,200,000 sats, which is around $1300 or $120/month. While I'm very grateful for all donations, it's sadly unsustainable.

So if you find this project useful, please send some sats, it would be really appreciated.

If you're a potential sponsor, feel free to contact me in Nostr !

Geyser Fund

Warning

This project is in a very early stage. Until more people look at the code and check the various computations in it, the datasets might be, in the worst case, completely false.

Instances

Structure

parser: The backbone of the project, it does most of the work by parsing and then computing datasets from the timechain.
website: A web app which displays the generated datasets in various charts and dashboards.
server: A small server which will serve both their resources

Setup

Requirements

1 TB of free space (will use 60-80% of that)
A running instance of bitcoin-core with txindex=1 and rpc credentials
Git

Docker

Coming soon

Manual

First we need to install Rust (https://www.rust-lang.org/tools/install)

curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh

If you already had Rust installed you could update it just in case

rustup update

Optionally, you can also install cargo-watch for the server to automatically restart it on file change, which will be triggered by new code and new datasets from the parser (https://github.com/watchexec/cargo-watch?tab=readme-ov-file#install)

cargo install cargo-watch --locked

Then you need to choose a path where all files related to kibō will live

cd ???

We can now clone the repository

git clone https://github.com/kibo-money/kibo.git

In a new terminal, go to the parser's folder of the repository

cd ???/kibo/parser

Now we can finally start by running the parser, you need to use the ./run.sh script instead of cargo run -r as we need to set various system variables for the program to run smoothly

For the first launch, the parser will need several information such as:

--datadir: which is bitcoin data directory path
--rpcuser: the username of the RPC credentials to talk to the bitcoin server
--rpcpassword: the password of the RPC credentials

Everything will be saved in a config.toml file, which will allow you to simply run ./run.sh next time

Here's an example

./run.sh --datadir=$HOME/Developer/bitcoin --rpcuser=satoshi --rpcpassword=nakamoto

In a new terminal, go to the server's folder of the repository

cd ???/kibo/server

And start it also with the run.sh script instead of cargo run -r

./run.sh

Then the easiest to let others access your server is to use cloudflared which will also cache requests. For more information go to: https://developers.cloudflare.com/cloudflare-one/connections/connect-networks/

Philosophy

Best: Replace Glassnode as the go to
Diverse: Have as many charts/datasets as possible and something for everyone
Free: Is and always will be completely free
Open: With a very permissive license
Trustless: You can verify and see exactly how each dataset is computed
Independent: Only one, easily swappable, dependency (Price API)
Educational: By providing many datasets that can be used to describe how Bitcoin works and why
Timeless: Be relevant and usable 10 years from now by being independent and not do address tagging
Sovereign: Be self-hostable on accessible hardware
Versatile: You can view the data in charts, you can download the data, you can fetch the data via an API
Accessible: Free Website and API with all the datasets for everyone

Brand

Name

kibō means hope in japanese which is what Bitcoin ultimately is for many, hope for a better future.

Logo

The dove (borrowed from svgrepo for now) is known to represent hope in many cultures.

The orange background is a wink to Bitcoin and when in a circle, it also represents the sun, which means that while it's our hope for a better future, we still have to be careful with our collective goals and actions, to not end up like Icarus.

Roadmap

More Datasets/Charts: If a dataset can be computed, it should exist and have its related charts
Simulations: Such as DCA simulation over a period of time with various strategies
Dashboards: For a quick and real-time view of the latest data of all the datasets
Nostr integration: First to save preferences, later to add some social functionnality
Descriptions: Add text to describe all charts and what they mean
Start9 support: By making the whole suite much easier to self-host, it's quite rough right now
API Documentation: Highly needed to explain what's what
Datasets by block timestamp: In addition to having datasets by date and height

Infrastructure

Here's the current infrastructure of the main instance and its backup.

It uses 2 servers, a full and a light one without the parser running but with still datasets syncronized via Syncthing.

Cloudflare is used for their tunnel + CDN services.

Though it's recommended to change to default Browser Cache TTL configuration from 4 Hours to Respect Existing Headers (in Websites / YOUR_DOMAIN / Caching / Configuration / Browser Cache TTL) and activate Always use https.

Iterations

A list of all the previous versions and ideas:

8.4 KiB Raw Blame History