Thread by @DvdSchwrtz

A gentle intro to building on @graphprotocol:

First, why build on the Graph?

Because censorship resistant, trusted contracts on #Ethereum is a good starting point but client files (ipfs) and app data also need to be served from decentralized sources to create unstoppable 1/6

client dapps. If you are building on #Ethereum, the Graph is a key part of your tech stack for storing and retrieving data.

There are 3 main files you will work with when building on the Graph:

1. Yaml file - config that tells the Indexers what data to read from #Ethereum 2/6

2. Graphql file - defines the data types for Indexers to store
3. Mapping file(s) - manipulates the raw #Ethereum data into the defined data types and saves these as records for Indexers to serve.

Each yaml/ graphql/ mapping file combo represents a specific subgraph. 3/6

A protocol might have a single subgraph or multiple depending on complexity. At @synthetix_io we have many subgraphs to capture a variety of data across synth rates, exchanges data, staking info, ether loans, liquidations, binary options, etc... 4/6

The yaml config file includes the addresses & abis for the contracts a subgraph should track.

The yaml file also has a path based reference to the graphql file which types the entities Indexers will serve and another path to the 5/6

mapping file where you use assembly script to manipulate data from #Ethereum events, calls & blocks data into the defined entities for indexers to serve.

The Graph is a key part of the web3 stack. If you would like to learn more, check out their docs 6/6 thegraph.com/docs/