L1 metagraphs are doing the initial data validation and can use custom algorithms to reach consensus.
It can consist of a public decentralized node network or it can be kept private and permissioned. They can scale & define a TPS according to their own requirements
3/14
Hybrid L0 metagraphs can be seen as an intermediary between an L1 metagraph and the global L0 hypergraph.
It generates "metagraph snapshots" which can be submitted to the global L0 for inclusion in a global snapshot.
4/14
The metagraph L0 validates on the shared layers - so both on L1 & L0 (hence "hybrid").
They validate the shared subset of operations offered by the L1, as well as confirming their L1 validations are consistent with total state updates and wrap it in a "metagraph snapshot"
5/14
The Global L0 hypergraph accumulates all the metagraph snapshots posted by each metagraph and stores this into a global L0 state by the main L0 chain.
6/14
This all sounds great but let's move this into a more relatable analogy.
Since HGTP is a transfer protocol, it makes sense to compare it with a postal service that distributes packages from one place to another.
7/14
Lets imagine I want to send some L1 validated data from one place to another and think of it as a physical item.
I would wrap up my validated data in some bubble wrap foil (L1 blocks) and I'm sending it to a postal office for further distribution (metagraph L0).
8/14
The postal office (metagraph L0) I'm sending my package to has a reputation (PRO) which impacts the fees I pay to distribute it.
I'm choosing one that has a high reputation and a lot of free bandwidth (staked $DAG), so I'm sure my package is processed as fast as possible.
9/14
The bigger or more complex my package (data) is, the more fees I'll need to pay as well.
I could pass these incurred fees that I need to pay, to the end user I'm sending my package to, using my own metagraph tokenomics model/currency, to recover my cost.
10/14
My package has now arrived in the postal office (metagraph L0), they are validating if it all checks out against both layers (local & global state).
If it all checks out, they take my wrapped L1 package and put it in a box (metagraph snapshot) for further distribution
11/14
The box is now given to a distributer which is delivering it to the final destination.
This could be directly to another metagraph if I want to privately share it or it could be stored in an L0 warehouse, where anyone can retrieve my package (and act as oraclized data)
12/14
The receiver can then reject or accept the box. A receipt gets sent back to the sender that shows whether or not the box was successfully accepted by the receiver.
If it wasn't accepted, then the L1 metagraph will need to take that into account in their process.
13/14
I hope this analogy & explanation makes sense and makes it a bit more tangible.
Let me know in case I missed some things or if I'm wrong about certain aspects of the network architecture. I always like to learn!
14/14
• • •
Missing some Tweet in this thread? You can try to
force a refresh