Thread by @orbisproject

There's a general consensus in the #Blockchain industry about the long-term viability of L2 scaling solutions...

#Bitcoin embraced lightning network

#Ethereum & #Tezos embraced rollups

And #Cardano focuses on Hydra

A thread on #Hydra & how it measures up to a zk-rollup: 🧵👇

If you look at the broader #blockchain landscape

#Cardano is more similar to #Bitcoin

than to any other #blockchain out there

• Ledger models (UTxO & E-UTxO)
• L2 scaling approach (Lightning & Hydra)
• A similar approach to "Inclusive accountability"

Basically,

#Cardano aspires to be a better #Bitcoin, offering smart contract capabilities

Without compromising on the basic principles #Satoshi believed in

So let's take a look at the scaling approach of #Cardano through Hydra

What's Hydra?

Hydra comes under the category of "state channels"

A state channel is a smart contract that enforces predefined rules for off-chain transactions

State channels enable participants to transact quickly and freely off-chain, then settle finality with Mainnet

But to be specific, Hydra implements "isomorphic" state channels

Isomorphic means making use of the exact same transaction format & contract code as the underlying L1

Meaning,

the transactions running on the L2 can be mapped to L1 transactions & vice-versa

Isomorphic state channels extend the L1 network of #Cardano to L2 in an organic way

This avoids the need to create a bridge into an independent layer 2 network

And there is no need to wrap native assets and Plutus smart contracts

There is a fundamental architectural difference between

The state channels on—for example—#Ethereum & Hydra

This comes from ways in which #Cardano & #Ethereum handle state

The localized nature of the state of the E-UTxO ledger model enables isomorphic state channels

But the Account model creates the "Global state"

in this complex distributed & dynamic computing environment

You cannot easily extend the exact underlying ledger to an L2 & execute the transactions

This would require wrapping native assets & solidity smart contracts

Hydra is not a monolithic protocol

It's a family of several subprotocols

Hydra can be divided into 4 components:

• Head protocol
• The tail protocol
• Cross-head-&-tail communication protocol
• A set of supporting protocols for routing, reconfiguration, & virtualization

Most of the components mentioned above are under development after years of R&D

The Hydra Head protocol matured over time as a proof of concept and in March 2022, it was implemented on a testnet

Here's a link to the Hydra roadmap:
github.com/orgs/input-out…

So let's take a look at the concept of ''Head protocol''

A Hydra Head is an isolated network

that creates an off-chain mini-ledger between a restricted set of participants

that works similarly but faster than the on-chain ledger(L1) and evolves in parallel to the main network

On the Hydra head the participants follow a different set of rules compared to the L1

Inside a Hydra head, all the participants need to agree on all transactions flowing through it

that means transactions are only valid when all the participants explicitly approve it

So let's take a look at the way it works:

Multiple parties could come together and open a Hydra Head

And the participants could commit funds to it

Meaning,

Moving the on-chain funds to a script address that locks them under specific rules

All participants have to sign all transactions to avoid foul play

Any participant may decide to quit the head by closing it any time

Then all participants walk away with the latest state they had agreed to off-chain

You could compare Hydra heads as ‘private poker tables’

Now let's look at the differences between Hydra and zk-rollups in the 3 core aspects of these protocols:

• Security
• Usecases
• Scalability
• Market Demand

1) Security

A Hydra head is basically an extension of L1

Which allows interaction between a limited number of participants

This allows Hydra head to be secured by the L1 at a protocol level

But trust is necessary between the participants of the Hydra head

A zk-rollup scales an L1 by providing a similar or different execution environment

That can bundle 100s of transactions into a single transaction & executed outside of L1

The transaction data gets posted to L1

That allows rollups to be secured by native L1 security

Thus, zk-rollups on #Cardano can provide a much better security environment than #Ethereum/EVM

Considering the advantages provided by the E-UTxO ledger model

that avoids the complications of the "Global state"

And secure programming environment based on functional programming

2) Usecases

Hydra is very similar to the Lighting Network of #Bitcoin

It offers way more capabilities than Lightning

The main use cases are:

• Poker game
• NFT-Auction
• Pay-Per-Use API
• Inter-Wallet Payments

For detailed info, click the link👇
hydra.family/head-protocol/…

In contrast to Hydra,

zk-rollups are fully general-purpose scaling solutions

Meaning,

• Anyone can even run a copy of L1 inside a rollup
• Allowing existing dApps & native assets to migrate to rollups
• With almost no need to write any new code

This allows zk-rollups to be as useful as the underlying L1 in this case #Cardano

This allows Orbis to provide all the use-cases that are possible on #Cardano

Including the deployment of application-specific L3 rollups

That could be optimised for specific use-cases

3) Scalability

Many Hydra heads could be randomly created concurrently on an L1

to allow users to execute transactions

The limitation is in the number of participants in a single hydra head

Inter-head interactions are possible, but they are still in R&D

Like Hydra, zk-rollups allow for faster execution of transactions,

When compared to the execution speed of the underlying L1

On top of it,

the scalability & use-cases provided by the L3 & L4 rollups

can make the L1 protocol a major settlement layer for billions of users

4) Market Demand

Solutions similar to Hydra (payment & state channels) are already live on #Ethereum & #Bitcoin for years now

For eg:

#Bitcoin has the Lightning network

#Ethereum has at least 4 of them:

• Perun
• Kchannels
• Raiden Network
• Connext Network

The limitation in user participation and use-cases has resulted 

in the low demand for scaling solutions similar to Hydra

What the market tells us is there is immense demand for "general purpose" scaling solutions like optimistic & zk-rollups

Credits: @l2beat

These rollups are seeing more user adoption day by day

Currently,

The market is showing more demand for optimistic rollups

But as zk-rollups mature, they would be orders of magnitude more scalable and secure than optimistic rollups

TL;DR -

• Hydra is an isomorphic state channel, which includes a family of subprotocols

• The Hydra head protocol is on testnet now

• Hydra supports faster transaction execution but has limited use-cases in comparison to zk-rollups

If you are looking to learn more about zero-knowledge proofs and zk-rollups—

Here’s a similar thread on "Functional Programming" that you might find useful:

https://twitter.com/orbisproject/status/1557402526852550659?s=20&t=1EtjYirBs__10D0CBpCt_g