Thread by @jJosjuaThreatt

Many of the significant crypto hacks have been characterized by bridge exploits. They often contain lots of money & have few trusted key holders. Hacks such as the wormhole exploit encouraged the creation of a #Ergo centric bridge put forth by @mhs_sam.
🧵👇🏾

Some of the advantages of the Rosen bridge are
⁃Ergo centric
⁃Other chains only need multi-sig to support Rosen
⁃No smart contract logic in other chains
⁃2 layer architecture
⁃Audits are only needed on ergo
⁃Easy configuration for other chains

Let’s take a high level overview of how the bridge works.

The first step is a user either sending funds to the multi-sig on ergo side or chain-x said depending on where they are bridging to.

Watchers, who periodically look for transactions in the multi-sig wallet find said transaction.

Once watchers have witnessed some arbitrary number of block confirmations they create an event box on the ergo side.

If a particular event(transaction) has an sufficient number of event boxes created for it by different watchers a watcher can then create a approved event box. Any watcher can create a approved event box

Guards then look for approved event boxes & independently verify the event. Guards also wait for multiple block confirmations to verify events

When a significant portion(m of n, 2 of 3, etc) of the guards have independently verified the event the event is broadcasted to the recipient chain.

The users tokens are then minted or released on the target chain & the transaction is confirmed

Guards then watch and ensure the transaction is mined on the #Ergo side & that enough block confirmations have occurred to be confident in the transactions finality.

- Let’s talk about watchers
Anyone can be a watcher, watchers must stake the “Rosen token” to get the same number “Rosen event tokens”. Every box needs one Rosen event token. You will lose your token if the event is found to be fraudulent. It will be sent back to you if not.

To avoid this problem of watchers watching other watchers & not looking at the event independently a honeypot(trap) can be created to punish(slash) watchers that report on fraudulent events. Other potential methods are also available, such as commit reveal schemes. twitter.com/i/web/status/1…

-let’s talk about Guards

- The guard set is predefined & consists of a few well known entities. Who these guards is very important.
- Guards independently verify events
- M of N guards are needed to submit transaction

Guards create final transaction, they then share fees with the participating watchers & distribute Rosen tokens to those watchers. Bridges with more volume will naturally have more watchers because watchers like profits. This also leads to a efficient natural distribution

Scaleable

- the 2 layered architecture allows for scaling, guards don’t need to increase with the number of chains. One just needs to add watchers to the new chain of higher volume chain when needed.

The saves computation resources for the guards as they don’t need to keep track of every transaction on every chain. Just the approved event boxes.

Big thanks to @ERG_Armeanio 😎I used slides from a video he did

Or higher volume chain*
These typos are killing me🧐😓