wen sentinels?
(Any views expressed in the below are the personal views of the author and should not form the basis for making investment decisions, nor be construed as a recommendation or advice to engage in investment transactions.)
Before we answer the question “wen sentinels” let’s first explore why Zenon Network needs Sentinels. We cannot ask wen if we don’t know why.
TL;DR
The community is speculating that Sentinels will add to network security in the form of PoW and incentivize users to operate public nodes. We do not know how this will be implementer or when.
Why don’t “the devs” give us an update?
No idea. But it is possible they are still designing the architecture around how sentinels will work. They could be reluctant to release a design only to have it change. In addition, the community is working hard to deliver the remaining SDKs, trustless bridge, and native wallet. Currently, there is no “real” activity on the network and we don’t need Sentinels at the moment. So why give an update?
Our developers are here because they want to be here. Not because some VC is paying them to be here. That means the developers will work at their own pace and give us updates when they want, not when we expect them. In all honestly, I’m accustom to continuous and frequent communication on my teams IRL. But I can understand when the work is done by volunteers, who are we to demand updates or timelines. The fact that the developers even write code should be enough for us to be thankful.
What are Sentinels?
According to a recent Medium posts, “Sentinels are full archival nodes that store, share, and passively validate the dual-ledger state”. And in the last AMA, the developers indicated “Sentinel nodes will need to assemble PoW links and will also have the possibility to run zApps, while Sentries will mainly act as observers and have lighter requirements”.
In case you missed the recent dialogue on Telegram about Sentinels,
georgezgeorgez posted some thoughts:
A while back community member @notafudderguy asked me to evaluate Saito network. Saito correctly identify that there are other network mechanisms that need to be incentivized besides block production
Such as actually broadcasting tx so they get to miners. So they have something called routing work. Where nodes add their own signature to a tx before broadcasting it.
When a miner includes a tx, these routing work signatures are put into a lottery to share part of the fees. They reward staking the same way, lottery. So the nodes are incentivized to share tx as much as possible.
I think our sentinels will be incentivized in a similar way. I think our pow links will be similar in goal to routing work, and I think actually have hash weight, instead of just a signature.
Sentinels have been said to be incentivized public nodes for IBD etc. And also seemingly for tx routing. (In the original whitepaper, sentinels were said to be a representative node which tx were sent to first)
And he goes on to conclude:
So I envision something like Sentinels competing with each other to make PoW-links and broadcast, so that their PoW links will be confirmed and then a per epoch reward calculation based on pow-links confirmed.
How could Sentinels work in practice?
- The user initiates a transaction
- The transaction is broadcasted to the other nodes in the network
- Sentinel nodes prove that they have received this transaction using Pow Links
- The sentinel node relays the transaction to other sentinel nodes
- Other sentinel nodes do the same computation (PoW)
- The process continues until it meets a specific weight threshold
- Once this is achieved, the Sentinel will send the transaction to a consensus node (Pillar) that is chosen at random
- After the consensus node has validated the TX, it will serve as a standard to discriminate against double spend.
How do Sentinels solve the problem of nothing at stake?
First, what is the nothing at stake problem? In a proof of stake network, there is no cost for validators (Pillars) to fork the network and keep both chains running. Bad actors could double spend and with nothing at stake Pillars could mine both chains resulting in a confirmed double spend.
However, Sentinels do have something at stake, and that is PoW. They will only work on the longest chain (just like a Bitcoin miner). Therefore, Sentinels could be the decider of which fork to keep.
How to incentivize Sentinels to stay on the same fork?
On Telegram, George asks us how to incentivize Sentinels.
- Should creating PoW links be assigned to Sentinels, like the Pillar consensus elections?
- Or should it be a free for all based on PoW?
This is all stupid and a waste of my time. wen sentinels?
If we want an update on Sentinels, maybe we should have an opinion on how to incentivize them. Let’s come to a consensus, and present an opinion to Mr. Kaine. Maybe then he will give us his thoughts.