[Discussion] Deployment Strategy

Preamble

With the crypto market heating up in recent months, new and old projects receive a lot of attention. This being also the case for EVM-compatible chains (L1), Sidechains to Ethereum and various Layer 2s. During these times, relevant information gets diluted with noise and reaching consensus around decisions of deployment on other chains or L2s gets harder.

For our protocol, it is in its best interest to focus on real data and analyse which deployments make sense. The two apparent extreme approaches are undesirable and would produce the worst outcome - this being to deploy everywhere (Shotgun or Sushiswap approach) or deploy nowhere. The optima lies somewhere in between.

This post is here to analyse relevant factors and suggest a framework to engage in a fruitful discussion with everyone.

Deployment Targets

First, we should discuss what are the current possibilities. For that, let’s set up categories and order them from the most relevant target to least:

  1. EVM compatible Layer 2 (e.g. Optimism, Arbitrum)
  2. EVM compatible Layer 1 (e.g. Binance Smartchain, Fantom, Avalanche)
  3. non-EVM compatible Layer 2 (e.g. StarkWare)
  4. non-EVM compatible Layer 1
  5. non-smart contracts platforms (Cardano lol)

Why this order?

Firstly, EVM compatibility should be the most important factor. A specialized Layer 2 or a Layer 1 that is not EVM compatible would add significant development hours.

Secondly, Layer 2s to Ethereum are the natural extension to the Ethereum ecosystem in which mStable is rooted and where mStable users are based in and therefore the prefered choice. Bridges enable natively easy access.

Framework

What should we consider as points when it comes to evaluating deployment opportunities? mStable as a protocol needs some prerequisites. Without it, it can’t add value to the ecosystems. Let’s analyse what are the most important ones.

  1. Are there at minimum 3 similar priced assets available (e.g. Stablecoins pegged to USD) to create a basket?
    – Similar in market cap to avoid mispricing.
  2. Is there an established lending market for these assets (e.g. Aave) to generate some yield for SAVE?
    – Ideally, 2 lending markets that are considered safe (Tier 1 and Audited)
  3. Ideal timing: Is there an established AMM Swap market but no Stableswap?
  4. Is there already an ecosystem to expand around (To avoid 100% of the mAssets being deposited in SAVE)?.

Conclusion

This post offers a framework approach to new deployment opportunities. This is intentionally kept high level to start a discussion and not to get lost in details.

I am very eager to hear what you think and if you think there should be more, less or any other considerations that are relevant.

Next steps: We should come to some consensus around this approach and start to evaluate the market conditions. Hopefully, then we can have a good framework upon which we can rely to evaluate the landscape.

1 Like

Thanks a bunch for kicking this discussion off @Dimsome, and I guess I’ll do the honours to be the first to comment :smile:

So, in no particular order, some blurbs I collected over the last couple of days reflecting on all of this, well knowing that we all share different opinions on this topic:

1.) Regarding deployment targets: I still believe that StarkWare might be a long-term better strategy than let’s say BSC or Fantom, simply because it is native to Ethereum, and inherits it’s security. We should not forget that each L1 requires it’s entire ecosystem to be present and deployed, which in many cases is still in its grassroots compared to some beefy L2 deployments. Agreed though that StarkWare is not an immediate goal down the line, and we should evaluate a correct time on when to consider it moving forward. A privacy-preserving technology like Aztec might well come before this is the case.

2.) We should carefully evaluate the net positive from deploying on any L1 other than Ethereum, simply because the additional workloads put on the different roles (Dev/Marketing/BD/DAO etc…) required might not at all justify the means, whereas on a L2 solution they can more easily be added due to their native fit on to existing workloads.

3.) What stance the core contributors, the DAO, as well as the Metanauts and community take on these deployments will very much reverberate across the Ethereum ecosystem, and should not be overlooked imho.

Some of the proposed solutions clearly sacrifice decentralization for higher throughputs, and we might want to ask ourselves in which way they would be clearly superior to other sidechain solutions like Polygon, on which we’re seeing and enjoying some healthy traffic already.

Some enthusiasts in the space would not take kindly towards us deploying on every single chain available, and with every move we collaboratively decide upon, we send a ripple and a message out into the ecosystem that carries some energy and vibration with it, especially when considering direct competitors to Ethereum like Avalanche, or completely centralized solutions like BSC.

4.) We should also consider here the discussion around cross-chain AMM capacities, similar to what Connext and Hop Protocol are building, and set into motion a discussion if we ever wanted to provide something like this ourselves, or merely click into existing solutions. I’d urge on the side of caution, and would prefer to collaborate with existing protocols, rather than do our own thing, as we’re already at capacity as it is.

One day we will no longer need to even move tokens at all back to a L1 chain, and will transact between L2s freely, and at minimum cost. This makes me believe that settling on a single L1 solution, and then leverage the L2 technology together with the cross-chain composability of other protocols to achieve the desired ripple effect discussed here makes the most sense for now, until further down the line a technology might emerge that easily allows for trustless L1-cross-chain transfers (that we’re ways from this is shown by the recent Thorchain hacks)

I’m all for moving mStable to more chains, but we should start where it matters most, which in my humble opinion is clearly Arbitrum and Optimism. These are the scaling tools that will enable mStable users to seemlessly interact with us, and also bridge to other solutions in the ecosystem easily in the short to medium term.

mAssets will also be able to move in a much better fashion, and we’re not opening ourselves up to the scrutiny of an already biased sentiment around our product that arguably needs more use-cases on the current deployment, rather than the same use-cases across multiple L1s.

Opinions are my own of course, and I know that many will disagree with my stance here, but that’s the beauty of it all as well, isn’t it? :blush:

1 Like

Thanks shubi, I think you make really good points. To your points:

  1. StarkWare is definitely interesting long-term, especially once generalized EVM is possible.

  2. I agree, that is why we need to carefully choose where to deploy and L2s are the obvious choice when it comes to workload.

  3. That is why we need to have this discussion here. We need to evaluate what is the aggregate opinion and what solutions have what existing ecosystem already. mStable is in a very fortunate position that we don’t have to be the first ones to deploy. This makes the choice somewhat easier.

One thing that I strongly agree with is this point:

Optimism is not open yet, but to add some data to the discussion, Artbitrum has seen insane growth. I think at this point, we can’t avoid Arbitrum anymore:

Also, I’ve seen this dune analytics dashboard floating around and it makes sense to add it to this discussion: Dune Analytics

And the hokey stick growth:


(source: Arbitrum – L2BEAT)

Thanks @Dimsome for starting off this conversation. I really believe scaling is one of our key current strategic challenge, as more users get onboarded into Defi.

mStable journey started on Ethereum and expanded to Polygon. Scanning other potential L1 / L2 makes sense now more than ever as they become stronger projects with important traction.
In my opinion, as L2 do not sacrifice decentralisation, they have a major role to play.

Please find below a framework that can be used to assess the different existing solutions, the maturity of their ecosystem and the potential fit we could find with mStable product suite.

L1-L2 assessement framework

1 Like

Thanks @Dimsome for starting this conversation. I would echo the points that Steph made above. In my opinion, focussing on Ethereum L2s is a vote for decentralization and is much more in line with mStable’s values. Arbitrum seems like the obvious first choice.

My technical knowledge is not as good as it could be so forgive me if I am thinking about this in the wrong way, but some big questions for me are:

  • How can we make sure that mUSD (and other mAssets) can be freely bridged between L1 and L2?

  • How do we think about governance on L2s?

As I understand it, there are issues around bridging assets directly since currently mUSD on Ethereum <> mUSD on another chain due to the different basket of assets backing the currency. Being able to seamlessly move mUSD around the ecosystem through protocols like Hop seems crucial to creating utility and demand for mUSD outside SAVE.

Rather than just doing a straight deployment on L2 and creating a copy of mStable on Arbitrum, for example, is there an opportunity to think about it differently and create products on L2 that aggregate deposits and bridge them back to a pool on L1?

Could MTA on L2 be used to vote on proposals on L1, perhaps through delegation?

Hi guys,

Following my post on the discord governance channel please find my point of view here on the forum.

I would personally love the project to stick with Ethereum based solutions even if the hype is on other L1 chains these days (higher APY yes, but for how long?). They will inherit Ethereum security and will be more sustainable (my guess), it would also accrue the value of ETH itself in the end. I haven’t checked at all any of the other chains so I have nothing against them so far except centralised ones that we should avoid (my opinion). Other chains are always good for competition though.

Arbitrum seems to be a good fit at first with several projects on mainnet. Still in early phase but might offer a solid base to start with.
Read also good stuff on Optimism too but as there are less projects available it’s harder to gather feedbacks.

I’m not a technical guy, just very interested by mStable and the next steps for the project to remain up to date in the game, in the long run.

Regarding the governance votes it would be great that the MTA from each chains are recognised. Balancer is facing the same issue at the moment. It could be interesting to exchange with them about how they plan to move forward on this topic (they are now also on Arbitrum distributing BAL tokens on the 3 chains).
Some projects are already doing this across Ethereum, BSC and Polygon.

In a later step we might also discuss about the governance token staking across L1 and L2 (assuming the Ethereum based solution has been adopted).