Is There a Trade Off Between Stake Pools and Decentralization?

Vasiliy Shapovalov (Tech Lead, Lido), F.P. (Co-Founder, Socean) and Ella Kuzmenko (Product Manager, Stake Pools & Delegation Program, Solana Foundation) chat with Anatoly about the complexity and game theory surrounding stake pools, decentralization and censorship resistance.

00:10 - Intro

01:38 - Collaterals, maximizing censorship resistance

07:40 - APYs and investors

09:31 - How to get penetration across DeFi

14:58 - Governance in a liquid stake pool

18:23 - Automation vs. programmatic on-chain governance

20:44 - Factors in selecting validators

29:27 - Growing the validators set

32:21 - Stake pool token in DeFi

35:09 - Liquidity fragmented between too many pools

41:01 - Who controls the network?

44:46 - Increasing decentralization

00:39 - Intro

01:51 - pencilflip’s background
03:30 - Working at facebook vs. web 3.0

07:31 - How pencilflip got into crypto

08:52 - Views on NFTs

10:45 - Getting into Solana

15:29 - Experience working in lower level

17:56 - What was his method to learn Solana?

21:01 - What’s the hardest concept on Solana?

23:53 - How fast did he move from Rust to Anchor?

27:35 - Building on Solana

33:24 - Advice to people moving to Web 3.0

DISCLAIMER

The information on this podcast is provided for educational, informational, and entertainment purposes only, without any express or implied warranty of any kind, including warranties of accuracy, completeness, or fitness for any particular purpose.

The information contained in or provided from or through this podcast is not intended to be and does not constitute financial advice, investment advice, trading advice, or any other advice.

The information on this podcast is general in nature and is not specific to you, the user or anyone else. You should not make any decision, financial, investment, trading or otherwise, based on any of the information presented on this podcast without undertaking independent due diligence and consultation with a professional broker or financial advisor.

Anatoly (00:10):

Hey folks. This is Anatoly and you're listening to The Solana Podcast. I have a super exciting episode today, it's all about Stake Pools and decentralization and censorship resistance. And I have a group of guests that I'm going to let them introduce themselves, just to make it a little easier. So Ella, why don't you start first?

Ella (00:29):

Sure. Hey guys. Ella, I'm a Product Manager of Censorship Resistance at The Solana Foundation working on Stake Pools and the delegation program.

Vasiliy (00:39):

Hey. I'm Vasiliy, I'm tech lead at Lido. Honestly, I think that the person who should be here instead of me is someone, of course like Felix or [Uto 00:00:49] or maybe Brian, but they couldn't make it, so I'm here instead as a second best option.

Anatoly (00:58):

Awesome to have you. We'll take the second best.

FP (01:03):

Hi, guys. I'm FP. I'm co-founder and CEO of The Socean Stake Pool. Nice to meet you guys all today.

Anatoly (01:11):

Awesome. So censorship resistance Stake Pools, I've been pounding the table on this for two years as the most important thing and proof of stake networks, because I have this crazy belief that if we have liquid staking as collateral and DeFi than financial analyst to analyze systemic risk and these things, we'll actually prefer collateral that maximizes censorship resistance. And that is a crazy thing, because it would tie incentives for maximizing censorship resistance in the network to its actual use and primary use being DeFi. Is this real or not? Is this going to happen?

Vasiliy (02:00):

I probably got some experience to tell here because we were in production longer, not on Solana, but in general, longer than most liquid staking pools. And I can say that it's less pressure to decentralize than I thought it would be on one hand. On the other hand is much more pressure than we usually have as a stake provider, as node operator. I come from a stake provider, P2P.org that is pretty big itself. So about 4 billion stake of fire down depends on phase of the moon, the day.

And people who usually stake, there is the kind of weak, very weak, but it's a prisoner's dilemma when people are incentivized to stake with best node operator. And when there is no clear best, they go by brand but there is a number of pretty good node operator that people are incentivized to stake with because these good node operator don't lose the mistake and give them good profits and stuff like that. And basically it leads to centralization because they are not incentivized very much to the centralized stake. And it's probably on goodwill and many stakers don't give enough thought to goodwill, but stake and pools always do. Basically, they're professionally obliged to do this and better holding up to hold the node operator accountable.

I think in Lido, we have a better monitoring system for node operator around for us. And most of the big stakers like changes and funds and stuff that we're monitoring people who stake for us way better than most stakers. And I can say that the trade of is real here, when that liquid stake can token hold us, are not putting a lot of pressure on us, but they're putting some and we are well equipped to react to that. And we would honestly welcome more pressure on this front.

Anatoly (04:14):

So, Ella, has this been easy to convince people that censorship resistance matters or is it like they're just learning about it for the first time?

Ella (04:24):

Yeah, that's an interesting question. I think I definitely would second what Vasiliy was saying about how it's surprisingly harder than you would think. People definitely will follow where the rewards are. And I don't think that is surprising. I think there's an interesting opportunity for Stake Pools to play with that idea and give rewards while also touting the benefits of censorship resistance. So, "Hey, we will give you great rewards, but you can also get governance tokens and you can help us build the future together."

And I think there's an interesting way that you can frame that discussion where you don't really have to pick one or the other. And I think to put a maybe crazy idea out there, I think we're only seeing the beginning of what can be built on top of Stake Pools. So it's pretty standard to take your Stake Pools tokens and you go stake them and then you earn some additional yield there. But I don't think we've really unlocked the potential of realizing that the underlying asset that you're staking will continue to accrue value every epoch and you should be able to build crazy financial things on top of that, that actually give you way better rewards than staking with an individual validator will ever do.

And as the product person, I just put crazy ideas out there and wait for other people to build them. But I think we're at the very early stages of that. And so I'm super excited for a year from now, what crazy things people have built, where the rewards are actually way sexier in Stake Pools. And you don't even have to care about censorship resistance by the fact that you participate in Stake Pools, you will be helping that. So that's the future that I'm really excited for.

Anatoly (06:09):

What do you think FP?

FP (06:12):

So the first question was, what do we think about the efforts towards decentralization and I think we're getting there, but I think it's still early days. If you add all of our Stake Pool operators together, we may have 10 minutes all between us and that's less than ever stake. That's less than one validator. So there's still a long way to go. And they charge 8% fees. What's going on? So definitely it is not a rational choice. It's more of a possibly just like inertia sort of thing.

And then I would say, to me, there seems to be a little bit of a trade off between Stake Pools and decentralization. And what I mean by that is even between Stake Pools, there are Stake Pools that decentralize more and there are Stake Pools that decentralize less. And in some sense, there is a trade off here because if you stake with too many validators, then you don't get good APY and people don't want to stake with you. And of course, if you only stake with the best ones, then you're not really doing your job as a stake pool. So there's a little bit of a delicate balance here, but I like what Ella said in the sense that there's interesting financial instruments you can build on top, which should make the APY discussion, it just falls out.

Anatoly (07:30):

So the APY is between all the pools and validators are pretty close. They don't really deviate by more than like 10%. Do investors actually optimize for that right now or participants? Are they actually looking at that or are they making a decision once and not even thinking about it later for months on end? What kind of behaviors do you guys see both as a normal stake operator and a pool operator?

Vasiliy (08:00):

As a stake operator, I can say that there is a lot of people who absolutely look at returns. We usually, when we go into network, we prepare profit reports for them and show them they are staking with us and we get better returns and stuff like that because that's one of the points that node operator can actually differentiate on. And there is not a lot of them, basically. We offer the same service to people.

But as a liquid staking protocols, there is a lot more of thing that can be a differentiator, way lot a lot. The node operator selection is one thing. Other thing is the opportunities to use your stake token in DeFi and CeFi and financial use for it. And this stuff beats these 0.1% point difference squarely. People don't care about the 0.1% point difference. But when they can actually use your token in 10 more protocols than the other person. So I think like, that's going to play as a serious factor way, way in the future, not for the few first years of stake, the liquid staking.

Anatoly (09:20):

So this is the difference between stable coins. Is how much penetration they have across DeFi protocols or exchanges even. Do you think exchanges are going to start having liquid staking like Lido, so Lido token?

Vasiliy (09:38):

Yes. I know it'll happen. It's not the matter of I think, I know it'll happen. It'll be inevitable. It'll start with smaller changes that don't have capacity to develop their own stake, liquid staking and don't have the network effect to make it a good option for people to use their exchange liquid staking. And then it comes to basically everywhere, I think. There is a pretty serious trading volume on liquid staking tokens right now and it's growing bigger month by month. So eventually, it'll be stupid not to waste them.

Anatoly (10:19):

FP, is that what you guys are most worried about or most working on? How do you get penetration across DeFi?

FP (10:26):

Yeah. I think so. Something that worries me is a lot of the protocols giving out emissions and the TVL is growing and all that. But I just wonder how much of it is organic growth because Stake Pools are very different from AMMs like ORCA or trading Texas, Mango where whereby in ORCA, they make their revenues from you doing stuff, from you trading or doing stuff. But in a stake pool, you want to do nothing. I mean, what we want our users is just literally put the SOL in us and just do nothing.

So it is a little bit of a different incentivization. And I wonder whether these incentives are sustainable, because look, if you're chasing the people who are farming short-term yield, these are not the people that you want in your stake pool anyway. You want people who are in it for the long haul. So I'm a little bit worried about this.

Ella (11:17):

Yeah. To piggyback off of that. I think something that's uniquely interesting for Stake Pools that is not true for staking to an individual validator is yes, you want them to just hold their stake tokens in your pool, but you also do want them to participate in the broader project. And what I mean by that is when you have of governance tokens, you have the ability to actually impact where the project will go. And you have the ability to be active in a way that you can't be, if you are, let's say, in CeFi buying an index fund from Vanguard. They're not going to ask you, "Hey, do you have opinions about where Vanguard should go next?" And I think similarly, if you're staking to an individual validator, like sure, they might be earning you great rewards. That's very important, obviously. But I think at some point, everybody gets to a point where they say, "Hey, more rewards would be great, but what I really want is a community."

And so I think Stake Pools that lean into this idea of, hey, we're going to give you this governance token, yes, hold your tokens. Do whatever you want on DeFi. But more than that, tell us what you want to see in the community and where you want the future of this project to go. I think that's a very unique power to Stake Pools that will organically grow. We just have to figure out how to market that in a way that's appealing to people who are institutional investors, retail investors, total crypto newbies, who don't even know what a Dow is. Don't know what governance tokens are, don't know what a stake pool is. So there's a lot of work to do there, but I think we have our work cut out for us because it lends itself to this very unique dynamic between all of the stakers.

Vasiliy (12:56):

The way I think about that is it will be a lot more market driven than participation in governance doing. People are usually who are staking as node operator and provided most of them, don't care to make governance decisions. You can actually look at how it will works with Cosmos and other proof stake blockchains, where governance is a part of staking. And you can see that most people don't vote apart from how they validate the votes, where they do.

They select basically a company that is aligned to them or maybe select the person that give them best returns. And then they don't take a look at governance usually. That's not true for all people, but that's a clear majority that delegates the governance power and it'll be pretty much the same with Stake Pools with liquid staking protocols. They won't be able to even to connect with most of the holders of the staking tokens, because they won't be like passionate enough to connect back, to understand what they want. So it'll be very indirect.

There will be staking pools that gouge some of the governance decision from stakers, but not from all of it. Not even from most of them, like from 10% of them, by volume and not by number. By number, it'll be like probably not 10% like about 0.1%, but they will take much more or maybe about the same pressure from protocols that uses staking token from the stakeholders in the blockchain ecosystem that don't use a liquid staking token by important like develop teams, develop clients and researchers as an ecosystem and stuff like that. And liquid staking pool will be a nexus of governance that will try to combine all this pressure in the single direction from stakers, from protocols, from major participants in the ecosystem.

Anatoly (14:58):

What is governance in a liquid stake pool? What is the function of it for the community that owns the token? What should they be looking at?

FP (15:08):

First and foremost, the delegation strategy. I think the community needs to decide the delegation strategy. I don't think this should be left to the founders or the creators of the stake pool. It should be democratized. I think another thing is fees. So I think the community should decide the fees that a stake pool should charge. And the last thing I would say is, we would like a lot of the associated infrastructure to be run by the community as well.

So for instance, the program, the upgrade authority is already given to the community. Treasury decisions are already given to the community, but there are still things like the front end or paying for a custom RPC note and things like that right now is centralized. And we would like that all to be on chain eventually. So I think that's all quite important.

Vasiliy (16:02):

My thought here is that the role of governance in a good liquid staking protocol is to drive itself to extinction. So it won't be easy or it won't be fast, but essentially liquid staking is walking in the outermost part of the security of the protocol. It touches the most important parts of the protocol like censorship, resistance, and decetralization and security and all of that. And if it gets a significant power in this parts and if it's not credibly neutral, it's like a great thing.

It should be credibly neutral and you can't be credibly neutral for long when your governance is overpowered. It's a natural thing for all governance to take too much power and use it in not a great way. So it basically has to, in order to be accepted by stakers and ecosystem as a ligand liquid stake protocol. The ligand part of staking, it should be self-depreciating to a point that where governance power are time locked and very light and mostly algorithm driven.

Anatoly (17:27):

This is interesting point because I think the goal of governance of a layer one is also to obsolete itself. Is how do we build the structures? And part of the reason of building out Stake Pools was because the foundation was running its own delegation program. And it really felt like why don't we get the community to do its own delegation programs. And then how do we get zero to one thing working, how do we now go want to earn? And that's always a way to disintermediate yourself from the governance work and then eliminate it all together. I think it's interesting that like inherently there isn't a drive to eliminate it from the community. We just want to push it out of the foundation and have you guys figure out what does that fine line between automation and having everything be programmatic to on chain governance?

Vasiliy (18:32):

Well, not yet. It's a work in progress. We are working on maybe systemizing the ways we can... What inputs do we have, is this programmatic governance, to understand where we can get the signal from, what we can use as a strong signal. We can't get rid of the governance entirely. We can just make it in a way that... Well, like I said, the role of governance in the mistaken is to take all this input from protocols and ecosystem and stakers and the outside water is large and fabricator of consensus out of it.

So part of this can be automated because we can have the signals in bits and bites and we can use algorithm to aggregate this signals into party of decision maybe. Right now, we're looking at stuff like what is objectively good characteristics of a node operator for example, for selecting node operator like up time and special risk and the reputation that is proxy by amount of stake can all the other protocols that they are staking.

And this is a strong signal. We can look at like time of operation within Lido, which is roughly correlate with reputation and outside Lido as well. We can look at stake as preference and the stake token can hold the preference to understand what they want, which is also a proxy for reputation, which I don't have. The things I don't have a good solution for getting into account, what people who run protocols think and what people who are major in the ecosystem think, because it's not directly correlated to a stake in stake pool. And we don't have a good way to get these signals yet, maybe ever.

Anatoly (20:29):

You guys like Lido and FP have two different approaches from what I can tell in terms of building out the validator set and the delegation strategy. FP, what are your thoughts on this? What are you guys driving most as the number one factor in selecting validators?

FP (20:48):

So I think it's important not to have a white list of validators because I think this is exclusionary. I think it's important not to dictate what fees validators should charge, because I think fees are only important in so far as they affect performance. So in some sense, we don't want to control validators. I think we shouldn't. We shouldn't dictate how validators... That being said, of course performance over time is very important. I think if not the most important. Yeah.

And the other thing I would say is, the decentralization, obviously we shouldn't be staking to nodes that are in the MSG, they have too much stake or nodes that are in one of the data centers that is in the MSG. So one of the top three data centers. But that being said, there also a middle ground. You don't want to spread your stake among, let's say, 600 validators, for example. And the reason why you don't want to do that is because then you can't make a meaningful difference in decentralization. You want of do want to reward validators that are doing well, that are also out of the security group. So yeah, I would say it's a bit of a balancing act here.

Anatoly (22:13):

Vasiliy, you guys have a totally different approach. I'm excited too, why did you guys come up with that system? And what is the Lido way?

Vasiliy (22:19):

To expand a bit on what the system is, we've got a wide list of node operator that run with Lido and charge the same commission and get the flat amount of reward. What the reason behind this, the whitelist selection is done by basically a peer review. We've got a lot of node operators, already validating Lido in different protocols in Ethereum, in Terra and now in Solana. And we have a submission process where people submit, they want to stake for Lido and we get the node operators. They took a look at them at the setup they have and historical performance in Solana and other blockchains, especially in Solana and stuff like that. And community participation and select that the next five or so participant of the whitelist when we need to expand.

The why we do that because we want to have good stake distribution that will be good for Solana and that's not the best, but it's easily achievable way to do that. Because that way we can guarantee that node operator are good because they're selected by the community of node operator essentially. And we can guarantee that they have enough stake to run the operations and have enough profit that say that. So they really want to keep this good business going. That's a good business for them. That's what they want to do. They are not arranged by scrap. They are paying their DevOps engineers handsome salary and stuff like that, so that they can afford to be honest.

It's not great in the sense that it's a process that allows us to select the distribution folks, but it doesn't allow people to come in fresh and grow. And that's not great. But as a temporary thing, when there is a good community of node operator that are just like not selected yet, it works, I think very well.

FP (24:36):

I think part of the reason why Lido does it is from what Vasiliy said, it's meant to make sure that the node operator are reliable and performant. And I would put forward that there's a very easy way to look and to see if a node operator is performing, just look at their API. So in some sense, I mean, I don't want to make any implications, but I believe this peer review process is a bit nepotistic. It's like if you're in our secret cabal and if we know you and and we like your DevOps engineers and blah, blah, blah, then will onboard you. Of course, that's not the case, but it's what it seems like.

Anatoly (25:15):

This is the most controversial Solana podcast we've ever had.

Vasiliy (25:21):

I wouldn't say that's not true. It really does not allow newcomers to come in easily because there is a community of node operator that been through thick and thin via market, like Greeks through this days, when we all worked like in the red four years, that was what happened. We used make way less money than we earned, like with P2P, which was a pretty big one even this time. Like I said, it's not great, with this process, we can't get in people who didn't build this reputation and track record and stuff.

What I don't agree with you that you can easily estimate how good is node operator, but looking at their performance, that's just not true. That's not how you estimate a node operator. You don't only evaluate performance. You also evaluate tailor risks. And tailor risks, you can't evaluate by performance. You should understand that these folks have bus factor of more than one. They don't have a single guy running all this stuff because if this guy gets sick, your validators get stuck.

You should understand that they will stay up at night when there is an upgrade. You should understand that if there is a via market, they will stay to the blockchains they're running and they don't all run on Hetzner. So because that's, at least used to be the easiest way to get APR is to run the same data center as everyone. That's how skip rates they used to work in Solana.

There is way more nuance in selecting a good set of node operator than just looking at performance. The geographical distribution, the jurisdiction distribution, the track record, other blockchains, which runs the reputation and community participation being in discord or running projects for Solana and stuff like that. There is way more stuff about node operator that is not easy to understand from just on chain metric. On chain metric is like the 20, 30% important stuff of choosing validator, because there are a lot of validators with good on chain metrics, but there are differentiated by stuff that is not seen by most people at all.

Ella (27:44):

I would say if somebody is staying up all night to make sure that their validator is running and they do restarts within the first five, 10 minutes, they're going to have better rewards. So I would say, it's more than 20%. I agree that being decentralized and being in data centers that are different from other people are doing community projects is super important. But I do think that rewards are a good proxy for how active the validator is actually running their node.

Vasiliy (28:13):

You can say that, that's a prerequisite. If you have good bad performance, you're not a good validator. That's true. That's not what makes your excellent node operator because excellent node operators run explorers, for example. And there are certs basically, for example. You can't say that this guy has the same performance cert, so they're as good. That's not true.

Ella (28:35):

But I mean, I would say there are maybe like 10 community members who run dashboards and different tooling. And I think there are way more than 10 stellar validators. So sometimes it's just not within their area of expertise. They could be excellent DevOps people and run validators across many blockchains, but they're not a web developer. That's just not their skillset, but I wouldn't say that they don't contribute to the community.

Vasiliy (29:01):

Yeah. What I'm saying just there is much more nuance, especially when you don't have 300 places for a node operator, you don't have enough money to pay them for 300 validators and you need to select 15 or 20 or 50.

Ella (29:18):

Unless a hundred million SOL gets stake to Stake Pools, then you can expand that list to 3,000 validators and everyone will be profitable.

Anatoly (29:26):

So this is the challenges. How do we grow the validator set? And it almost in my mind is like, you need both, you need people that are driving, we need higher quality. We need due to proof points that you know how to manage keys, but we also need people that are like, okay, just on board and figure it out and try it. Yeah. This is a tough problem. And I think part of the reason of not wanting the foundation to do it and push out this technology, a stake fulls is because we don't know. You guys are both sound very much validator operator focused, but these things like, I think are some form of financial, like DeFi application too. How much of your time are you thinking about like how these things actually work in DeFi?

Vasiliy (30:22):

I think I'd say a lot. That's what makes or breaks the liquid staking, the whole point of liquid staking is that it's liquid and usable in finance. I actually don't think a lot about a lot of time about node operator because I used to work here. I'm working as the staking provider since like 2020, early 2020. So I'm just have strong opinions because I do it right. But I have to think a lot about DeFi because that's uncharted, it's new.

FP (30:58):

So yeah. I mean, I think as Toly points out, I think the validator operator stuff is important, but really it really is just a baseline. And I think what we do with it next is the thing that's more important. So the question was, how do we think about how it's composed with DeFi? It's just the beginning. So right now what are the main things that you can do with your stake pool token? So you can put it in an AMM and provide liquidity that way, you can do lending and that's about it, I think.

I mean, there's lots of stuff you can do and you want to use the stake pool stake SOL in any occasion where you can use regular SOL. So whether it's just buying from a marketplace or doing some more exotic stuff, like options trading, that sort of stuff and not just putting it in liquidity pool or borrowing or lending or leverage yield farming. So yeah. I basically want to expand the ways in which stake pool tokens can be used. And I think that's going to be a big draw for people to start staking with us.

Anatoly (32:18):

How much work is it to get that adoption or to have a specific stake pool token used in a DeFi?

FP (32:27):

I think integration takes time. I mean, it really depends on the partner which you're integrating with. And I think some things just haven't been built out yet actually. So Ella and I have been talking about how we can use these stake pool tokens in the NFT marketplaces, for example. But none of this stuff has been built out. So, yeah. So we'll get that, but it's not there yet, I would say. So we have to build it.

Vasiliy (32:57):

There is two parts to the answer. One is how long does it take to build. The other is how long does it take to convince people to build? The first is, faster than usual for financial products in traditional finances, but still long because we know that shipping is hard and convincing is also can be pretty, pretty complicated. For example, we started the integration process on MiCA, I think in February this year. And we only now getting an executive at least take things on MiCA, I think around next week or so. That's how long it cost with MiCA. And it's very similar amounts of time with a major protocols on Ethereum that are by now pretty conservative. Solana is not conservative yet. Most of the protocols on Solana make fast and break things, move fast and break things. So I don't think it'll take like this long stake Solana tokens to be a major participant of DeFi, but it's still time.

Ella (34:12):

Yeah. I would say the technical integrations, they're not technically challenging, you're integrating an SPL token. So that part is pretty easy or not as challenging as you would imagine. I think in the early days, when the TVL was very small, it was maybe hard to convince platforms that they should care about this weird stake pool token thing. Now that TVL is close to $2 billion US dollars. They maybe will now take those meetings and be like, "Oh, okay. Yeah, let's integrate all the stake pool tokens."

And maybe whereas before they would have some liquidity requirements say, prove that users actually want this on our platform. Why should we spend the time integrating it? I think hopefully the script will flip and they'll be like, "Hey please, can we integrate your stake pool token?" But I think it just, realistically it takes a couple of months to get at that traction. And hopefully we have some momentum now and we can push forward more of those integrations.

Anatoly (35:08):

Is there kind of danger of liquidity being fractured between too many pools?

FP (35:14):

Hey, I seem to recall asking you this exact question on discord back in September, Toly. Yeah. I wondered this myself to be honest because I think there is a happy medium. You don't want one stake pool taking all of it because there are protocol risk there as it ends points out. Yeah. And if they fail it, that's dangerous. Well, on the other hand, it's going to be really difficult to integrate a hundred different stakes pools.

Although that being said, there are things we can do to mitigate it. One of which is to enforce some sort of standardization. So one good step would be, for example, to use the Solana reference recommendation instead of... Maybe it's too late now for some of the existing Stake Pools. But that being said we were talking about adapters. I don't know if you recall some sort of adapter, some sort of layer that makes sure that the Stake Pools can all interoperate with one another. I think that would be really good.

Vasiliy (36:14):

I think it's inevitable that a single representation of staked Solana to be the major player here. So that's basically Lido thesis and I'm seeing it play out in the Ethereum and in the LUNA and in the entire ecosystem. So I think it's going to happen. It doesn't necessarily mean that it'll be one stake pool, but the alternative here is just another layer of aggregation. One thing for example, was proposed by Michael from Curve where like basically a stable pool of multiple liquid staking tokens was used and LP token from this pool was proposed as a basically unit of account. I'm not sure that it will happen, but I am pretty sure that there will be one aggregate stake Solana token, that will take the majority of the market.

Anatoly (37:20):

I actually think that these things are far more fluid because it's all people based at the end of the day. And people will do promotions and get communities together and have fun or get excited about a thing that some innovation and you will see liquidity move from one thing to the other simply because it's exciting. And it feels like it's just a little too static for there to be only one token. This is not how normally people operate, but we'll see.

Yeah. It's at any given moment one winning token maybe is a better way to put it. So it doesn't mean there will be one token for eternity, but at any given moment, there will be a clear winner except maybe the moments of flipping that's how I see it.

FP (38:14):

So I worry a little bit about that actually. I worry because we talk about increasing decentralization. And that was the reason why Stake Pools were created in the first place. And it's true that if you have one stake pool controlling all the stake, that solves a particular kind of centralization, Nakamoto coefficient. But then it introduces a new kind of centralization. And maybe there are risks that can be mitigated that way, but still this worries me a little bit. So I'd rather have an ecosystem with a good number of different Stake Pools.

Ella (38:51):

That's where the education piece comes in. You got to let your delegators know the importance of censorship resistance and decentralization so if there is a sexy new aggregated stake pool token, they don't just gravitate towards it because it looks good without thinking about the consequences of that.

Anatoly (39:09):

But the yields are so high.

Vasiliy (39:15):

I don't like the dynamic at all, that there will be one lean stake token, but I think it's inevitable. And what we can do is not oppose it, but we can build protocols that will be a net good for the system anyway, even if this happens, hence the self depreciate of governance and in liquid stake and stuff like that, that's all flowing from there.

Ella (39:40):

I wouldn't say I'm oppose to it, I think in an instance where you have 10 really small stake pool operators, let's say universities decide, hey, we want to run the Yale stake pool and the UDab stake pool. And they have very fragmented liquidity. I think it makes total sense for there to be an aggregated university stake pool token, support university students help them get their pizza and ramen. Great. That's a fun way to do it. But that's a very specific use case where you're trying to make sure liquidity isn't fragmented.

But I think every stake pool today has more than 600,000 sold deposited into it Solana. So I wouldn't say that's a huge fragmentation. It seems like people have chosen the pool that they like and they're happy with the performance, with the project. And that's the one that they picked. And so I don't know that they would be attracted to something that tries to average everything out and is just a generic token, but I could be wrong.

Vasiliy (40:38):

Yeah. It's so very interesting to see the play out.

Anatoly (40:40):

So are you guys worried about if these are used collateral, like liquidations rapidly moving stake from the lenders to the people collecting to the traders, is that going to change the dynamic of the makeup of who controls the network over the long-term?

Vasiliy (41:05):

Not reallY. How I've seen it work by, in liquidation that happened in Terra and similar ones that there were not exactly strictly liquidation, but more of fire sale events in Ethereum when the price of weather went down and people were going out of stake teeth as well. The dynamics here is that people who have low time preference are selling at low prices and people who have higher time preference, they are buying. So then they went of the liquidation, the price goes down and people with more foresight and more patience are getting the discounted stake token. So if anything, that looks like stake token getting in the hands that smarter and are in for a longer game, usually. So not always the case, but very much looks like this.

FP (42:05):

I don't know. I mean, that being said, when you have all these incentive programs and emissions coming out. That doesn't that see to incentivize people who jump around pools, trying to find the best ones. And they're getting rewarded by lots of governance tokens at the end of the day. So what do we think about that?

Vasiliy (42:25):

I don't think that it's something to really think about, I don't know. Jumping around and getting this governance tokens and it is a natural way to get some money for people who like money. I don't know. That's not a bad thing. If you like some juicer smart contract risks and rockeries in your life, that's a very exciting way to spend time.

Anatoly (42:55):

Yeah. There is I think a danger, but I don't know how big it is in that normally for like a validator to receive more stake, the best they can really do is offer 0% commission and then they can start bribing people. And it's hard to bribe people, but with liquid staking, it's a lot easier. You can just simply say, when you stake with this pool, you get so many more rewards than you do anywhere else, because you can min this new reward token. And is that a dangerous, scary thing that could result with a third or more of the stake, all moving towards this hot, shiny thing? I don't know.

This is the part of where I think it's very critical for DeFi to mature and to have real analysts and people analyzing these things and looking deep and giving a ton of pushback on things that look a little fishy. It naturally happens, but only happens on crypto Twitter and still so much stuff sneaks through.

Anatoly (44:06):

So we'll see what happens, but thank you guys for joining. Super excited to have this actually being live now and making so much headway and growing so rapidly. Honestly, if we actually get to a point where DeFi is incentivizing censorship resistance, we're kind of done. We built it. We can actually take a break. So I'm looking forward to that.

FP (44:41):

Is that the biggest concern for you as a creator of the layer one, the increasing this decentralization, would you say that's the biggest concern?

Anatoly (44:49):

Yeah. This is the thing that I'm most worried about, because I think to do it in a sustainable way, it means that you need to have a use case which benefits from decentralization. You need to have external users that have a benefit that exceeds the cost of running the network. It can't just be self-sustaining tokens moving around. So to truly succeed there means that, we build something useful to the world. And that's the ultimate goal.

What else are you an engineer if not to build something useful? If that's what you care about, then you should be an artist and that's a totally different thing. Yeah. Awesome to chat with you guys. Thank you for being on the show and thank you for all the hard work everyone is doing, Ella, Vasiliy, FP. Just thank you guys.

FP (45:48):

Thank you so much for having us today.

Vasiliy (45:49):

Thank you.

Ella (45:49):

Thank you.