Small pools are dying...what do you think?

nOpt a.k.a. k, means optimal number of saturated pools in the Nash equilibrium (i.e. k number fo fully saturated pool). You’re making statements at the 6th epoch (i.e. myopic view as it stated in the ledger/design spec).

Anyway, if k is 150, and number of pools >1000, what do you think would happen to the other 850 (assuming very small tailing)? The current a0 is very small pool friendly, we expected higher, in which case a much-much more small pools would suffer much harder for long term.

k in other hand, affects rewards, the current 150 was a bit underestimated, and we expected 250-350 for boot strap, and around 350-600 after a half year and ~1000 if it’s sustainable at all (cos it affects rewards) after a year or few years. They’re going to increase it, I would expect between 350-500 cos of the current crasiness. But, anyway, what do you think would happen to the rest or them who are not ranked in that k or nOpt number? And what if your pool is not in that golden-ticketed one?

And, also, what does small pool means? Small, by delegation or stake?

I always referring to a small pool as small pledged pool, and not the stake delegated to them.
If you do not know why than that’s ok.

We’are social animals and behave like animals, and stupid enough to justify our actions as it would rational.

5 Likes

As always, I promote Cardano as a future protocol which does not give any sh*t about our wet dreams (driving lambo /w some blonde on the side as a part of the deal etc.), but for common good sometimes in the future, and if my pool won’t survive (very-very likely) I won’t screaming and blaming anything else, but sitting down and be satisfied that it could achieve what was expected to be done, eliminate the weakest link of the chain, us the humans, and running a sustainable long living protocol. If not, then it’s failed.

As it should not be a pool agnostic but simply human agnostic protocol.

7 Likes

This is a bit extreme. I don’t think expressing concerns about the network tending towards centralization is necessarily driven by “wet dreams”. Speaking for myself, I participate in these discussions with “sustainability” in mind, not lambos and “blondes” (which by the way, is also an unhealthy stereotype).

If it’s not sustainable for the little pools, real decentralization will remain just a dream, as the network will be dominated by no one else aside from the very few who are already at the top from the start.

Humans are behind these pools. It’s probably not a good idea to decouple the operators from their pools.

6 Likes

Nice video. Good to see the faces behind the stake pool. I would say that it is too early for the small pools to decide to quit. Give it a try at least to 4 ~ 6 months to work out (if you can afford the operational cost), then make your balance.

2 Likes

There are no small pools in the target equilibrium state. Just k (almost) fully saturated pools. All other pools will die or just run on their own pledge. Since we do have pledge, besides pool fees which will be pretty low as well, the top k pools will be the ones with the most pledge. This will ensure decentralization since a pool like 1pct won’t be able to have 10 pools with such a high pledge. And if they have, fine, all the more incentive for them to keep the network alive and well.

1 Like

Yep, that’s what we all agree on. That’s where we want to go. But with the way things are right now, we’re not heading that way.

You probably haven’t seen this thread:

A system that is dominated by and benefits only a select few, will not last very long. Just a few hundred years ago, monarchs and dictators ruled. How many of them do you see today who are still actually in power?

The remaining monarchs are now just serving as symbolic figureheads, unable to really exercise any power. And the remaining dictators aren’t exactly popular either. This is because we’ve all learned by now that the best way to move forward is to have more people work together.

The more participants there are who help maintain the network, the better. If we are to allow power and control of the network to be consolidated, we shouldn’t even be talking about a distributed/decentralized system in the first place.

6 Likes

Totally agree! I just don’t think that we can draw conclusions regarding the parameters at the moment. Daedalus ranking is not yet working as intended, people are just learning how to stake, and of those 1k stake pools, a good bunch is probably not performant enough in the long run.

The problem of having some kind of oligarchy governing Cardano is an eventuality I also see. The way the stakes are distributed favors people who bought in at the ico or traded very early. Not saying the shouldn’t be rewarded for the risk they took back in the day but it makes cardano more of a corporation or conglomerate which offers its services to us, the users and only let’s us participate in a small portion. I don’t have any data on this but I speculate that among all people active in the forum, reddit and other channels maybe own 30% of circulating supply. I’m sure someone has made a more detailed analysis about this.

Edit: another problem might become voting decisions based on short sighted greed, just like you see with many stock corporations.

5 Likes

This is exactly what I saw when I took a look at EOS in the early days of looking at Cardano, and decided that EOS had no future by comparison. I tried to work with EOS for a couple of things & it was like AWS / Amazon Web Services: only with a more finely distributed architecture, with more complicated rules, with no real decentralisation or representation.

I knew that there had to be something different, and believed Cardano was it. The consolidation of influence you’re describing is definitely possible but only at the cost of the promises of “pushing power to the edges” that were made by CH from the beginning and believed by the rest of us all along. It will be more than just a disappointment not to have those promises come true… it will the loss of Cardano’s unique edge over EOS and other competitors.

8 Likes

I think the “power is pushed to the edges” because people can use the the services that will run on top of cardano and don’t have to rely on some entity in between to transfer their value or proof their identity and so on. It’s not meant as anyone can run a successful stake pool. And actually I’m less concerned regarding the stake pools and more regarding how “the whales” will influence all the voting. Sure their incentive is to grow the network value, but this still doesn’t mean they’ll be making the “right” decisions.

Only time will tell. Until then it’s a giant 2.5 billion dollar experiment.

2 Likes

Yes, let’s see how the “unbanked” and disenfranchised of the world take it when they realise they’re free to participate as consumers but not as producers.

3 Likes

I am also running a small pool and I completely agree with your statement. What I don’t like is that there is a big delta between the current situation and what has been advertised by IOHK.

  1. There was so much talk about “skin in the game” and now our pledge means nothing. One should be aware that people had been increasing their pledge and maybe even stretched to the point of their financial limits or beyond to be more competitive and now they are not. Everybody can join and be as much competitive with the pledge or without it. We can’t get rid of the feeling that everything was just a marketing move on the expense of early adopters so the media could write about enormous success of shelley with more than 1000 pools.
  2. In the ITN, IOHK and the foundation were giving out so called golden tickets. A pool would receive 10, 20 or more millions stake. Since they never said what the criteria to receive the golden ticket was we actually never knew if we were going to get it or not. But those of us who got it were implicitly led to believe that we could expect some support to kickstart the pool on the mainnet too. I am aware that late comers wouldn’t agree with me but if you think for a while, those of us who run pools in the ITN have been committed to the project for almost a year, some even longer. People invested heavily in infrastructure, we invested a lot of time into the project. We’ve been running nodes, testing, reporting bugs etc. since then. Would be really unfair for us to receive some support on the mainnet?
  3. Because of how things were conducted during the ITN majority of pool operators were expecting some support from IOHK on the mainnet. Instead of supporting all/majority of pools (not just ITN pools) with at least 3 M stakes as you suggest, they started their own pools. And we are not talking about 2 or 5 pools. We are talking about 20 IOHK pools and emurgo pool(s) and now Charles says that he will start his own pool (another 200M gone from the community pools!). Everyone is saying “wait, it’s only the beginning” and so on. This mantra is pure hopium! It’s is completely wrong! It may be the beginning for you, but some of us have been investing time into the project for a year. We should demand numbers, dates. But even that is not enough. We need transparency! When IOHK declared that k would be set to 150, they did not say they were planning to start 20 of their own pools. We could say the k is 130 not 150. Was that transparent? No, it wasn’t!

We’ve been waiting for IOHK to deliver for a long, long time. We, the early adopters were the ones supporting them from the start and during the hard times when there were more sceptics around. Showing some respect to community through transparency is not too much to ask I think. If my pool doesn’t make it, so be it. What I want is to have transparency and not the abuse of power that we are seeing now - keeping everyone silent and giving them hope that they will receive their golden ticket and keep telling them that it’s just the beginning.

Remember saying and doing are two different things. Once Google’s corporate phrase was: ‘Don’t be evil’ but we all know how that turned out.

Don’t get me wrong. I still believe that Cardano is technically one of the best projects out there and I am sure it has great potential to succeed but I believe those things need to be addressed.

IOHK, keep your zillions of ada but please, just be more transparent.

Sorry for the mistakes, I am not a native English speaker.

15 Likes

Interesting insight! I was not involved in any way with the ITN besides delegating so I can’t comment on what was communicated to the people actively participating.

Even though cardano continues to be one of the most transparent blockchain projects I agree with you that some topics as you mentioned could be communicated more clearly. AMAs and all are great, but not everyone has the time to watch those and so I for one at least would appreciate some iohk staff chiming in here more often (maybe this happens on the Telegram channel, idk).

Speaking for myself, what I mean when saying we’ll have to be patient for a while longer is that due to the desirability ranking not working properly at the moment it’s hard to tell what effects a0 and k actually have.

If you have enough “skin in the game” your pool should be able to win at least one slot to make your performance factor > 0 which I guess is the first big hurdle to take in order to attract delegators. A higher k value won’t change that only a lower d. It’s probably to late for that now, but if it were possible that pools won slots as if d was zero but the federated nodes could still override a percentage the blocks based on d it would be easier for small pools to bootstrap into the system at this early stage.

Regarding iohk running their own pools, this wouldn’t be that big of a problem if almost all Ada was staked, in this case they’d only control around 6% of the active stake. CF also should have roughly 6% and I hope they delegated it more or less evenly. Once exchanged start delegating this should also boost smaller pools since an exchange would look to maximize profits and thus delegate to the most desirable pools. Of course they could also run their own pools, but suspect this is not part of their business model as they make much more money by trading and lending. Also they’d need to lock up a big amount of Ada as pledge which could pose a liquidity risk for them.

Having said all that, as long as d is still > 0.5 I think it wouldn’t hurt to try out higher k values and see how that goes (big pools won’t like that though ;)). If it turns out that most pools are not up to a certain standard k could be decreased again.

PS : not being a native English speaker myself I didn’t notice that you might not be.

Few or no mistakes language-wise, I would say. :slight_smile: A well-written post, and similar to many of my thoughts in my long post today at Running multiple pools is more profitable than just a single one -> this does not lead to true decentralization! . Since I’m not an ambassador, there’s not really anything to do, there, but despite weeks of work researching, packaging, and testing Cardano Node (including filing debug diagnostics multiple times during build failure issues via Cabal with Cardano Node 1.18.1 (prior to being ‘pulled’) and 1.19.0 (I’m running a patched version, now—almost all I which I feel was unacknowledged by maintainers), and in ultimately open-sourcing my independent take on packaging Cardano Node in Docker specifically from a high-availability point of view (including block-producing nodes), I’m actively considering how much longer I’ll want to stick around running CRAB pool, since it’s so far from profitable, what’s gone to launch isn’t what I understood would happen, and I’m beginning to feel a bit like I’ve sacrificed weeks of full-time work and infrastructure—for what?

I’ve been an active open-source contributor for years, and that’s fine, but often those projects I’ve been involved with haven’t been backed by a huge company, or been explicitly finance-driven. I ultimately am very supportive of the technology behind Cardano, and I agree with many (although not all) of the goals, and I certainly think it and ADA will be around for some time to come. But that is very different to it making business sense to operate a stake pool, since the understanding I had of an experienced devops/programmer/architect with all the ‘required experience’ checkpoints and far more, being able and willing to operate a stable, reliable, highly-available stake pool, and with a large financial investment as well with regards to pledge (measuring in the thousands of Euros, which to me, at least, is a significant sum), appears to have been a huge misunderstanding.

Ultimately, all that now appears to matter is huge amounts of stake and marketing, and even then, I see pools 1 or 2 orders of magnitude larger than mine, that still stand little chance as things stand at breaking into a market dominated from launch by numerous IOG pools and now other sets of pools seemingly with a single operator, too. It’s really very disappointing. :slightly_frowning_face:

3 Likes

Do people think the IOG pools are here to stay? I’m sure I saw it explicitly stated somewhere that they’re only intended to help get everything up and running and will be retired. The Emurgo ones too, I think. Don’t remember where I saw that I’m afraid.

2 Likes

That’s what I am talking about. No transparency.

1 Like

Thanks @adante for bringing me here. I read through most of all the replies on this thread. It’s a difficult situation indeed for many. I think the best way to look at this problem is the same way you would any other business. What is your business plan, what are you goals, how much risk is involved and how long are you willing to take those risks and cover the losses before you throw in the towel and declare the business a total loss and move on. That is a personal choice for each of us.

After carefully reviewing “Design Specification for Delegation and Incentives in Cardano” document released by IOHK July 23rd, 2020, specifically Section 5.6.1, it became clear to me that the present conditions we are struggling with are by design. The ranking problem in Daedalus is also by design, they are following design spec per 5.6.1. “We predict that pools with rank ≤ k will eventually be saturated, whereas pools with rank > k will lose all members and only consist of the owner(s).”

What I also noticed was Section 5.9 Information in Daedalus. In that section they concede their design spec to an assumption that Delegators will always have “ALL” the information they need to make the best choice about delegation. This is what the wallet ranking system is failing to do by keeping saturated pools highly ranked. A correction here is needed ASAP. It won’t fix the whole problem but it will move us in the right direction.

All in all, like many have said, this is a long game, we are dealing with probability game theory functions. They will be harsh. Not everyone will survive, many will lose money trying to. The best thing I believe we can do now is simply bring as much value to delegators and the project as a whole as possible. That also includes winning a community seat at the CF table for better representation of SPOs. That also means raising “k” sooner than later to put more pressure on saturated pools to outflow, even if that means those same SPOs just spin up more pools.

Decentralization itself as an approach to solving the problems we currently face in our antiquated and corrupt monetary systems is largely incumbent upon human chaos regulating itself to an optimal function for the goals of the project. It will take a lot of time, many will perish, and others will join. This is a ride that will push us all to our limits and then, with some luck, those limits will translate directly to a collective success for us all.

5 Likes

I don’t recall this ever being true. There was no “need” for IOG1-20 or Emurgo pools on mainnet when we have the d variable which will slowly increases the blocks made by stake pools over time anyways.

The intent was always for IOG and Emurgo to run their own stakepools in some way. Cardano Foundation had stated previously that they did not plan to run their own pools, but as of now I don’t know if they are staking on the mainnet - and this may have changed.

I do recall that there were plans to have some of the IOG ADA be staked to other stake pools at some point, but we don’t know if/when this will happen. If/when we can stake to different pools from the same wallet and k (saturation) decreases, we may see IOG and others stake to other pools… or they may just spin up IOG21-99. It remains to be seen what IOG’s plan is.

We, Aloha Cardano, were hoping that a clear roadmap which explained the plans for all the variables (d (decentralization), k (saturation), and a0 (pledge)) was going to be laid out together. However, for right now the focus seems to be purely on d (decentralization). Unfortunately, currently there are too many questions and not enough answers.

Aloha Cardano
(ALOHA)

1 Like

In all organizations, conflict of interest is taken care but here it is not.

  1. IOG run pools - They have advantage over anyone else.

  2. Emurgo runs it - They have advantage over anyone else.

  3. Pool operators are having multiple titles; like Ambassadors, Explorers owners like adapools.org - all run pools. - Obviously at least some of them will have advantage but unfortunately technical ambassador pools are suffering due to less marketing.

  4. And finally ITN tickers where you have first mover advantage…

Its not a fair world :innocent: just to go with our technical capability. Most of the people survive with marketing skills.

Anyways some good news I heard is Charles wants to have a massive increase in K. That may happen soon but there is no discussion about a0. May be IOG wants to support some good dedicated guys with smaller pledges now?

Its too early so I still have faith in Cardano.

2 Likes

He did say though that a0 and k are heavily related, so I suppose both paramters will be adjusted at the same time.

I heard like once K is increased then they will revise a0. I think the first preference is for K and later a0.

If a0 is revised now, many pools will be less desirable due to very less pledge amounts.

1 Like