Hi,
I have been attempting to understand how pools will get there ranking, I think I have a pretty good grasp of it, but Sybil is giving me trouble.
It seems that Sybil by lowering everyone’s reward potential is a tax. This Sybil tax is overcome to a degree by the amount of pledge one can commit to a pool.
What I find peculiar is many PoS use a “Bond to Staking Ratio,” so if I work from that as a reasonable Sybil defense level, 12% or 3,733,498 ADA, leaving room for 27,378,986 delegated ADA balances with other PoS cryptos.
I do not have 3,733,498 ADA to pledge, which means that any stake pool I offer is not Sybil resistant.
Am I looking at this correctly?
Is Sybil a tax on all of us, operators and delegators? It seems that is is by lowering everyone’s reward for participating. Between a Sybil tax of 17% & a treasury tax of 20% on top of the operator cost and margin fees, I must be missing something.
Help.
I should also include that I operate a stake pool for another PoS crypro, which I know very well, Cardano not so much.
I think you have it pretty much correct. I don’t look at Sybil resistance as a tax. It’s somewhat of an unfortunate consequence of the game theory of POS and there to protect the network at this stage.
1 Like
It’s not really a tax, as I don’t think the reward you miss out on by being small goes to somewhere else like a treasury. If you start a stake pool with little pledged ADA then the lower potential reward will make you less attractive, but doesn’t prevent you from being able to operate.
I believe many smaller stake pool operators are looking for other creative ways to attract people and of course as you earn you could put that straight back in to grow your pledge over time. The article also mentions possible future changes to “boost” your reward by being a reliable pool.
All in all it seems to make sense, otherwise a Sybil attack would be too easy. It might be frustrating for smaller actors, but at the end of the day a global financial operating system that may have millions of users holding their currency on should be operated with a degree of professionalism by people that have something to lose.
We are moving out of the proof of concept phase as a technology that previous blockchains have been thought of as being like by most of the world.
2 Likes
Thank you @Donnybaseball & @phil.lewis,
I’m disappointed that I will not be operating a pool, I get why, but it is still disappointing.
To bad Cardano does not use the concept of pledge to Sybil delegation ratio.
Thanks again.
i am unclear still as to what you are asking. you are seeking for a tax be applied, because you familiar with such a tax from other systems? … but because Cardano is not using the same method as these other chains, then you cannot run a pool for cardano once PoS?
they are using a different method against an attack, is that is enough to sway you against running a pool?
i could not follow your example, sorry. i do not think that example you suggest will be the way used for cardano.
also, the article is over a year old now, and with Shelley right around the corner, there will be months of testing and a phased shift to a main-net ahead. we will see what happens over the next 6 months.
Hi @qardano,
Thanks for letting me know Shelley will not be here until the end of the year.
A tax is not the correct word, but Cardano does lower a pools reward through Sybil implementation. All PoS have to deal with Sybil, I run staking pool for another PoS crypto that has a “bond to staking ratio” lower bond less people can delegate to you, but you can still operate a small pool that is Sybil resistant.
Cardano choose to define the number of pools which is interesting, but makes it expensive to operate a staking pool because each pool capacity is identical, making the Sybil requirement quite large.
It is my understanding that Cardano has no minimum pledge, but Sybil protection still remains even if they are not saying how much pledge to be Sybil safe, likley a pool will need millions pledged to become a saturated pool, which it seems is the point in Cardano papers on this.
I do wish they specified mathematically how much ADA to be Sybil protected, but I see why they have not as Cardano pool selection is based mainly on rewards.
i don’t see shelley as a defined date, if you’ve been keeping up, it’ll be a process, with a beginning, middle, and end.
i’m having a hard time trying to follow, you seek sybil resistance, yes? does not cardano have it?
"you can still operate a small pool that is Sybil resistant."
how is a single pool “Sybil resistant”? isn’t the idea that all the pools together are making for the resistance?
"I do wish they specified mathematically how much ADA to be Sybil protected"
do you mean percentage of ADA staked vs. the whole? for the network to be protected against a Sybil attack?
Cardano targets the number of pools dividing up all the ADA equally. A Sybil attack is having more to gain in hurting the network than in helping the network.
All PoS cryptos have a target number, a ratio that establishes how much a pool can take before it is saturated. Cardano has all pools equal so all pools should have millions of pledged ADA to be Sybil resistant, pool operators have more to gain by supporting the network than harming it.
I cannot expect to run a successful pool without being Sybil safe, I do not have millions of ADA to do this.
It is not based on % of ADA staked, it is based on a pools capacity, in Cardano all pools are equal in this.
i’m still trying to piece together where the problem arises.
even if you yourself are not staking “millions” of ADA as your own pool operator, the network itself, by those others doing so (all pools being equal), they (and you) would still protect the network,… no?
those smaller pools will ‘fill up’ as the larger pools start off saturated, regardless of the owners stake. isn’t this the intent?
as for the sybil resistance, we also need to recognize that the shelley release will be in phases, and not a simple “flip of a switch” and some sort of “land rush” to fill up the 1000 pool slots. it will be a phased transfer from a federated network, to a decentralized one, over time.
This is not a problem, it is how Cardano is meant to operate. I did not understand Sybil in Cardano at first, I do now.
The majority of ADA will have to find its way naturally into pools with a lot of pledge, they being the top pools. I do not see pools with few ADA making it into the list as has been pointed out to me, the people with the most at stake should be the ones protecting the system.
Cardano is an enterprise crypto, pool operators will also have to be enterprise capable.
This is really exciting, not like any other PoS in the charts.
At first I was disappointed that I could not operate a pool, now that I understand better how Cardano works I’m happy that I will not be running one. 
i think running pools will be an effort that will pay off for some, and be a hobby for others.
one could run the numbers, once we have a clear idea of total staked ADA. a pool operator could hope to ‘skim’ a fee of x% of 1/1000th of the total staked ADA, if all 1000 pools reach a level of saturation and grow organically up wards from there. ie, all the staked ADA is staked in beneficially maximized pools, they would sit at 1/1000th of the total staked, no?
would it be more beneficial to be a ‘small’ stake holding operator of a large pool, with a fair fee. … or better yet, many such pools, with such a fee. … then the operator could most efficiently “double-dip” on their ADA holdings, getting both their full fair/optimal stake reward (let’s say 3%/yr), but also maximize the pool operator skim (which could net 0.1%-2% of 1/1000th of the total staked ADA’s over time, regularly) by operating efficient pools until near saturation, then seeding another model of that pool and doing it again, letting the previous pool churn along reaping % in fees minus costs, while the ‘next pool’ climbs up the ladder. (or perhaps even model different pools for different user bases and groups of ADA holders, seeding those and maintaining them.)
if one is the majority/only holder of their own large already saturated pool,… all they will be getting is their own part of the total stake they deserve, they will get no operator ‘fee’, as the will be their only ‘customer’!
I think the point is, that it will be hard once the whales that decide to pledge their ADA to pools have finished saturating pools (e.g. IOHK said they will probably have about 80 pools) for small operators to differentiate themselves to attract people to delegate to them based on the remaining number of possible pools (i.e. 1000 - number of whale operated pools).
If for example we say that whale pools saturate half of the total number (e.g. 500/1000), if there are more than 500 small operators, not everyone will be able to operate a viable pool. Sure there will be some jostling for a while (in our example maybe 750 operators may try to fill the remaining 500), but eventually some will drop out until an equilibrium is established. Over time you may find that k is usually plus or minus 5% and never exact, as pools come and go. At the end of the day it is a protocol designed to to be secured based on those with wealth being interested in protecting it. I am not sure how any PoS system can be secured without this as an underlying condition. It is also why I posed the question about what would attract people to delegate in this thread Do you plan to run a stake pool in 2019? [Anonymous Survey] - #5 by phil.lewis.
I think that for those who have been following the Cardano project and are passionate about supporting it due to how it differentiates itself from other blockchain projects, it is exciting to think that it is getting to a point where they can start to participate in a way other than just holding ADA.
I also think that when there is talk about how a stake pool can be run on something as cheap as a Rock Pi it gives people the impression that it is within reach of anyone. While that is technically true, it isn’t a practical reality, because while some of the technical limitations may be removed (i.e. you don’t need expensive hardware like ASIC miners), the protocol design makes it more difficult for some than others.
I think this comes back to some of the messaging that comes from the project teams and maybe assumptions being made about the level of knowledge that the audience has. There have been people in the “Cardano Stake Pool Best Practices Workgroup” Telegram channel that have been asking how they can participate as a pool operator, after saying they have no experience in implementing and operating IT systems and applications. Sure we could provide documentation to follow the bouncing ball and get a stake pool running, but as I said in my previous post
what would those people do if something went wrong with their node and didn’t know how to diagnose the problem? I appreciate that the protocol is designed to allow for node failures, but ultimately we need the majority of the network to be operated professionally.
Why use science and formal methods to build a professional product if it isn’t going to be operated professionally. It would be like Boeing using science and formal methods to build their planes and then letting learner pilots fly them.
1 Like
Thank you for your thoughts, I for one am looking forward to delegating my ADA to the most desirable pool that appears in the list. Cardano is doing the pool screening work for all of us, amazing when you work through it.
1 Like
"what would those people do if something went wrong with their node and didn’t know how to diagnose the problem?"
their node/pool would not be able to claim the reward if any of the addresses in their pool were chosen as the slot leader while that node/pool was down. … such metrics like % uptime, and what/where the server/node is running will be selling points as well.
some will want the security of an AWS instanced pool for that ability to never miss their chance if/when chosen, while others might understand that 99.1% uptime on a home-baked Rock Pi running on solar panels and batteries, would appeal to those seeking a more ‘decentralized’ alternative.