Jul 27, 2023 | Voltaire era: Parameter committee intermediate state

Hi all,

Here are the notes from the previous meeting, held on July 27th:

The Parameter Committee (interim state) meets on a tri-weekly basis. It discusses all parameters relating to the Cardano protocol including network, technical and economic parameters, providing technical advice and recommendations on updatable parameter settings.

Updated list of the Members (interim state):

Chairs:
Chair: Kevin Hammond
Vice- chair: Alex Moser
Vice- chair: Vijay Bhuvangiri

Advisory Group Heads:
Network group: Neil Davies
Economic group: Samuel Leathers
Technical group: Markus Gufler

Advisory Group Members:
Network group: Karl Knutson, Matthias Sieber, Marcin Szamotulski
Technical group: Ruslan Dudin, Michael Peyton Jones
Economic group: Sergio Sanchez, Philip Lazos, Giovanni Gargiulo

Other:
Communications: Tommy Kammerer, Nathaniel Acton, Addie Girouard, Matthew Capps
Domain experts: Martin Lang, Andrew Westberg
Observers: WIP
Secretary: Joaquín López

Remit

The purpose of the Parameter Committee is to provide technical advice and recommendations relating to the updatable protocol parameters, taking into account economic, security, network and other considerations, as appropriate.

Agenda and Updates:

Agenda:

  • Proceed with the recommendation for PCP-001.
  • Discuss agreement for a change request and timelines.
  • Communications alignment and follow up (PC setup, scope, goals, etc).
  • Review outstanding requests and discuss potential recommendations for future PCPs.

Updates:

  • The Parameter Committee considered comments and inputs from the community, provided in Cardano Forum, on-chain SPO poll, and other sources.
  • The Parameter Committee recommends to retain targetNumberOfStakePools (“k”) at 500, and to reduce minPoolCost to 170 ADA.
  • Based on the recommendation, the Parameter Committee will produce a formal Cardano Change Request to be considered for approval by the signing authorities. Wednesday, August 23 is the proposed target date for enactment.
  • The Parameter Committee agreed to publish some informative blogposts to support the PCP-001 recommendation.
  • Future PCPs will be considered in subsequent meetings.
7 Likes

What does this report mean?

1.Doesn’t the parameters change immediately just by submitting PCP-001 on August 23rd?
2.If the signing authorities approve this PCP, on what schedule will the parameters be changed?
3.Has there been a community consensus on this proposed change?

4 Likes
  1. if signed by genesis key holders: yes

  2. halving minPoolCost is a one-step change. It is a non mandatory new threshold. Every SPO can freely decide to set new lower fixed costs gradually or immediately

  3. yes. See SPO poll results

1 Like

SPO poll has many problems and I don’t think it deserves a reference for changing parameters.

In the SPO poll, each parameter was selected from two options prepared by the Cardano Foundation.
However, the actual parameter change does not necessarily have to be decided from the extreme options of maintaining the status quo or halving (or doubling).

Changing parameters based on the results of the SPO poll which are determined according to the two extreme choices provided by the Foundation for each parameter, feels like too much intervention by the Foundation.

For example, I prefer to keep the k parameter as it is and gradually lower the minPoolCost.
However, there was no such option for the SPO poll, so I chose None.
Among other SPOs, I think there are a fair number of people who have individual, specific ideas, but have made compromises because of the extremely limited number of options.

It is difficult to say that the format of the SPO poll adequately reflects the will of the community.

And there were missing things to explain when conducting the SPO poll.
・Explanation that it will be treated as a clear basis for parameter changes
・Explanation of when the parameter change will be implemented
・Explanation on what criteria to judge, such as whether to judge by stake value or simply by the number of votes
etc

I am not going to dispute the authority that the Parameter Committee has, and that the three organizations have the final say. (at least on this occasion)
But I think there is a problem with thinking that you have the consensus of the community.

3 Likes

interesting views…

actually this is a significant difference between k and minPoolCost.

  • significantly (eg doubling) increasing k is enforcing all Pools into a new situation
  • significantly (eg halving) lowering minPoolCost is not enforcing any pool owner to follow this new lower threshould. It is a free choice for every individual operator to keep, gradually or instantly lower.

If I understand you right, you don’t like the CF-influenced options, but want another option who does not allow other Pool owners to immediately lower their mPC to halve of now? right?

Wrong! please read the announcements and try to understand to scope and processes between the Poll, the Parameter Committee and the genesis key holders.

please read the poll question till the very last word :wink:

Why do you think judging on the number of votes, or anything else then stake weight is relevant other than having some statistic reports?
Block production and consensus in delegated proof of stake is entirely, fully and only based on cumulated stake weight. If 15000 very small pools decide something against the interest of the top300 largest pools (and all their delegators), why should these 300 accept it and start running software to hardfork into something they don’t want?

PoS is PoS nothing else.
Especially as long as nobody comes up with a viable alternative on how to count votes, wallets or heads in a reliable way.

The results for minPoolCost have a quite clear preference, while the results on k are very ambiguous. Clearly no consensus on that, and much more an open question about participation and majority thresholds.

What is your definition of consensus?

  • majority threshold
  • 50%+1lovelace
  • …of participating or delegated stake

how would you determine consensus in the future on more complex parameter changes with multiple directly related and influenced parameters?
If you poll them individually, how to ensure the voted results fit togeter, make sense and are not resulting in a complete disfunctional state?
If you poll them all together as “parameter-sets”, how to define them to not end up on socials then with people complaining about “my perfect preference was not given, so this is not valid and there is no consensus” ?

1 Like

In politics, proper framing of the question can determine the vote outcome. Put another way: Providing options that will split the community vote, will result in a split vote and no clear consensus.

However, the more I think about this, the less critical I become of how the vote was structured. As much as I would like there to be a clear consensus amongst small pool operators to set minPoolFee to something very low or zero, I just don’t think there is a clear consensus. For example, I often hear small pool operators with total stake of less than 1M arguing for minPoolCost to remain at 340 Ada for them to survive and grow their pool. :confused:

Consequently, I think the best way forward might be to simply have rolling votes over time about whether to halve minPoolCost or double it. And separate votes about doubling or halving K. After a number of such cycles we will end up where the SPO community ultimately agrees.

1 Like

Thank you for your reply.

If I understand you right, you don’t like the CF-influenced options, but want another option who does not allow other Pool owners to immediately lower their mPC to halve of now? right?

My thinking is that lowering mPC may cause low fee competition among pools, and it is unknown what impact this will have, so we should lower it in stages and wait and see carefully.
However, what I want to argue now is that there should have been more options, not my thoughts on the parameters.
For example, the opinions of those who think that “mPC should be 0” could not be reflected in the voting this time. I think the opinion of such people should be reflected in the vote.

Wrong! please read the announcements and try to understand to scope and processes between the Poll, the Parameter Committee and the genesis key holders.

I apologize for any confusion caused by my wording.
What I wanted to say was that there was a lack of explanation that the results of SPOpoll had such a strong impact that it would be a basis for claiming that “the parameter changes recommended by the Parameter Committee have gained community consensus.”
If the explanation was sufficient, I think there would have been more criticism of the voting format beforehand.

please read the poll question till the very last word :wink:

Is it the “from Q3 2023” part?
This was incorrect in my assertion that “explanation was missing”. I am sorry.
(I wish there had been a more specific implementation timeframe specified if possible.)

PoS is PoS nothing else.
Especially as long as nobody comes up with a viable alternative on how to count votes, wallets or heads in a reliable way.

I have no objections to decisions based on PoS.

how would you determine consensus in the future on more complex parameter changes with multiple directly related and influenced parameters?

I think there are various methods, but at least the criteria should be clarified before the vote takes place.
In addition, it is necessary to consider so that individual thoughts are reflected as much as possible, such as preparing more options.

The voting format in this round, which involved selecting each parameter with a two-choice option, does not seem to have taken into consideration reflecting individual thoughts.
In this case, I don’t think there would have been a problem even if there were many voting options, as long as the voting results were used as a reference by the parameter committee.

If the criteria for judging the results of the voting were clarified in advance and it was possible to reflect their intentions in the voting,
I think the results should be respected by everyone.

1 Like

Thanks!

…the most important part in my opinion is, to see

  1. the design was never aimed to make it as perfect as possible.
    Instead the goal was to give a hands on experience on how such surveys and votes could happen in the future, and also some practical triggers from suboptimal properties

  2. the poll happened in May, but the most important learning effects are still happening, especially these days now.

What if someone in the future is
Submitting a poll without any dates?
Or if someone submit a vote for next epoch execution, but some software is not ready?
Wouldn’t it be good to think and get in touch with it, now in 2023, before Voltaire is here?

What if I told you, k and mPC are probably the less important parts of the whole poll?

I just don’t get how you could have missed this for 6 months, when you are so passionate about?

2 Likes

I agree. They were the least important parts of the whole poll. I personally hope that people learned a lot from the process because there are going to be much more important polls in the future.

2 Likes

??? k and mPC were the only two things in the poll. If they were the least important part: What was the important part?

2 Likes

How it all played out:

  • How the community handled things and didn’t handle things.
  • How SPOs understood the implications of any changes.
  • How SPOs communicated their understanding to their delegators.
  • How delegators cared or didn’t care.
  • How SPOs negotiated with each other.
  • How SPOs handled vote options that were not structured to their liking.
  • How the vote outcome has been interpreted.
  • etc.
3 Likes

Thank you for your reply.

I feel the same way and that is why I expressed my opinion.

I had the same thought.
So it was a bit unexpected that the results of the vote would be treated as an important reference for changing the parameters, as was the case this time.

1 Like

I don’t agree. I think one thing that was pretty clear from the poll is that there is majority support for halving minPoolFee. I don’t think we need another poll to prove that.

The problem is that this debate has been going on for several years. SPOs are tired of waiting for something to be done. It is pretty clear to many that the minPoolCost is acting as an anti-competitive regulation which benefits OG incumbents and handicaps newcomers. Arbitrary regulations like this is exactly what blockchain is NOT supposed to be about.

Even if you like the concept of arbitrary regulation, remember that minPoolCost supposedly reflects a real world cost. IE. expenses in fiat currency. When minPoolCost was set at 340, Ada was trading at 5 or 6 US cents. Now Ada is trading at 30 US cents and it was up to 3 dollars. If the minPoolCost is still supposed to reflect a fiat amount then it should be more than halved.

3 Likes

In general I agree with your reactions in this thread.

Just the last paragraph in last comment seems to be simplifying the situation too much. One of the things you should take into account are node HW requirements which at least tripled. An average SPO is running 3-5 nodes (1-2 block producers, 2-4 relays) if not participating in any testnet.

So in my opinion, if you would like to take into account ADA price, you should calculate server hosting primarily in various cloud environments (or find a standard cloud provider - AWS?). Hosting nodes is a fix cost and minPoolCost was in fact meant to cover this.

Now consider that a small SPO, let say 1M delegated ADA, gets a block in roughly 66% of the epochs. There are 6 epochs each month, let say these SPOs mint at least 1 block in 4 epochs. So currently such SPO is earning 1.360 ADA (4 x 340) x 30 US cents is about 400 USD.

How many nodes with at least 30GB RAM can be run for this amount of money? We should also include some manhours for server monitoring and maintenance and of course some time to keep watching protocol/Cardano news. This is just the technical side of thing.

And this is the real world cost you mentioned, but in fact was omitted in that paragraph. The ADA price in USD is just one part of the puzzle.

In the end you will come to an opinion that, when it comes to costs, small SPOs do not generate much profit (if any) throughout the crypto cycle. Of course this is very different during bull runs, but same like for any other crypto holder buying during the winter or accumulation phase or running a DCA strategy.

Sorry for a longer reaction I hope it makes sense as I was little bit interrupted during the writing.

1 Like

@lunarpool.io your arguments absolutely make sense, but you are thinking about things the wrong way.

The simple fact is that a pool with 1M delegated Ada is not competitive. It is not competitive for it’s operator (owner) and it is not competitive for it’s delegators. It unfortunately needs to go out of business according to Ouroboros.

Have a good read of the Ouroboros papers and you will see that the protocol has been deliberately designed to create the incentives that drive the network towards 500 (K) equally sized, saturated pools. There is no mention of this minPoolFee regulation in the Ouroboros design. The evidence I have is that this “feature” was added later because someone thought it a good idea.

The problem I have with this “feature” (bug) is that it is anti-competitive. Newcomers are not on an equal footing with OG incumbents that have already built a collection of delegators. This makes delegators unusually sticky because there is no way for the operator of a small pool to build size without somehow convincing delegators to act against their own economic interests. Basically a small pool has to convince people to be financially stupid. And, I don’t think that is the right way to behave.

This is what I say on my own website because I want to be honest:

I do not recommend you stake with Terminada pool yet because the protocol does not allow setting the fixed fee any lower than 340 Ada.

With a fixed fee of 340 Ada you will lose too much of your rewards to fees until the pool size is over 10 million Ada.

I should be able to compete by lowering my minPoolCost for a period of time so that people can choose my pool without being economically harmed. Then if I am successful in building a large enough pool I can become profitable. Tricking people so they lose rewards in order for me to benefit is not my thing.

In fact, I am aware that some pools manually pay the minPoolCost 340 Ada amount back to delegators because they are so desperate to work around the problem in order to have a chance of honestly joining the over 10M club.

Unfortunately, the way things are currently makes it impossible for a small pool operator to honestly compete with OG incumbents. That situation is not what blockchain is supposed to be about.

5 Likes

Hello, I just wanted to follow up. I was recently told by a member of the Parameter Committee that K=1,000 is still on the table. I’m not sure if this is official communication or official stance of the committee since none of the notes specifically say that K=1,000 will be part of the committee’s future recommendations. If so, can this be added to these notes or the notes for the next meeting?

The notes posted from July 6th say:

" The committee recommended decreasing the minPoolCost parameter to 170 Ada per epoch in line with the results of the SPO poll, and monitoring the effects of changing this parameter. Once the effects on the network and stake pools have been analyzed, further changes to either or both parameters can be considered."

Can you share what will be monitored? What are the metrics being watched? Over what timeline will they be watched or how long must they remain stable? What are the target numbers for these metrics that would trigger the committee to put in a recommendation for K=1,000?

Also, I would like to follow up on this post from July 6th Notes, on a request to add disclosures, which I believe another member of the committee said were in progress, unless I misunderstood them:

Lastly, I was wondering is replying to the notes here the preferred method of communicating with the Parameters Committee? Or is there a different method? Also is there a main contact from the committee that we should be in contact with?

Thank you!

5 Likes
  • Based on the recommendation, the Parameter Committee will produce a formal Cardano Change Request to be considered for approval by the signing authorities. Wednesday, August 23 is the proposed target date for enactment.

It’s 24th August now (at least where I am). Is there any update on this?

4 Likes

Just to update for anyone who happens across this thread in particular. An answer to this question and further has continued in the thread of the latest PC meeting minutes release: Aug 17, 2023 | Voltaire era: Parameter committee intermediate state - #7 by jeremyisme

1 Like

FYI: there is a clarification note in the latest update on https://forum.cardano.org/t/sep-7-2023-voltaire-era-parameter-committee-intermediate-state/122115:

*Clarification note: on the meeting notes from July 27 we said that “The Parameter Committee recommends to retain targetNumberOfStakePools (“k”) at 500, and to reduce minPoolCost to 170 ADA.” We would like to clarify that the Parameter Committee meant to FIRST reduce minPoolCost to 170. THEN after reconsideration look at k again and maybe recommend to increase it. This is in line with the meeting notes from June 6 “The committee recommended decreasing the minPoolCost parameter to 170 Ada in line with the results of the SPO poll, and monitoring the effects of changing this parameter. Once the effects on the network and stake pools have been analyzed, further changes to either or both parameters can be considered.”

1 Like