Author Topic: RDPOS - Recommended Delegated Proof of Stake  (Read 10922 times)

0 Members and 1 Guest are viewing this topic.

Offline gamey

  • Hero Member
  • *****
  • Posts: 2253
    • View Profile
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #30 on: July 24, 2014, 03:12:24 am »
This is just great!!  +5%

The big structural benefit I see is that it makes use of the social net the delegates now have to form.
The discussion has shown that the byzantine's general's problem is not completely solvable. Economies of scale and uneven distribution of stake lead to centralized systems. The general has to trust the few miners/frogers/delegates even if they are (well) incentivized to be honest.
What is added with your suggestion to the overall system is "social control" if you will: Delegates have to know each other. And the pressure to be honest is much stronger than before, because all delegates have to know each other and are not good delegates if they miss judge.
In other words, further specialization: It takes the judging work away from the shareholder and gives it to specialized delegates. Delegates specialized on this can make far greater due diligence than shareholders ever could! They can get to know the delegates in person etc...


Social net ?  Why is that good ? The basic problem here is there is this assumption that delegates have the network's long term interest at heart.  Which is somewhat true, but maximizing the number of fellow friendly delegates is going to outweigh that.  Much like miners in POW, there is no reason delegates necessarily care about the network's long term health.  We assume they will, but there is no requirement to have a large stake to be a delegate.  Delegates will first and foremost be profiteers (nothing wrong with that) and their slate selection will most likely reflect this.

There will be lots of backdoor dealing and crap that will leave a bad taste.   

Are politics better with political parties ?  Maybe they are....

As far as the technical reasons, I won't begin to comment as I'm sure BM+team have it figured out so I can't weigh that in.

If BM goes with this though, do not rename it to RDPOS...  Just leave it as DPOS and consider this a feature.

It also means that people will be required to become part of slates and deal with all that, instead of giving that time/effort directly back to improving the network.  More time will be spent glad handing in PMs etc. 

edit - ok ok ok.. I popped off a bit before reading all the responses.  I see that the slates are additive etc.  This will greatly mitigate my concerns.  I would like to see it implemented in a UI.  If you can add slates to each other, then that goes a long ways.  I mean, it is kinda obvious it should be like that, but ... one never knows.

edit-2
Quote
Interesting comment! Well worth thinking about. Summary:
- delegates can provide self serving slates. I think that would be easily obvious. And if he creates a lot of names that are not obviously his name but operated by him. All those strangers would not be very trustworty anyway.
- "Such and such was removed off slate GOOD GUYS for petty reason #4923" <- needs un-emotional discussion style, true!
- "gamification" (ways to game the system) - in which way?

It won't necessarily be obvious they are self-serving.  If we can't tell who is self-serving individually, then how will you tell that the least trustworthy 50% of a slate aren't just put in there for reasons unbeknownst to us ?

It isn't so much the discussion style, as just how parties/cliques play out.  Mud slinging, negative campaigning, etc.

Gamification - The more options you give a system, the more gaming will be done.  Now we have this extra layer between the voter and the delegate.  I could probably sit here and come up with a dozen scenarios that previously could not have happened.  Granted, a lot would not be the most plausible, but I'm sure I could come up with some decent ones.  Look at how the electoral college screws things up in the US !

Edit 4 - If you do allow readily additive slates, then I think slates  will be sufficiently randomized that most blockchain space savings will be lost.  So I'm not sure what to think there..  I assume there would also be negative slates ?  Oh my.
« Last Edit: July 24, 2014, 03:26:09 am by gamey »
I speak for myself and only myself.

Offline Agent86

  • Sr. Member
  • ****
  • Posts: 471
  • BTSX: agent86
    • View Profile
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #31 on: July 24, 2014, 03:33:27 am »
I think not knowing who to vote for is not the reason for low participation; I suspect it's people haven't even got the wallet yet or imported private keys.   People with a huge stake are not attacking the chain.

With RDPOS I can already see the horsetrading between delegates:  You add me to your slate and I'll add you to mine, I'll only add you to my slate if you help me fund my "pet project" etc... 

Anyone is free to make recommendations or publish a list but encouraging people to vote blindly for a list they haven't looked at doesn't help.  There are other ways of making the process easy.

Offline bytemaster

Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #32 on: July 24, 2014, 03:35:34 am »
I think not knowing who to vote for is not the reason for low participation; I suspect it's people haven't even got the wallet yet or imported private keys.   People with a huge stake are not attacking the chain.

With RDPOS I can already see the horsetrading between delegates:  You add me to your slate and I'll add you to mine, I'll only add you to my slate if you help me fund my "pet project" etc... 

Anyone is free to make recommendations or publish a list but encouraging people to vote blindly for a list they haven't looked at doesn't help.  There are other ways of making the process easy.

Think of it more as following twitter feeds... and all users publish slates, not just delegate.s
For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

Offline toast

  • Hero Member
  • *****
  • Posts: 4001
    • View Profile
  • BitShares: nikolai
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #33 on: July 24, 2014, 03:38:26 am »
It is inevitable because it requires no changes to the blockchain but I agree turning it on by default might be premature

Sent from my SCH-I535 using Tapatalk

Do not use this post as information for making any important decisions. The only agreements I ever make are informal and non-binding. Take the same precautions as when dealing with a compromised account, scammer, sockpuppet, etc.

Offline gamey

  • Hero Member
  • *****
  • Posts: 2253
    • View Profile
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #34 on: July 24, 2014, 03:41:49 am »
You're also going to have negative campaigning now.  Previously having a negative campaign against person would have limited benefits, so it wasn't likely that we would see much of it.  Now with a slate, it all at once becomes a tactic that has a lot more strength behind it and so I would expect that we see it once delegate slots become competitive.

For example-
My group of 10 guys, only 7 are making it as delegates.  Well...  maybe being a halfway decent tactician, I decide it makes more sense to have those 3 guys negatively campaign against the delegates sitting on the edge approval.

This is just one example...
I speak for myself and only myself.

bitbro

  • Guest
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #35 on: July 24, 2014, 03:42:21 am »
Delegates are our politicians; slates are our political parties........... ...


Sent from my iPhone using Tapatalk

Offline bytemaster

Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #36 on: July 24, 2014, 03:46:04 am »
Delegates are our politicians; slates are our political parties........... ...


Sent from my iPhone using Tapatalk

Unfortunately so...  the primary thing I want to have everyone consider is this:

1) Whether we support a particular system (parties) or not... they will form
2) If we make it hard to join/use/combine parties by attempt to "stomp it out" then the organized minority will vastly outmaneuver the disorganized majority.

If you cannot beat it (political parties) then do as much as possible to embrace and extend the concepts in a controlled manner.   

People want to avoid thinking and want to trust others.  They will seek this out or abstain... abstaining is worse than allowing them to at-least defer to someone they trust easily.
For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

bitbro

  • Guest
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #37 on: July 24, 2014, 03:54:40 am »
One slate will dominate in the long run, no? And won't this provide an impetus for others to hard fork and launch a new company?


Sent from my iPhone using Tapatalk

Offline maqifrnswa

  • Hero Member
  • *****
  • Posts: 661
    • View Profile
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #38 on: July 24, 2014, 04:07:12 am »
One slate will dominate in the long run, no? And won't this provide an impetus for others to hard fork and launch a new company?


Sent from my iPhone using Tapatalk

that's a good point

I guess it'll be like the board of directors of a corporation. The shareholders elect the board of directors, but the board of directors recommends their slate to the shareholders. Almost always, the shareholders go along with the slate - but every once and a while they hold a revolt.

I think a hard fork to launch a new company won't be as successful as putting together enough votes to get elected. Because if you can't get enough votes to get elected, how do you think you can build enough support behind your new company? but you're right, I don't know if it will consolidate into a oligarchy or not.

I guess it'll be more like the web of trust. well connected individuals will be on the most slates and have the highest chance of being elected while high-value loners won't be able to push their own through as easily.
maintains an Ubuntu PPA: https://launchpad.net/~showard314/+archive/ubuntu/bitshares [15% delegate] wallet_account_set_approval maqifrnswa true [50% delegate] wallet_account_set_approval delegate1.maqifrnswa true

Offline Agent86

  • Sr. Member
  • ****
  • Posts: 471
  • BTSX: agent86
    • View Profile
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #39 on: July 24, 2014, 05:00:59 am »
I don't think "political parties" are at all inevitable with pure approval voting; I don't see the incentives for them.

Voting for delegates should be easy because delegates have a VERY simple and verifiable job.  It's not rocket science; vote for a bunch of people who come across as trustworthy and if you pick a bad one it's no big deal as it will soon be apparent.

We are complicating things because we are overpaying delegates.  There is no reason to give fees for things like registering an asset to the delegate that processed that block.  Delegates should be paid a fair amount for the service provided.

Offline bytemaster

Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #40 on: July 24, 2014, 05:05:09 am »
I don't think "political parties" are at all inevitable with pure approval voting; I don't see the incentives for them.

Voting for delegates should be easy because delegates have a VERY simple and verifiable job.  It's not rocket science; vote for a bunch of people who come across as trustworthy and if you pick a bad one it's no big deal as it will soon be apparent.

We are complicating things because we are overpaying delegates.  There is no reason to give fees for things like registering an asset to the delegate that processed that block.  Delegates should be paid a fair amount for the service provided.

You are not thinking 4th dimensionally....

For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

Online Shentist

  • Board Moderator
  • Hero Member
  • *****
  • Posts: 1601
    • View Profile
    • metaexchange
  • BitShares: shentist
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #41 on: July 24, 2014, 05:31:46 am »
1. at the moment i choose some delegates with little fees, but honestly i don't know i 5% or 100% is required to operate the delegate. maybe you can add how much fees this delegate will earn with his fee structure in 1 month or so. then maybe everyone could see if 100% or 5% is required or are 100% delegates just greedy.

2. why not force the network user to vote?

"at the moment you picked zero delegates. please be aware that to pick  trustworthy delegates are an importent job to secure the network. your transaction will be delate for x time as long as you are not voting"

something like that?

Offline liondani

  • Hero Member
  • *****
  • Posts: 3719
  • Inch by inch, play by play
    • View Profile
    • My detailed info
  • BitShares: liondani
  • GitHub: liondani
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #42 on: July 24, 2014, 07:02:48 am »
What happened to the idea of having automatic votes based on past performance as default?

Light weight nodes have no means of independently verifying past performance and even evil nodes can appear to be performing well.

Could this proposal be included with yours in parallel?

You said that one of the biggest challenge is to make voting easy for the average user , and you are absolutely right...  I will add to that,  that a big challenge also is to make the average user  to vote in the first place!  Of course with your suggestion the percentage of voters will increase dramatically, but always there will exist a percentage of users that will never vote. So my proposal is this:

As long as the user has not voted yet, their client should by default automatic vote the most popular slate id* of delegates or at least a default slate id that is focused more on  network security than on any other aspect (charity, marketing, etc.) until of course they pick up manual  one by them self!


* or better, to pick up randomly a slate id that is on top x%  of the slate popularity list
   so for example if x=30 and the differend slate ids are 1000 we know that all users that didn't' vote
   have picked up by default one of the best first 300 slate ids
« Last Edit: July 24, 2014, 09:21:56 am by liondani »

Offline santaclause102

  • Hero Member
  • *****
  • Posts: 2486
    • View Profile
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #43 on: July 24, 2014, 01:20:32 pm »
Quote
Gamification - The more options you give a system, the more gaming will be done.  Now we have this extra layer between the voter and the delegate.  I could probably sit here and come up with a dozen scenarios that previously could not have happened.  Granted, a lot would not be the most plausible, but I'm sure I could come up with some decent ones.  Look at how the electoral college screws things up in the US !

Dont think that the electoral college is a good analogy. The analogy I thought of is shareholders = shareholders, slate selectors = directors, delegates = executives. Differences: Shareholders do not directly select executives which is possible with RDPOS. With RDPS it is not a yes or no decision for the delegates/executives. With RDPOS there are many slates you can select or you dont have to select a slate at all and vote for 101 delegates (only at 101 approved delegates you vote has the full effect) of your own choice or just a few you know personally. Then all the different votes are proportionally weighted as opposed to the electoral college where a relative majority approves a canditate. The executives/directors analogy is also not perfect as shareholders here only vote with yes or no on one proposed slate of directors and dont vote at all for the executives.
With RDPOS there is also no http://en.wikipedia.org/wiki/Interlocking_directorate

What ways to game the system could you come up with?

Delegates are our politicians; slates are our political parties........... ...

Sent from my iPhone using Tapatalk
Unfortunately so...  the primary thing I want to have everyone consider is this:
1) Whether we support a particular system (parties) or not... they will form
2) If we make it hard to join/use/combine parties by attempt to "stomp it out" then the organized minority will vastly outmaneuver the disorganized majority.
If you cannot beat it (political parties) then do as much as possible to embrace and extend the concepts in a controlled manner.   
People want to avoid thinking and want to trust others.  They will seek this out or abstain... abstaining is worse than allowing them to at-least defer to someone they trust easily.
+5% absolutely realistic. People are minimizing effort and maximizing benefit /effect. Maximizing their positive influence on the system as they are interested in it's success as they are shareholders. And minimizing effort by trusting others, delegating effort, appreciating specialization.

Quote
I guess it'll be like the board of directors of a corporation. The shareholders elect the board of directors, but the board of directors recommends their slate to the shareholders. Almost always, the shareholders go along with the slate - but every once and a while they hold a revolt.
As far as I remember executives suggest the director slate? Could be wrong though.

Quote
I don't think "political parties" are at all inevitable with pure approval voting; I don't see the incentives for them.

Voting for delegates should be easy because delegates have a VERY simple and verifiable job.  It's not rocket science; vote for a bunch of people who come across as trustworthy and if you pick a bad one it's no big deal as it will soon be apparent.
The problem is that unless you approve 101 delegates you have not used your full voting power. The shareholder has to decide whether it is better to vote only for those he knows are good delegates or whether he should vote for a slate that has been suggested by a trusted delegate. The latter might make sense because a bad player can only be voted out effectively if everyone uses his full voting power.

Quote
We are complicating things because we are overpaying delegates.  There is no reason to give fees for things like registering an asset to the delegate that processed that block.  Delegates should be paid a fair amount for the service provided.

What does it have to do with the delegate pay?

Quote
As long as the user has not voted yet, their client should by default automatic vote the most popular slate id* of delegates or at least a default slate id that is focused more on  network security than on any other aspect (charity, marketing, etc.) until of course they pick up manual  one by them self!
This is just randomly amplifying good or negative voting tendencies.

BM, what about automatically firing delegates? Is that implemented or still planed?

Offline CLains

  • Hero Member
  • *****
  • Posts: 2605
    • View Profile
  • BitShares: clains
Re: RDPOS - Recommended Delegated Proof of Stake
« Reply #44 on: July 24, 2014, 01:35:47 pm »
Basically we got two unintended, inevitable consequences of DPOS:

1. People can delegate the job of voting to a slate.
2. Delegates can pay the individuals who vote for them.

We are now mitigating the risks these schemes pose.

Like Delulu I am most concerned about the Cynical Economy Threat: People will vote for delegates and slates that pay the most for individual votes (not via burn that appreciates everyone, but directly).