Author Topic: RDPOS - Recommended Delegated Proof of Stake  (Read 18918 times)

0 Members and 1 Guest are viewing this topic.

Offline Pheonike

They would have to be options that the delegate is also given a chance to select and broadcast in as well.l

Offline liondani

  • Hero Member
  • *****
  • Posts: 3737
  • Inch by inch, play by play
    • View Profile
    • My detailed info
  • BitShares: liondani
  • GitHub: liondani

Instead of having pre-made slates by the delegates, why can't we make our own slates. I would like a system where I can can pick and choose from a select group of categories and have my client randomly pick delegates who fit that criteria on a time cycle bases. For example I should be able to have ten votes that I want use with following criteria

1) burn rate % range
2) charity %
3) Investment %
4) Time cycle in days
5) Missed blocks

and some other criteria. Of course I can manually change the votes at any time I want.


how would the system identify 2) & 3) ?

Offline Pheonike


Instead of having pre-made slates by the delegates, why can't we make our own slates. I would like a system where I can can pick and choose from a select group of categories and have my client randomly pick delegates who fit that criteria on a time cycle bases. For example I should be able to have ten votes that I want use with following criteria

1) burn rate % range
2) charity %
3) Investment %
4) Time cycle in days
5) Missed blocks

and some other criteria. Of course I can manually change the votes at any time I want.


Offline gamey

  • Hero Member
  • *****
  • Posts: 2253
    • View Profile
If I vote for you 'Agent86 delegate' & you have 'Gamey delegate' on your slate & Gamey turns out to be bad so you remove him from your slate, because I voted for you doesn't it mean that my shares also stop voting for Gamey? If so that is good. Or have I misunderstood?

Edit: oh ok if it is like you're saying I have misunderstood
No your shares wouldn't stop voting for Gamey in that scenario.  It wouldn't be good to implement it the way you are thinking.

Ok what do you think of this. Having a table of selection criteria that you can select from and then the system votes for the 101 delegates that most fit those criteria?

If you run a full client that isn't "thin" then my understanding is you could have a slate created solely on empirical metrics.  I think there would also be some advanced option/fork that lets you weigh slates.  Having 2 orthogonal (lol college) slates seems like it would cut down bs candidates.  There would still be a non-deterministic element in the wallet to suggest the winner of ties etc.  I dunno.. neat stuff. 

Lol @ me being example bad guy.  How'd people figure it out ?!
I speak for myself and only myself.

Offline Empirical1

  • Hero Member
  • *****
  • Posts: 884
    • View Profile
If I vote for you 'Agent86 delegate' & you have 'Gamey delegate' on your slate & Gamey turns out to be bad so you remove him from your slate, because I voted for you doesn't it mean that my shares also stop voting for Gamey? If so that is good. Or have I misunderstood?

Edit: oh ok if it is like you're saying I have misunderstood
No your shares wouldn't stop voting for Gamey in that scenario.  It wouldn't be good to implement it the way you are thinking.

Ok what do you think of this. Having a table of selection criteria that you can select from and then the system votes for the 101 delegates that most fit those criteria?

Edit: What I mean is, it seems that RDPOS is just trying to simplify the process of electing a good group of delegates. But obviously this introduces a whole group of new potentially bad incentives into the system. If you could select criteria (performance and other) on a checklist and then have the system select delegates. It seems that would be a solution to simplifying selection of delegates you like without having to deal with the potential issues of RDPOS as it is now.
« Last Edit: July 25, 2014, 04:51:50 am by Empirical1 »

Offline Agent86

  • Sr. Member
  • ****
  • Posts: 471
  • BTSX: agent86
    • View Profile
If I vote for you 'Agent86 delegate' & you have 'Gamey delegate' on your slate & Gamey turns out to be bad so you remove him from your slate, because I voted for you doesn't it mean that my shares also stop voting for Gamey? If so that is good. Or have I misunderstood?

Edit: oh ok if it is like you're saying I have misunderstood
No your shares wouldn't stop voting for Gamey in that scenario.  It wouldn't be good to implement it the way you are thinking.

Offline Empirical1

  • Hero Member
  • *****
  • Posts: 884
    • View Profile
Ok I can see the need for something like this now that I'm thinking of voting myself.

I like to keep a portion of my stuff in cold storage & I don't want to have to access my funds just to change my vote every few weeks. Voting for a few people I really trust that maintain good slates passes the burden to my favourite delegates to actively police the system and as a result I'd only have to get involved in an ermegency. (Also it's unlikely an avg. shareholder can be expected to form an opinion on more than +-15 candidates I would guess.)

Though I agree with Gamey that from a marketing perspective it 'feels' like a negative. (Also enjoying reading your thoughts Gamey on how it could change the incentives.)

Personally I'd like to be able to actively change votes of BTSX I have in cold storage but it doesn't seem like that would be technically possible.
Empirical, you are not understanding the proposal.  You still can't change your votes without making a transaction.  You vote for a particular slate but if the person whose slate you chose changes their preferred slate, the slate you voted for does not automatically update.  You would have to vote again.

If the proposal allowed you to just defer your voting to someone else and let them vote on your behalf it would be an even worse idea than it is already.

If I vote for you 'Agent86 delegate' & you have 'Gamey delegate' on your slate & Gamey turns out to be bad so you remove him from your slate, because I voted for you doesn't it mean that my shares also stop voting for Gamey? If so that is good. Or have I misunderstood?

Edit: oh ok if it is like you're saying I have misunderstood
« Last Edit: July 25, 2014, 04:16:09 am by Empirical1 »

Offline Agent86

  • Sr. Member
  • ****
  • Posts: 471
  • BTSX: agent86
    • View Profile
Ok I can see the need for something like this now that I'm thinking of voting myself.

I like to keep a portion of my stuff in cold storage & I don't want to have to access my funds just to change my vote every few weeks. Voting for a few people I really trust that maintain good slates passes the burden to my favourite delegates to actively police the system and as a result I'd only have to get involved in an ermegency. (Also it's unlikely an avg. shareholder can be expected to form an opinion on more than +-15 candidates I would guess.)

Though I agree with Gamey that from a marketing perspective it 'feels' like a negative. (Also enjoying reading your thoughts Gamey on how it could change the incentives.)

Personally I'd like to be able to actively change votes of BTSX I have in cold storage but it doesn't seem like that would be technically possible.
Empirical, you are not understanding the proposal.  You still can't change your votes without making a transaction.  You vote for a particular slate but if the person whose slate you chose changes their preferred slate, the slate you voted for does not automatically update.  You would have to vote again.

If the proposal allowed you to just defer your voting to someone else and let them vote on your behalf it would be an even worse idea than it is already.

Edit: sorry I don't mean to be short.  I think I'm just feeling frustrated and wish I had a little more control over some of these decisions.
« Last Edit: July 25, 2014, 04:18:18 am by Agent86 »

Offline Empirical1

  • Hero Member
  • *****
  • Posts: 884
    • View Profile
Ok I can see the need for something like this now that I'm thinking of voting myself.

I like to keep a portion of my stuff in cold storage & I don't want to have to access my funds just to change my vote every few weeks. Voting for a few people I really trust that maintain good slates passes the burden to my favourite delegates to actively police the system and as a result I'd only have to get involved in an ermegency. (Also it's unlikely an avg. shareholder can be expected to form an opinion on more than +-15 candidates I would guess.)

Though I agree with Gamey that from a marketing perspective it 'feels' like a negative. (Also enjoying reading your thoughts Gamey on how it could change the incentives.)

Personally I'd like to be able to actively change votes of BTSX I have in cold storage but it doesn't seem like that would be technically possible.



Offline liondani

  • Hero Member
  • *****
  • Posts: 3737
  • Inch by inch, play by play
    • View Profile
    • My detailed info
  • BitShares: liondani
  • GitHub: liondani
(RDPOS, Recommend DPOS)

We shouldn't name  "Recommend Delegate Proof Of Stake" : RDPOS to hard to remember and to many consonants, less vowels ...

I recommend to pick one of them:  ReDPOSReD.POSREDPOSRED.POS
from "Recommend Delegate Proof Of Stake"




PS It looks perfect next to your red avatar bytemaster!

PS2 RED       
1. In China and many other Asian cultures, it is the color of happiness.
2.Since the French Revolution, the red flag has been the symbolic color of revolution
3.In the 20th century, red was the color of Revolution; it was the color of the Bolshevik Revolution in 1917 and of the Chinese Revolution of 1949, and later of the Cultural Revolution
4.Red is the color that most attracts attention.

http://en.wikipedia.org/wiki/Red

RED=REVOLUTION=ReDPOS="Recommend Delegate Proof Of Stake"
« Last Edit: July 24, 2014, 10:45:20 pm by liondani »

Offline gamey

  • Hero Member
  • *****
  • Posts: 2253
    • View Profile

Another problem that I don't think anyone has brought up, is you can have infinite slates that are roughly equivalent.  In fact, if I was clever and wanted to win, I'd make up 10 different slates with different marketing angles.  The union of those slates would be all me all day.

There is an example of gaming in its purest form.  Now how do we do enough due diligence to make sure such a thing doesn't happen?

There are assumptions in here that this is similar to meatspace situations where there is a cost to creating a slate has a cost.

One slate per account... user interface doesn't change much from what we have today.  You follow people not slates.  You would have to convince people to use the recommendations of all your sock puppets.   In reality what would happen is people would simply follow their friends and well known people in the industry.

I didn't read the OP close enough, so now I am understanding it a bit better...

I agree about people typically following friends or people in the industry, but you can still game things and create sock puppet type delegates and completely detach your real motivations.  That is the nature of the crypto world.  Having combinations of slates available and encouraged is absolutely key.

By making combination of slates readily accessible you dilute the power of individual slates enough to decentralize. There will be a lot of negativity involved with this system and things we do not wish to see or be involved with. 

The main problem is not simply how do you get people to vote, but how do you get them to toggle on 101 people in a meaningful manner.  Sigh, I guess  this is probably the best way to enforce network security even if I hate all the negative crap that seems inevitable.  good job.
« Last Edit: July 24, 2014, 09:47:14 pm by gamey »
I speak for myself and only myself.

Offline bytemaster


Another problem that I don't think anyone has brought up, is you can have infinite slates that are roughly equivalent.  In fact, if I was clever and wanted to win, I'd make up 10 different slates with different marketing angles.  The union of those slates would be all me all day.

There is an example of gaming in its purest form.  Now how do we do enough due diligence to make sure such a thing doesn't happen?

There are assumptions in here that this is similar to meatspace situations where there is a cost to creating a slate has a cost.

One slate per account... user interface doesn't change much from what we have today.  You follow people not slates.  You would have to convince people to use the recommendations of all your sock puppets.   In reality what would happen is people would simply follow their friends and well known people in the industry. 
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 gamey

  • Hero Member
  • *****
  • Posts: 2253
    • View Profile

Another problem that I don't think anyone has brought up, is you can have infinite slates that are roughly equivalent.  In fact, if I was clever and wanted to win, I'd make up 10 different slates with different marketing angles.  The union of those slates would be all me all day.

There is an example of gaming in its purest form.  Now how do we do enough due diligence to make sure such a thing doesn't happen?

There are assumptions in here that this is similar to meatspace situations where there is a cost to creating a slate has a cost.
I speak for myself and only myself.

Offline santaclause102

  • Hero Member
  • *****
  • Posts: 2486
    • View Profile
Delegate pay is relevant because if it is excessive then voters have to balance voting for delegates they trust to perform the job of delegate (easy to do) vs. keeping track of how all the delegates spend money (hard to do)
That is a good argument. But a seperate issue from the OP suggestion because the due diligence work didn't (overall) become (much) less. 

Apart from the RDPOS vs DPOS question:
Pro seperating workers and delegates: It is easier for people to make a decision because they dont have to weigh the two different qualities.
Contra seperation: Shareholders then have to vote on two issues. 

Agent, have a look at https://bitsharestalk.org/index.php?topic=5868.msg82124#msg82124
+
Quote
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.

Offline Empirical1

  • Hero Member
  • *****
  • Posts: 884
    • View Profile
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).
I'm not too concerned about that. I know miners vote for Ghash.io for example, even to the point it threatens Bitcoin, but now that the people with the power are the shareholders, I don't think we'll see the same results.

Anyone want to walk me through this? I don't understand it, le why

Miners are more like employees & shareholders are owners.

You can tempt employees with better incentives & wages and they will have a limited concern with how it effects the underlying company. Whereas you'd have to provide really amazing incentives to tempt shareholders to accept deals that threaten their larger underlying investment, so much so that it I imagine in theory, it should essentially work out the same cost for an attacker as just buying the shares. 

Also with DPOS there's no advantage to big pools or centralisation of delegates, with mining it provides more consistency (finding blocks more frequently) so pools get get bigger without even needing to offer amazing incentives etc.
Edit: Oops centralisation of delegates would make the system cheaper to run so there is an advantage.
« Last Edit: July 24, 2014, 06:57:53 pm by Empirical1 »