Bitcoin Forum
November 09, 2024, 01:18:26 PM *
News: Latest Bitcoin Core release: 28.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 2 [3]  All
  Print  
Author Topic: Quark investors - Quark information on cycles and the push to move towards PoS.  (Read 3159 times)
DeathAndTaxes
Donator
Legendary
*
Offline Offline

Activity: 1218
Merit: 1079


Gerald Davis


View Profile
May 31, 2014, 12:27:48 AM
 #41

Quote
PoW coins with low/no block reward have been shown to not work. A recent failure is Coin2, which was under massive attack and had to implement PoS after that. The attacker even posted in the ANN thread and claimed that he earned 20+BTC by 51% attacks. Considering Coin2 only has a very small market cap, 20BTC is a huge number. Quark could be the next Coin2 if we blindly believe Quark is unique and shouldn't be changed in anyways, and don't take any measure to increase network security.

Not the best example.  Coin2 PoS was 51% attacked.
digitalindustry (OP)
Hero Member
*****
Offline Offline

Activity: 798
Merit: 1000


‘Try to be nice’


View Profile WWW
May 31, 2014, 09:28:46 AM
 #42

Quote
PoW coins with low/no block reward have been shown to not work. A recent failure is Coin2, which was under massive attack and had to implement PoS after that. The attacker even posted in the ANN thread and claimed that he earned 20+BTC by 51% attacks. Considering Coin2 only has a very small market cap, 20BTC is a huge number. Quark could be the next Coin2 if we blindly believe Quark is unique and shouldn't be changed in anyways, and don't take any measure to increase network security.

Not the best example.  Coin2 PoS was 51% attacked.

ha ha..

- Twitter @Kolin_Quark
digitalindustry (OP)
Hero Member
*****
Offline Offline

Activity: 798
Merit: 1000


‘Try to be nice’


View Profile WWW
May 31, 2014, 09:31:25 AM
 #43

I understand people have concerns - but we have to do full due diligence and look at all the data -

- Twitter @Kolin_Quark
IPCoinz
Member
**
Offline Offline

Activity: 63
Merit: 10


View Profile
May 31, 2014, 11:12:20 AM
Last edit: May 31, 2014, 11:38:18 AM by IPCoinz
 #44


The problem is fundamental and will stay long-term, so the solution must be fundamental. I think the only viable option is Proof-of-Stake (PoS), the form of mining when a probability of block generation increases dramatically with increase of coins in the wallet of miner. TO day this technology has already proven itself with a several years of operating in probably a hundred of coins. I know no serious issues reported or acknowledged security concerns.

Moreover, PoS is viable economically, as it reward long-term coin supportes and provide income on capital. Also PoS stimulate people to maintain full nodes which favorably affects network integrity and security.

Still, PoS is not ideal because it give a significant share of block and coin generation to coin bags with large wallets. But it could be adjusted to reduce such factor. For example the network could provide even tiny amount of coins generated this way and the process will be still continuing since the costs is almost zero. Or amount of PoS generated coins and/or probability of block finding may be depend nonlinear from wallet. Anyway we could find some intelligent and economically justified solution.



What is needed is an algo which will assess a penalty (a drain) against any wallets holding excessively large stakes in the coin. This algo would constantly evaluate and be aware of wealth distribution on the whole network based upon activity recorded in the block chain (pseudo-anonymously).  When the size or advantage of a wallet, or a quorum of wallets, or a "neighborhood" of wallets, reaches a certain threshold or relational structure, then the algo will trigger the "fission" of the coin balance (a small fraction of it) to occur in the designated penalty wallet(s), while the network makes corresponding microdeposits in another designated set of wallets, the beneficiary wallets.

A Fission event may (A) create new coin and/or (B) deduct existing coin.  A Fission event occurs, for example, when one or more stakes becomes "too big," or an oligarchy of stakes becomes too entrenched.  Fission causes the dispersal of coin from the big stake(s) into and among wallets deemed worthy of being beneficiaries.  That opens a wide door, doesn't it?

Fission may accomplish wealth redistribution, or not.  Its up to the devs and the market to decide which fission regimes are the best.  

My preference is to have fission deduct from top stakes and drizzle the harvest into every other wallet on the network in statistically-defined patterns and portions.  

This "fallout" or dust caused by fission can be distributed randomly in all wallets or apportioned methodically into the appropriate beneficiary wallets.  Different coin developers can compete with different coins defining different ways to determine which "standards" and wallets are "appropriate."  It would open up a whole new generation of PoS coins.  

To sum up: In the proposed "fission" mechanism, whenever the algo determines that a critical mass exists in one or more wallets, or that the time has come to forceably reshape the wealth distribution profile of the network, the stakes in the one or more wallets undergo "coin fission," and each stake blasts a calculated sum of coin out into the community to be received however the devs have arranged it. The dynamics of any coin's fission rules can be customized and finely tuned.

And importantly, stakeholders can reasonably predict the consequences of holding a fissile coin, because the terms and criteria are published; however, stakeholders can never predict the precise timing and exact cost of any one fission event (so they cant avoid it). Thus they must acquiesce to the "stake tax" when investing in the coin or they must go find another coin to parasitize.  Still, in a fissile coin, the largest stakeholders do gain enough steady growth overall to compensate and reward them for remaining dominantly invested in the coin.  

My vision of Fission is that it tickles the big wallets more than it mauls them.

Thus, the above concept is a suggestion for implementing a "coin fiission" algo that might save us from the current doldrums of the "gotto go PoS" days of altcoins, where the imperfect infrastructure of the digital economy is selecting strongly for PoS algos even though we may be better off without much PoS.  

The proposed Fission algo will operate upon wallet balances that reach a "critical mass," or throughout a network when it starts dying for lack of liquidity due to excessively large stakeholdings (e.g., high coin age).  Fission imposes a size-dependent cost upon the largest and laziest stakeholders and it makes offsetting deposits in other wallets according to variable rules and goals.  This will promote liquidity in PoS and will smooth out the rough edges of wealth-concentration patterns that would otherwise characterize a mature PoS network.  


Got it?  Great.  Have a beta version in my inbox by tomorrow...

Your comments and criticisms are most welcome...
IPCoinz
Member
**
Offline Offline

Activity: 63
Merit: 10


View Profile
May 31, 2014, 11:35:14 AM
 #45

And actually, one could include considerations of coin age in the Fission protocol too.  Coin age is a parameter that will pinpoint oligarchs so they can be nuked.
KoinMaster
Sr. Member
****
Offline Offline

Activity: 323
Merit: 250


View Profile
May 31, 2014, 02:05:03 PM
 #46

I understand people have concerns - but we have to do full due diligence and look at all the data -

Kolin, you keep boasting about how great Quark's distribution is. In fact it's really bad. 30-40% of all Quark coins are owned by 3-4 people...

http://bitinfocharts.com/top-100-richest-quarkcoin-addresses.html


 
 
           ▄████▄
         ▄████████▄
       ▄████████████▄
     ▄████████████████▄
    ████████████████████      ▄█▄                 ▄███▄                 ▄███▄                 ▄████████████████▀   ▄██████████

  ▄▄▄▀█████▀▄▄▄▄▀█████▀▄▄▄     ▀██▄             ▄██▀ ▀██▄             ▄██▀ ▀██▄             ▄██▀                   ██
▄█████▄▀▀▀▄██████▄▀▀▀▄█████▄     ▀██▄         ▄██▀     ▀██▄         ▄██▀     ▀██▄         ▄██▀        ▄█▄          ▀██████████████▄
████████████████████████████       ▀██▄     ▄██▀         ▀██▄     ▄██▀         ▀██▄     ▄██▀          ▀█▀                        ██
 ▀████████████████████████▀          ▀██▄ ▄██▀             ▀██▄ ▄██▀     ▄█▄     ▀██▄ ▄██▀                                       ██
   ▀████████████████████▀              ▀███▀                 ▀███▀       ▀█▀       ▀███▀      ▄███████████████████████████████████▀
     ▀████████████████▀
       ▀████████████▀
         ▀████████▀
           ▀████▀
║║


║║
.
.

║║
██
║║
.
.

║║
██
║║
.
║║


║║
digitalindustry (OP)
Hero Member
*****
Offline Offline

Activity: 798
Merit: 1000


‘Try to be nice’


View Profile WWW
May 31, 2014, 02:43:47 PM
 #47

I understand people have concerns - but we have to do full due diligence and look at all the data -

Kolin, you keep boasting about how great Quark's distribution is. In fact it's really bad. 30-40% of all Quark coins are owned by 3-4 people...

http://bitinfocharts.com/top-100-richest-quarkcoin-addresses.html


yeah i do boast about it - because unfortunately you seem like you are new here so you need education likely, so let me help you out:

( i will keep repeating my self i guess)  (i will copy paste this in a notepad so i don't have to keep writing it)

- Quark is fully distributed
 - it has no mining monopoly.
- it can't be price controlled by a monopoly of mining hardware.
 - It can only get more distributed not less  as time moves forward.
- all other currencies can only get more monopolized until full distribution.
- you can't know how to measure how distributed a crypto is because an one person can make 100's of addresses this is basics you need education.

- The best way to measure distribution is to try to do that though a "network effect"
- Quark has a very healthy network effect, a large community and lots of activity  -
- probably about 70 people own most of the BTC that exists in % terms.

and lets not forget.

- people are stupid.

- so the people that accumulated the large wallets and put them in a single address we can reasonable assume they did that for a motive to try to show that Quark is centralized, as all the other Crypto that are monopoly owned by a hand-full of people they (quite naturally) split the units they own up into multiple addresses to fake a healthy distribution.

- but you have only to look and see that there is no activity going on much past this forum on nearly all of these currencies, Bitcoin is the exception because people know that the fiat system looks very shaky  so want a "golden" fall back.

- Quark is the other exception because it has a better economic design than Bitcoin one that encourages activity and distribution, where as Bitcoin has Mining Monopoly inaccessibility and increasingly centralized strict fixed reward - (which i've even seen tards try to deny ??)

even the "cool kids" of crypto "elite" (read semi autistic) agree with the Quark design of the EQ reward.


oh the last one:

^ most people know everything i wrote is common knowledge - if you don't understand this you are probably the person losing money here.

- Twitter @Kolin_Quark
victzhang
Full Member
***
Offline Offline

Activity: 193
Merit: 100


View Profile
May 31, 2014, 03:47:03 PM
 #48

Quote
PoW coins with low/no block reward have been shown to not work. A recent failure is Coin2, which was under massive attack and had to implement PoS after that. The attacker even posted in the ANN thread and claimed that he earned 20+BTC by 51% attacks. Considering Coin2 only has a very small market cap, 20BTC is a huge number. Quark could be the next Coin2 if we blindly believe Quark is unique and shouldn't be changed in anyways, and don't take any measure to increase network security.

Not the best example.  Coin2 PoS was 51% attacked.

ha ha..

That simply shows the fact that implementing PoS and not asking people to open their wallets do not work.
However, it is WAY easier to encourage people to open their wallet than to mine. For Quark I don't think it's a problem because there is a strong community behind it.
victzhang
Full Member
***
Offline Offline

Activity: 193
Merit: 100


View Profile
May 31, 2014, 04:21:48 PM
 #49


The problem is fundamental and will stay long-term, so the solution must be fundamental. I think the only viable option is Proof-of-Stake (PoS), the form of mining when a probability of block generation increases dramatically with increase of coins in the wallet of miner. TO day this technology has already proven itself with a several years of operating in probably a hundred of coins. I know no serious issues reported or acknowledged security concerns.

Moreover, PoS is viable economically, as it reward long-term coin supportes and provide income on capital. Also PoS stimulate people to maintain full nodes which favorably affects network integrity and security.

Still, PoS is not ideal because it give a significant share of block and coin generation to coin bags with large wallets. But it could be adjusted to reduce such factor. For example the network could provide even tiny amount of coins generated this way and the process will be still continuing since the costs is almost zero. Or amount of PoS generated coins and/or probability of block finding may be depend nonlinear from wallet. Anyway we could find some intelligent and economically justified solution.



What is needed is an algo which will assess a penalty (a drain) against any wallets holding excessively large stakes in the coin. This algo would constantly evaluate and be aware of wealth distribution on the whole network based upon activity recorded in the block chain (pseudo-anonymously).  When the size or advantage of a wallet, or a quorum of wallets, or a "neighborhood" of wallets, reaches a certain threshold or relational structure, then the algo will trigger the "fission" of the coin balance (a small fraction of it) to occur in the designated penalty wallet(s), while the network makes corresponding microdeposits in another designated set of wallets, the beneficiary wallets.

A Fission event may (A) create new coin and/or (B) deduct existing coin.  A Fission event occurs, for example, when one or more stakes becomes "too big," or an oligarchy of stakes becomes too entrenched.  Fission causes the dispersal of coin from the big stake(s) into and among wallets deemed worthy of being beneficiaries.  That opens a wide door, doesn't it?

Fission may accomplish wealth redistribution, or not.  Its up to the devs and the market to decide which fission regimes are the best.  

My preference is to have fission deduct from top stakes and drizzle the harvest into every other wallet on the network in statistically-defined patterns and portions.  

This "fallout" or dust caused by fission can be distributed randomly in all wallets or apportioned methodically into the appropriate beneficiary wallets.  Different coin developers can compete with different coins defining different ways to determine which "standards" and wallets are "appropriate."  It would open up a whole new generation of PoS coins.  

To sum up: In the proposed "fission" mechanism, whenever the algo determines that a critical mass exists in one or more wallets, or that the time has come to forceably reshape the wealth distribution profile of the network, the stakes in the one or more wallets undergo "coin fission," and each stake blasts a calculated sum of coin out into the community to be received however the devs have arranged it. The dynamics of any coin's fission rules can be customized and finely tuned.

And importantly, stakeholders can reasonably predict the consequences of holding a fissile coin, because the terms and criteria are published; however, stakeholders can never predict the precise timing and exact cost of any one fission event (so they cant avoid it). Thus they must acquiesce to the "stake tax" when investing in the coin or they must go find another coin to parasitize.  Still, in a fissile coin, the largest stakeholders do gain enough steady growth overall to compensate and reward them for remaining dominantly invested in the coin.  

My vision of Fission is that it tickles the big wallets more than it mauls them.

Thus, the above concept is a suggestion for implementing a "coin fiission" algo that might save us from the current doldrums of the "gotto go PoS" days of altcoins, where the imperfect infrastructure of the digital economy is selecting strongly for PoS algos even though we may be better off without much PoS.  

The proposed Fission algo will operate upon wallet balances that reach a "critical mass," or throughout a network when it starts dying for lack of liquidity due to excessively large stakeholdings (e.g., high coin age).  Fission imposes a size-dependent cost upon the largest and laziest stakeholders and it makes offsetting deposits in other wallets according to variable rules and goals.  This will promote liquidity in PoS and will smooth out the rough edges of wealth-concentration patterns that would otherwise characterize a mature PoS network.  


Got it?  Great.  Have a beta version in my inbox by tomorrow...

Your comments and criticisms are most welcome...

However most top big wallets belong to exchanges... See some good discussions blow:
http://www.reddit.com/r/QuarkCoin/comments/25rt89/52_of_quarks_belong_to_only_25_addresses/
http://www.reddit.com/r/QuarkCoin/comments/24c1nq/the_4_richest_quark_address_turns_out_to_be_the/
BainbridgeJ
Full Member
***
Offline Offline

Activity: 163
Merit: 100


View Profile
July 11, 2014, 10:53:00 PM
 #50

Interesting dilemma!
digitalindustry (OP)
Hero Member
*****
Offline Offline

Activity: 798
Merit: 1000


‘Try to be nice’


View Profile WWW
July 12, 2014, 06:01:15 AM
 #51

Interesting dilemma!

Ive enjoyed the discussion so far.

- Twitter @Kolin_Quark
Pages: « 1 2 [3]  All
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!