Bitcoin Forum
June 17, 2024, 02:30:10 AM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Poll
Question: Which PoS approach will win out?
Peer-coin (Hybrid Pow/PoS mining)
Nxt (Transparent forging)
Bitshares (DPoS)

Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 [13] 14 15 16 17 »  All
  Print  
Author Topic: Which Proof of Stake System is the Most Viable  (Read 25725 times)
ChuckOne
Sr. Member
****
Offline Offline

Activity: 364
Merit: 250

☕ NXT-4BTE-8Y4K-CDS2-6TB82


View Profile
June 10, 2014, 03:36:11 PM
 #241

IMO, the cap means its very different, as I can see the same pooling problem with TF as in PoW. But it seems not many agree with this viewpoint.

So, I ask again: what is this cap good for? Where is the technical difficulty in creating a new pool, when my first pool is capped?

Yeah the cap doesn't mean much IMO, the "downvotes" are the more important part.
The important part is that as a pool operator, you can only pay dividends to all stakeholders and not just to your voters, otherwise you will get voted out. This alleviates economies of scale which lead to centralization.

Interesting. I will think about that.

I still would like to see an answer from sumantso. Smiley
spazzdla
Legendary
*
Offline Offline

Activity: 1722
Merit: 1000


View Profile
June 10, 2014, 08:02:18 PM
 #242

Id say PPC but it's 100% based on the fact I am mining it soo... ya... lol.
David Latapie
Hero Member
*****
Offline Offline

Activity: 658
Merit: 503


Monero Core Team


View Profile WWW
June 14, 2014, 11:26:28 AM
 #243

The poll doens't take into consideration PPC-derived Novacoin's dynamic inflation control (NVCS), which paves the way for very high inflation rate for early adopters (rewarding risk-taking).

Monero: the first crytocurrency to bring bank secrecy and net neutrality to the blockchain.HyperStake: pushing the limits of staking.
Reputation threadFree bitcoins: reviews, hints…: freebitco.in, freedoge.co.in, qoinpro
sumantso
Legendary
*
Offline Offline

Activity: 1050
Merit: 1000



View Profile
June 15, 2014, 10:19:51 AM
 #244

IMO, the cap means its very different, as I can see the same pooling problem with TF as in PoW. But it seems not many agree with this viewpoint.

So, I ask again: what is this cap good for? Where is the technical difficulty in creating a new pool, when my first pool is capped?

Yeah the cap doesn't mean much IMO, the "downvotes" are the more important part.
The important part is that as a pool operator, you can only pay dividends to all stakeholders and not just to your voters, otherwise you will get voted out. This alleviates economies of scale which lead to centralization.

Interesting. I will think about that.

I still would like to see an answer from sumantso. Smiley

Thats because you can in PoW, say, put up two pools and grab control. Here you need to set up some 50 odd delegates. Of course, the developer may decide that its popular and its needs more and then the number of delegates can be increased.

IMO, trying to grab control with 50 puppets instead of 2 is much more difficult.

ChuckOne
Sr. Member
****
Offline Offline

Activity: 364
Merit: 250

☕ NXT-4BTE-8Y4K-CDS2-6TB82


View Profile
June 15, 2014, 10:30:14 AM
 #245

IMO, trying to grab control with 50 puppets instead of 2 is much more difficult.

I know. You said that. But I was asking of the why? Where is the additional difficulty?
clout (OP)
Full Member
***
Offline Offline

Activity: 207
Merit: 100


View Profile
June 15, 2014, 04:26:17 PM
 #246

IMO, trying to grab control with 50 puppets instead of 2 is much more difficult.
I know. You said that. But I was asking of the why? Where is the additional difficulty?

Because people have to vote for the delegates. Also in what situation would anyone profit from a manipulation of 51% of the delegates? Or for that matter in what situation do you believe that anyone would profit from Ghash.io  performing a 51% attack on the bitcoin network?
ChuckOne
Sr. Member
****
Offline Offline

Activity: 364
Merit: 250

☕ NXT-4BTE-8Y4K-CDS2-6TB82


View Profile
June 15, 2014, 05:19:41 PM
Last edit: June 15, 2014, 05:33:11 PM by ChuckOne
 #247

IMO, trying to grab control with 50 puppets instead of 2 is much more difficult.
I know. You said that. But I was asking of the why? Where is the additional difficulty?

Because people have to vote for the delegates.


That makes no sense. Why would this make it more difficult?

Also in what situation would anyone profit from a manipulation of 51% of the delegates? Or for that matter in what situation do you believe that anyone would profit from Ghash.io  performing a 51% attack on the bitcoin network?

I do not understand.
clout (OP)
Full Member
***
Offline Offline

Activity: 207
Merit: 100


View Profile
June 15, 2014, 05:37:50 PM
 #248

IMO, trying to grab control with 50 puppets instead of 2 is much more difficult.
I know. You said that. But I was asking of the why? Where is the additional difficulty?

Because people have to vote for the delegates.


That makes no sense. Why would this make is more difficult?

Also in what situation would anyone profit from a manipulation of 51% of the delegates? Or for that matter in what situation do you believe that anyone would profit from Ghash.io  performing a 51% attack on the bitcoin network?

I do not understand.

In order to have 51% of the delegates under your control you either have to have 51% of the stake or you have to accumulate 51% of the votes from the shareholders. If you can do that how have you manipulated the market? All actions are voluntary and people will only vote for you if you provide the best service as a delegate.

Also people that talk about theoretical attacks do not understand the practical implications of such an attack. There is no action without its cost and consequences. If you perform a 51% attack please explain how you can reasonably profit from it? I have never heard anyone making a practical argument for how a 51% attack is profitable.


ChuckOne
Sr. Member
****
Offline Offline

Activity: 364
Merit: 250

☕ NXT-4BTE-8Y4K-CDS2-6TB82


View Profile
June 15, 2014, 06:11:59 PM
 #249

In order to have 51% of the delegates under your control you either have to have 51% of the stake or you have to accumulate 51% of the votes from the shareholders. If you can do that how have you manipulated the market? All actions are voluntary and people will only vote for you if you provide the best service as a delegate.

The initial and still unanswered question was: why would it be more difficult to gain X% via Y hubs than to gain X% via 1 hub?
clout (OP)
Full Member
***
Offline Offline

Activity: 207
Merit: 100


View Profile
June 15, 2014, 06:36:08 PM
 #250

competition
ChuckOne
Sr. Member
****
Offline Offline

Activity: 364
Merit: 250

☕ NXT-4BTE-8Y4K-CDS2-6TB82


View Profile
June 15, 2014, 06:37:00 PM
 #251

competition

Same goes for 1 hub.
clout (OP)
Full Member
***
Offline Offline

Activity: 207
Merit: 100


View Profile
June 15, 2014, 06:56:35 PM
 #252

Oh I think I get what you are saying, although I am not entirely sure. But the notion of using one delegate was used in early testing however it was problematic from a technical standpoint. Theoretically speaking there is very little difference as long as the single highest voted delegate or multiple highest voted delegates can be voted out easily if they do not perform their task as prescribed by the networks constitution or shareholder consensus. However, more specifically an attempt to attack the network would be carried out in different fashions given the different network structures. If there is one delegate theoretically I believe it is easier to discern an attack on the network and immediately fire the delegate. If there are multiple delegates it is harder to initiate and coordinate an attack as the order that delegates sign blocks is reshuffled after N blocks.

ChuckOne
Sr. Member
****
Offline Offline

Activity: 364
Merit: 250

☕ NXT-4BTE-8Y4K-CDS2-6TB82


View Profile
June 15, 2014, 07:22:05 PM
 #253

Thank you. I think I got what you mean. Smiley
idlecore
Newbie
*
Offline Offline

Activity: 6
Merit: 0


View Profile
June 16, 2014, 02:02:19 AM
 #254

IMO, trying to grab control with 50 puppets instead of 2 is much more difficult.
I know. You said that. But I was asking of the why? Where is the additional difficulty?

Because people have to vote for the delegates. Also in what situation would anyone profit from a manipulation of 51% of the delegates? Or for that matter in what situation do you believe that anyone would profit from Ghash.io  performing a 51% attack on the bitcoin network?

They can be sued/hacked/coerced by governments, banks or other corporations. Preventing an entity from profiting from a 50% attack isn't enough.
clout (OP)
Full Member
***
Offline Offline

Activity: 207
Merit: 100


View Profile
June 16, 2014, 09:29:11 AM
 #255

IMO, trying to grab control with 50 puppets instead of 2 is much more difficult.
I know. You said that. But I was asking of the why? Where is the additional difficulty?

Because people have to vote for the delegates. Also in what situation would anyone profit from a manipulation of 51% of the delegates? Or for that matter in what situation do you believe that anyone would profit from Ghash.io  performing a 51% attack on the bitcoin network?

They can be sued/hacked/coerced by governments, banks or other corporations. Preventing an entity from profiting from a 50% attack isn't enough.

If anything prevents a delegate from doing his job (ie signing a block at the designated time) then he will be voted out and replaced by a delegate that is not being sued hacked or coerced by governments. Delegates have a simple task and do not have that much power. Shareholders will vote for the most diverse set of delegates in order to maintain a resilient network. Global competition is the most robust solution to the problems you described.
idlecore
Newbie
*
Offline Offline

Activity: 6
Merit: 0


View Profile
June 16, 2014, 10:44:35 AM
 #256

IMO, trying to grab control with 50 puppets instead of 2 is much more difficult.
I know. You said that. But I was asking of the why? Where is the additional difficulty?

Because people have to vote for the delegates. Also in what situation would anyone profit from a manipulation of 51% of the delegates? Or for that matter in what situation do you believe that anyone would profit from Ghash.io  performing a 51% attack on the bitcoin network?

They can be sued/hacked/coerced by governments, banks or other corporations. Preventing an entity from profiting from a 50% attack isn't enough.

If anything prevents a delegate from doing his job (ie signing a block at the designated time) then he will be voted out and replaced by a delegate that is not being sued hacked or coerced by governments. Delegates have a simple task and do not have that much power. Shareholders will vote for the most diverse set of delegates in order to maintain a resilient network. Global competition is the most robust solution to the problems you described.

The point made was that the possibility of attack will be there whether it is profitable or not.
clout (OP)
Full Member
***
Offline Offline

Activity: 207
Merit: 100


View Profile
June 16, 2014, 11:35:21 AM
 #257

Ok... so let someone incur the cost of "attacking" the network. If the shareholders are not complacent then they will just for the network and the attacker will have control of a worthless network. You simply are not considering the economic cost of attacking the network. What you consider an attack is not actually an attack at all.
eeee12344
Member
**
Offline Offline

Activity: 73
Merit: 10


View Profile
June 21, 2014, 06:45:41 AM
 #258

No doubt
NXT is great。。。
FandangledGizmo
Legendary
*
Offline Offline

Activity: 1138
Merit: 1001


View Profile
July 05, 2014, 01:49:09 AM
 #259

IMO, trying to grab control with 50 puppets instead of 2 is much more difficult.
I know. You said that. But I was asking of the why? Where is the additional difficulty?

Because people have to vote for the delegates. Also in what situation would anyone profit from a manipulation of 51% of the delegates? Or for that matter in what situation do you believe that anyone would profit from Ghash.io  performing a 51% attack on the bitcoin network?

I'm a big supporter of BitShares but in terms of who would profit from a Ghash.io 51% attack on Bitcoin...

TPTB behind the traditional banking system have $ trillions on the line, Bitcoin and crypto-currencies are a big threat, especially as I see capital controls and deposit confiscations in a lot of Western countries futures. 

Performing a 51% attack on the Bitcoin network via Ghash.io,  if it could decimate the price as well as significantly damage public faith in crypto-currencies would be extremely valuable to them imo, no?





ChuckOne
Sr. Member
****
Offline Offline

Activity: 364
Merit: 250

☕ NXT-4BTE-8Y4K-CDS2-6TB82


View Profile
July 05, 2014, 01:16:20 PM
 #260

Sure.
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 [13] 14 15 16 17 »  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!