Light Weight Clients & POS

[Rune]

Ooh I love the auto-firing mechanism. That makes it at least as secure as POW

[pc]

If I make some very wrong assumptions about delegate behavior, and if you think there is a 1% chance that the next delegate will be in a collusive group, then you can wait one minute (or two if chains compete?) and have a 1 in a million chance of this next string of delegates colluding to reverse history. That's still better than the confirmation you get from a bitcoin thin client after 10 minutes.

The problem I see is that a number of colluding delegates can make an honest delegate look bad (and possible get him to be automatically fired). This is dangerous.

[eagleeye]

Its good the light weight client is on the top of the mind of bytemaster.

Really good.

No complacency with him.

Just remember there will be slow times and there will be fast times.

[xeroc]

Hi Toast. Is the light wallet project started now? What's the plan?

Yep .. they are working on it already .. IIRC it's gonna be a JavaScript implementation .. check the BitShares repos on github!

[lastagile]

If I make some very wrong assumptions about delegate behavior, and if you think there is a 1% chance that the next delegate will be in a collusive group, then you can wait one minute (or two if chains compete?) and have a 1 in a million chance of this next string of delegates colluding to reverse history. That's still better than the confirmation you get from a bitcoin thin client after 10 minutes.

Hi Toast. Is the light wallet project started now? What's the plan?

[toast]

If I make some very wrong assumptions about delegate behavior, and if you think there is a 1% chance that the next delegate will be in a collusive group, then you can wait one minute (or two if chains compete?) and have a 1 in a million chance of this next string of delegates colluding to reverse history. That's still better than the confirmation you get from a bitcoin thin client after 10 minutes.

[CLains]

Great 

What does that mean for me, user, investor, product, hype wallets, future, creativity, vision?

[alt]

With Bitcoin light weight clients rely entirely on proof-of-work validation in combination with a trusted node.  So long as the cost of faking the proof-of-work is higher than the value of the transaction you can generally trust that your transaction is valid and included in the official chain.

On the other hand POS coins whether peer coin or Nxt have no such metric.  The only way to validate the chain is to have a full copy of the chain or to fully trust someone who does.

Fortunately with DPOS we once again have a means of deterring and punishing those who could potentially lie to a light weight client.  If the delegate signs a statement that the transaction was valid and confirmed and it turns out to be false, then that signed statement can be used to unilaterally fire the delegate.   This means that light weight validation of small transactions can now be validated with the same weight as POW based light weight validation.

Great 
if delegate can get profit by statement a invalid translation ?

Correct... signing that something was valid, when in fact it was invalid would be grounds to provably and immediately fire a delegate.    In fact, we could allow anyone to enter this business provided they post a bond.

Great 

[lzr1900]

Hi Bymaster, just wondering if you are working on  BTS Light Weight Client . I hardly open my windows version bts wallet due to I hardly find any wallet is worse than   BTS.   

With Bitcoin light weight clients rely entirely on proof-of-work validation in combination with a trusted node.  So long as the cost of faking the proof-of-work is higher than the value of the transaction you can generally trust that your transaction is valid and included in the official chain.

On the other hand POS coins whether peer coin or Nxt have no such metric.  The only way to validate the chain is to have a full copy of the chain or to fully trust someone who does.

Fortunately with DPOS we once again have a means of deterring and punishing those who could potentially lie to a light weight client.  If the delegate signs a statement that the transaction was valid and confirmed and it turns out to be false, then that signed statement can be used to unilaterally fire the delegate.   This means that light weight validation of small transactions can now be validated with the same weight as POW based light weight validation.

Great 
if delegate can get profit by statement a invalid translation ?

Correct... signing that something was valid, when in fact it was invalid would be grounds to provably and immediately fire a delegate.    In fact, we could allow anyone to enter this business provided they post a bond.

bump

[myhometalk]

Hi Bymaster, just wondering if you are working on  BTS Light Weight Client . I hardly open my windows version bts wallet due to I hardly find any wallet is worse than   BTS.   

With Bitcoin light weight clients rely entirely on proof-of-work validation in combination with a trusted node.  So long as the cost of faking the proof-of-work is higher than the value of the transaction you can generally trust that your transaction is valid and included in the official chain.

On the other hand POS coins whether peer coin or Nxt have no such metric.  The only way to validate the chain is to have a full copy of the chain or to fully trust someone who does.

Fortunately with DPOS we once again have a means of deterring and punishing those who could potentially lie to a light weight client.  If the delegate signs a statement that the transaction was valid and confirmed and it turns out to be false, then that signed statement can be used to unilaterally fire the delegate.   This means that light weight validation of small transactions can now be validated with the same weight as POW based light weight validation.

Great 
if delegate can get profit by statement a invalid translation ?

Correct... signing that something was valid, when in fact it was invalid would be grounds to provably and immediately fire a delegate.    In fact, we could allow anyone to enter this business provided they post a bond.

[bytemaster]

With Bitcoin light weight clients rely entirely on proof-of-work validation in combination with a trusted node.  So long as the cost of faking the proof-of-work is higher than the value of the transaction you can generally trust that your transaction is valid and included in the official chain.

On the other hand POS coins whether peer coin or Nxt have no such metric.  The only way to validate the chain is to have a full copy of the chain or to fully trust someone who does.

Fortunately with DPOS we once again have a means of deterring and punishing those who could potentially lie to a light weight client.  If the delegate signs a statement that the transaction was valid and confirmed and it turns out to be false, then that signed statement can be used to unilaterally fire the delegate.   This means that light weight validation of small transactions can now be validated with the same weight as POW based light weight validation.

Great 
if delegate can get profit by statement a invalid translation ?

Correct... signing that something was valid, when in fact it was invalid would be grounds to provably and immediately fire a delegate.    In fact, we could allow anyone to enter this business provided they post a bond. 

[BTSdac]

With Bitcoin light weight clients rely entirely on proof-of-work validation in combination with a trusted node.  So long as the cost of faking the proof-of-work is higher than the value of the transaction you can generally trust that your transaction is valid and included in the official chain.

On the other hand POS coins whether peer coin or Nxt have no such metric.  The only way to validate the chain is to have a full copy of the chain or to fully trust someone who does.

Fortunately with DPOS we once again have a means of deterring and punishing those who could potentially lie to a light weight client.  If the delegate signs a statement that the transaction was valid and confirmed then that signed statement can be used to unilaterally fire the delegate.   This means that light weight validation of small transactions can now be validated with the same weight as POW based light weight validation.

Great 
if delegate can get profit by statement a invalid translation ?

[xeroc]

The "light weight" you mean, as I understand, is not to store full copy of block chain.

Yup.

Why light weight clients of POS coins like PPC or NXT need to fully trust some full node, what the possible faking way?

I think I don't get the point why cost of faking POS node is much smaller then POW?

I think it is easier to generate a secret pos chain with just like 5% of all stake. If you see the secret chain and are only connected to the one node it looks like the main chain!

[HackFisher]

The "light weight" you mean, as I understand, is not to store full copy of block chain.

Why light weight clients of POS coins like PPC or NXT need to fully trust some full node, what the possible faking way?

I think I don't get the point why cost of faking POS node is much smaller then POW?

[bytemaster]

With Bitcoin light weight clients rely entirely on proof-of-work validation in combination with a trusted node.  So long as the cost of faking the proof-of-work is higher than the value of the transaction you can generally trust that your transaction is valid and included in the official chain.

On the other hand POS coins whether peer coin or Nxt have no such metric.  The only way to validate the chain is to have a full copy of the chain or to fully trust someone who does.

Fortunately with DPOS we once again have a means of deterring and punishing those who could potentially lie to a light weight client.  If the delegate signs a statement that the transaction was valid and confirmed then that signed statement can be used to unilaterally fire the delegate.   This means that light weight validation of small transactions can now be validated with the same weight as POW based light weight validation.