Pages:
Author

Topic: The state of crypto - The only serious thread on the subforum - page 5. (Read 19463 times)

legendary
Activity: 1008
Merit: 1007
Delegated proof of work, which Bitcoin is, doesn't.  It's no more fault tolerant than DPoS.

Hashcash POW is subject to sybil due to ASICs, this is clearly something that satoshi did not foresee, I agree. However, that does not rule out all other types of POW - there exists a near idealised POW in which one CPU really is one vote.

DPOS has no white paper and has not been proven to solve the byzantine problem, much like most other consensus designs.
legendary
Activity: 1154
Merit: 1001
Thank you r0ach & others for comments, this thread is a really good read.

If I understand correctly, in Bitcoin/PoW, any miner can opt for not delegating his hashing power, and yet remain an active participant. I know most people do delegate their hashing power, for various practical and economical reasons, but still, it is not a mandatory position. The miner has full authority over its own hashing/voting power, and unilaterally decides (if he delegates to a pool or not).

Is DPoS comparable, in that the holder/staker has authority over the decision to delegate or not delegate his hashing/voting power, and remains a participant in equal standing regardless of his position towards delegation? I'm definitely not familiar enough with DPoS, would appreciate any pointer to a meaningful intro on the subject.
legendary
Activity: 1596
Merit: 1027
Great thread!

Pretty informative and it reveals a new forward thinking about cryptos and its impact. I really enjoyed your appreciation.
legendary
Activity: 1260
Merit: 1000
edit: only POW provably solves the byzantine generals problem in the face of sybil attack

Delegated proof of work, which Bitcoin is, doesn't.  If 70% of the hash rate is in china owned by three pools, you have no way of knowing these pools aren't owned by the same person (sybil).  The only way is to audit them yourself, which is the purpose of the voting mechanism in DPoS, to audit the block validators for sybil.  The only difference is, the audit mechanism is built into the protocol of DPoS and excluded entirely from DPoW (delegated proof of work).

Seriously, read my first and second post in this thread, I'm not bullshitting here, it's painfully obvious this is the case:

The Bitcoin consensus mechanism is incorrectly labeled Proof of Work

https://bitcointalksearch.org/topic/the-bitcoin-consensus-mechanism-is-incorrectly-labeled-proof-of-work-1176835
legendary
Activity: 2968
Merit: 1198
I don't remember his exact language used to describe what was needed to prevent the system from failing, which is why I said "correct me if I'm wrong".  I thought I remember seeing the 200% specification somewhere.  I doubt it's 1000% (order of magnitude) since this would just give you the same centralized pool structure as Bitcoin now.

Here it is:

Did you just reveal that mining has a proof of burn involved because I don't see any other way for that statement to work otherwise.

I already told you that if users are required send PoW with each transaction and they don't care about their insignificant electrical cost then PoW will be unprofitable, especially if the CPU-only hash has been well designed to be within an order-of-magnitude of the potential ASIC optimization. In my recent archives (July?), smooth and I estimated this order-of-magnitude for Cryptonite at between 1 and 2, and I believe mine may be slightly better than Cryptonite.

There is your first example of me doing something technical you formerly thought is impossible. And there will be many more such cases. Stay tuned...
legendary
Activity: 2968
Merit: 1198
And why Ethereum is in such huge trouble that nobody seems to acknowledge.  It took Satoshi a long time to create the consensus mechanics of Bitcoin, now you have Vitalik trying to create one on the fly after the coin was already released.  They can't copy standard proof of stake due to all of it's negatives.  He's either going to have to beat Satoshi, copy DPoS, or a far more experimental method that might end in critical failure like a DAG chain.

It would be absolutely insane for Etherium to actually go through with changing the core consensus mechanism of the chain. Like it or not, there really is no viable alternative to POW at the moment.

edit: only POW provably solves the byzantine generals problem in the face of sybil attack

I actually think it is possible that Ethereum will just stick with PoW (or maybe a hybrid PoS+PoW). They've always couched the discussion of PoS in terms of "if it can be made to work". Obviously Vitalik is a fan of PoS and he criticizes PoW based on what he calls its devastating environmental impact but I also think he is at times intellectually honest enough to just admit that it can't be made to work and conclude "we tried our best, but ultimately we decided to stick with PoW."

This would be a bruising political battle given the obvious desire by existing Etherium stakeholders to cut the coin supply from infinite to premine+18 month of mining or whatever, but I don't view it as impossible (nor am I predicting it).
legendary
Activity: 2968
Merit: 1198
would just give you the same centralized pool structure as Bitcoin now.

Unlikely if other aspects of the system are radically different which seems likely from how he described it. You could possibly get some other failure, but it wouldn't likely be the same as Bitcoin at all.
legendary
Activity: 1008
Merit: 1007
And why Ethereum is in such huge trouble that nobody seems to acknowledge.  It took Satoshi a long time to create the consensus mechanics of Bitcoin, now you have Vitalik trying to create one on the fly after the coin was already released.  They can't copy standard proof of stake due to all of it's negatives.  He's either going to have to beat Satoshi, copy DPoS, or a far more experimental method that might end in critical failure like a DAG chain.

It would be absolutely insane for Etherium to actually go through with changing the core consensus mechanism of the chain. Like it or not, there really is no viable alternative to POW at the moment.

edit: only POW provably solves the byzantine generals problem in the face of sybil attack
legendary
Activity: 1260
Merit: 1000
As for Anonymint's coin, correct me if I'm wrong, but I think he's relying on people not being able to create an ASIC with more than 200% efficiency of a CPU or the entire model fails?

1. He said order of magnitude. In normal English that usually means 10x, though binary order of magnitude -- 2x -- also means something occasionally. I'm not sure what 3x means.

2. I don't think he said the entire model would fail otherwise, so that may be your made up conclusion. I don't really remember though, he may have suggested that.

I don't remember his exact language used to describe what was needed to prevent the system from failing, which is why I said "correct me if I'm wrong".  I thought I remember seeing the 200% specification somewhere.  I doubt it's 1000% (order of magnitude) since this would just give you the same centralized pool structure as Bitcoin now.
legendary
Activity: 2968
Merit: 1198
As for Anonymint's coin, correct me if I'm wrong, but I think he's relying on people not being able to create an ASIC with more than 200% efficiency of a CPU or the entire model fails?

1. He said order of magnitude. In normal English that usually means 10x, though binary order of magnitude -- 2x -- also means something occasionally. I'm not sure what 3x means.

2. I don't think he said the entire model would fail otherwise, so that may be your made up conclusion. I don't really remember though, he may have suggested that.
legendary
Activity: 1260
Merit: 1000
bitcoin and everything todate related is simply an overkill in complication for a p2p currency.

And yet bitcoin still remains the most simple consensus design of all of them. This gives you some idea just what a hard problem distributed, trustless consensus is.

And why Ethereum is in such huge trouble that nobody seems to acknowledge.  It took Satoshi a long time to create the consensus mechanics of Bitcoin, now you have Vitalik trying to create one on the fly after the coin was already released.  They can't copy standard proof of stake due to all of it's negatives.  He's either going to have to beat Satoshi, copy DPoS, or a far more experimental method that might end in critical failure like a DAG chain.

As for Anonymint's coin, correct me if I'm wrong, but I think he's relying on people not being able to create an ASIC with more than 200% efficiency of a CPU or the entire model fails?  So even if this coin is theoretically perfect, nobody will even know if it's valuable or not until it's been on the market for several years, reach 100 million in market cap, then attract ASIC vendors to attempt to full scale commercialize the coin?
legendary
Activity: 1008
Merit: 1007
bitcoin and everything todate related is simply an overkill in complication for a p2p currency.

And yet bitcoin still remains the most simple consensus design of all of them. This gives you some idea just what a hard problem distributed, trustless consensus is.
legendary
Activity: 1876
Merit: 1000
the most sensible thing to be stated in this thread is that BTC is a rube goldberg machine. bitcoin and everything todate related is simply an overkill in complication for a p2p currency.

the blockchain idea leads to too much complications trying to plug its obvious flaws, leading to the biggest flaw of them all complication itself.


as i say its a currency not a mission to mars.

and the extant financial technology isn't a rube goldberg machine?

I mean, I'll give yah that, they are both rube goldberg machines - the difference is that with bitcoin everyone knows how it works (well, one can know how it works). The extant financial system..... not so much.

and, IMO, currency is much much more complicated than a mission to mars. That's merely getting from point A to point B. Currency is a monstrous technology in comparison - a device we use to store value, transfer time, exchange intangibles. We will, someday, get to mars. But we won't get there without currency providing a means to allow civilization to network.

yes but your comparing apples and oranges here on so many fronts, where to begin;

1) average joe's use fiat (without fully understanding it) because they have to, no one has to use bitcoin.

2) average joe's whether (unfounded or not) trust gov backed fiat (partly also because they have too), bitcoin your trusting yourself (well and also an anon dev with a billion $ stash), too trust bitcoin you kinda have to understand it, to trust fiat you can do it blindly (and again really have no choice).




hero member
Activity: 770
Merit: 504
sr. member
Activity: 419
Merit: 250
the most sensible thing to be stated in this thread is that BTC is a rube goldberg machine. bitcoin and everything todate related is simply an overkill in complication for a p2p currency.

the blockchain idea leads to too much complications trying to plug its obvious flaws, leading to the biggest flaw of them all complication itself.


as i say its a currency not a mission to mars.


I agree with this, too.

Instead of using our brainpower to find a way to answer the question, "why crypto" to Joe Blow, we are making it harder for him to understand everyday.

Or let's spend our energy figuring out how to market a crypto, which seems to be a lot harder then we ever expected.
legendary
Activity: 1260
Merit: 1008
the most sensible thing to be stated in this thread is that BTC is a rube goldberg machine. bitcoin and everything todate related is simply an overkill in complication for a p2p currency.

the blockchain idea leads to too much complications trying to plug its obvious flaws, leading to the biggest flaw of them all complication itself.


as i say its a currency not a mission to mars.

and the extant financial technology isn't a rube goldberg machine?

I mean, I'll give yah that, they are both rube goldberg machines - the difference is that with bitcoin everyone knows how it works (well, one can know how it works). The extant financial system..... not so much.

and, IMO, currency is much much more complicated than a mission to mars. That's merely getting from point A to point B. Currency is a monstrous technology in comparison - a device we use to store value, transfer time, exchange intangibles. We will, someday, get to mars. But we won't get there without currency providing a means to allow civilization to network.
legendary
Activity: 1876
Merit: 1000
the most sensible thing to be stated in this thread is that BTC is a rube goldberg machine. bitcoin and everything todate related is simply an overkill in complication for a p2p currency.

the blockchain idea leads to too much complications trying to plug its obvious flaws, leading to the biggest flaw of them all complication itself.


as i say its a currency not a mission to mars.
legendary
Activity: 1260
Merit: 1000

DPoS is the future.  Here's my DPoS proposal if it's required to scale past 101 delegates for increased redundancy:

A super delegate system - Re-inventing the Roman Republic

https://bitsharestalk.org/index.php/topic,18457.0.html
legendary
Activity: 1008
Merit: 1007
legendary
Activity: 2114
Merit: 1090
=== NODE IS OK! ==
PoS is the future
Pages:
Jump to: