Topic: Which Proof of Stake System is the Most Viable - page 13. (Read 25775 times)

Sounds like he want to pick DaT's brain.

If he had a solid solution he would be releasing without the fanfare.  Just my opinion.

Bitshares and NXT are both exciting. I lean towards Dpos because the development is more open.  Bitshares also seems to have a fuller support for DACs.

The forging pools with nxt equal the delegates in dpos. This is what makes them equal.

Our main developer wants to answer you. But he does not use BTT anymore;

https://twitter.com/comefrombeyond/statuses/470997528604049408

I think it will be a battle between nxt @ bitshares
Nxt has the advantage that has released their "product" first, security wise in my mind the winner is bitshares but only on the playfield they will proove it.
Both coins are still undervalued and I am personaly invested more in favor of bitshares and I continue to invest there because I think they have a greater potential.
Hope the devs and founders from the 3 "coins" will share their good ideas and "work together" so all communities will win a share from the success of this innovative industry! We are all on the same neighborhood 

As Stan (bitshares founder) said:
"Each of us will drive more awareness to each other than they will take away from each other.  It's the same reason competing car dealers and hardware stores tend to cluster on the same street.  Get the traffic to come to your neighborhood and then compete!"

In NXT you can also form forging pools and may lead to Bitcoin style problems.

Gavin recently stated the network is "over secure"...meaning there is too many
miners...if later on, miners need to drop out because there is no more profit,
there may be fewer miners.  There will have to be a balance between security
and fees.   

It largely depends on how big Bitcoin gets.  If a billion dollars
a day is transacted, then a million dollars a day in fees is 1/10th of a percent.

The code as released doesn't solve the nothing at stake problem.  It also can be hard forked by creating an alternate chain longer than 720 blocks.   NXT promises there is some magical code (EC and TF) which will resolve these problems but the world isn't ready for it yet.

https://bitbucket.org/JeanLucPicard/nxt/src

Care to elaborate?

It'll be 3% of what it is now. We can hope that 1 BTC doubles in real-world value at least every 4 years to compensate, and it might for a while, but that rate of increase can't continue forever. I am thinking in longer timescales than 20 years here.

DPOS and NXT are practically identical, the only difference is which mechanism is used to pick block producers (both are essentially by stake-vote)

The same is true with PoS.  Yes you have proven that 1% of a large number is larger than 1% of a small number.

Blackcoin have announced that they are about to release an improved PoS. Details are still unclear, so not sure if it will solve the issues that the Peercoin model currently has.

After talking with the Nxt developers it is fairly clear that Nxt and DPOS are very similar.  They both deterministically select the next block producer, operate on a fixed absolute block production schedule and support DELEGATING your stake to others.  

The difference lies in how the delegated stake is managed.  In DPOS everyone is delegating at all times and all clients observe behavior.  In DPOS it is possible to vote AGAINST a bad actor and thus nullify their own votes.   This prevents someone from buying up 10% or more of the network and then abusing their 10% in some way.  

Where Nxt has block production proportional to balance, DPOS is 1 block per delegate with a changing slate of delegates that are voted on proportional to stake.   There are limits that prevent concentration in one delegate and the wallets automatically rebalance votes from delegates that get too much.  

Lastly the difference is that all shareholders are paid from dividends without having to do anything.  The delegates take a small cut of the transaction fees.  

I think that Nxt / DPOS will be the future as Peercoin has too many issues.

20 years from now, daily reward is still 112 blocks...should be more than enough to keep fees low.

Bitcoin uses vote by hashpower and thus everyone is trusting less than 10 super nodes (mining pools).  Try to look at the true nature of what these systems are doing rather than at the marketing spin the BTC community has put on things.

With bitcoin you cannot vote a miner or pool out.  The coin holders have no say at all and only a super minority actually mine bitcoin these days.   

The system is still governed by math, it just operates as 99 small mining pools where the users can redirect their hash power at will and where all mining pools have equal say and take turns.   

In DPoS, you vote when sending in a transaction. You don't have to keep your wallet open.

You should keep your client online, though as it monitors the network and alerts you in case any delegate is misbehaving.

Although there's no improvement in security in the long run. When someone designs an improved ASIC, the network becomes vulnerable to the danger that an attacker will deploy the new technology first. As the new ASIC is adopted by miners, we return to the status quo. The miners need to keep upgrading, just to stay in the same place both with regard to their revenues, and with regard to network security. As long as technology keeps improving, miners will need to pay for new hardware, and the community will have to pay for that hardware, as well as their electricity costs, either through block-reward inflation or through fees.

PoS does not intrinsically need inflation. Nxt is a PoS coin that has no inflation. There's no block reward, just fees, so the number of coins is constant.

That's true while the block reward dominates. As the block reward reduces, and transaction volume increases, transaction fees will come to dominate in Bitcoin. The cost of security will be equal to the sum of fees, and the community will get greater or lesser security by paying higher or lower fees.

This is why I care about Proof of Stake. It seems to me that either Bitcoin fees will end up very high, to pay the miners for the work they do securing the network; or else the fees will end up comparatively low, and the network will be insecure. If PoS works (a big "if", I concede), then it's inherently more efficient and will require lower fees for the same security, and that could lead users to shift over to it in the long run. (Much as lower fees is often cited as a reason for merchants to prefer Bitcoin over credit cards.)

Also if Sunnys involvement in PeerShares is the clincher, you might want to reevaluate your position. In a recent chat discussion between Sunny King and Bytemaster on peercointalk it seems he was not as involved with peershares as you suggested.

Peercoin is not complete in my view and has some remaining issues, before it can be considered a viable pos-solution, most glaring one being their checkpointing requirement.

I really want to like NXT and even though it is not strictly a technical issue with their particular solution, but I do not like their cloak and dagger behavior with the source-code and only releasing it to peer review after they "feel" the world is ready.

Clout do you think adding cpos to the discussion and/or poll would be advisable or is that proposal still too theoretical?

One little addendum regarding the whitepapers by Daniel Larimer, part of the "security model" in his theories is that of many competing chains and free market choice between them, instead of the other seemingly isolated central blockchain models. While I understand the philosophical ideal of wanting security solely through algorithms, I'm not convinced that this mathematical Elysium is achievable in practice, besides it all hinges on market evaluation by humans anyway. So while not being limited to tapos/dpos I do think the free choice between chains implied in those whitepapers (ByteMaster always mentioned the competing chains when I've heard him explain his concepts) is more versatile than other models.

PS
What's up with all the references to the scores on IQ-tests and drawing conclusions from them? To me those things were more about how long can you fight the boredom and keep focus. Most of the ones I had to fill in even used multiple choice, talk about ridiculous. Multiple choice makes it just about impossible to give the wrong answer.