[VoterApathy] How To Make 51% Attacks A Lot Harder (without more people voting)

[monsterer]

Precisely.  That particular gambit is fixed by the ability to fork out the bad guys and continue on.

Again, you are either missing or misrepresenting the seriousness of this issue. A hard fork to fix something like this would likely destroy the value of the currency completely; it is not the same as a hard fork which adds a feature, and may well need to rollback a whole load of transactions as well, which may have knock on effects on regular users or exchanges, depending on how insidiously the attacker behaved.

At any rate, bitshares should be stronger than 13% attack resistent, don't you think?

[CoinHoarder]

So it seems the OP went right over everyone's head. Admittedly I am not certain it would work, but it seems like no one even got the concept. I was hoping BM would read the OP and comment as to whether it is feasible. I probably described it wrong, but I feel like it is a possible solution. Here is the most recent thread on Bitcointalk about the idea, along with a link to the first thread. I was against it at first because I didn't want Litecoin to be Bitcoin's "bitch", but I eventually came around to the idea... as you can see in the original and subsequent thread. Those threads explain the idea in more detail if you're having trouble getting what I was trying to say in the OP.

https://bitcointalk.org/index.php?topic=677783.0
https://bitcointalk.org/index.php?topic=176556.0

That would required the MUSE chain to also validate transactions of the BTS chain to use the hash of the right fork.

Well, yeah... that's the point of the proposal. Each Bitshares chain would work together to make each separate chain more resilient.

Furthermore, BitShares already favours the longest leg (which happens to be the leg with highest witness participation and thus "highest degree" of consensus
Maybe I still don't get the magic behind the idea

So does Bitcoin/Litecoin (for which the idea was originally conceived for)...  the idea is that when an attacking fork occurs that there be an option to NOT favor the longest leg and instead favor the honest leg that is recorded on the other chains.

51% is a gross overestimate. Try a 13% attack; that's what bitshares is currently vulnerable to.

Well, duh.... but if I say 13% attack hardly anyone knows what I'm talking about, yet if I say 51% attack then everyone knows what I am talking about.


For the record I like DPoS, and I think it is one of the features that makes Bitshares attractive. It is definitely unfortunate that hardly anyone votes. I think we should force centralized exchanges to vote by taking our business elsewhere if they refuse to vote. That is one easy thing we could do.

[Stan]

A well informed vote from someone who has done the hard work to understand the issues is highly valuable.

A random (e.g. machine generated) vote is useless - it just increases the noise floor without changing the outcome.

Alas, rewarding voting yields mostly the latter, not the former.

[Stan]

And if it did that, then all the inactive voters would come out with torches and pitchforks to fix it.

Until things start going in a direction they don't like, many of us are content to let it ride.

A decision not to vote is a vote for the status quo.  That is not the same as being centralized.

I'm afraid it is much more serious than you think.

This one account could vote in all its own delegates which then sit there refusing to process transactions. Votes are transactions, so you would never be able to vote them out.

This would bring the entire chain to a standstill forever until the dev team introduce a hard fork to rewind the chain.

Precisely.  That particular gambit is fixed by the ability to fork out the bad guys and continue on.

The most efficient solution is one run by an impeccably trustworthy benevolent dictator.  That Day is coming, but until then we can allow just about anybody to serve in the role as long as their feasible actions are strictly limited and we have the ability to detect, correct, and block any misbehavior.

Until that time, a valid voting decision is to tolerate the current set of dictator(s) as being a suitably constrained most efficient solution.  Trust is hard to earn and easy to lose, so any modicum of actual impeccability and trustworthiness they may possess are reinforced.

When considering the suitability of DPOS, you need to consider the whole design, not just the voting mechanism in isolation.

[jakub]

Voting has its usefulness (as it increases the overall security of the network) so it should be rewarded.
What if we introduced a special dividend, something like 2-3% p.a. for those who do vote for at least 10 witnesses? (either directly or indirectly via a proxy).
It would have to come from dilution but it will only hurt those who don't care anyway.

If we say decentralization is important for us then we need to pay for it. Problem solved.

[Chronos]

Good luck convincing many friends to buy shares in a centralized blockchain with shareholder directed dilution 

Good point. In fact, BTS is in a bit of a stranglehold with Cryptonomex, through no fault of its own:

• The majority of voters can vote in new worker pay via dilution
• Cryptonomex (and/or like-minded founders) controls a majority of "active" votes, partly due to voter apathy
• Therefore, Cryptonomex can vote in worker pay for itself (which it does)

So, as a matter of opinion, you might conclude that significant trust must be placed in Cryptonomex not to over-dilute, in the same manner that citizens trust their central bankers.

It's not easy to "fix" this.

[TravelsAsia]

I agree that this is not widely understood, but that is not a knock on DPOS, just on our ability to overcome false narratives that occur when even our long term members get careless about what they say.

My only thought on this matter is that if we had more ways to get voting better communicated to end users, we would see more participation.. right now its just an area that has been less of a priority in development over other things.

That might work for people that spend a lot of time on the forums, I would argue that most people don't want to be that engaged. I think there's too much of an assumption that everyone is as hardcore as the people that are always posting on the forums.

 I like where Monsterer is coming from.

[monsterer]

And if it did that, then all the inactive voters would come out with torches and pitchforks to fix it.

Until things start going in a direction they don't like, many of us are content to let it ride.

A decision not to vote is a vote for the status quo.  That is not the same as being centralized.

I'm afraid it is much more serious than you think.

This one account could vote in all its own delegates which then sit there refusing to process transactions. Votes are transactions, so you would never be able to vote them out.

This would bring the entire chain to a standstill forever until the dev team introduce a hard fork to rewind the chain.

[BunkerChainLabs-DataSecurityNode]

I agree that this is not widely understood, but that is not a knock on DPOS, just on our ability to overcome false narratives that occur when even our long term members get careless about what they say.

I am stating a hard truth here in the hopes that it will get the powers that be to think about the problem more clearly.

I'll say it again:

At this very moment, 1 single bitshares account could vote out all the delegates and install its own for whatever purposes they liked. That means 1 account controls the entire network, which is far, far more centralised than bitcoin, nxt or any other cryptocurrency except maybe ripple.

The hard truth is that anybody and everybody can buy more BTS and change this at ANY TIME.

We have proxy voting now and a lot of people have choosen to do this instead of being more engaged. Again.. its the stakeholders CHOICE. NOBODY forced them to take or not take those actions. No ONE OR TWO OR THREE holders could force them to take those actions.

Half the people arguing for changes in this thread to voting have already admitted to giving up their stake in BTS so it's kinda moot to be taking advise from non-stakeholders on how to get voters to be more engaged.

My only thought on this matter is that if we had more ways to get voting better communicated to end users, we would see more participation.. right now its just an area that has been less of a priority in development over other things.

Sadly part of the mandate for dposhub was to help improve this. We had plans that included having users when they 'liked' and update it would literally cast their vote for that worker, or witness, or committee member.

Engagement like this needs to become common place where the content and the action are on the same screen.

Coming up with other crazy schemes is just unnecessary. We just need a better delivery method for what we got that we know can and does work.

[Stan]

I agree that this is not widely understood, but that is not a knock on DPOS, just on our ability to overcome false narratives that occur when even our long term members get careless about what they say.

I am stating a hard truth here in the hopes that it will get the powers that be to think about the problem more clearly.

I'll say it again:

At this very moment, 1 single bitshares account could vote out all the delegates and install its own for whatever purposes they liked. That means 1 account controls the entire network, which is far, far more centralised than bitcoin, nxt or any other cryptocurrency except maybe ripple.

And if it did that, then all the inactive voters would come out with torches and pitchforks to fix it.

Until things start going in a direction they don't like, many of us are content to let it ride.

A decision not to vote is a vote for the status quo.  That is not the same as being centralized.

[Stan]

Buy and hold more BTS.

Vote with it.

Tell your friends to do the same.

Problem solved.

Good luck convincing many friends to buy shares in centralized blockchain with shareholder directed dilution 

Far less centralized and far less dilution that Bitcoin.

2 years and you still fail to grasp the elementary difference between 'immutable' and 'shareholder directed' in terms of trust required

Edit, Also perception is important. I think the general perception is that BTC is more decentralized than it is, among other things because there is a belief that hashers will quickly switch pools if they need too.

As we learned back in the ProtoShares days, hashers will continue to support a slightly more profitable pool even if it blatantly refuses to include any transactions.

The max dilution rate of both is immutable, but BTS shareholders can dial it to zero.  The max rate of Bitcoin is also the min rate and it all gets wasted, while whatever the much smaller BTS rate becomes gets spent on what the shareholders think is most useful.

There is no comparison.

I agree that this is not widely understood, but that is not a knock on DPOS, just on our ability to overcome false narratives that occur when even our long term members get careless about what they say.

I agree with you about the hashers, this is not the widely held perception though.

The problem is if BTS is perceived as centralized & those centralized shareholders can direct and increase dilution then that requires a lot more trust.

As we saw with the merger, diluting BTSX for DNS and Vote lost us value, so we can see shareholders can make poor decisions with dilution and to date haven't been able to use it for anything that's been able to rapidly grow BTS vs. more 'wasteful' competitors like many hoped.

The same perceived centralized power group also had to pay themselves salaries to carry on working on BTS, including yourself, however BTS continued losing value during that time, so even though you and others were working far in excess of the value of that salary, it wasn't translating into added value for the DAC/shareholders. So from that perspective, dilution in a DAC still has to prove itself as well.

So when a small group, due to voter apathy/other has the power to direct dilution including towards mergers, this requires a great deal more trust than a dilution rate that has already been pre-defined and given BTS have thus far failed to make gains vs. competitors, trusting a fairly centralized group with that power, when you feel you are an insignificant sized shareholder,  requires even more trust going forward imo. 

I agree that this is not widely understood, but that is not a knock on DPOS, just on our ability to overcome false narratives that occur when even our long term members get careless about what they say.

I think you also get careless sometimes, primarily out of enthusiasm and as a result create a credibility problem for BTS.

A long term member like me, could be more supportive & focus on the positives, but unfortunately it's in my nature to focus on things I see as weaknesses and I realize I probably rank around the newmine's and tonyk's of this world that criticize & find fault too much at times.

Some good points. 

It might be good to distinguish between "voter apathy" and "voter satisfaction".  If I'm happy with how something is being run, or even neutral, I may not bother to vote.  I only come out of my cave to vote when something has me riled up. 

We all sit somewhere on the continuum between pessimistic on the left and optimistic on the right.  So anyone to the left of my position seems overly pessimistic and vice versa.  Entrepreneurs are inherently optimistic or we would never attempt anything. 

We all need to recognize that the more constructive and optimistic we are, the better for everybody's net worth.

Success is about attracting new entrepreneurs and users and a positive outlook is helpful in doing that.

I've never seen many successful coaches give a pessimistic "pep talk" in the locker room at half time.  Slamming doors and throwing chairs however, seems to be fair game...

This forum is our team's locker room.

[monsterer]

I agree that this is not widely understood, but that is not a knock on DPOS, just on our ability to overcome false narratives that occur when even our long term members get careless about what they say.

I am stating a hard truth here in the hopes that it will get the powers that be to think about the problem more clearly.

I'll say it again:

At this very moment, 1 single bitshares account could vote out all the delegates and install its own for whatever purposes they liked. That means 1 account controls the entire network, which is far, far more centralised than bitcoin, nxt or any other cryptocurrency except maybe ripple.

[Empirical1.2]

So, I don't think it's a case that people are happy with the current level of centralization but that they don't feel they can't individually help change it or they would rather sell and invest in other options they perceive as not being as centralized.

So in other words, you assume that those who sell are those who would like to vote but never did as they got disappointed that others don't vote.
Quite risky hypothesis.

This one looks to me more likely:
Shareholders don't vote because they don't perceive decentralization important enough.

Not necessarily like to vote, they may be apathetic too, but want a system that provides greater decentralization by design without them having to do anything.

Also people don't like investing in the first place in 'decentralized' blockchains that are perceived to be 'centralized'. So for example some people would say NXT was only given to 73 people, I'm not investing in that. Or that had a huge premine, I don't want to invest in that...

So it stands to reason that once they perceive the blockchain they are currently invested is/has become too centralized they may choose to sell. 

Especially because in BTS the possibility exists to pay yourself a salary or pay yourself a contract, perhaps grossly overcharging for it.
Whereas in other systems they often receive the same pre-defined pay relative to work so centralization is fairly moot from that perspective.

I usually think through my answers a bit more,  just responding very quickly today  

[jakub]

So, I don't think it's a case that people are happy with the current level of centralization but that they don't feel they can't individually help change it or they would rather sell and invest in other options they perceive as not being as centralized.

So in other words, you assume that those who sell are those who would like to vote but never did as they got disappointed that others don't vote.
Quite risky hypothesis.

This one looks to me more likely:
Shareholders don't vote because they don't perceive decentralization important enough.

[Empirical1.2]

Buy and hold more BTS.

Vote with it.

Tell your friends to do the same.

Problem solved.

Good luck convincing many friends to buy shares in centralized blockchain with shareholder directed dilution 

Far less centralized and far less dilution that Bitcoin.

2 years and you still fail to grasp the elementary difference between 'immutable' and 'shareholder directed' in terms of trust required

Edit, Also perception is important. I think the general perception is that BTC is more decentralized than it is, among other things because there is a belief that hashers will quickly switch pools if they need too.

As we learned back in the ProtoShares days, hashers will continue to support a slightly more profitable pool even if it blatantly refuses to include any transactions.

The max dilution rate of both is immutable, but BTS shareholders can dial it to zero.  The max rate of Bitcoin is also the min rate and it all gets wasted, while whatever the much smaller BTS rate becomes gets spent on what the shareholders think is most useful.

There is no comparison.

I agree that this is not widely understood, but that is not a knock on DPOS, just on our ability to overcome false narratives that occur when even our long term members get careless about what they say.

I agree with you about the hashers, this is not the widely held perception though.

The problem is if BTS is perceived as centralized & those centralized shareholders can direct and increase dilution then that requires a lot more trust.

As we saw with the merger, diluting BTSX for DNS and Vote lost us value, so we can see shareholders can make poor decisions with dilution and to date haven't been able to use it for anything that's been able to rapidly grow BTS vs. more 'wasteful' competitors like many hoped.

The same perceived centralized power group also had to pay themselves salaries to carry on working on BTS, including yourself, however BTS continued losing value during that time, so even though you and others were working far in excess of the value of that salary, it wasn't translating into added value for the DAC/shareholders. So from that perspective, dilution in a DAC still has to prove itself as well.

So when a small group, due to voter apathy/other has the power to direct dilution including towards mergers, this requires a great deal more trust than a dilution rate that has already been pre-defined and given BTS have thus far failed to make gains vs. competitors, trusting a fairly centralized group with that power, when you feel you are an insignificant sized shareholder,  requires even more trust going forward imo. 

I agree that this is not widely understood, but that is not a knock on DPOS, just on our ability to overcome false narratives that occur when even our long term members get careless about what they say.

I think you also get careless sometimes, primarily out of enthusiasm and as a result create a credibility problem for BTS.

A long term member like me, could be more supportive & focus on the positives, but unfortunately it's in my nature to focus on things I see as weaknesses and I realize I probably rank around the newmine's and tonyk's of this world that criticize & find fault too much at times.