We Don't Want Democracy, We Want Consensus!

[Peter R]

i'm telling you what the  "consensus rules" are,
Nodes express their acceptance of a new block by mining on top of it and the longest chain composed of valid transactions is Bitcoin.
this is from the white paper. and it wouldn't make any sense any other way.

Bitcoin has evolved since the whitepaper was written.  It gives a good introduction to the overall concepts, but the whitepaper gets many of the details wrong.

I think in this case, you are mistaken.  At the moment, the majority of the economic power is in agreement with the hashpower.  Perhaps this will always be true, but it doesn't have to be.

Imagine for a moment a case where 53% of the hash power chooses a particular set of consensus rules, and 47% chooses a different set of consensus rules.  Imagine that these consensus rules are such that neither side sees any of the other side's blocks as valid.  Now imagine that 98% of the users (merchants, consumers, charities, etc) all choose to use the same consensus rules as the 47% of the miners.  Which is the "real bitcoin?  I think that most people would say that the 47% fork is the "Real" bitcoin since that is what nearly everyone is using as bitcoin.  The fact that 53% of the miners have chosen to contribute their hashpower to some other chain doesn't really matter.

Now, there are some market forces in play that make this scenario VERY unlikely.  Since most people are using the 47% fork, it will be difficult for the miners on the 53% fork to find anyone that wants to buy the coins they are mining.  This will significantly reduce their ability to sell their coins and pay for their operating costs.  Therefore, there will be significant financial pressure on them to switch back to mining on the other fork.

This is what is meant when people talk about the "majority of the economic power". There is enough financial pressure on miners that they will tend to follow the desires of the economic power.  Therefore, in reality it is the economic power that decides which is the "real" bitcoin, and not the miners.

This state can't really persist, though.  The chain that only has 2% of the users behind it will rapidly lose value (assuming people can trade between chains), miners will then stop mining it, and it will be orphaned by the longer chain.  So I think you two are saying nearly the same thing. 

[DannyHamilton]

So I think you two are saying nearly the same thing. 

Perhaps, but unless I'm mistaken, adamstgBit seems to be implying that the miners can (without the consent of the economic power) choose and force a set of protocol rules on the users.  That the miners will get to choose which fork is "bitcoin", and that anyone that wants to use "bitcoin" will have to use the fork that the majority of the miners choose.

I'm saying that's backwards.  The economic power gets to choose a set of protocol rules, and the miners that wish to be profitable have to use the fork that the economic power chooses.

You are correct in that we are both saying that vast majority of the miners (those that wish to mine profitably) end up mining the "real bitcoin", but I think he and I are disagreeing on which is the cause and which is the effect.

[turvarya]

I haven't read the whole thread yet, but I have to point some things out.

I live in a country, where politics follows a consensus path, I even have some Power Point slides from some University Courses about Politics in Austria, where the word consensus is all over the place. So, I know, what consensus means in reality. It does not mean, that everybody wins. It often means nothing is happening at all(there are some reforms, which are talked about for at least 10 years even when the matter is urgent).
It also often means, that a bad compromise is made, nobody is really happy about.

I am not promoting any other kind of decision making, but the whole cheering about, how great a consensus based model is, is just purely naive, I live with the downsides of it since I am born.

[toknormal]

Mike Hearn has made it clear that he does not believe in the Consensus Building Process and is doing nothing to help in it's success. His suggestion seems to be to put in place a CENTRAL TRUSTED AUTHORITY who will make decisions for Bitcoin. To me, this goes against everything Bitcoin stands for and indicates that Hearn does not really understand the philosophy of consensus and decentralization that underlies Bitcoin.

At the risk of having to put the "chicken wire" up, I'd respectfully suggest that Mike Hearn totally understands "the philosophy of consensus and decentralization" and it's the OP of this thread who doesn't.

For a start, there are only 2 forms of "consensus" in cryptocurrency that matter:

[1] - mining consensus (decides what code gets accepted onto the network)

[2] - market consensus (decides what the result is worth)

That's it.

Bitcoin is an open source project and open source specifically implies diversity - not consensus. In other words, developers are free to create, promote and implement diverging versions of the codebase and have the stakeholder consensus mechanism (above) arbitrate amongst their offerings.

The idea of "developer consensus" alluded to in this thread by the OP is not a consensus at all but a cartel.

It appears to me that Mike Hearn understands this and the OP does not.

[OgNasty]

Bitcoin is an open source project and open source specifically implies diversity - not consensus. In other words, developers are free to create, promote and implement diverging versions of the codebase and have the stakeholder consensus mechanism (above) arbitrate amongst their offerings.

The idea of "developer consensus" alluded to in this thread by the OP is not a consensus at all but a cartel.

It appears to me that Mike Hearn understands this and the OP does not.

I disagree with your thoughts on Bitcoin and this being how open source projects work.  I've heard this argument many times and I disagree with it in this case.  The reason is simple.  

What Mike is doing isn't simply taking the code and making his changes and then releasing it on his own.  He is attempting to hijack control of the Bitcoin blockchain with his implementation.  He could have gone about releasing his altcoin like everyone else and starting a new blockchain with the Bitcoin code he took and altered like everyone else before him.  Trying to simply make himself the new Satoshi so he can make the rules as he sees fit is a dangerous precedent and I don't think it should be supported by anyone.  He's attempting a hostile takeover of a multi-billion dollar economy that was billed as being decentralized and providing freedom.

[poeEDgar]

I don't think "my way -- exponential scaling -- or the highway" is an acceptable approach. Cutting Core developers out the picture entirely and including only those who share this limiting view of what bitcoin is... this is not something I can support.

XT is above all a path toward establishing new leadership. I asked Andresen whether, if XT were to achieve full acceptance, he would then include all the earlier Bitcoin core devs in the new XT team. He replied that “[XT] will have a different set of developers. Part of the reason for forking is to have a clear decision-making process for the software development.”
http://www.newyorker.com/business/currency/inside-the-fight-over-bitcoins-future

[catch.me.if.you.can]

You dont want democracy? Who do you think you are? A dictator?

[funkenstein]

Relevant consensus tool:

http://coin-vote.com/poll/55e8a33299baadff0a34acb7

[toknormal]

What Mike is doing isn't simply taking the code and making his changes and then releasing it on his own.  He is attempting to hijack control of the Bitcoin blockchain with his implementation.  He could have gone about releasing his altcoin like everyone else and starting a new blockchain with the Bitcoin code he took and altered like everyone else before him.  Trying to simply make himself the new Satoshi so he can make the rules as he sees fit is a dangerous precedent and I don't think it should be supported by anyone.  He's attempting a hostile takeover of a multi-billion dollar economy that was billed as being decentralized and providing freedom.  It's an absolute disgrace if you ask me.

Please spare us the melodrama.

Neither MH nor any other developer are in any position to hijack anything. If he was, the entire Bitcoin economy wouldn't be worth the price of a couple of asics.

Whats happening with the divergent developer agendas is consistent with the principle of the technology - why else do you think a mining majority is required for adoption of revisions ? Developer consensus is not, was never, and never will be something that investors can implicitly expect. You can't accuse someone of attempting to "hijack" the blockchain just because they are in a minority of the developer population who have commit access. Thats ludicrous because it would imply that the commercial, investor and mining community is not at liberty to adopt a revision that doesn't have majority developer support. You're basically making the rules up as you go along.

The consensus in bitcoin is by mining majority, not developer majority. Live with it.

[DannyHamilton]

- snip -
Trying to simply make himself the new Satoshi so he can make the rules as he sees fit is a dangerous precedent and I don't think it should be supported by anyone.
- snip -

So, instead of Mike Hearn becoming "the new Satoshi so he can make the rules as he sees fit", you prefer that Wladimir van der Laan remain "the new Satoshi so he can make the rules as he sees fit".

Meh. Six-of-one, half-a-dozen of another. Different words to say the same thing. You aren't changing the concept, just which name you're currently willing to accept in the position.  Either way you have a single person with their hands on the reigns of "a multi-billion dollar economy that was billed as being decentralized and providing freedom."

Maybe I should fork my own software that builds on the Bitcoin blockchain and then I can become "the new Satoshi so I can make the rules as I sees fit".  I'll implement ONLY the 8 MB increase when/if the XT change triggers and none of Mike's other changes (including skipping the "double every 2 years" change). I won't place any votes for XT and I'll report exactly the same as Core.  That way users can continue to support the Core rules, but they won't initially be left behind if the majority join the XT bandwagon.  It'll take some of the power away from Mike, and if enough people choose to switch to my implementation after XT triggers, then perhaps it'll put pressure on Mike to remove some of the more controversial stuff from his implementation.

[adamstgBit]

So I think you two are saying nearly the same thing.  

Perhaps, but unless I'm mistaken, adamstgBit seems to be implying that the miners can (without the consent of the economic power) choose and force a set of protocol rules on the users.  That the miners will get to choose which fork is "bitcoin", and that anyone that wants to use "bitcoin" will have to use the fork that the majority of the miners choose.

I'm saying that's backwards.  The economic power gets to choose a set of protocol rules, and the miners that wish to be profitable have to use the fork that the economic power chooses.

You are correct in that we are both saying that vast majority of the miners (those that wish to mine profitably) end up mining the "real bitcoin", but I think he and I are disagreeing on which is the cause and which is the effect.

i agree, hashing power will fallow economic power and not the other way around.

in this particular case tho i was stating that bitcoin is a sort of "democracy" by definition, because bitcoin is whatever majority of users ( hashing power or BTC hodlers ) think it is.

brg444 and I were having a torllishous discussion. lol

[adamstgBit]

The consensus in bitcoin is by mining majority, not developer majority. Live with it.

ya but miners have always mined whatever the devs tell them to mine....

had all the dev been in agreement with 8MB from  the start miners would have upgraded no questions asked, like every other BIP that has come b4 this one.

in reality no one group "controls it."

miner, investors, large bitcoin entities (Exchanges, etc.. ),  devs,  all have alot of influence.

these groups are all dependent on one another , "control" is hard to pin point.

[adamstgBit]

i'm going to build a website where we can watch in real time, the blocklimit debate destroy bitcoin, by quantifying the level of disagreement among these groups.   

[johnyj]

Good points. You can never totally get rid of bad actors that intentionally disturb the decisoin making process
- snip -

I'm not talking about "bad actors".  The fact that someone has a different goal than you doesn't necessarily mean that your goal is "good" and theirs is "bad".

The point is that a "Consensus Building Process" can't work if participants goals don't align, especially if the participants are passionate about their goals.

Sure you can reduce the influence that you allow to those that disagree with the majority, but that isn't "consensus", that's "majority wins" democracy. It's just another form of "exile".  Instead of immediately rejecting their input, you slowly remove their input until their input becomes meaningless.

I'm not talking about the right or wrong of each proposal, by bad I mean someone intentionally disrupt the consensus building process, e.g. insists on his own idea without making effort to modify it or make some compromise to reach consensus

In consensus decision making model, anything that is against the consensus building is negative, the correctness of each proposal itself is irrelevant. We suppose that each proposal is correct from its specific viewing angle, but since the community as a whole must move forward, all of the proposal must be evaluated and eventually merged towards one direction which best present the overall direction

In a centralized democracy system, the minority will have to accept the rule that winners make. But in a decentralized consensus system you always have the right to leave and play your own game. If you think you are right and the rest 90% is wrong, you can just fork your own coin and make your own exchange and start to persuade merchants to accept your coin, etc...

But you might find out that doing so will be extremely difficult without the support from majority of the community, your altcoin most likely will be buried among thousands of other coins and make no sense. So when you make a decision, you will also weigh the value of community support, and since you will get maximum community support when your proposal are not refused by the majority, your proposal will be more cooperative instead of competitive

[allthingsluxury]

Everyone needs to be open to suggestions, some are just simply deeply fixed in their predetermined ideas.

[kelsey]

The consensus in bitcoin is by mining majority, not developer majority. Live with it.

ya but miners have always mined whatever the devs tell them to mine....

had all the dev been in agreement with 8MB from  the start miners would have upgraded no questions asked, like every other BIP that has come b4 this one.

in reality no one group "controls it."

miner, investors, large bitcoin entities (Exchanges, etc.. ),  devs,  all have alot of influence.

these groups are all dependent on one another , "control" is hard to pin point.

yes you can say any group has control from a certain perspective, all are dependent on each other.

[toknormal]

I'm not talking about the right or wrong of each proposal, by bad I mean someone intentionally disrupt the consensus building process, e.g. insists on his own idea without making effort to modify it or make some compromise to reach consensus

Out of interest, how do you objectively define which party has "intentionally disrupted" the consensus, or even what the consensus was ?

If two developers launch a code revision and three refuse to endorse it, how do you determine which is the "disrupting" group ? By pure majority ? Does that mean that if one person changes their mind, the other group is now the disruptive element by definition ?

This is just crazy logic because the reality is that in open source there is no such thing as "consensus" in the general case because in theory the code is open to everybody for forking. You might only have one fork available or you might have multiple. If it's the latter then the network reaches a consensus about which fork to adopt but it's amongst the NETWORK PARTICIPANTS that the consensus resides because the concept has a formal and objective definition in that context. It is a non-existent concept in the context of the developer community other than as a loose agreement to refrain from creating multiple forks which, as we've seen, isn't worth the paper it's not written on.

[uxgpf]

Maybe I should fork my own software that builds on the Bitcoin blockchain and then I can become "the new Satoshi so I can make the rules as I sees fit".  I'll implement ONLY the 8 MB increase when/if the XT change triggers and none of Mike's other changes (including skipping the "double every 2 years" change). I won't place any votes for XT and I'll report exactly the same as Core.  That way users can continue to support the Core rules, but they won't initially be left behind if the majority join the XT bandwagon.  It'll take some of the power away from Mike, and if enough people choose to switch to my implementation after XT triggers, then perhaps it'll put pressure on Mike to remove some of the more controversial stuff from his implementation.

Otherwise good idea, but BIP 101 doesn't skip every two years. It grows linearly between the checkpoints. http://imgur.com/a/VFNT9

I'm all for more options, just in case Mike does something reckless. He seems so eager to implement stuff, that I'm not sure if everything is well thought out. (and Wladimir seems to be the opposite of that). Too bad Gavin got enough of being a maintainer, I think he fit that role well.

[DannyHamilton]

Otherwise good idea, but BIP 101 doesn't skip every two years. It grows linearly between the checkpoints.

I love the fact that I still learn new stuff about bitcoin even after years of reading about it and participating in discussions.

I was about to correct you, and when I went to quote BIP101:

https://github.com/bitcoin/bips/blob/master/bip-0101.mediawiki

The maximum size shall be 8,000,000 bytes at a timestamp of 2016-01-11 00:00:00 UTC (timestamp 1452470400), and shall double every 63,072,000 seconds (two years, ignoring leap years), until 2036-01-06 00:00:00 UTC (timestamp 2083190400).

I noticed the very next sentence that I somehow managed to overlook every other time until now:

The maximum size of blocks in between doublings will increase linearly based on the block's timestamp.

Until now, I've only quickly skimmed over BIP101 to get a general idea of what it is suggesting.  This new knowledge has motivated me to carefully read through the entire thing and read up a bit more on other writings about it.  Thanks for pointing out my misunderstanding.

Clearly my idea of creating a fork of Core/XT that joins the XT network on the trigger, but freezes at 8 MB blocks wouldn't work.  It would have to accept the slowly growing blocks at least until it had support of the majority of the economic power in the system.

[BitProdigy]

Relevant consensus tool:

http://coin-vote.com/poll/55e8a33299baadff0a34acb7

This is interesting but it seems like people have incentive not to risk making a mistake with their private keys by not voting at all, and also even if everyone did vote, the more wealth equals more votes which I'm not sure is a great thing, but perhaps a necessary evil to endure.

The people who hold the most stake in bitcoin have the most weight in a decision about bitcoin. Seems to hold water.