EB94 – Gavin Andresen: On The Blocksize And Bitcoin's Governance

[tvbcof]

Bitcoin, however, is a peer-to-peer IT community, so the process of problem-solving involves a lot more email shouting, forum harrumphing, social media trolling, barista bating and generally singeing each other’s ironic Edwardian beards.

It's an upgrade from having decisions made in the cloistered and opaque bowels of The Bitcoin Foundation.  Thankfully someone leaked the discussion Hearn provoked about coin tainting and that little gem started being bandied about in public.  If not, the whole rotten organization may not have crumbled.

[meono]

That was a great interview. A lot of good point have been made. Worth watching.

I was happy to hear Gavin say that he wants to "get to the point where there will be multiple robust implementations of the core protocol."  Based on the opposition I've received here to similar ideas I've proposed, I thought that perhaps I was missing something.  I'm more confident now that I'm not.  We need to decentralize development.  

This gave me a new talking point: "Let's kill Bitcoin Core and allow the green shoots of a garden of new implementations to grow from its fertile ashes"

More info on the idea: http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-September/010802.html

Yup bitcoin core is dead. Many newbies here still can't grasp this simple concept.

Forking is by design feature of bitcoin. Bitcoin core is nothing but an implementation of the protocol. It is NOT the protocol itself.

[brg444]

That was a great interview. A lot of good point have been made. Worth watching.

I was happy to hear Gavin say that he wants to "get to the point where there will be multiple robust implementations of the core protocol."  Based on the opposition I've received here to similar ideas I've proposed, I thought that perhaps I was missing something.  I'm more confident now that I'm not.  We need to decentralize development.  

This gave me a new talking point: "Let's kill Bitcoin Core and allow the green shoots of a garden of new implementations to grow from its fertile ashes"

More info on the idea: http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-September/010802.html

Yup bitcoin core is dead. Many newbies here still can't grasp this simple concept.

Forking is by design feature of bitcoin. Bitcoin core is nothing but an implementation of the protocol. It is NOT the protocol itself.

 

How is it "dead"? Who's taken its place?

[meono]

Which won't before getting consensus. Meanwhile it's just an alt-client.

No, it's not an alt-client.  Electrum, or Multibit or MyCelium or Synala or BitWasp are alt-clients.  They all still follow the Bitcoin protocol.

XT on the other hand is creating their own blockchain, hence it's an alt-coin.

By the same logic, what you're using now is not bitcoin dumbass.

[Peter R]

That was a great interview. A lot of good point have been made. Worth watching.

I was happy to hear Gavin say that he wants to "get to the point where there will be multiple robust implementations of the core protocol."  Based on the opposition I've received here to similar ideas I've proposed, I thought that perhaps I was missing something.  I'm more confident now that I'm not.  We need to decentralize development.  

This gave me a new talking point: "Let's kill Bitcoin Core and allow the green shoots of a garden of new implementations to grow from its fertile ashes"

More info on the idea: http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-September/010802.html

Yup bitcoin core is dead. Many newbies here still can't grasp this simple concept.

Forking is by design feature of bitcoin. Bitcoin core is nothing but an implementation of the protocol. It is NOT the protocol itself.

 

How is it "dead"? Who's taken its place?

Nothing is dead.  What is slowly dying, however, is the idea that the implementation called "Core" is somehow core to Bitcoin.

Core is locked-up; Wladimir won't integrate changes without Core Dev consensus. With all the arguing, I think there's now too much bad blood around Core.  Why not let it go?  Let's not reach consensus with Bitcoin Core.  This will then provide impetus for new implementations to fork from Core (like XT did) and implement whatever scaling solution they deem best.  The users will then select the winning solution simply based on the code they choose to run.  The other implementations will then rush to make compatible changes in order to keep their dwindling user bases.  

This is the decentralized spirit of Bitcoin in action.  Creative destruction.  Consensus formed simply by the code we run.  

[luv2drnkbr]

This gave me a new talking point: "Let's kill Bitcoin Core and allow the green shoots of a garden of new implementations to grow from its fertile ashes"

We have those.  They're called alt-coins, and they all suck.

Not to mention, they're all based off Bitcoin Core code.

Alt-clients =! alt-coins

There is a distinction.

Alt-client =! Bitcoin XT

There is a distinction.

Nope, my bitcoin XT is compatible with your bitcoin Core. We can send bitcoins to each others.

Otherwise in which category you would put XT?

XT is attempting to fork the network and change the consensus rules. It is no longer simply "just another implementation".

By that logic, every patch and fix to Bitcoin is an alt-coin.  The original limit of putting in the 1mb cap was an alt-coin from the original.

In fact, XT is *less* of an alt-coin than that.  Even if XT gets its 75%, if everybody then suddenly decides "nevermind I'm sticking to 1mb", those 1mb blocks will still be valid according to it, and it will still follow the long chain proof of work.  So this change is actually less incompatible than numerous soft and hard forks of the past, yet you consider all of those "Bitcoin" but somehow XT is an alt-coin.

"Bitcoin" is whatever the community determines it to be.  If BIP101 gains traction and becomes what people use, then "Bitcoin" will be that, and everybody else in the world except you will consider your 1mb chain to be an alt-coin.

XT doesn't change a damned thing unless and until the community agrees that the change needs to be implemented.  It doesn't activate unless and until there's already consensus.  To claim that it's a contentious fork, or that it's an alt-coin, is just factually incorrect.

[meono]

That was a great interview. A lot of good point have been made. Worth watching.

I was happy to hear Gavin say that he wants to "get to the point where there will be multiple robust implementations of the core protocol."  Based on the opposition I've received here to similar ideas I've proposed, I thought that perhaps I was missing something.  I'm more confident now that I'm not.  We need to decentralize development.  

This gave me a new talking point: "Let's kill Bitcoin Core and allow the green shoots of a garden of new implementations to grow from its fertile ashes"

More info on the idea: http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-September/010802.html

Yup bitcoin core is dead. Many newbies here still can't grasp this simple concept.

Forking is by design feature of bitcoin. Bitcoin core is nothing but an implementation of the protocol. It is NOT the protocol itself.

 

How is it "dead"? Who's taken its place?

Nothing is dead.  What is slowly dying, however, is the idea that the implementation called "Core" is somehow core to Bitcoin.

Core is locked-up; Wladimir won't integrate changes without Core Dev consensus. With all the arguing, I think there's now too much bad blood around Core.  Why not let it go?  Let's not reach consensus with Bitcoin Core.  This will then provide impetus for new implementations to fork from Core (like XT did) and implement whatever scaling solution they deem best.  The users will then select the winning solution simply based on the code they choose to run.  The other implementations will then rush to make compatible changes in order to keep their dwindling user bases.  

This is the decentralized spirit of Bitcoin in action.  Creative destruction.  Consensus formed simply by the code we run.  

Thats the beauty of permissionless system.

These damn fools somehow believe there must be a "permit" (which they call "consensus" without understanding wtf that means) to implement a new protocol change.

Not to mention there is NO way to enforce what they're preaching but go on forums and attack everyone who agrees with the change. Heck even using character assassination against the coder as a way to judge his code/idea

Lucky humanity only contains small amount of these fools.

[meono]

This gave me a new talking point: "Let's kill Bitcoin Core and allow the green shoots of a garden of new implementations to grow from its fertile ashes"

We have those.  They're called alt-coins, and they all suck.

Not to mention, they're all based off Bitcoin Core code.

Alt-clients =! alt-coins

There is a distinction.

Alt-client =! Bitcoin XT

There is a distinction.

Nope, my bitcoin XT is compatible with your bitcoin Core. We can send bitcoins to each others.

Otherwise in which category you would put XT?

XT is attempting to fork the network and change the consensus rules. It is no longer simply "just another implementation".

By that logic, every patch and fix to Bitcoin is an alt-coin.  The original limit of putting in the 1mb cap was an alt-coin from the original.

In fact, XT is *less* of an alt-coin than that.  Even if XT gets its 75%, if everybody then suddenly decides "nevermind I'm sticking to 1mb", those 1mb blocks will still be valid according to it, and it will still follow the long chain proof of work.  So this change is actually less incompatible than numerous soft and hard forks of the past, yet you consider all of those "Bitcoin" but somehow XT is an alt-coin.

"Bitcoin" is whatever the community determines it to be.  If BIP101 gains traction and becomes what people use, then "Bitcoin" will be that, and everybody else in the world except you will consider your 1mb chain to be an alt-coin.

XT doesn't change a damned thing unless and until the community agrees that the change needs to be implemented.  It doesn't activate unless and until there's already consensus.  To claim that it's a contentious fork, or that it's an alt-coin, is just factually incorrect.

A simple way to put this is: bitcoin become bitcoinXT .... If they really care about the name.

[brg444]

That was a great interview. A lot of good point have been made. Worth watching.

I was happy to hear Gavin say that he wants to "get to the point where there will be multiple robust implementations of the core protocol."  Based on the opposition I've received here to similar ideas I've proposed, I thought that perhaps I was missing something.  I'm more confident now that I'm not.  We need to decentralize development.  

This gave me a new talking point: "Let's kill Bitcoin Core and allow the green shoots of a garden of new implementations to grow from its fertile ashes"

More info on the idea: http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-September/010802.html

Yup bitcoin core is dead. Many newbies here still can't grasp this simple concept.

Forking is by design feature of bitcoin. Bitcoin core is nothing but an implementation of the protocol. It is NOT the protocol itself.

 

How is it "dead"? Who's taken its place?

Nothing is dead.  What is slowly dying, however, is the idea that the implementation called "Core" is somehow core to Bitcoin.

Core is locked-up; Wladimir won't integrate changes without Core Dev consensus. With all the arguing, I think there's now too much bad blood around Core.  Why not let it go?  Let's not reach consensus with Bitcoin Core.  This will then provide impetus for new implementations to fork from Core (like XT did) and implement whatever scaling solution they deem best.  The users will then select the winning solution simply based on the code they choose to run.  The other implementations will then rush to make compatible changes in order to keep their dwindling user bases.  

This is the decentralized spirit of Bitcoin in action.  Creative destruction.  Consensus formed simply by the code we run.  

I think you're just trying to "teach the controversy" and turn the whole nature of this debate around.

I think the bad blood you are describing is nothing but generally healthy debate and a necessary conservative attitude toward a very fragile consensus system.

Let go of what? What do you propose replaces Bitcoin Core?

The idea you describe is the natural result of the most technically-able developers having regrouped around was was originally the reference implementation.

There are already multiple different implementations of Bitcoin. Gavin himself states in his interview we shouldn't need more than 3-4. Do you understand why some are more popular than others? No one is stopping anyone from creating different implementation so I'm not sure I understand the apparent need for a change in attitude here?

I think s7r had an absolutely down to earth response to what I can only qualify as your "flavor-of-the-day" concern trolling on the Bitcoin dev list: http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-August/010796.html . If you are really being honest (which I'm increasingly beginning to think you are not) then I think it's worth considering your "concerns" are overblown. Stop buying everything Mike Hearn tries to sell you.

[brg444]

That was a great interview. A lot of good point have been made. Worth watching.

I was happy to hear Gavin say that he wants to "get to the point where there will be multiple robust implementations of the core protocol."  Based on the opposition I've received here to similar ideas I've proposed, I thought that perhaps I was missing something.  I'm more confident now that I'm not.  We need to decentralize development.  

This gave me a new talking point: "Let's kill Bitcoin Core and allow the green shoots of a garden of new implementations to grow from its fertile ashes"

More info on the idea: http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-September/010802.html

Yup bitcoin core is dead. Many newbies here still can't grasp this simple concept.

Forking is by design feature of bitcoin. Bitcoin core is nothing but an implementation of the protocol. It is NOT the protocol itself.

 

How is it "dead"? Who's taken its place?

Nothing is dead.  What is slowly dying, however, is the idea that the implementation called "Core" is somehow core to Bitcoin.

Core is locked-up; Wladimir won't integrate changes without Core Dev consensus. With all the arguing, I think there's now too much bad blood around Core.  Why not let it go?  Let's not reach consensus with Bitcoin Core.  This will then provide impetus for new implementations to fork from Core (like XT did) and implement whatever scaling solution they deem best.  The users will then select the winning solution simply based on the code they choose to run.  The other implementations will then rush to make compatible changes in order to keep their dwindling user bases.  

This is the decentralized spirit of Bitcoin in action.  Creative destruction.  Consensus formed simply by the code we run.  

Thats the beauty of permissionless system.

These damn fools somehow believe there must be a "permit" (which they call "consensus" without understanding wtf that means) to implement a new protocol change.

Not to mention there is NO way to enforce what they're preaching but go on forums and attack everyone who agrees with the change. Heck even using character assassination against the coder as a way to judge his code/idea

Lucky humanity only contains small amount of these fools.

Stop trying to kill strawmans. No one ever said anyone should need a permit to implement protocol change.

What we ask is to let code stand on its own. Gavin has been publicly lobbying the industry to support its BIP101 and maintained a steady PR campagin in the last couple months using blog posts, MSM interview to try to converge support for its "solution" by steering the masses using populist ideas and appeal to authority.

That is what's "wrong" about the whole XT fiasco.

[Peter R]

Let go of what? What do you propose replaces Bitcoin Core?

Probably 3 or 4 forks from Core, XT being one of them.  Another probably implementing BIP100.  

The idea you describe is the natural result of the most technically-able developers having regrouped around was was originally the reference implementation.

But two of these "most technically-able developers" have now regrouped around XT.  This regrouping seems like a useful method to resolve impasses (e.g., XT has now implemented several features that were blocked by Core).  

There are already multiple different implementations of Bitcoin. Gavin himself states in his interview we shouldn't need more than 3-4.

I agree.

No one is stopping anyone from creating different implementation so I'm not sure I understand the apparent need for a change in attitude here?

I agree. But I thought you didn't.  Didn't you say that what Gavin and Mike are doing is some how a coup and is unethical?

I think s7r had an absolutely down to earth response to what I can only qualify as your "flavor-of-the-day" concern trolling on the Bitcoin dev list: http://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-August/010796.html

He said that he agreed that multiple implementations would be helpful, but he was worried that Bitcoin Core was the only really stable one at the moment.  I pointed out how that's not a problem, since the new implementations would fork from Core.  

He also said that the fact that anyone can contribute code means Bitcoin Core is decentralized.  I disagree.  We already know that ideas that were allowed in XT were blocked from Core.  Therefore, we need mulitple implementations run by people of various ideologies, to gain more diversity.  

Lastly, he didn't address the main problem with the current centralized development model: "With this level of centralization, it may be possible in the future for a group of coders to prevent important changes from being made in a timely fashion (e.g., should their interests no longer align with those of the larger Bitcoin community)."  

[brg444]

But two of these "most technically-able developers" have now regrouped around XT.  This regrouping seems like a useful method to resolve impasses (e.g., XT has now implemented several features that were blocked by Core).

For absolutely valid reasons. Are you for trust-adding censorship lists Peter? You realize these did not make Core because there was a majority consensus against them and not because a "rogue" dev exercised his veto.

I agree. But I thought you didn't.  Didn't you say that what Gavin and Mike are doing is some how a coup and is unethical?

I absolutely did because it is. You "thought I didn't" because you suffer from the same sick tendency to build strawmen suggesting we are somehow against permissionless innovation instead of understanding and recognizing our arguments. To quote:

What we ask is to let code stand on its own. Gavin has been publicly lobbying the industry to support its BIP101 and maintained a steady PR campagin in the last couple months using blog posts, MSM interview to try to converge support for its "solution" by steering the masses using populist ideas and appeal to authority.

That is what's "wrong" about the whole XT fiasco.

Lastly, he didn't address the main problem with the current centralized development model: "With this level of centralization, it may be possible in the future for a group of coders to prevent important changes from being made in a timely fashion (e.g., should their interests no longer align with those of the larger Bitcoin community)."  

This would imply all of the independent devs working under core have been compromised. Highly unlikely. If a consensus is reached that an important change needs to be made action will be taken and those publicly working against consensus will be evicted from the decision process or the consensus of developers will move to a new implementation if the git repo has somehow been hi jacked.

Now, assuming your suggestion to dissolve core makes any sense at all what do you propose happens of the dozen of developers currently maintaining it? They just go their merry way?

I mean.. seriously Peter get a grip. Please don't go full retard. You know you're about to cross that line when you post such a stupid topic to the Bitcoin dev list :

Let's kill Bitcoin Core and allow the green shoots of a garden of new implementations to grow from its fertile ashes

Do you seriously expect people to take you seriously?

[Peter R]

Now, assuming your suggestion to dissolve core makes any sense at all what do you propose happens of the dozen of developers currently maintaining it? They just go their merry way?

Not at all.  They are talented coders who love doing what they are doing.  I suggest they form groups of like-minded people and fork Core into their own competing implementations.  Based on their reputations and good-will, they will no doubt attract strong user bases.  These people are assets to the Bitcoin community.  

Brg444, are you going to the Scalability Conference in Montreal?  I'd like to meet you.  

[S4VV4S]

Bitcoin doesn't need governance: Bitcoin is governance.

PS: Ph0rk you Gavin!

Lessons in propaganda: teaching the controversy.

"This is not about the block size but really about Bitcoin governance. It doesn't suit populist interests and we should change it!"

Hmmm, interesting how people must politicize everything, to satisfy their egos and make themselves feel good and important, even though the things they are trying to politicize weren't meant to get politicized and we're invented exactly to fight against this politicization.

True, but like you said, people must politicize everything, to satisfy their egos and make themselves feel good and important.
Think about it.
Bitcoin is open source, therefor anyone can modify the code.
However, if they become it's "politicians", then they have control (over something that is not mean to be controlled - ironic).

[brg444]

Now, assuming your suggestion to dissolve core makes any sense at all what do you propose happens of the dozen of developers currently maintaining it? They just go their merry way?

Not at all.  They are talented coders who love doing what they are doing.  I suggest they form groups of like-minded people and fork Core into their own competing implementations.  Based on their reputations and good-will, they will no doubt attract strong user bases.  These people are assets to the Bitcoin community.  

Brg444, are you going to the Scalability Conference in Montreal?  I'd like to meet you.  

Then you should understand it is inevitable they all reconcile under the same implementation because it happens that although they sometimes have disagreements they are like-minded people and have shown an ability to work relatively well together. I haven't seen any sign of interest from any developer to jump-start their own implementation given the gigantic task this implies. Maybe what you want is for them to change the name (Bitcoin Core)?

I am still pondering the idea of going to the conference. I am concerned about my ability to provide sufficient insight given the apparently very technical focus. (I'm not a developer) That and the 150$ ticket. That said, I've never really had the chance to meet-up Bitcoin folks in person as the community around me is practically inexistant so I am somewhat curious. Guess I have a day or so to decide...!

[WhatTheGox]

We defo need more of the power in bitcoin decentralized.  We have 5 high priests? wow do we.. really like it if everyone was totally equal and computing power was the only winner.

[brg444]

We defo need more of the power in bitcoin decentralized.  We have 5 high priests? wow do we.. really like it if everyone was totally equal and computing power was the only winner.

Get away from my Bitcoin you socialist shill.

No one in this world is equal.

[Denker]

I'll tell you one thing:  That French guy in the advertisement needs to be popping that zit.  The thing is trophy class!

If he put the event on youtube he could probably make more in views than he could from running that wallet service thing he's got going.

Oh my god this is so mean what you said but I had to laugh about it to be honest. Damn now I have pictures in my mind where he is standing infront of a mirror using his two index fingers trying to pop that thing. God damn it!

[meono]

Now, assuming your suggestion to dissolve core makes any sense at all what do you propose happens of the dozen of developers currently maintaining it? They just go their merry way?

Not at all.  They are talented coders who love doing what they are doing.  I suggest they form groups of like-minded people and fork Core into their own competing implementations.  Based on their reputations and good-will, they will no doubt attract strong user bases.  These people are assets to the Bitcoin community.  

Brg444, are you going to the Scalability Conference in Montreal?  I'd like to meet you.  

Then you should understand it is inevitable they all reconcile under the same implementation because it happens that although they sometimes have disagreements they are like-minded people and have shown an ability to work relatively well together. I haven't seen any sign of interest from any developer to jump-start their own implementation given the gigantic task this implies. Maybe what you want is for them to change the name (Bitcoin Core)?

I am still pondering the idea of going to the conference. I am concerned about my ability to provide sufficient insight given the apparently very technical focus. (I'm not a developer) That and the 150$ ticket. That said, I've never really had the chance to meet-up Bitcoin folks in person as the community around me is practically inexistant so I am somewhat curious. Guess I have a day or so to decide...!

Ofcourse you cant add crap. From your dumb post i know you dont have any technical knowledge at all. Lets be real here. You came to bitcoin as a speculator/trader. I guess your hate toward Gavin and Mike is because you believed they cause the price to fall.

Your lack of understanding bitcoin economy is amusing. Like you think miners are the one who gives bitcoin value. When did you join the bitcoin world again? After the price soared i bet.

[johnyj]

IT guys do not pay enough attention to an important fact that a monetary system is not a software project, they have totally different priorities. Thus any kind of change that is normal for a software project can be devastating for a monetary system

Imagine that Federal Reserve change the USD version every 3 months, and even fork it once a while, so that USD from New York is not interchangeable with USD from California, just because New York FED and San Francisco FED have different view on QE, then the whole US economy will be in chaos like what happened in North Korea