Meni Rosenfeld on XT: https://www.reddit.com/r/Bitcoin/comments/3h9cq4/its_time_for_a_break_about_the_recent_mess/cu6udfe
Thanks for the that.
Quote
Bitcoin-XT represents a somewhat reckless approach, which in the name of advancement shatters existing structures, fragments the community, and spins the ecosystem into chaos.
A new enemy of Frap.doc and the Gavinistas emerges!
Let's see what kind of poo they fling this time.
If you're going to quote Meni at least have some respect and not take it out of context. here is the full post in it's entirety
Quote
MeniRosenfeldMeni Rosenfeld - Bitcoin Expert 172 points 9 hours ago*x2
Sorry for being the odd one out in theymos' list, not having my stance on the issue clear
. I will clarify it, but only after I explain at length why I reject the whole premise of the question.
This used to be a technical debate about block sizes. We were presented with two bad choices: Keep it at 1MB which is obviously too low, or increase to 8MB/20MB which is obviously too high (obvious to me anyway). As a community we've failed to reach a compromise, and I think that if more people pushed for a reasonable number like 3-4 MB in the short term (including also Gavin and Mike on one extreme, and Peter on the other), things would be different now.
Given that compromise failed, Gavin and Mike went ahead to push Bitcoin-XT, and now the real issue isn't about technology, it's about the philosophy of Bitcoin evolution. To me, Bitcoin-XT represents a somewhat reckless approach, which in the name of advancement shatters existing structures, fragments the community, and spins the ecosystem into chaos. Whereas Bitcoin Core represents a frigid approach, where no technology improvement will ever be made because consensus can't be reached, and where we can't do anything about the fact that we can't do anything, because of the delegitimization of attempts to change the status quo by forking and letting the best currency win (and make no mistakes, there will be many necessary technology improvements down the road; the block size limit pales in comparison).
Both choices are bad.
I had plans once to write a paper about the game-theoretic aspects of changing the Bitcoin protocol, the contingencies in case of a fork, and how the mere threat of a fork can create a Schelling point which would prevent it from happening. I regret never getting around to it, because it might have been illuminating in the current debate. (For that matter, the other paper I had plans for writing was about transaction fees and how they relate to things like a block size limit; I regret that too, but again the block size is no longer the real issue).
But anyway, I do strongly believe that the possibility of forking Bitcoin - even if at first it has no consensus - is vital to Bitcoin's health, growth and survival. It's the glue that holds everything together and makes sure the Bitcoin economy has a say in case something goes wrong with the development. Ideally a contentious fork would forever remain a theoretical possibility - but if it is possible it means it can happen, and that's what we're seeing right now. Rejecting a fork on the grounds that it's a fork is wrong.
Of course, there are grounds to reject Bitcoin-XT on the grounds that its timing and method are wrong. The objection to the technical change was too strong to just gloss over it. We're definitely not anywhere near the point where Pieter, Greg or Wladimir can be conceivably considered rogue and we should break away from them. Mike and Gavin have, quite arrogantly I would say, assumed that this is just like any other software change and that virtually everyone will just automatically follow them, where the reality is far from it. They didn't properly consider the consequences of making this move without enough support. (They might reply that they have been fighting for this for a long time and exhausted all other options, but I don't accept this - they made many attempts at persuasion, but not enough at compromise).
So here we are, having to choose between two bad alternatives. The mere act of choosing commits, in my opinion, the logical fallacy of privileging the hypothesis. There are millions of possible approaches, and "someone" out there restricted them to just 2. Most of the decision process (culling from millions to 2) was made for me and I'm left rubber-stamping one of the choices that remain. (See also http://lesswrong.com/lw/mi/stop_voting_for_nincompoops/).
But hey, even a noisy bit contains some information, and the question was asked, so...
Given the choice between short-term sticking with 1MB or going all the way to 8MB, I am in favor of going to 8MB.
Given the choice between sticking with Bitcoin Core or switching to Bitcoin-XT, I am in slight favor of sticking with Bitcoin Core, but that could change any time.
All that said, the parent post explaining theymos' policy makes no sense to me. As explained above, the possibility of forking is an integral part of Bitcoin. As long as Bitcoin-XT has non-negligible support as the true Bitcoin implementation, even with nothing resembling unanimous consensus, it is a part of what Bitcoin is, and of course is on topic on a Bitcoin subreddit.
Even if for some reason we take a purist approach that Bitcoin = Bitcoin Core, I'd imagine that most posts about Bitcoin-XT would compare it in some way to Bitcoin Core, and as such are on-topic (just like a post comparing Bitcoin and Litecoin would be on topic).
I'm considering upgrading the above to a post, but honestly, since it includes a discussion of Bitcoin-XT, I'm not even sure it passes the moderation rules...
Sorry for being the odd one out in theymos' list, not having my stance on the issue clear
. I will clarify it, but only after I explain at length why I reject the whole premise of the question.This used to be a technical debate about block sizes. We were presented with two bad choices: Keep it at 1MB which is obviously too low, or increase to 8MB/20MB which is obviously too high (obvious to me anyway). As a community we've failed to reach a compromise, and I think that if more people pushed for a reasonable number like 3-4 MB in the short term (including also Gavin and Mike on one extreme, and Peter on the other), things would be different now.
Given that compromise failed, Gavin and Mike went ahead to push Bitcoin-XT, and now the real issue isn't about technology, it's about the philosophy of Bitcoin evolution. To me, Bitcoin-XT represents a somewhat reckless approach, which in the name of advancement shatters existing structures, fragments the community, and spins the ecosystem into chaos. Whereas Bitcoin Core represents a frigid approach, where no technology improvement will ever be made because consensus can't be reached, and where we can't do anything about the fact that we can't do anything, because of the delegitimization of attempts to change the status quo by forking and letting the best currency win (and make no mistakes, there will be many necessary technology improvements down the road; the block size limit pales in comparison).
Both choices are bad.
I had plans once to write a paper about the game-theoretic aspects of changing the Bitcoin protocol, the contingencies in case of a fork, and how the mere threat of a fork can create a Schelling point which would prevent it from happening. I regret never getting around to it, because it might have been illuminating in the current debate. (For that matter, the other paper I had plans for writing was about transaction fees and how they relate to things like a block size limit; I regret that too, but again the block size is no longer the real issue).
But anyway, I do strongly believe that the possibility of forking Bitcoin - even if at first it has no consensus - is vital to Bitcoin's health, growth and survival. It's the glue that holds everything together and makes sure the Bitcoin economy has a say in case something goes wrong with the development. Ideally a contentious fork would forever remain a theoretical possibility - but if it is possible it means it can happen, and that's what we're seeing right now. Rejecting a fork on the grounds that it's a fork is wrong.
Of course, there are grounds to reject Bitcoin-XT on the grounds that its timing and method are wrong. The objection to the technical change was too strong to just gloss over it. We're definitely not anywhere near the point where Pieter, Greg or Wladimir can be conceivably considered rogue and we should break away from them. Mike and Gavin have, quite arrogantly I would say, assumed that this is just like any other software change and that virtually everyone will just automatically follow them, where the reality is far from it. They didn't properly consider the consequences of making this move without enough support. (They might reply that they have been fighting for this for a long time and exhausted all other options, but I don't accept this - they made many attempts at persuasion, but not enough at compromise).
So here we are, having to choose between two bad alternatives. The mere act of choosing commits, in my opinion, the logical fallacy of privileging the hypothesis. There are millions of possible approaches, and "someone" out there restricted them to just 2. Most of the decision process (culling from millions to 2) was made for me and I'm left rubber-stamping one of the choices that remain. (See also http://lesswrong.com/lw/mi/stop_voting_for_nincompoops/).
But hey, even a noisy bit contains some information, and the question was asked, so...
Given the choice between short-term sticking with 1MB or going all the way to 8MB, I am in favor of going to 8MB.
Given the choice between sticking with Bitcoin Core or switching to Bitcoin-XT, I am in slight favor of sticking with Bitcoin Core, but that could change any time.
All that said, the parent post explaining theymos' policy makes no sense to me. As explained above, the possibility of forking is an integral part of Bitcoin. As long as Bitcoin-XT has non-negligible support as the true Bitcoin implementation, even with nothing resembling unanimous consensus, it is a part of what Bitcoin is, and of course is on topic on a Bitcoin subreddit.
Even if for some reason we take a purist approach that Bitcoin = Bitcoin Core, I'd imagine that most posts about Bitcoin-XT would compare it in some way to Bitcoin Core, and as such are on-topic (just like a post comparing Bitcoin and Litecoin would be on topic).
I'm considering upgrading the above to a post, but honestly, since it includes a discussion of Bitcoin-XT, I'm not even sure it passes the moderation rules...
