Gold collapsing. Bitcoin UP.

[Aquent]

Unfortunately, gaming is always possible. It is impossible to be 100% certain about any thing.

The miners can game the classic signalling just as they can game BU's signalling. The miners could, for example, simply mod classic to have a 1mb block while seemingly communicating to all through version that they are mining bitcoin classic blocks, etc.

Their problem however is that they have to be sure 51% of the miners are taking part in this deception, a deception not limited by technicalities, but by this breakthough game theoretical solution to the generals problem.

Now if 51% of miners so deceive then we are in a different territory, but in its absence, any miner who lacks that 51% would only be lying to himself by modding classic or lying in the signed BU blocks.

However, I do agree that classic seems safer and has miner's backing, quite a few devs working on it, etc, so it does make a perception difference as far as miners are concerned, which is why I support it. But realistically, as far as BU is concerned, it makes no difference at all in my view whether miners are choosing some fixed protocol limit which only they run, or a gui menuable limit BU style, as long as the economy nodes move onto BU.

 

[cypherdoc]

my emphasis should have been on non mining full nodes whose variety of limits will inc orphaning risk for miners.

 

[Mengerian]

Interesting insight into the mind of Theymos in a reddit thread started by Jeff Garzik to discuss moderation policy and the definition of an "altcoin".

While it is technically/economically possible for a hardfork to succeed even despite controversy, this would mean that some significant chunk of the economy has been disenfranchised: the rules that they originally agreed on have effectively been broken, if they want to continue using Bitcoin as they were before. What happened to the dream of a currency untouchable by human failings and corruption? How will we know that the bitcoins we own today will be valid tomorrow, when the rules of Bitcoin have no real solidity? The 21 million BTC limit is technically just as flexible as the max block size; given this, why should anyone ever consider Bitcoin to be a good store of value if these rules are changeable even despite significant opposition?

Especially if these controversial hardforks succeed despite containing changes that have been rejected by the technical community, or if they happen frequently, I can't see Bitcoin surviving. How would anyone be able to honestly say that Bitcoin has any value at all in these circumstances?

Therefore, I view it as absolutely essential that the Bitcoin community and infrastructure ostracize controversial hardforks. They should not be considered acceptable except maybe in cases of dire need when there is no other way for Bitcoin to survive. It's impossible to technically prevent contentious hardforks from being attempted or even succeeding, but that doesn't mean that we need to view them as acceptable or let them use our websites for promotion.

Also, I feel like it's pretty clear that the 50BTC-forever fork was not Bitcoin by any reasonable definition, since Bitcoin has no more than 21 million BTC, so I disagree with your proposed classification scheme for that reason as well.

He appears to genuinely believe that the survival of Bitcoin, and its future value, depend on the ability of the community to socially ostracize anyone attempting a hard fork.

It is hard to imagine how one could hold this view, and still have any hope that Bitcoin will be successful.

 

[Aquent]

It would only increase risks if non mining nodes choose a lower limit than miners are willing to choose. However, non mining nodes do not run orphaning risks, therefore they are far less conservative than mining nodes. As such I think it is highly unlikely that non mining nodes would choose a limit that is lower than the limit chosen by mining nodes.

The chosen limit, especially by miners, would probably be just above demand, with non mining nodes prob being a bit more free with perhaps 4-8mb. However, all this is somewhat new, so, classic does a good service of kicking the can while actual long term solutions are tested etc.

 

[Zangelbert Bingledack]

Theymos:

The 21 million BTC limit is technically just as flexible as the max block size

That use of "technically" came straight from the mouth of Gmax in his debate with Peter R on the mailing list a few months ago.

As I mentioned on reddit yesterday, these guys just look up to the Core devs. They're probably not being paid off or anything, it's just natural hero worship.

 

[cypherdoc]

Garzik's onboard. that's frigging huge. 2 of the original Core devs now on Classic:

 

[Peter R]

there's a huge difference.

b/c with BU, no one will know with perfect certainty how many nodes or miners out there will have their limit set at 2MB unless we have developed "perfect" signalling and agreement, which assumes for the moment that there won't be gaming of the situation. and both of those are impossible.

thus miners will be skiddish about inc blocksize.

Like @Aquent already said, this issue applies with all implementations. Unlimited just woke people up to this fact. As @dgenr8 said, the only way to know for sure if your block will be accepted is to go ahead and try!

Miners won't attempt bigger blocks until (a) they see user-agent strings that would support them, (b) they hear statements from industry players like Coinbase and BitPay, (c) they get a feel from online discussions that people have increased their limits, etc. This applies regardless of whether the implementations used are Core, Unlimited, Classic, or a combination of all three.

@cypherdoc: it seems your support for Unlimited is waning. Do you not think Unlimited and Classic can coexist, perhaps even be symbiotically? Would you prefer all development to centralize around a single implementation again (e.g., Classic), or be spread over three of more implementations?

Lastly, I'm starting to really like the branding idea (from Gavin) that "Classic is for industry. Unlimited is for end-users. End users won't care about the block size anyway (or they shouldn't). Unlimited has a pretty GUI, Classic is linux-only...."

What do you think about that?

 

[albin]

Garzik's altcoin definition is what everyone with any sense already intuitively believes, and he's expressed it in a very helpful and precise way.

 

[Zangelbert Bingledack]

Major dev and small-block guy Alex Morcos also in favor of competing implementations. Who is left?

 

[bitcartel]

@Peter R

@cypherdoc: it seems your support for Unlimited is waning. Do you not think Unlimited and Classic can coexist, perhaps even be symbiotically? ...
Lastly, I'm starting to really like the branding idea (from Gavin) that "Classic is for industry. Unlimited is for end-users. End users won't care about the block size anyway (or they shouldn't). Unlimited has a pretty GUI, Classic is linux-only...."

Both Classic and Unlimited are in competition for limited community resources. Based on the branding you mention above, it's likely Classic would attract the best (subjective of course) software engineers who want to focus on solving hard problems. Meanwhile Unlimited as a pretty GUI would have to find developers familiar in QT willing to maintain cross-platform builds / quash platform-specific bugs. If Unlimited were to focus on the GUI, in the long run it would probably make sense to migrate the project to CoPay in order to attract front-end developers who can skin a pretty web UI on top of new features from Classic. Either way, the result will most likely be that consensus continues to be captured by one project.

 

[sickpig]

I was waiting for this one:

 

[YarkoL]

The chosen limit, especially by miners, would probably be just above demand, with non mining nodes prob being a bit more free with perhaps 4-8mb. However, all this is somewhat new, so, classic does a good service of kicking the can while actual long term solutions are tested etc.

I think in practice full (non-miner) node settings will reflect the limits of their processing power. If the miners converge to a high enough block size, lot of nodes will have to shut down operation, or upgrade their hardware. Because of this, I'd like to see BU provide some incentive for full nodes to stay in the game, as a decentralized counterbalance to the mining industry. But I'm not sure at this stage what that could be.

 

[albin]

Peter Todd is totally jumping the shark here, because nobody except btcd and libbitcoin is actually re-implementing anything consensus-related. XT/BU/Classic/etc. all make fairly trivial consensus code changes merely to blocksize, and btcd being a true re-implementation is developed by some serious pros who really appear to know their stuff (as opposed to being experts on social media pontification) and have contributed immensely to finding edge cases. I don't know much about libbitcoin, but it appears that it's really only prominently used for the (now largely abandoned?) Darkwallet and Airbitz's servers, so the systemic risk of consensus failure is pretty contained.

 

[rocks]

Garzik's onboard. that's frigging huge. 2 of the original Core devs now on Classic:

This is what happens when you circle the wagons and kick everyone out who disagrees from the party line. You are left with a small number of people, while the many who were kicked out band together.

It is still shocking to me that Maxwell thought this would work

Edit: Question, since we never saw the big Dec accouncment that was hinted at, do you think it was bitcoin classic or something else?

 

[Aquent]

@Peter R


Both Classic and Unlimited are in competition for limited community resources.

I don't think that's fully the case. I instead agree with @Gavin Andresen 's idea of a niche. That niche can be interpreted in many ways - a miner node vs user node - but the one I want to mention and think is more important is a short term solution vs a long term solution.

Classic is a can kick and nothing else. It's the last emergency option sort of thing ala bip 102. However, we can't be forking every time we need to go from 2mb to 3mb to 4mb or whatever. Bitcoin Unlimited therefore, in my view, may have the niche of a long term solution.

 

[Justus Ranvier]

btcd being a true re-implementation is developed by some serious pros who really appear to know their stuff

The fact that they've never been acknowledged as such is a real injustice.

 

[YarkoL]

I was waiting for this one:

one should put that Satoshi quote into context.
@Gavin Andresen, it seems, had just discovered the innermost heart of Bitcoin, the Script (which among other things makes possible to verify ownership of public keys), and was wondering if a slightly less complex version could be implemented. To which Satoshi retorted that the system depends on all nodes behaving in lockstep, i.e. getting identical results from the script execution using an identical procedure. So any alternative system that tried to cut the corner by replacing Satoshi's style scripting would soon find itself being a "menace" for the network. So this is a very specific topic that does not really apply well to the debates raging now.

However, two posts later Satoshi actually speaks in more general terms about "the menace", and it would have provided Todd more convincing ammo, but I guess it was too dense to quote in a single tweet:

A second version would be a massive development and maintenance hassle for me. It's hard enough maintaining backward compatibility while upgrading the network without a second version locking things in. If the second version screwed up, the user experience would reflect badly on both, although it would at least reinforce to users the importance of staying with the official version. If someone was getting ready to fork a second version, I would have to air a lot of disclaimers about the risks of using a minority version. This is a design where the majority version wins if there's any disagreement, and that can be pretty ugly for the minority version and I'd rather not go into it, and I don't have to as long as there's only one version.

What he is describing here, is kind of a situation that we have right now, the main difference being that maybe the current "majority version" has distanced itself so far from the "first version" that it may soon find itself the "minority version". Also, had the custodians of the "majority version" been content with "airing a lot of disclaimers" instead of outright censorship, things might not have turned so ugly.

 

[Inca]

Edit: Question, since we never saw the big Dec accouncment that was hinted at, do you think it was bitcoin classic or something else?

It maybe that the spontaneous emergence and excitement around BU delayed them slightly.

Either way today seems a step towards the complete isolation of Theymos and the remaining Core fanatics on the one side, with multiple implementations including the very exciting bitcoin classic on the other.

Now we just need to see the network start to shift nodes over to XT, BU and classic and you will see the behaviour of the Core dev's and Theymos suddenly snap towards the needs of the economic majority or face oblivion.

 

[cypherdoc]

one should put that Satoshi quote into context.
@Gavin Andresen, it seems, had just discovered the innermost heart of Bitcoin, the Script (which among other things makes possible to verify ownership of public keys), and was wondering if a slightly less complex version could be implemented. To which Satoshi retorted that the system depends on all nodes behaving in lockstep, i.e. getting identical results from the script execution using an identical procedure. So any alternative system that tried to cut the corner by replacing Satoshi's style scripting would soon find itself being a "menace" for the network. So this is a very specific topic that does not really apply well to the debates raging now.

However, two posts later Satoshi actually speaks in more general terms about "the menace", and it would have provided Todd more convincing ammo, but I guess it was too dense to quote in a single tweet:

What he is describing here, is kind of a situation that we have right now, the main difference being that maybe the current "majority version" has distanced itself so far from the "first version" that it may soon find itself the "minority version". Also, had the custodians of the "majority version" been content with "airing a lot of disclaimers" instead of outright censorship, things might not have turned so ugly.

What does this say about the wide open scripting system SW proposes?

 

[Justus Ranvier]

one should put that Satoshi quote into context.

tl;dr, he was lazy and didn't want to fix the bugs and inconsistencies in his code that an independent implementation would uncover.

Or, less unkindly, the software engineering resources required for such an effort simply did not exist in Bitcoin back then.