Gold collapsing. Bitcoin UP.

[BldSwtTrs]

When could we reasonably expect to see this scaling controversy be terminated?

Months have passed and it seems we are still at the same point. I know a lot of discussions have taken place. As expected, this have gone nowhere.

Are we doomed to wait indefinitely until something bad happens or is there any hope that this issue will be solved within the foreseeable future?

For example, by reading this thread I know that Bitcon Unlimited is in preparation. When will it be released? Is it a serious contender? (ie. is there a serious chance influential people will support it?)

 

[Peter R]

@_mr_e

That's pretty interesting! With Monero's ring signatures and work like this on Directed Acyclic Graphs, I'm beginning to have a new appreciation for certain alt-coins.

"I am also wondering if it would also ever be possible to actually hand out portions of the reward based on who mined the weak blocks? Wouldn't this have the effect of completely decentralizing mining pools?"

I think that's possible, but with subchains the fees in each Δ-block effectively do the same thing. They add to the "pot" such that the

(effective block reward) = (actual block reward) + (sum of fees in subchain).

Each time a Δ-block is found we get a new black dot on the chart, never less than the block reward. This effect will become more pronounced in the future when fees are no longer small compared to the block reward.

[doublepost=1450643623][/doublepost]@_mr_e :

On second thought, I think I see your point. Subchains don't reduce variance of mining rewards, but sharing the block reward would. Hmm...need to think more about that.

 

[BldSwtTrs]

Last week I have met a guy who is working for one of the biggest bank of the world as a blockchain consultant, by doing so, he makes something like 14x the median salary of my country.

He was fervently against small blocks and think small blockist are totally nonsensical. It was interesting to see a guy technically very competent against the stance of the so called "experts".

 

[cypherdoc]

 

[VeritasSapere]

When could we reasonably expect to see this scaling controversy be terminated?

Months have passed and it seems we are still at the same point. I know a lot of discussions have taken place. As expected, this have gone nowhere.

Are we doomed to wait indefinitely until something bad happens or is there any hope that this issue will be solved within the foreseeable future?

For example, by reading this thread I know that Bitcon Unlimited is in preparation. When will it be released? Is it a serious contender? (ie. is there a serious chance influential people will support it?)

To attempt to answer this question, which really is most likely an unknown to most people including myself, unless there is a group of insiders planning something behind the scenes, which anecdotal and circumstantial evidence might suggest.

Ultimately it is up to the participants of the Bitcoin economy to bring about change, from miners, businesses and users. We all have an influence, I do my part by running nodes, mining and participating in the forum discussions. I do think that if we see Core continue to delay, we will see Bitcoin fork, this situation can not continue forever. However the truth is that nobody really knows how this is going to end, considering that this is completely unprecedented in history.

I think that this is a test of the governance mechanism within Bitcoin, I can not be sure if it will work but I believe that it will. My hope is that soon there will be an industry announcement pledging support for BIP101 or another implementation that increases the blocksize. If this does not happen it will have to be a community lead fork instead. I am prepared to do this even if we are in the minority, which I suspect we are not. The only way to find out however will be to take this action, for the moment this would still be unwarranted, until January at least, which is when BIP101 will be able to first be activated. Even if the majority of the industry is planning to do this it is possible that they have not revealed their intention clearly in order not to cause undue panic in the market.

In short, wait till January, see what happens, if BIP101 does not fork the network, we can initiate a community lead fork instead. I am also itching to see Bitcoin unlimited released as well, I am sure the theZerg is working hard on it, I am expecting it be released in a few days? Not sure how long such a development is supposed to take? I think it is very important that it is released soon though, we need more alternatives for people to choose from, this will help further decentralize development and effectively help preserve the freedom inherent within the protocol.

 

[Peter R]

Last week I have met a guy who is working for one of the biggest bank of the world as a blockchain consultant, by doing so, he makes something like 14x the median salary of my country.

He was fervently against small blocks and think small blockist are totally nonsensical. It was interesting to see a guy technically very competent against the stance of the so called "experts".

It's because most of them are not Bitcoin experts--and I hope the community is finally starting to recognize that. I would say that they are skilled technicians when it comes to the low-level implementation details of the Bitcoin Core codebase, and some are skilled at cryptography and computer science too. But that represents perhaps only a third of what it takes (IMO) to be a Bitcoin expert.

I've always been annoyed with the attitude that some engineers and scientists have of being "bored" with lower-level implementation details, viewing technicians as second-class problem solvers. In my opinion, you don't really fully understand a problem or design until you get into the messy real-world details. When I joined the community in 2013, I was surprised that we had a sort of "reverse prejudism" instead, holding people that actually do the coding in higher regard intellectually (you don't even GitHub??).

We design a lot of circuit boards in my non-bitcoin life, and a good PCB layout guy is amazing. For example, they know all this stuff about keeping noise down and telling you how the ground currents will likely flow--and they explain it to me and I'm like "wow, that totally makes sense from what I know about physics but I would have never thought about that!" But that deep expertise really only holds for this narrow sub-topic. When the conversation deviates outside of it, it becomes crystal clear that their understanding of physics or engineering is actually fairly superficial. Well yeah! Because they specialized in one particular topic and have tons of hands-on experience with it, whereas the generalist might have studied engineering or science in a formal setting for a decade or longer. To work effectively, both specialist and generalist need to understand the limitations of their knowledge.

Someone like Greg is a classic example of a talented technician with a chip on his shoulder. I'll preface this by saying that talented technicians are awesome and can make more money and be more useful than many of the hands-off "generalist" types. I used to think Greg's understanding of Bitcoin was infallible--because so much of what he says amazes me and turns out to be correct once I dig into the implementation details. But during the last year--as I've got up to speed myself--it's become increasingly clear that he actually has a fairly superficial understanding of large swaths of computer science, information theory, physics and mathematics. But he presents himself in such a way as though he's an expert at all of them. And I won't even get into his superficial understanding of economics.

In my opinion, someone like Gavin Andresen is a genuine Bitcoin expert.

 

[Zangelbert Bingledack]

I am also wondering if it would also ever be possible to actually hand out portions of the reward based on who mined the weak blocks? Wouldn't this have the effect of completely decentralizing mining pools?

My thoughts exactly. This would be huge if it could be done.

It's because most of them are not Bitcoin experts--and I hope the community is finally starting to recognize that.

Knew I'd like this post from the very first line Reddit version anyone?

Also want to say your graphics kick ass, both in how polished they are and the overall design, packing in information in novel ways (area of the circle shows the cost to doublespend, wow).

 

[rocks]

I recently completed a new paper analyzing @rocks's "mini-blockchain" idea. I renamed this idea "subchains," and examined some of its properties related to fees, equilibrium hash rates, orphaning risk, and the security of zero-confirmation transactions. I believe I've shown that security of zero-confirm transactions should be significantly enhanced (possibly neutering RBF in the process) and that fees will directly contribute to the PoW cost, refuting Maxwell's belief that “the fact that verifying and transmitting transactions has a cost isn’t enough, because all the funds go to pay that cost and none to the POW ‘artificial’ cost.” This last result is also important concerning the block size limit debate, because it was an argument used by small-block proponents in favor of a tight limit to drive up fees for security purposes.

The "subchain" technique is just one possible application of weak blocks—but it’s possibly the easiest to analyze mathematically, and it has the very nice property of adding measurable security to zero-confirmation transactions.

This is an excellent paper @Peter R. The nested-subchain concept is a great extension of the concept. I originally thought miners could make an orphan risk vs. propagation time trade-off decision by mining on X levels back in the subchain (i.e. sacrifice the last 3 subblocks or 30 seconds of fees for near instantaneous header-only propagation), but nested-subchains seem much more elegant and seem to increase the safety of including the most recent transactions.

Subchain's as a method to neuter RBF is another great angle to explore, but it depends on how the implementation used works. If subchains are implemented with the same confirmation guarantees as the main chain, then yes subchains effectively neuter RBF since the RBF double spend would be rejected.

However, if subchains are implemented simply as a communication mechanism to communicate which transactions will be in the next block (which was what I originally thought), then it is still possible for miners to include RBF double spends in later subblocks. Here they would agree to only include the 2nd RBF transaction in the real main block and ignore the original transaction. I think most here would like the guarantees of the first version to strengthen zero confirm transactions, but there are other implementations possible and we should make sure the right one becomes the accepted one.

I won't have time to give the paper the proper read it deserves for a week or two, but will try to do so soon. Thanks again for the great analysis.

 

[Inca]

The gulf in the quality and content of material here and on other fora is remarkable. I am seeing lots of new users - welcome!

I simply don't have the time to read and absorb everything! But I will read Peter R's paper.

Different noises coming from Peter Todd again suggests internal struggle and a change of direction at Core may be on the cards. The ground has shifted imperceptibly has it not?

 

[rocks]

we, as part of ecosystem, are putting political pressures. VCs on both side are exercising political pressure. A lot of others actors are involved and are making their part.

I see nothing wrong with it, to me such activities are the way the market is made.

Sure a system could become rigged as tim pass but I see it as a evolving phase through which the things move to the next iteration.

Peter Todd is of course using emotionally charged and misleading language. As @Zarathustra said, the idea that "users walking away from Core = political pressure" is a manipulative frame and we should reject it completely every time it arises.

Precisely nothing can be correctly called politics in this space, because there are no central authorities, so the term should never be used as it guarantees confusion and derailment. Would Microsoft say it was under "political pressure" to revamp Internet Explorer because users are complaining, people are switching to Chrome/Firefox and techies are advising that IE will lose its dominant market share? No, it was under market pressure.

The very fact that Maxwell and Todd use the word "political" reveals that they're stuck in the paradigm of centralized control. The frame must be thrown off as soon as it is introduced, else every subsequent statement we try to make unwittingly supports them by reaffirming the notion that real centralized control even exists in Bitcoin.

Another frame is the overarching notion that "Core = Bitcoin," and it's one big blockers fall prey to all the time. Pressure on Core is not pressure on Bitcoin. The fact that Core could succumb to pressure from governments is all the more reason not to have one monolithic implementation, ensuring there is not even the appearance of being able to exert government pressure on Bitcoin itself.

These guys' bogus ideas arising from their having one foot still stuck in the central authority paradigm will readily sway onlookers if not called out, or at least if allowed to infect even the big block / dev decentralization side of the debate, because humans are so predisposed to centralist thinking.

Decentralization of development is not politics as Maxwell tried to claim on the mailing list in response to @Peter R; it's a rejection of politics and an embracement of the market. Core is experiencing *decentralization* pressure. I won't even say competitive pressure, because the frame then turns to "competition for what?" For the Iron Throne? The One Ring? Cast it into the fires. No, in fact it is a throne of paper, or a ring of paper, crumbling as soon as you dare to look at it.

So in the end we have found another BU supporter

Agree that "politics" is a charged word and it is actually a positive source of change in today's situation.

What I was trying to express is I think a user led revolt that results in an economic majority leaving the existing system and switching to a new system, is a stronger outcome than a change being pushing into the core client under pressure and everyone still using the same core client.

The advantage of a user led revolt that results in the ecosystem switching to a new client that meets their needs better, is it demonstrates changes to Bitcoin cannot be forced it's users. This means that if governments one day try to push X change in, there is a precedent of the user base to say no thank you and switch to something else that meets their needs.

The disadvantage of the core client simply being pressured to accept BIP101, is we are still left with the idea that Bitcoin is whatever the core client says it is, and have also introduced the idea that the maintainers of the core client will include changes they disagree with due to pressure. In my mind that makes it easier for one day a government to try to push some change into the core client and have the change be accepted.

 

[Zangelbert Bingledack]

@rocks

I agree that Core getting forked off would be the Holy Grail of outcomes here. It seems we either get the small victory of Core giving in on blocksize sooner, or the bigger victory of them sliding into irrelevance, though probably later. Considering the timing, it may even be a wash. My sense is that on blocksize they'll crumple rather than cede power.

include changes they disagree with due to pressure.

I think there's a distinction to be made here with regard to "pressure":

• Faced with market pressure, the Core team's choices are (a) include the change and satisfy the market or (b) be forked off and let some other team satisfy the market. Either way the change happens.*

• Faced with government pressure, though, while presumably they would indeed have to include the change, Core would just get forked off since the market wouldn't like it. Government pressure would merely serve to force them into retirement or anonymity.

Of course, we'd all probably feel a lot safer if the precedent for forking away from the "reference" implementation had already been set before any government tries to exert such pressure. Core adapting is smoother in the short term, but creates more inertia in the long term.

*The dog wags its tail:

 

[rocks]

I am also wondering if it would also ever be possible to actually hand out portions of the reward based on who mined the weak blocks? Wouldn't this have the effect of completely decentralizing mining pools?

If you take that approach (and it was my first inclination as well), then subchains start to look equivalent to simply changing the protocol to have a very short confirmation time (i.e. blocks every 2 seconds), which improves speed but has security trade-offs.

An advantage of subchains is they provide a method to have the security benefits of longer confirmation times (we almost never see two competing branches more than 1 depth long) with the speed of shorter confirmation times. This is because subchains use shorter confirmation times for communication only, and longer confirmation times for security.

If you force fee sharing from subchain blocks at the protocol level, then the idea becomes very close to simply being a fast confirmation time. Since block rewards are a lottery that happens every 10 minutes, it seems the system is fine with the winner claiming the past 10 minutes worth of fees.

Another thing to consider is miners are not being paid for sending and receiving transactions or for building a block. All miners send and receive the same transactions over the same 10 minute period, but only one of them "wins" and the winner captures 100% of the rewards.

So it is the same to consider it part of a miner's job to build subchain blocks as part of the price just to participate in the lottery, but in the end the winner is the 1 who found the right hash.

 

[cypherdoc]

Back on the mining/energy topic, Lopp's "The Future of Bitcoin Mining" is a good read: https://medium.com/@lopp/the-future-of-bitcoin-mining-ac9c3dc39c60

And looks like the ASIC-fired water-heater coils are already under development:

great article.

Crucial to this is the idea that bitcoin generated by embedded mining is more convenient — and hence more valuable — than bitcoin bought at market price and manually moved over to the site of utility.

as proof of this, one could move the small mined amounts to a phone wallet to pay for foldapp.coffee.com coffee.

The reason I find embedded mining to be such a powerful concept is that if it catches on we will be incentivizing the general public to mine at a net loss in fiat terms. If this sounds illogical, consider the perspective that the miner will essentially be paying for some other sort of utility, but they will be doing so by “spending” electricity. That is, you can sidestep setting up a billing relationship with a new service provider and instead just add a slight bump to your electricity bill - this reduction in friction can be valuable. If a sufficient number of devices are mining a sufficient portion of the global hashrate at a net loss, this could put for-profit industrial scale miners out of business.

i find this to be a very viable concept. these little 21BC's mine quietly in the house with no need for additional cooling. they're stable and don't need constant attention yet produce a steady stream of satoshis for micropayments and coffee. the electric bill is hardly noticeable, the continuous stream calming and useful, AND are portable (meaning i can take them to Starbucks for the afternoon of studying). once these things get small enough, they can be plugged in anywhere in range of a wifi (not necessarily yours).

Many miners are treating their operation as a long term speculative play. From my conversations with miners, most do not immediately convert all of their earnings into fiat — they’ll convert what is necessary to pay for ongoing operating costs, but profits are likely to be saved in bitcoin.

as i've said many times here, this is my impression as well. i've referenced the straw poll taken by Michael Terpin at Hasher's United last year where all the miners in the audience indicated they hold as many coins as possible generated from mining. i know my own experience of 3yr of mining where i've never sold a coin. this is a good thing b/c it makes miners, esp Chinese miners, vested interests and helps validate my concept of Governing Dynamics. it also strengthens the arguments i've been making about how i believe that it IS possible to convince Chinese miners that 101 or BU is in their best interests in terms of increasing the exchange price of their holdings by inc users & merchants worldwide which should cause a huge windfall to their profits allowing them to perhaps pass the baton onto the next generation of miners outside of China (i think the miners aren't so stupid as to not realize they can be shutdown in a heartbeat) who are unconstrained by bandwidth or a repressive communist regime. or those windfall profits could allow them to wield considerable more economic influence to take down their specific GFW restraints so they could compete on an equal footing with new more bandwidth robust Western miners. remember that none of us, even the Chinese, should want Bitcoin to be constrained to the least common denominator of bandwidth technology. that is asking for trouble as Bitcoin needs to stay on the cutting edge of new developments.

the other reason the Chinese miners should want to get the exchange price up of their holdings ASAP is that 21co is coming hard and furious. as the article says, decentralization in mining is the result of embedded mining chips. i think this is a wave to be reckoned with in my short experience with them.

 

[Justus Ranvier]

The reason I find embedded mining to be such a powerful concept is that if it catches on we will be incentivizing the general public to mine at a net loss in fiat terms. If this sounds illogical, consider the perspective that the miner will essentially be paying for some other sort of utility, but they will be doing so by “spending” electricity. That is, you can sidestep setting up a billing relationship with a new service provider and instead just add a slight bump to your electricity bill - this reduction in friction can be valuable. If a sufficient number of devices are mining a sufficient portion of the global hashrate at a net loss, this could put for-profit industrial scale miners out of business.

This one of the reasons that it's far too early to worry about Bitcoin's security in 50 years when the block subsidy is negligible - we have no idea who is going to be mining then or why.

All we do know is that if Bitcoin still exists 50 years from now, there will be a very large number of people (possibly all of humanity) with very strong incentives to make sure it keeps working. They'll figure it out.

I can imagine scenarios where all mining is done at a nominal loss and no users pay transaction fees. Imagine, for example, Walmart operating in an environment where they need Bitcoin to operate in order for their cash registers to work, and to pay their suppliers and employees. They may very well decide to devote a fixed percentage of their gross to mining and just consider it the cost of doing business - a form of insurance that helps guarantee that money continues to work. Other companies might do the same thing, and they'd for sure exploit their contributions for all the advertising and PR benefit they could get from them.

Alternately, they might put that money use some kind of locking opcode to create a series of bounties miners can claim that can only be redeemed at incrementing block heights. This would replicate the double-spend disincentive that the block reward currently provides by making sure there's an opportunity cost associated with not extending the chain.

Maybe in the future, the retail industry will take on the task for paying for the network on their own because they determine that it's better for them if customers don't pay transaction fees at all.

Or, maybe it really will look like what we have now but with higher transaction rates. We really won't know until it happens.

 

[Peter R]

So I figured either (a) the Blockstream Core Devs will hate subchains and accuse me of pseudo-science, or (b) they'll like subchains and accuse me of taking credit for their work. Looks like it's leaning towards (b) at the moment:


Edit: Freudian slip

 

[rocks]

@Peter R Well if they take path b) then the question to ask back is if they already thought of this, why haven't they implemented it already? Block propagation is the main limiting factor today, this and IBLT together significantly improve block propagation.

If they know of working solutions but do not make progress on them and instead focus on other (trivial) projects, then that is worse as it shows they actively do not want to scale bitcoin.

Every single core dev that tells you they already thought of this should be strongly pushed back on with questions on why they haven't started work on it then. Subchains are a "soft fork" after all, its the type of solution that works great within their constraints.

 

[solex]

@Peter R
Surely, all Kanzure has achieved is giving you a list of references which may be usable. A collection of blog and forum posts, emails and irc is clearly an incoherent and incomplete treatment. A paper like yours is the first in academic literature, and sorely needed. This is especially as full RBF is being given an airing lately, risking the elimination of 0-conf, as according to Friedenbach and Todd "it was always broken". They are incorrect and anyone acting on this fallacy is making a colossal error in eliminating a major Bitcoin use-case.

The P2Pool software does have its share-chain as a subchain, but not the breakthrough nesting concept with fractal qualities.
https://en.bitcoin.it/wiki/P2Pool

edit: breakthrough

 

[Justus Ranvier]

This is especially as full RBF is being given an airing lately, risking the elimination of 0-conf, as according to Friedenbach and Todd "it was always broken". They are incorrect and anyone acting on this fallacy is a making colossal error in eliminating a major Bitcoin use-case.

One possibility is they are incorrect.

The other is that they are lying about their motivations, and intentionally sabotaging this use-case in order to force all Bitcoin users into a different use case that is more profitable for them.

Eventually the accumulation of evidence will more strongly support one possibility more than the other.

 

[solex]

Actually JR, you are almost certainly right about "intentionally". I do assume unbiased professional integrity from the get-go, but the evidence keeps mounting against it.

 

[rocks]

One possibility is they are incorrect.

The other is that they are lying about their motivations, and intentionally sabotaging this use-case in order to force all Bitcoin users into a different use case that is more profitable for them.

Eventually the accumulation of evidence will more strongly support one possibility more than the other.

What I am confused by is according to their current "rules" don't all the main devs need to agree to a change in order to put it in?

If Gavin or one other said no to RBF, then shouldn't it be dead? Period.

I can't imagine there is no disagreement with RBF and everyone is on board. Why does that not kill this horrible idea.