Topic: Gold collapsing. Bitcoin UP. - page 594. (Read 2032266 times)

That would be a best-case scenario to me (with the caveat that I would accept a modest and simple increase based upon quality forward looking analysis.)

Relatedly, and even more importantly, it would allow Bitcoin core to basically freeze as a codebase.

As an engineer, I appreciate well constructed, well documented, and elegant code.  Also as an engineer, I've lived with utterly shit code (often enough of my own creation) because of priorty and risk profiles, and because it worked for it's intended purpose and was reasonable well understood.

I was a fan of alternate implementations in earlier phases of Bitcoin's existence.  I even provided modest funding to the guy who was dicking around with an erlang implementation/joke.  Times have changed, and a realistic expectation of sidechains are THE main change.  This relegates the need for experimentation and elegance to the dustbin.


Likewise, and I think that is a great thing.  This because they are, as best I can tell, experiance professionals many of whom have fought in the trenches on the right side of the historic battles.  Fairly demonstrably so.  To a large degree their instincts and mine seem to align on a lot of fronts and this one in particular.


Wrong again Bob.  With sidechains, native Bitcoin is available to anyone who wants to pay what it's worth.  Anyone!  What you want is to model the ubiquitous welfare state structure that abound with plebs being subsidized by powerful institutions who leverage economies of scale to drop their own costs while milking the herd for different kinds of value streams.

What I see sidechains doing is giving everyone the opportunity to enjoy the kinds of freedom that current Bitcoin allows, but in an honest and robust manner rather than in the sneaky subsidized manner that logically must be the end-game when trying to make native Bitcoin be the one-size-fits-all solution under anything remotely resembling the current system of support.

That would be great, if it was possible. However any given sidechain will always require more trust than the main chain. I don't believe the white paper made the claim that sidechains will be zero trust, did it?

The worst case scenario for sidechains is that they are used as an excuse to keep Bitcoin's transaction rate capped forever.

Based on the composition of the Blockstream team, I feel pretty safe assuming this is a goal which they will actively pursue.

In that scenario we do, in fact, end up replicating the legacy banking system. There will be a real Bitcoin which only a privileged elite can access, and the rest of humanity will be relegated to transacting in money substitutes.

well seemingly you, for articulated reasons that we're exploring, would like to restrict people from sending their coins to a sidechain.

I agree people should be able to do what they want with their bitcoin and send them where they want.

Another way to look at sidechains btw is that a federated peg is a multi-sig (with modest parameters eg 5 of 10 and some trustworthy security competent bitcoin interested businesses) and a SPV peg is a bigger federated peg with a 5000 of 10000 multisig with dynamic membership (ie whoever is mining the chain right now).  The spv peg op_code is an opcode to understand those dynamic membership multisigs.  The dynamic membership multi sigs (DMMS for short) are written about in the sidechain white paper http://www.blockstream.com/sidechains.pdf and are a different way to look at bitcoin mining, though its actually the same thing.

You can also combine DMMS sigs with regular multisigs, its just an op code so you can program with it.  eg IF ( 0.5*DMMS + 0.5*multisig(5,10) ) THEN spend coin.  Or IF ( hashrate > 75% AND DMMS ) OR ( hashrate <= 75% AND multisig(5,10) ) THEN spend coin can react to hashrate drops.  Different people could even use different peg rules on the same chain depending on their security tradoff preferences.

I dont see that as some how evil or dangerous relating to offchain transactions (we've seen them cause system shocks mtgox etc) or federated pegs or voting pools (then you are depending on the 5 of 10 of their security etc its better but its still non-zero risk) ... its just the next evolution in that direction - more decentralised, more things enforced by the network.  For example read Nick Szabo's recent blog post about fiduciary code http://unenumerated.blogspot.com/2014/12/the-dawn-of-trustworthy-computing.html what do you think he's talking about?

You complain about incentive, but I talked about the details of that and there are multiple secure and workable parameter choices with usable tradeoffs , and off-chain transactions and voting pools have incentive too - the holder just steals them for free with no effort!!  Thats pure human trust.

Incentive discussion was here

Also for the non-voting pool version the statistics are horrible it used to be that 50% of exchanges that ever existed went down with loss of user funds.  Yes its improving but lets ask you a question: do you store investment amounts of bitcoins on an exchange?  Or a web wallet?  Or an offline/paper wallet with backups?  I imagine you do the latter.

Its also (I said this before on this thread) basically embarrassing that bitcoin is repeating the banking governance failures of the last century and of 2008 etc when the entire point of smart-contracts and programable trust is that users can have direct control of their funds and not have to trust third parties.

One of the interesting motivations for wanting the extensibility that sidechains provides is to be able to make that zero trust a reality for more bitcoin transaction types.

Adam

btw (I know you were predicting JustusRanvier) but I dont think one can characterise the op_spv is to the disadvantage of blockstream competitors.  Its an op code, anyone can use it to make sidechains or other things. 

Other things: you can also use the op_spv opcode to enhance security and efficiency for SPV clients like smartphone wallets completely separately from sidechains.  And to fast catchup headers also, the compact SPV proof is work-preserving.

If for example OpenTransactions sees a use for it, or ethereum wants to switch out ethers for bitcoin (i imagine the developers are attached to their premine for now) but perhaps Mastercoin or something thats market cap is falling, seeing less use and the ICO people probably mostly dumped by now (if thats a fair characterisation, dont follow it closely) maybe they might want to migrate to a sidechain, thats all expected and good.  Permisionless innovation is the point of sidechains.  You could even one-way peg the remaining mastercoins into a sidechain to allow their residual tradability.

If you mean it allows bitcoin to more easily incorporate features, well most alts dont really have features, or not meaningful or useful/non-broken ones, but there are some feature coins and bitcoin 2.0ish coins that do.  I dont see how extending bitcoin is unfair - they're all leeching off and copying bitcoin, which was the actual innovation - why cant bitcoin take little bits of innovation back too?

Also alt-coins or feature coins can have sidechains off them also for beta versions, or secondary functionality if they were actively developed.

Adam

With respect to Bitcoin's development road map versus the world's vast array of financial relationships, in order to receive universal acceptance it usually best to keep language and protocol as simple as possible with enough functionality to serve the majority. Overly complex or superfluous language makes some people suspicious. Projects that offer Turing-complete complexity are empowering, but not always welcome. Having said that, there is nothing stopping meta applications from improving Bitcoin's footprint. In defense of Justus' pit bull defensiveness, sometimes strength of conviction is enough to clear the smoke from a conversation mired in nuance. While I often disagree with him on many things, I admire his brevity and focus.

i agree completely.  there's too much FUD being slung around about problems with offchain services.  if anything, it's getting better as in the case of exchanges and wallets.  not perfect, but better esp for the bigger ones.

and who is anyone to restrict where ppl send their BTC?

I dunno, it seems part of the bitcoin social contract is to be deliberately (and ideally enforceably) agnostic about what people do with their bitcoins. It that means sending them to some crazy off-chain system with voting pools or whatever other silly idea, so be it. How is that anyone's business any more than spending them on wikileaks donations?

If bitcoin can't handle people doing crazy things off chain like voting pools or MtGox, then it has some pretty big problems. It seems fairly proven that it can survive MtGox at least.

This includes the so-called federated model of sidechains.

I still see a gargantuan difference between that and changing the protocol.

Seems like a good time for me to chime in here. Maybe a year ago I used to hear this sort of thing about Ripple, that somehow Ripple and Monetas were competitors. But we are not, since our software does different things. This will become much more evident as our software rolls out.

The same is true of Blockstream. If you generalize enough, you can find some sort of overlap and then assume we must be competitors. But the reality is that no one else is building what Monetas is building, and no one else will be able to do the things that our software does.

That's nothing negative on Blockstream, it's just different -- that's all. Anyone who thinks there is competition between us lacks an understanding of what we're building and how we plan to monetize it.

I welcome Blockstream's entry into this space and I think alt-coins are the ones most concerned about what they represent.

------------------------------------------

re: Justus, Adam Back

The reason Monetas contracted Justus to help us build the voting pools is because he is a Bitcoin purist. Justus keeps us honest. And I think that is the same role he is playing here in this thread. Bitcoin is his highest value.

And I think what Justus is saying is, that a framework needs to be developed for any changes to Bitcoin. I think he's right that it's not good enough for changes merely to be "open source, open IP." There is a potential conflict of interest, and there is an easy way to put any concerns to rest.

Justus said it best:


I really don't think that's much to ask. Do you, Adam?

I think Adam is a good guy with good intentions. Certainly he deserves respect -- he has earned it. All around. He's a brilliant guy and he's been working for good for a long time. We all owe him our thanks at the very least for his invention of hashcash, his work at Zero Knowledge, etc. And I know Justus can be blunt; Adam has been very graceful in this thread.

Yet it is true that Monetas is not asking for any forks in the Bitcoin protocol. Blockstream is. That's why these questions are being raised. There is an easy way to put them to rest.

I am not so sure... If the change to the protocol means not having their revenue cut 50%, I bet that most miners, even those who are not in the cartel, would eagerly support it.

Merchants who "accept bitcoin" would not care and would not take sides, since they receive dollars and, at worst, may lose some customers if bitcoin collapses.  Payment processors, exchanges, and the like would have to choose a side, but since siding against the cartel could mean losing their income for months, they will probably upgrade before the deadline; and that will force their clients to upgrade, either before the deadline or before accessing their services again.  Note that, up to the deadline, users and miners who choose to upgrade will continue to interoperate with those who don't.

I seems that when people discuss a 51% attack they assume some entity that wants to either destroy bitcoin, or pull some quick scam (like double-spending) and run away.  But cartels usually employ their power to maximize net revenue over the long run, and are careful not to kill their cash cow.

At the current BTC price, the change I described would mean 600 k$/day for 2 years for the whole network, which is ~440 M$.  If the cartel has >50% of the power, that means >220 M$ of extra revenue just for the cartel.

Depending of their costs, that could mean the difference between making 100 M$ profit (say) if they try and succeed, or having to turn off their equipment and lose all their capital, as a reward for their "good behavior", if they do nothing.   At worst, if they try and fail, they will lose only a few weeks of revenue; and from then on they would get the same "reward" they would get if they did nothing.

Moreover, I don't believe that many people would want to leave bitcoin just because of that change.  Not even the long-term holders should change their minds about bitcoin's future because of a change to a parameter that was clearly chosen quite arbitrarily.  (People in countries with high inflation or bank deposit haircuts do not stop accepting their currency; they just try to spend it as fast as they can, or convert it to investments that are resistant to inflation.) The price may even go up, if the change is properly marketed as preserving the network's unique strength etc.

The cartel would have to change the difficulty level and the difficulty adjustment algorithm to keep the new chain flowing at an acceptable rate during the transition period.   The cartel can move its power back and forth between the two chains, according to what the non-cartel miners are doing, in order to always have 51% of cooperating hashpower on both chains.  As noted above, most of the non-cartel miners would have strong incentive to cooperate with the cartel, and upgrade to the cartel's protocol, before  the deadline; they would then put any rebel miners at an even bigger disadvantage.


I still do not understand what sort of "big red button" could do that.  Changing the protocol to make the ASICS useless would force any remaining rebel miners to give in to the cartel, and would require that the rebel users upgrade to a different protocol (which is supposed to be a catastrophe) leaving them on a junk coin with a minuscule network, no service infrastructure, and a minority of the former users.  But they would still have the option to upgrade to the cartel protocol and find all their coins waiting for them in the cartel chain, apart from any transactions that they made in the junk chain after the deadline.


I must be missing some essential detail.  I understood that each sidechain would be just a black box from the viewpoint of the bitcoin protocol, and its designers/developers would be free to choose whether and how to peg its operations to the bitcoin chain -- without the bitcoin network having to know about it.   Is there more to the 'sidechain' concept than that?

Could a sidechain be a BTC/USD exchange providing sub-second trades with cryptographic certificates? It would not be possible to peg every such trade to the blockchain, right?

Thanks for your response.  Yes, I'm the lead dev on btcd.

a and b) First, both of these are completely irrelevant.  No one knows who Satoshi was/is, yet that isn't a problem for Bitcoin Core.  So, even if every single contributor were unknown (they're not as you can find the names of every contributor on github), why would it be a problem for btcd when it isn't for Bitcoin Core?  Second, the defense bit is not true, but even if it were though, I'd venture to say with the number of contributors on Bitcoin Core, at least one or more have affiliations in one way or another with some organization that you, or somebody else, doesn't like.  The code is completely open for all to see, so it would be obvious to see if something nefarious were going on as you are implying.

c) It has been publicly stated many, many times that it is being coded because the ecosystem severely needs multiple diverse implementations if Bitcoin is ever going to truly flourish to the level we all want it to.  Thinking that a single implementation controlled by a handful of people is a good scheme for a system that is supposed to take over the World's monetary supply is just not very forward thinking.  This has been covered ad naseum.

d) This is a complete red herring and exists just as much with Bitcoin Core as it does with anything else.  The exact same thing can easily happen (and already has as you've noted) with Bitcoin Core.  This is a real problem with the technology that needs to be solved independent of the number of implementations on the network.  In fact, working toward solutions which address this fundamental issue is necessary for Bitcoin Core too since it means mistakes there don't lead to undesirable consequences either.  The current approach is "well let's just hope it doesn't happen!" and frankly, that just isn't good enough.

e) You've been misinformed.  Here is a blog post I made on the very first package we released back in May of 2013 which was only the wire protocol that clearly states "Here at Conformal, we are keenly aware of how important it is that any full-node implementations agree on the exact same set of rules for what blocks get included into the block chain as minor differences can lead to chain forks and double-spends." (https://blog.conformal.com/btcwire-the-bitcoin-wire-protocol-package-from-btcd/).  The Bitcoin Core devs weren't even aware we were working on the project at that point in time.  There was no "convincing" needed or involved.

f) This is referring to d and e, see those.

Fair enough.

It is true that if Open-Transactions could be use to transact in Bitcoin substitutes, and that would suck fee revenue away from Bitcoin.

It is also true that outcome is 100% avoidable.

The only reason OT would suck transaction fee revenue away from Bitcoin is if the Bitcoin blockchain is forbidden for processing those transaction.

If Bitcoin won't process the transactions, they'll be processed somewhere else.

It's up to the core developers and Bitcoin community at large whether or not OT ends up sucking fee revenue away from Bitcoin mining.

I'm not a core developer, so I can't stop them from making changes to Bitcoin that would increase the blockchain's transaction processing capability.

In fact, it would be extremely out of character for me to even argue against such a change given that I've spent three years arguing for more transactions to happen on chain, to the point of saying that's the only way Bitcoin can economically survive in the long term.

It's also a perfectly viable future for OT to handle financial instrument contracts (BTC-denominated and otherwise) that aren't appropriate for the blockchain anyway while leaving all monetary transactions to the blockchain. It might even be the case that a high percentage of OT transactions end up causing a blockchain transaction anyway so that OT transaction volume helps drives Bitcoin transaction volume.

But if real monetary transactions aren't allowed to happen on the Bitcoin blockchain, then Bitcoin substitutes will trade in OT.



As far as the voting pool concept itself goes, I've always thought (and said many times) that people shouldn't leave their Bitcoins in the custody of a third party such as an exchange.

However, since they're going to do it anyway no matter how many times I say that you don't own Bitcoins if you don't control the private key, and since I'm tired of seeing people lose money to incompetent and/or corrupt exchanges, I guess I have to fix the exchanges.

Well there is the federated peg that doesnt need miner support.  There is the one way peg that also doesnt (but is more limited than 2wp).  And also most of the miners we've talked to informally seemed pretty keen on the idea.

I mean for example namecoin gets a pretty high Merge Mine rate and thats not really actively used even.  Mining profits are quite thin right now with difficulty reaching equilibrium for the price range so another percent or two makes all the difference.  I got the idea they were also interested in future potential for more transactions and more types of transactions as they considered this a source of bitcoin price upside, and they are mining bitcoins and living off the margin.

Plus a price uptick is a big deal to current miners, its a kind of derivative on bitcoin price.  They get to cash in for 3months at a higher price with too low difficulty until the manufacturers can ramp new manufacturing pipelines and get lots of new equipment to market.

I think the bigger picture though is its premature - we have not yet released code, nor BIP draft for community discussion and picking apart and redesigning etc before there is something to merge mine.  Its also useful (maybe you said it yourself also I think) for a federated peg to operate for a while in parallel with that open design discussion for people to see how it works.  You can view the federated peg as a protocol adaptor - there can still be mining occuring on the sidechain with the federated peg, just the return peg is translated into a multisig for bitcoin by the federation servers.

After that if the community can settle on a op-code for extensibility everyone is happy with people including us can try to stand up a few sidechains.  If no one likes them then they wont get mined.  So as you can see sidechains and the op_spv really depend on community and economic majority approval, and thats a good thing.

Adam

Agreed.  But you're also not completely off the hook.  There are also economic implications from offchain and voting-pools version of offchain.  People also have a right to ask about that similarly.  There are actually economic implications.  If OpenTransactions sucks a bunch of transactions off-chain that deprives bitcoin of transaction fees, and so it lowers bitcoin security.  I am not saying this is a bad tradeoff, just point out that nothing is free of implications.


Was that really called for?  I am trying to be responsive here...

I said its a very valid question.  I tried several times to answer.  Feel free to pick apart the answer but dont just say I didnt answer, or I cant really reply short of reiterating as I wont know what it is you found unclear or unconvincing about the argument.

Adam

hmm, it took me until about 2013 when i first heard about SC's to realize that one of the beauties of Bitcoin is its simplicity.  

i mean, look, we have unencumbered BTC units traveling freely on an ultra secure blockchain whose value involves no counterparty and no other asset.

now, you're asking us to accept an unprecedented change to the protocol which will allow BTC units to intermingle with all sorts of assets limited only by our imaginations all while traveling on a less secure SC.  

serious question.  what kind of revolutionary tech, ie spvp, relies on "convincing" an entire industry called "mining" to support and secure itself?  b/c in the absence of close to 100% MM'ing support, i see SC's failing miserably from attacks.  there is no way you'll get 100% of miners to MM one, let alone dozens of SC's.  that's a huge risk to the entire Bitcoin system.

Hmm so it took me a while to see this, but actually most economic differences can be mathematically modelled and opted into on a sidechain.  For example demurrage could be implemented.  Or different block intervals.  Or different tapering of block reward, or ongoing block reward.  Those define a deterministic calculable difference.

Sidechain pegs do not have to be 1:1 nor silly things like 1:1000 just mean sidecoins are milli bitcoins, but also the rate can be deterministically time-changing.

Now the even more interesting realisation once you see this, is it provides a formula to value altcoins.  If you wouldnt opt into the sidechain with that behaviour, if you had to pay above par, you probably are just seeing more clearly that you similarly shouldnt buy that altcoin.  An alt-coin is a watermarked bitcoin sold above par with a marketing story.


Using the really high level definition that bitcoin is an abstract model for converting electricity into proofs of work, where in all of these proofs globally define the total electrical expenditure, which then adjusts the J/proof to hold block interval on target at 10mins.  When you look at it like that, and if bitcoin cant quite do that its an implementation limitation, that maybe we can fix with quantum sci-fi in the future, then you can see that mining an alt-coin is sort of like watermarking a bitcoin and calling it a foo and expecting people to be suckered into paying above-par for it.

Or like gold bars but making them into triangular shapes or stamping them with a doge and expecting people to pay more.  A gold atom is a gold atom.  A proof of electrical consumption is a proof of electrical consumption.  That not many people are stamping doges on their gold bars doesnt mean doge stamped bars are more valuable, once the fad wears off, someone will realise they are being ripped off for gold or buy their own doge stamping machine and buy basic gold.

Adam

no, you missed the point.  take Dogecoin which is an inflationary alt.

try building a SC with an inflationary coin; no Bitcoiner would move their BTC to it b/c they don't believe in inflation.

Is that you're still not off the hook.

You've acknowledged that Blockstream is trying to change the protocol.

You've acknowledged that other companies which are not trying to change the protocol aren't subject to the same levels of legitimate scrutiny as those which are.

You haven't given a better answer than "trust us to to the right thing" for why Bitcoin users shouldn't be concerned about your plans for the protocol.

Which basically means we're at the same point today we were this morning, and at the months leading up to today, and where I expect things will remain for the foreseeable future.

I think thats wrong, lets say bytecoin with its ringsigs.  That provides some interesting privacy properties, not quite zerocoin, but safer crypto and more compact.  I could imagine a number of people would be interested to use that as a sidechain.

Adam

ps checkout cryptonote https://cryptonote.org/whitepaper.pdf and http://cryptonote.org for explanations (seems they've been busy making a nice web site with explanations).

Nah Justus is the good guy.  He just flames a little, so what.

Adam

Oh boy, thats another can of worms.  You dont happen to work for conformal do you btw?

So (it a long story) but a) no one knows who's funding them; b) most of them seem to have ex-defence contractor profiles; c) no one knows why they are coding it; d) if there is one off by even a single bit interpretation bug in it it breaks consensus and forks bitcoin, which can likely be systematically abused, to create a massive accounting/double-spend mess that will be too expensive to repair as last time - that could create a mega fork worse than the leveldb bug, and kill bitcoin or force some really harsh choices to best effort cleanup where there are innocent losers; e) it apparently took a lot of effort by the core devs to persuade them that this was a problem; f) it still has that problem.

I think that about covers it.

Adam