Amir Taaki wants to bring complete anonymity to Bitcoin. First he has to demonstrate it on other blockchains. The SC developers can demonstrate it on an altcoin and if it works, then Bitcoin can adopt it. If they come up with a two-way peg that works, then the trade-off for whatever benefit the SC offers may be worth some sacrifice, like security or functionality. It's unlikely it can be more secure than Bitcoin AND have other benefits. If they did, then just hard fork Bitcoin.
One way in which a side-chain would not be identical to bitcoin is it has different features. Unless we have provable security the added complexity would tend to make a side-chain less secure. (And thats a reason for side-chains .. so that people can get access to more features without exposing other people to the risks relating to their feature). For now there are also security tradeoffs in the 2wp mechanism, so that is another source of difference. But in the future, if for example snarks or some other innovation becomes a proven secure cryptographic construct, then we can have security-equal sidechains (and security-equal smart-phones). And then it may be possible to have a more secure side-chain: make a simplified side-chain that doesnt include bitcoin-script, p2sh, etc ie strips a bunch of stuff for pure cold storage. Of course you also have the incentive compatibility issue, however over time the difference erodes as volume & hence fees increase and reward decreases as we get through more halvings. Adam |
|
|
|
he points out flaws in some arguments that are commonly used to dismiss the risk of a 51% attack, e.g. that with cloud mining one could rent 51% of the power for 1 hour at much smaller cost than actually buying that much mining equipment.
Well there are three answers to that: a) cloud mining at scale on short contract is a bad idea; b) the cloud miner shouldnt rent out mining power to attack the chain any more than a country would rent out nukes to mentally unstable or psychopathic individuals; c) mining can be separated from voting, and some of the artificial centralisation reduced. For (a), [...] the attacker could rent for the minimum period allowed by the company, mine honestly for most of that time (thus presumably recovering most or more of the rent paid), and launch the attack during one hour. This is why it is dangerous for the mining company: they are sitting on $100m worth of equipment, on-demand renting it is dangerous to the network and to the hope of making a profit on their $100m investment. For (b), how could the cloud mining companies tell that some entity that already controls X% of the network's hashpower is now renting 51-X% from them? This entity could pretend to be N diferent clients. And I bet that no company requires miners to undergo psychological fitness tests.  Again a reason they should not do it. Whichever way you look at it renting out cloud mining is a bad idea because the reason to care about the meta-incentives (of bitcoins security, reputation, functionality etc) is lost - the $100m investment. The solution is to sell mining contracts so that the user actually owns the equipment, and often has also prepaid for the electricity as well. Its flash on-demand renting that is bad as then there is no investment incentive. This is actually part of the meta-incentive: a cloud miner should be able to forsee the problems of flash renting and realise as the equipment owner his capital is at risk. I suppose that (c) would require a substantial change in the bitcoin protocol, correct?
No its possible today. Luke-jr is working on it. Thats for the reverse problem that when you do buy a (long term) mining contract, you should control the vote. Otherwise we get articificial centralisation where the location of the physical mining data center artificially controls other peoples votes. Adam |
|
|
|
For a thing to be a sidechain I guess it would need to first be a chain. Can you point me to Bitstamp's blockchain?
They have a ledger but it isn't a block chain. so why does the WP, Odalv, and Adam refer to these federated server models as an implementation of SC's? which by my definition means involving a blockchain? Because the federated peg is a protocol adaptor - it translates the hash majority decision of the sidechain into something bitcoin can understand. It's not obvious, but it is the case that issued assets on the sidechain will already have full node security (up to the MM hashrate, and immune from any take-all coins attacks). For the purposes of issued asset trading, the sidechain works fine in this model, if we assume trade is issued - issued crosses; for that use pegged bitcoin is just transaction fees. Adam |
|
|
|
Adam's thought experiment depends on an omniscient AI. It's a common fallacy.
The sci-fi version tracks proof of work in alt-coins using any algorithm (via non-local communication and a cryptographic AI etc) but even the current bitcoin the only reason people are mining alts is because they hope to sell above par, humans can evaluate the factors, as I said "The universal AI is called common sense:" The universal AI is called common sense: changing hash functions does not change the joules expended per coin, [...] The 2015 Bitcoin doesnt yet know about users mining coins stamped with creative logos is because it lacks quantum non-local communication, we'll fix that one day, but in the mean time humans can convert one supply function to another with simple math and measure average electrical efficiency and when measured this way people are paying way over par, no rational entity would put money into marked coins, never mind a cryptographic universal AI.
I'm finding the analogy holding up surprisingly well, though it does depend on where you focus on the value lying, like gold: people discovered gold, it had useful properties, so then they invested in it and a network effect built up around it, and because of that they developed a keen interest in assaying gold, being 100% sure they have the real thing. You could argue bitcoinium (aka bitcoin) is the real thing, and imitations composed of proof of work, that are being sold above par (far above proof of work production cost) are the equivalent of gold plated lead. Adam |
|
|
|
Asking "[w]ho really rules?" researchers Martin Gilens and Benjamin I. Page argue that over the past few decades America's political system has slowly transformed from a democracy into an oligarchy, where wealthy elites wield most power.http://talkingpointsmemo.com/livewire/princeton-experts-say-us-no-longer-democracyyou guys have to ask yourselves why they have so much power. the answer, if you're reading this, is that you already know: control of Money. with Bitcoin, as it is, we are slowly taking back those reins of power. and now, we are being asked to change Bitcoin into a SC-enabled asset trading platform? speculative asset trading is NOT needed and distracts us from Bitcoins main function: Sound Money. People already are speculating with BTC on cryptsy, havelock (BTC denominated shares), just they are doing it in a less safe way because its not on-chain. Lets imagine gold or USD it was legally forbidden to use it for share trading, but share trading is very much legal. Do you think share trading would happen or not? Do you think that would be good for gold or the USD? If share trading creates more BTC (or scBTC) transactions buying/selling with BTC, or paying transaction with BTC that increases utility and hence price of BTC. Adam |
|
|
|
Indeed the paper had too much rhetoric and boldface for academic good taste. But he points out flaws in some arguments that are commonly used to dismiss the risk of a 51% attack, e.g. that with cloud mining one could rent 51% of the power for 1 hour at much smaller cost than actually buying that much mining equipment.
Well there are three answers to that: a) cloud mining at scale on short contract is a bad idea; b) the cloud miner shouldnt rent out mining power to attack the chain any more than a country would rent out nukes to mentally unstable or psychopathic individuals; c) mining can be separated from voting, and some of the artificial centralisation reduced. I did not bother to read his proposed fix to the longest-chain rule because it seems to involve timestamping transactions, and I suspect that such thing is even harder to get right than achieving consensus about the chain. I will wait until someone tells me that that section is worth reading.  checkpoints are a bad idea, they are a form of centralisation. Bitcoin is working to remove the limited very slow anti-DoS check pointing that it has. Adam |
|
|
|
2. philosopically, do you see Bitcoin as Money or as an economic "system" for trading assets of all types?
Personally I guess I see [bitcoin] as a kind of virtual commodity which can be used as sound money. But bitcoin is programable and as I wrote somewhere on the thread I think the programability, smart-contracts are also a big deal as well as the sound-money which is by itself awesome if thats all bitcoin could ever do. i think that's wrong. i view it as money. digital cash. digital gold. i think this is the biggest source of our disagreement. I am not sure how "money. digital cash. digital gold." != "virtual commodity which can be used as sound money" Adam |
|
|
|
What do people think of this paper: On The Longest Chain Rule and Programmed Self-Destruction of Crypto Currencies Nicolas T. Courtois (Submitted on 2 May 2014 (v1), last revised 10 Dec 2014 (this version, v11)) http://arxiv.org/abs/1405.0534His arguments are not backed up by his own citations and the rhetoric got ahead of the facts. He is a capable symmetric key cipher cryptographer, but he's been known to go on fact disconnected rants. bitcoiners seem to conclude he maybe thinks he'll get bitcoin consulting work if he claims the sky is falling in a loud voice. Either that or he just enjoys ranting. Apparently even academics can troll. (Present company excepted of course  Adam |
|
|
|
I'd evaluate #of transactions, mining fees generated, and # of BTC moved to the SC as primary indicators.
I am curious about how Blockstream get paid out though, just consultant fees for dev time?
I said a few things about that and so did Greg on reddit (how we expect to make money). Greg Maxwell (nullc on reddit) wrote some about how blockstream plans to make profit. https://www.reddit.com/r/IAmA/comments/2k3u97/we_are_bitcoin_sidechain_paper_authors_adam_back/clhoo7dI dont think making a profit is a bad thing - to hire developers & QA and UX designers and maintain software and design protocols and figure out how to use smart-contracts and find business partnerships to make those available to users all takes money. As those are good outcomes, and require more money, you have to have a profit to fuel it, you cant rely on investors to keep putting in more rounds! Its quite feasible to make money without being controlling, proprietary, centralising or evil. We certainly aim to try. and You should view blockstream as a sort of hybrid. We are developing FOSS open IP much as a not-for-profit would. But we are also aiming to make a profit by selling services, doing partnerships, advising integrators etc this is all complicated stuff and people need help to make it work. Like was said its kind of like Mozilla.
and Bitcoin has a lot to offer, and some of those things are not possible for the legacy systems to mimic. Particularly sound money, no counter-party risk, irreversable transactions (seizing and freezing basically prevent that outside of paper cash, though even that is partly relying on fungibility laws or it could have reversibility problems). Smart-contracts that are strengthened by no counter-party risk and irreversibility are one of the most interesting advantages I think. Without irreversibility and no freezability a "smart-contract" isnt smart, its just an electronic contract and we already have those. Ultimately if you combine it all you could rearchitect the financial system to largely remove systemic risk, add competition legacy systems cant react to (they intrinsically need their governance costs). This is why people gave us $21mil. Bitcoin all-in is a big deal. Sound-money is cool, but its only part of the picture.
If [blockstream profit plan] is something more arcane, it may be a cartel worthy of investing in, just in case they are able to run away with all the marbles.
Before we closed the seed funding round there were people who wrote asking how to buy in (I think some of them were confused and thought it was a "hot" new alt-coin ICO they could get it on the "ground-level" of). But also there were people who understood what a sidechain is. We do have one bitcoiner (non VC, tho qualified investor) who is an investor, and a number of the investors own BTC also. What I said to people on twitter was "you dont need to buy into side-chains, you're already in, its called bitcoin"; well really we're invested in bitcoin & the value of stock representing a stake in our ability to profit as described above, but you get the point. So for now I would buy more bitcoin Also in the (mining) incentive section of the white paper there was a concept of a time-shifted fee. Adam |
|
|
|
That really is bitcoinium and that has amazing implications...
...which would be more amazing if bitcoinium was mined via useful PoW (like protein folding). And if it is then mining of bitcoinium may become more profitable than bitcoinium itself, which in turn will drive value of bitcoinium to zero (we don't value by-products much). If its useful like sellable as a service, then that acts just like Adam Chirilluc cup of tea: Heating your cup of tea with a resistive heating element satisfies none of those properties. (Yes you could heat your tea on top of a stack of ASICs, and maybe someday people will do that; but the extent to which you extract use from the waste heat causes the average cost of production to fall and hence difficulty to rise to compensate).
an oversupply of the service will cause its price to fall also, and we'll still have the exact same Joules spent on bitcoin mining (minus the revenue from providing the service). Bitcoin price is set by people investing in it and utility demand. Production cost follows that, not the reverse, otherwise we wouldnt have difficulty falls as we saw last month. There's an economic principle that if a commodity is producible at below par, people will expend more effort (ie the difficulty will go up). Paul Sztorc explains the economic theory better in this blog article: http://www.truthcoin.info/blog/pow-and-mining/Adam |
|
|
|
Satoshi invented a new element called bitcoinium, and all alt-coins are made of it too; if you pay above par for bitcoinium you're getting ripped off.
Euro is a piece of paper with EUR printed on it. Dollar is a piece of paper with USD printed on it. A4 is a piece of paper with nothing printed on it. Well that those things are producible for a tiny fraction of par is the very problem bitcoin solves, in electronic form that gold solves in physical form. If you pay for either of them more than the price of paper you are getting ripped off.
Yes you are. In small parts via inflation. Adam, branding matters. The exact same suit produced in the exact Chinese factory sells for 100$ or 1000$, depending on whether Armani is stamped on it's side.
I dont think branding of digital artefacts that can be cheaply copied matters. Bitcoin is FOSS software, if someone comes up with something useful, it can and will be copied. That was what the pacman game analogy was about. However they were still fractions of actual bitcoins because the proof is universal due to the cryptographic AI, there was nothing Bob could do that could change that fact, no supply parameter changes, hash function used, software feature, not even a retro pacman game (loaded into the FHE processor in the coins), branding etc would change that because the universal cryptographic AI was measuring Joules expended, and unlike humans was not easily swayed by marketing and logos: a proof of joules mined is a proof joules mined, whatever letter or logo you stamp on the coin!
[...]
The pacman game doesnt change things either because if it was useful to play pacman on bitcoin, someone would fork the code and add it; an arms race of cutting and pasting each others code doesnt create value. Chances are the reason bitcoin doesnt have a pacman game is it isnt that useful or you dont need bitcoin to play pacman. The 2015 Bitcoin doesnt yet know about users mining coins stamped with creative logos is because it lacks quantum non-local communication, we'll fix that one day, but in the mean time humans can convert one supply function to another with simple math and measure average electrical efficiency and when measured this way people are paying way over par, no rational entity would put money into marked coins
Homo economicus doesn't exist, bits are not bits, and 1J used to heat my tea is not the same as 1J used to mine bitcoin, otherwise the price of bitcoin would closely follow the price of electricity.
I argue it does exist, and bitcoinium is bitcoinium. You're just missing a property of bitcoinium which is the electricity that went into it must have no other use, and that the use must be cryptographically verifiable in a decentralised way, and behave like a statistical poisson process (people around the world trying to do it and on average one mining a block of coins every 10mins). Heating your cup of tea with a resistive heating element satisfies none of those properties. (Yes you could heat your tea on top of a stack of ASICs, and maybe someday people will do that; but the extent to which you extract use from the waste heat causes the average cost of production to fall and hence difficulty to rise to compensate). I'm not defending alt-coins, they are scams, but it doesn't follow that all coins should be priced in joules, for the same reason that the price of a computer or of a suit is not the bill-of-materials cost.
Its a bit more than a Joule though. Its a time-adjusted Joule that you can cryptographically prove and that has no side-uses. That really is bitcoinium and that has amazing implications. Adam |
|
|
|
4. If Bitcoin fails to scale then I-Can-Scale-Coin will incorporate the necessary changes and the ecosystem will move across to that alt instead. Sidechains do not help with scaling because SC volume still needs to be handled somewhere.
It can be that different security and assurances are required for different uses. You could probably afford to lose the price of a cup of coffee. Eg imagine pettycoin on a sidechain. It has sharded validation. He thinks he can get to 100k TPS. I really dont think you want that on the main chain because its weaker, and he cut down a lot of bitcoin features to get it. But its useful. Eg take a look at Rusty Russell's pettycoin (its not an alt its a micropayment network bitcoin auxiliary chain aiming at scaling to 100k tx/sec.) He has a video up about pettycoin chain sharding https://www.youtube.com/watch?v=yzst_gChOr8. Peter Todd is also trying to figure out tree-chains which is a sort of hierarchical sharding idea. Snarks can solve the problem too but are novel bleeding edge crypto, and so far have a key gen trapdoor also. You probably for now dont want any of those things on the main chain if and until someone can firewall them in the chain or prove very robustly that they work. If bitcoin rejects such things, it maybe that innovation moves to alts. That would be a sad day to me because I think it could be the end of bitcoin and even cryptocurrency period as a store of value anyway. Quote from: JorgeStolfi on January 02, 2015, 02:47:31 PM Since the BTC rewards and fees are now worthless, most miners stop mining BTC and keep mining GNC only. Only a few persist, for sentimental reasons. The BTC block rate drops to near zero for months, until the difficulty gets readjusted. Then someone, not connected to Blockstream, GNC, or any other bitcoin entity, creates a new altcoin SuperShibaCoin (SSC).... (Summary of what JorgeSolfi is saying: if one alt-coin overtakes bitcoin, people will lose confidence in cryptocurrency because it will probably happen again, and a series of popping bubbles is not a good store of value). Yeah thats one of the reasons I am not keen on alt-coins. If an alt-coin took over bitcoin it might be the end of artificial scarcity (aka cryptocurrencies) in general, is my assertion too. You might enjoy this post: https://bitcointalk.org/index.php?topic=911339.msg10012730or the short tldr; twitter version: https://twitter.com/adam3us/status/550841397927235584Adam |
|
|
|
The SC's WP was disjointed to me. On the one hand, parts of it complain about alcoins, but them turns right around and gives a whole section to Freicoin, of all things. Talk about a screwed up economic policy. Adam even mentioned it again just the other day. This tells me they are not on board with Bitcoins sound or hard money principles. Probably different guys on their team contributed different sections or thi was a bone thrown to one of them. I seem to recall that Poelstra might be the one into demurrage.
No Jorge Timon. The section on demurrage as I recall was just an "it would be possible" in the section on possible incentives there were multiple others. You could it turns out implement opt-in bitcoin denominated demurrage on a sidechain (or interest, or other things - ie the argument that you cant do economic model feature coins on a side-chain is also false; its just that the cost is entirely bourne by those who opted into it, so whatever interest or demurrage or additional inflation there is, is bourne by the people who opted-in), so that creates a disincentive to opt in. Btw a number of bitcoin exchanges have turned fractional (inflation) without disclosing it, and then gone under when it was discovered. Same for some big poker companies - they were using customer stake to operate the company when market moved against them. Moving more things on chain is a way to avoid those, and side-chains gives the developer part of the community the flexibility to move things on chain. At least you can say a hypothetical side-chain with some unattractive economics wont be a surprise as it is with the normal way these failures happen, so you can avoid opting-in in the first place. The most obvious unattractive economics is - the operator takes your bitcoin. If you're interested in demurrage there's an economic theory by Gesell, the arguments for opting-in are actually complex. Perhaps a topic for a freicoin thread. Adam |
|
|
|
SCs have the potential of giving people a false hope of it being equivalent to Bitcoin but in fact not having all of the elements of MC Bitcoin.
True someone could make a hostile sidechain. But so can people make hostile multisigs (eg freeze your coins and extort you a % of them to get them back), obviously IOU offchain we've seen no-end of offchain things be scams, thefts, or incompetence failures. Or they could make defective wallets (thats happened a number of times also). The solution as with everything else with bitcoin is understand what you're doing and/or use software that is signed, on machines without malware, and from people who are competent. If you cant tell ask someone more competent (not you, but the general novice user). I wrote about that here: its caveat emptor, you shouldnt put money into a chain unless there is some assurance that security & bitcoin protocol knowledgeable people have audited it. People could certify chains (like sign them - "my name is blah and I'm a security researcher with reputation and I and my buddies audited this code and its good") or wallets could etc. Its good and a feature that people can opt to use uncertified chains. You want a situation where there is real open possibility for technical innovation & competition in chain features.
You also want no central control so no chains can get black listed.
Adam |
|
|
|
Adam the reason Bitcoin is considered successful thus far is real people have invested in giving it value, running SVP proof on the Bitcoin blockchain just put it all at risk.
Its not clear why you think adding an spv-multisig is worse than multisig or IOUs with offchain holdings (as most exchanges are doing). That seems reversed to me. What does it put at risk? Adam |
|
|
|
The spv verify opcode looks to me like to me to be in the class of a generically useful inclusions which anyone is free to use for a variety of developments or optimizations of existing developments.
Right. Do you suppose that anyone involved with Bitcoin dev had any ideas or involvement in efforts which could have made use pay-to-script-hash when it was integrated? I don't remember you pitching a bitch about that.
Well its interesting you should say that - there were discussions that the script language could have been extended during the p2sh discussion at the time. eg there was no reason the script inside p2sh couldnt have been a different script language. Lets say that script had been done, then sidechains might've been possible without a separate change. Also as I mentioned a few times its already very close to possible to implement a compact SPV verify in bitcoin script. Anyone of dozens of minor changes to slightly improve expressiveness of bitcoin script would allow it. It seems close to inevitable that sooner or later an expressiveness or refactor or clean up of bitcoin script will happen, for its own sake and reason: to make it easier or safer to write smart-contracts etc. About cypherdoc particularly he said this: I am not sure if you are aware sidechains are nearly possible with zero changes to bitcoin. Its already programable via the script language. It may even be doable with zero changes with some chained contorted big script to validate compact SPV proofs.
great, then do it if it doesn't involve a source code change. i have no problem with that. So that seems a little inconsistent to me. ie if this is really a bad idea why would cypherdoc not have a problem with it regardless of whether it required changes or not. I mean if its a principle you'd be arguing to please not do it even though its possible. Or to remove something from the language to prevent it, or put a technical defense preventing it if such a thing existed (seems unlikely but I havent explored it much). Not saying "I have no problem with that." fundamentally, none of the technical enhancements to Bitcoins money function couldn't be done on the MC. not easy to gain consensus but it CAN be done. i'm all for that. this is what we have testnet and federated server SC's for. experimentation. esp when we're only at the $4B market cap.
While its possible, its very much harder, and when changes are made its very much riskier and less secure. If you care about the security of your coins you should be for having a firewalled live beta and firewalled extension mechanism (if you support improvements "i'm all for that." you said). it also looks like CP and colored coins will bring us other assets to MC. that's good too.
CP = CounterParty. That doesnt bring anything to the chain, other than bloat, its a layered consensus system with its own alt-coin. If you valued the price of bitcoin, probably you'd be better pushing for sidechains than CP because its bitcoin denominated and increases demand and features for bitcoin. (Not sure whats up with bitcoin price - maybe its time for reverse psychology: big negative news, generates lots of free press, and there's no such thing as bad press?) who knows how much further MC achievements might have been accomplished if BS core devs were spending all that time working on Bitcoin Core that they undoubtedly have been dedicating to the spvp for the last year and a half. forget that shit and get behind Gavin and increase blocksize. now is the time to do this.
No they spent more time on core than before, because they quit their full time jobs/occupation and Mark said somewhere else on reddit he figured they'd spent 50% of their time at blockstream on core. (Unrelated to sidechains most of it .. eg in Pieter's case the headers-first speed up you were mentioning, though he's been working on that for a long time). You could check by looking at bitcoin github, there's a stats page. fundamentally, i think allowing an offramp for BTC units over to insecure SC's is economically and technically flawed for all the Sound Money reasons i've already articulated.
I fail to see the connection between sound money and an ability to freeze coins with an spv-multisig instead of a multisig. Bitcoin is protected with a firewall from features on the sidechain. The bitcoins never leave the chain, they're just frozen in an spv-multisig instead of a multisig. Lots of people are using multisigs, daily, to effectively do the same thing, its more secure to do it with an spv-multisig. Lots of people are not even doing that, they're using pure offchain in a shared wallet, for reasons that in time could be fixed on sidechains and then with a year of live testing with $1b on it kind of assurance, ported back into bitcoin main. improve the MC; it's not that hard.
Yes, it is that hard. Maybe I dont know do you follow bitcoin-dev, look at github, follow wizards and dev irc? Making changes to a $4b system is very risky. op-spv can avoid and reduce that risk, once it is done. Adam |
|
|
|
I fail to see the point of this rant.... nothing useful or particularly insightful...
Satoshi invented a new element called bitcoinium, and all alt-coins are made of it too; if you pay above par for bitcoinium you're getting ripped off. Adam |
|
|
|
Aethereum is probably more useful as a sidechain (rather than a spin-off1) and launching it as such would certainly help make the point that the value of a cryptocurrency comes from the network effect rather than from its features (which can be forked). I think spin-offs are less useful than sidechains but conceptually still important because they help shift people from thinking that the value comes from the tech, to realizing that the value comes from the distribution of wealth encoded in the ledger (i.e., the 'money-is-memory' point of view). Spin-offs challenge the notion that a technically-superior alt-coin could supersede bitcoin, since all its superior features could be cloned and then re-launched using Bitcoin's ledger. (Spin-offs also challenge the IPO-model of launching a cryptocurrency.) I always suspected that if Aethereum launched as a spin-off with some fanfare along with Ethereum, that Aethereum would slowly leach liquidity from Ethereum, but that both coins would eventually end up dead. On the other hand, Aethereum launched as a sidechain could survive indefinitely (as it's tied to Bitcoin's pool of liquidity rather than floating) and add value to the bitcoin network by offering features that aren't possible using the main chain (and I still suspect Ethereum would end up dead). 1Yes, the value of aether would float and Bitcoin users had a perpetual right to claim their share of coins. Very cool, fully agree. Anyway I was applauding and telling everyone about Aethereum at the time you released it. Hilarious. I know there are some people who disagree, but I think Gavin's proposal to increase the blocksize limit inline with the historical rate of growth of internet bandwidth is a reasonable way to allow the network to scale without significantly affecting centralization.
Yeah I'm not disagreeing other than to suggest you maybe want to take it easy increasing it, eg do it a little and then do more later and monitor the situation. My concern with the "do a little and then do more later" approach is the huge political effort required to implement multiple hard forks. If Bitcoin's market cap is 10 - 100X bigger, are we really going to be able to hard fork the protocol again? I don't think so, and thus scaling will happen in sidechains rather than the main chain by default (rather than because it's the best thing to do). Yeah I dont know. Maybe there's a way to have it reactively grow - if you talk to GMaxwell he has another idea involving rolling mean as a cap or something fancy like that (to avoid gaming to drive up the blocksize to lock out people with less bandwidth). Why not open up the Blockchain according to Gavin's proposal (a single hard fork to address scalability), while working towards methods of price discovery for P2P bandwidth / blockchain space? Worst case it's a soft-fork to reduce the blocksize limit in the future.
A problem with Gavin's proposal is the fixed schedule - what happens if bandwidth doesnt make the growth curve, then we get bandwidth driven centralisation. Soft-fork reduction ... maybe thats another answer yes. GMaxwell may have the winning idea. (He has a habit of doing that). Its also not clear every transaction needs to go on the mainchain - with or without sidechains. Eg you can do things with time-locks like payment channels and payment channel hubs without third party risk also for micro-payments. There ought to be a way to cryptogaphically back coins that can scale for different uses, or with sharding. Eg take a look at Rusty Russell's pettycoin (its not an alt its a micropayment network bitcoin auxiliary chain aiming at scaling to 100k tx/sec.) He has a video up about pettycoin chain sharding https://www.youtube.com/watch?v=yzst_gChOr8. Peter Todd is also trying to figure out tree-chains which is a sort of hierarchical sharding idea. Snarks can solve the problem too but are novel bleeding edge crypto, and so far have a key gen trapdoor also. Maybe chain pressure does something else: drive scaling innovation. You just dont want it to become disruptive short term as its hard to react bitcoin code fast, nor push people into weak offchain systems. There maybe an aspect of doing something simple and fast enough to get confidence of to take the pressure off while people figure out these more complex approaches. Adam |
|
|
|
So not a single thought on the SuperNet model eh Adam?
I dont know much about it (so these are just questions): So what is it? Whats the TLDR; version. Is it an alt-coin market + plug-in framework? Is it decentralised? Or central-checkpointed PoS? Does it include a floating alt-coin? Does it incentivise developers to make example apps by paying them from a pre-mine? Adam |
|
|
|
|
Show Posts
