Ethereum Anti-ASIC fork, is it the right time for bitcoin too?

[Stedsm]

--snip--

--snip--

Bitcoin is an open, permissionless system. Anyone can participate in the system in any way and/or capacity that they want. You can have your opinions, but you cannot judge another person, or impose how they should participate in it.

Well, it's true that nobody can put a "cap" over how much one can attain and what capacity they need to keep with them; but then, what about the 51% attack and even more hash power than that gained by any one entity or institution? What if their intentions are not good? That won't affect the network alone, but the value in everyone's mind for BTC is just because it can't be attacked that damn easily and be used as a borderless currency, what would happen if everyone's faith will be shaken up if some stupid statement comes up from big shitheads like Craig Wright or Roger Ver? Sometimes, these people don't try to manipulate everything including hashrates to market values on their own, but rather influence people on doing what they want these people to do.

We need to discuss everything in details with prominent Core figures and have their support

I was thinking about this.

Why do you need the Core developers' permission? The beauty of cryptocurrencies is it's open and permissionless. You have your right to freely make your own thing and release your own software. Linus Torvalds did not need to ask permssion to develop and release his own version of a unix-like OS did he.

Well, then what would come up would be called an alt like Bitcoin Cash which is not Bitcoin itself. It doesn't really need any Core devs' permission, but we need to come up with that consensus thing over the decision on whether any/many of the miners would actually support and run your developed software or would they continue with the ongoing one? This only gave birth to many alt (or I should say shit) coins due to having no support from people during the beginning and even now. Remember, it's us in the end who are going to use it, so decision over supporting such proposals need to be taken with too much care. Else, asking other devs to come up with an alt to show what they've got (through testnet first) would be highly preferential.

[Wind_FURY]

--snip--

--snip--

Bitcoin is an open, permissionless system. Anyone can participate in the system in any way and/or capacity that they want. You can have your opinions, but you cannot judge another person, or impose how they should participate in it.

Well, it's true that nobody can put a "cap" over how much one can attain and what capacity they need to keep with them; but then, what about the 51% attack and even more hash power than that gained by any one entity or institution? What if their intentions are not good?

Like the intentions of Jihan Wu?

Where is he? What happened to him?

That won't affect the network alone, but the value in everyone's mind for BTC is just because it can't be attacked that damn easily and be used as a borderless currency, what would happen if everyone's faith will be shaken up if some stupid statement comes up from big shitheads like Craig Wright or Roger Ver? Sometimes, these people don't try to manipulate everything including hashrates to market values on their own, but rather influence people on doing what they want these people to do.

They "manipulate" yet they are the altcoin.

Bitcoin is the people. What makes Bitcoin "Bitcoin" is the social construct, not the code, or the network, or the developers.

We need to discuss everything in details with prominent Core figures and have their support

I was thinking about this.

Why do you need the Core developers' permission? The beauty of cryptocurrencies is it's open and permissionless. You have your right to freely make your own thing and release your own software. Linus Torvalds did not need to ask permssion to develop and release his own version of a unix-like OS did he.

Well, then what would come up would be called an alt like Bitcoin Cash which is not Bitcoin itself.

aliashraf will deploy on a testnet.

It doesn't really need any Core devs' permission, but we need to come up with that consensus thing over the decision on whether any/many of the miners would actually support and run your developed software or would they continue with the ongoing one? This only gave birth to many alt (or I should say shit) coins due to having no support from people during the beginning and even now. Remember, it's us in the end who are going to use it, so decision over supporting such proposals need to be taken with too much care. Else, asking other devs to come up with an alt to show what they've got (through testnet first) would be highly preferential.

Where were you this whole thread? Read 5 pages back. Everything has already been discussed and debated.

Conclusion. A hard fork for a POW upgrade will never get consensus.

We need to discuss everything in details with prominent Core figures and have their support

I was thinking about this.

Why do you need the Core developers' permission? The beauty of cryptocurrencies is it's open and permissionless. You have your right to freely make your own thing and release your own software. Linus Torvalds did not need to ask permssion to develop and release his own version of a unix-like OS did he.

Obviously, permission is the most ridiculous thing one would ask for in this context  

But as I've mentioned above thread, they are carrying bitcoin history and what I want is inheriting this history not just the code but the full history and the culture.

You want the Core developers' validation so that the community will listen to you, is that what you want?

[aliashraf]

We need to discuss everything in details with prominent Core figures and have their support

I was thinking about this.

Why do you need the Core developers' permission? The beauty of cryptocurrencies is it's open and permissionless. You have your right to freely make your own thing and release your own software. Linus Torvalds did not need to ask permssion to develop and release his own version of a unix-like OS did he.

Obviously, permission is the most ridiculous thing one would ask for in this context  

But as I've mentioned above thread, they are carrying bitcoin history and what I want is inheriting this history not just the code but the full history and the culture.

You want the Core developers' validation so that the community will listen to you, is that what you want?

No, I want them to join and collaborate because I want the history. Bitcoin is disruptively new tech, very high tech, yes, but it is just a system, a decentralized system. Like any system bitcoin needs to evolve and adopt. A lot of things happened since 2009 and it is the time, I believe.

I'll launch the testnet nevertheless, but the actual hardfork won't happen until we could get Core devs onboard because a row of other important issues are in the agenda as well:

-We need to take care of pooling pressure flaw and getting rid of pools.
- SPV nodes should be eliminated and replaced by fast bootable pruned nodes.
-We need to improve performance like 10 times at least

and the fork should take place once by an all-in-one package to have a stable next version coin for like another ten years. It is not simple job and not gonna happen overnight or without vast contribution.

[Stedsm]

Bitcoin is the people. What makes Bitcoin "Bitcoin" is the social construct, not the code, or the network, or the developers.

Bitcoin is not people, it is for the people. And it's an open-source code only that's being worked out as a digital currency just because it is a token that has value, and that value is given by people. Sorry to say this, but IMHO, you're wrong in your words, if social construct is what makes Bitcoin ^Bitcoin^, then the code (which makes it able to show those numbers correctly in your wallet and stop double spending), the network (that keeps the transactions going smoothly and securely without any intervention) and the devs (who are responsible as well as working hard behind BTC's back to keep it secure and let it remain the number 1 cryptocurrency are all equally responsible in the making of this beautiful crypto.

[aliashraf]

- SPV nodes should be eliminated and replaced by fast bootable pruned nodes.
-We need to improve performance like 10 times at least

I think you meant SPV node & wallet, but regardless it's tall order due to wallet on mobile devices/low-end computer and i doubt it's as fast as SPV wallet unless we can make people realize the importance of full nodes or/and weakness of SPV wallet

Ofcourse, SPV wallets are nodes anyway.

Suppose we have a pruned node that maintains like 1000 blocks height and is setup to work with minimum bandwidth ( Blocks-only mode,  maximum connections set to 5 and -listen = 0 ) with the exception of storage and bandwidth requirements for  bootstrapping  phase,  such a node would be as light as most spv wallets and still it is highly secure and trustless and directly connected to p2p network and It would be easy to make such a node more secure by switching its peers more frequently or increasing its connections in critical circumstances or when there is spare bandwidth available.

Now the only remaining step would be fixing bootstrap problem which is absolutely feasible and I've discussed it in many occasions. For instance:
 https://bitcointalk.org/index.php?topic=5046152.msg46692966#msg46692966 this post and my other posts on the topic are mostly focused on bringing the feature to bitcoin with a soft fork which is not necessary anymore because once we are getting our hands dirty with a hard fork it would be rather straightforward.

[Wind_FURY]

I might be nitpicking, but define "Bitcoin node". A true node, in my definition, is something that should validate transactions and blocks, then relay them to its peers if they follow the consensus rules.

SPV wallets or "light nodes" can't be called "nodes" by that definition.

Bitcoin is the people. What makes Bitcoin "Bitcoin" is the social construct, not the code, or the network, or the developers.

Bitcoin is not people, it is for the people. And it's an open-source code only that's being worked out as a digital currency just because it is a token that has value, and that value is given by people. Sorry to say this, but IMHO, you're wrong in your words, if social construct is what makes Bitcoin ^Bitcoin^, then the code (which makes it able to show those numbers correctly in your wallet and stop double spending), the network (that keeps the transactions going smoothly and securely without any intervention) and the devs (who are responsible as well as working hard behind BTC's back to keep it secure and let it remain the number 1 cryptocurrency are all equally responsible in the making of this beautiful crypto.

What the people believe is Bitcoin, by social consensus, will be Bitcoin in my opinion. It is why Roger Ver's debate that "Bitcoin Cash is Bitcoin" won't win.

[aliashraf]

I might be nitpicking, but define "Bitcoin node". A true node, in my definition, is something that should validate transactions and blocks, then relay them to its peers if they follow the consensus rules.

SPV wallets or "light nodes" can't be called "nodes" by that definition.

yes, you are nitpicking  

What you are calling "true node" is a full node operating in full functional mode. You can tune full nodes to operate in blocks-only mode and to use limited number of connections (both inbound and outbound) and most importantly to prune historical chain data.

SPVs are nodes just not full nodes and as part of my hardfork wishlist, I think they should be eliminated from bitcoin and replaced by pruned full nodes with built-in fast bootstrapping using UTXO commitment techniques.

[DooMAD]

SPVs are nodes just not full nodes and as part of my hardfork wishlist, I think they should be eliminated from bitcoin and replaced by pruned full nodes with built-in fast bootstrapping using UTXO commitment techniques.

What about wallets on smartphones and other portable devices?  I wouldn't want to see people forced into using custodial webwallets because there was no option to utilise SPV.  Even a testnet is unlikely to convince me that's a viable concept.  At least until technology progresses to the point where even people in developing nations had mobile devices capable of being pruned nodes, anyway.  That feels like a fairly long way off at present.

[aliashraf]

Now the only remaining step would be fixing bootstrap problem which is absolutely feasible and I've discussed it in many occasions. For instance:
 https://bitcointalk.org/index.php?topic=5046152.msg46692966#msg46692966 this post and my other posts on the topic are mostly focused on bringing the feature to bitcoin with a soft fork which is not necessary anymore because once we are getting our hands dirty with a hard fork it would be rather straightforward.

I don't understand much about SDUC, but do you think it's bootstrap/sync speed could rival with current SPV wallet/nodes such as Electrum? The idea is great, but if the bootstrap/sync time took few minutes or have high minimum requirement (hardware) people won't use it. As DooMAD and i mentioned earlier, your idea isn't realistic for wallet on smartphone or any low-end devices with low storage capacity.

Sure it is realistic. How would it possibly not be? let's discuss it in more details:

A bitcoin (Core) pruned node in blocks-only mode and with 1000 blocks height needs  4 GB storage for the blockchain and another 4 GB for the UTXO , it doesn't maintain a mempool at all (not interested in transactions) still it is able to participate actively in consensus by full validation  of blocks. Such a node utilizes very low bandwidth if other considerations are taken properly (maximum number of connections set to 5-8, -listen = 0 ) totally feasible for a mid-range mobile device to afford almost seamlessly.

As of fast sync, a SDUC compatible node with 1000 chain height would boot as fast as a conventional spv plus necessary time for downloading a 8GB (an average of 1000 seconds for typical 4G)  secured UTXO and blockchain plus the  time needed to validate the chain (~1000 seconds) it is like 2000 seconds (~30 minutes) for already old generation  technology.

I need to emphasis here: such a node is provably as secure as any bitcoin full node as far as we are discussing resistance to long range attacks up to 1000 blocks deep which is a huge security level breakable only by attacks that cost more than 50 bln dollars with current prices.  

Another point to mention: SDUC (Soft Delayed UTXO Commitment) was proposed by me  with soft forking in mind as the name implies, with a hard fork there would be no need for a few considerations as the protocol enforces commitments and validation of commitments as well.

[mixoftix]

We need to discuss everything in details with prominent Core figures and have their support

I was thinking about this.

Why do you need the Core developers' permission? The beauty of cryptocurrencies is it's open and permissionless. You have your right to freely make your own thing and release your own software.

the act of discussing more something is about picking options more wisely.. and when a thread asks for more prominent people's opinion about something critical like BLOCKING-ASICS, this could observe the TRUST of people, more and more. and being permission-less doesn't mean you can begin coding and after 6 months of hard working, you suddenly understand that it was all about wasting of time.

so this is about modeling before implementation [1] that we rarely see in crypto-world. after having an idea we somehow used to jump in implementation and don't care about modeling. sometimes I ask myself, why don't we have a DFD [2] approach within BIP steps!? introducing a distributed modeling language may be the next improvement in decentralization efforts.

(Linus Torvalds did not need to ask permssion to develop and release his own version of a unix-like OS did he.

but in crypto-world, a coin carries a certain amount of value not a process. you can manipulate a process so you innovate, but what would we call it if we manipulate a value?

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

reading the history of Nokia could be useful for the future of Bitcoin. a shift in technology and user experience out of sights of Nokia simply lead them to bankruptcy. Nokia CEO ended his speech saying this "we didn't do anything wrong, but somehow, we lost" [3] and this thread is going to discuss it, if such a thing is going to happen for Bitcoin too, how we could prevent it.


[1] http://agilemodeling.com/essays/modelingApproaches.htm
[2] https://en.wikipedia.org/wiki/Data_flow_diagram
[3] http://fun.putidea.info/2016/06/nidokidos-nokia-ceo-cries-during-his.html

Conclusion from http://fun.putidea.info/2016/06/nidokidos-nokia-ceo-cries-during-his.html:

The advantage you have yesterday, will be replaced by the trends of tomorrow. You don't have to do anything wrong, as long as your competitors catch the wave and do it RIGHT, you can lose out and fail.To change and improve yourself is giving yourself a second chance. To be forced by others to change, is like being discarded. Those who refuse to learn & improve, will definitely one day become redundant & not relevant to the industry. They will learn the lesson in a hard & expensive way

[Wind_FURY]

I might be nitpicking, but define "Bitcoin node". A true node, in my definition, is something that should validate transactions and blocks, then relay them to its peers if they follow the consensus rules.

SPV wallets or "light nodes" can't be called "nodes" by that definition.

yes, you are nitpicking  

What you are calling "true node" is a full node operating in full functional mode. You can tune full nodes to operate in blocks-only mode and to use limited number of connections (both inbound and outbound) and most importantly to prune historical chain data.

SPVs are nodes just not full nodes and as part of my hardfork wishlist, I think they should be eliminated from bitcoin and replaced by pruned full nodes with built-in fast bootstrapping using UTXO commitment techniques.

Hahaha sorry. I just want to make it crystal clear for all the newbies who are reading the thread that "any node that does NOT validate and relay blocks, transactions are NOT nodes".

The best example I can come up with are the Ethereum "light nodes". Those poor users might be fooled into believing that they are "nodes" because they thought they are doing a service for the network. They're not.

[mixoftix]

I'm looking forward to see a working PoC.

then may I have your (and other skilled contributors) opinion and reply with the idea of flash-back-pinning:

https://bitcointalk.org/index.php?topic=5089384.0

I think such improvement could lead us to another stage of co-existing with ASICs and mitigate the threat.

[Wind_FURY]

What? I believe that Moore's Law's "death" doesn't have anything to do with Proof of Work. I might have misunderstood, because if ASICs or GPUs have reached the most density, and efficiency for the most processing power it can make, how does it kill POW? Mining will not stop.

Mining never stops, but centralization happens.

In your theory.

the Moore's Law means: "the number of transistors on a chip doubles every year while the costs are halved."
now the death of Moore's Law means: "the number of transistors on a chip reaches the saturation point."

so at the saturation point, while people like miners only could have access to chips and regular processors, Quantum or Optic,... computer manufacturers (that do not follow the Moore's Law) could totally defeat them in a processing (and therefore mining) race and become the majority. even an ASIC manufacturer could utilize his new devices in a mining competition and after a complete ROI for his R&D costs, now send them to the market for sale. this means regular miners are always one step back. before ASICs involvement in mining process, a fairness existed over all miner entities. this fairness is important for the existence of hobby miners that constantly support their coins in good and bad days, not industrial miners that turn off their facilities with the first drop of price.

please see the future..

Another theory would be, Moore's law's saturation point might make Bitcoin mining's competitiveness become more moderate, and more economical, causing hashing power to be more distributed.

[mixoftix]

What? I believe that Moore's Law's "death" doesn't have anything to do with Proof of Work. I might have misunderstood, because if ASICs or GPUs have reached the most density, and efficiency for the most processing power it can make, how does it kill POW? Mining will not stop.

Mining never stops, but centralization happens.

In your theory.

the Moore's Law means: "the number of transistors on a chip doubles every year while the costs are halved."
now the death of Moore's Law means: "the number of transistors on a chip reaches the saturation point."

so at the saturation point, while people like miners only could have access to chips and regular processors, Quantum or Optic,... computer manufacturers (that do not follow the Moore's Law) could totally defeat them in a processing (and therefore mining) race and become the majority. even an ASIC manufacturer could utilize his new devices in a mining competition and after a complete ROI for his R&D costs, now send them to the market for sale. this means regular miners are always one step back. before ASICs involvement in mining process, a fairness existed over all miner entities. this fairness is important for the existence of hobby miners that constantly support their coins in good and bad days, not industrial miners that turn off their facilities with the first drop of price.

please see the future..

Another theory would be, Moore's law's saturation point might make Bitcoin mining's competitiveness become more moderate, and more economical, causing hashing power to be more distributed.

wrong analysis..

the first ASIC emerged in bitcoin in 2011 [1][2] but ASICs widely got used in 2014 [3] and your first graph shows how harmful are ASICs. your graph of 2019 is decentralized because of crash in market and price level of bitcoin, so industrial mining farms shut down their facilities. the next increment in price level changes the distribution of process again.. ASICs role as Sword of Damocles.


[1] https://thenextweb.com/hardfork/2018/02/02/a-brief-history-of-bitcoin-mining-hardware/
[2] https://en.bitcoin.it/wiki/List_of_Bitcoin_mining_ASICs
[3] http://blog.zorinaq.com/bitcoin-electricity-consumption/

[MysteryMiner]

By definition it is always possible to implement software routines in ASIC. And ASIC will always outperform other forms of hardware. All that is possible is to make current ASICs obsolete.

[aliashraf]

By definition it is always possible to implement software routines in ASIC. And ASIC will always outperform other forms of hardware. All that is possible is to make current ASICs obsolete.

Not true. I have provided explanations and reference materials up-thread, take a look.

[Wind_FURY]

wrong analysis..

the first ASIC emerged in bitcoin in 2011 [1][2] but ASICs widely got used in 2014 [3] and your first graph shows how harmful are ASICs.

Because the technological advantage of one group greatly surpassed the rest of the other groups.

your graph of 2019 is decentralized because of crash in market and price level of bitcoin, so industrial mining farms shut down their facilities.

Hahaha no. The hashing power in the network didn't fall by much. In fact, it's rising again now.

As one group's technological advantage diminishes, causing mining's competitiveness to moderate, hashing power will be more distributed.

the next increment in price level changes the distribution of process again.. ASICs role as Sword of Damocles.

Then let's see, and talk again.

[aliashraf]

@ٌWind_FURY, the diagrams you've included have nothing to do with hashing power and its distribution, pools are intermediary entities and belong to another class of centralization threats and I've read something about Bitmain hiding its dominance behind fake names.

[Wind_FURY]

@ٌWind_FURY, the diagrams you've included have nothing to do with hashing power and its distribution, pools are intermediary entities and belong to another class of centralization threats and I've read something about Bitmain hiding its dominance behind fake names.

But it doesn't change the debate that Bitmain's technological advantage is not as dominant as what it was before. Making of ASICs is more distributed today, and hashing power is also more distributed.

A hard fork to change the mining algorithm would be an opportunity for a highly capitalized company to be a dominant force in making ASICs for the new algorithm. It would be back to the same problem again, and with one company having the technological advantage.

[aliashraf]

@ٌWind_FURY, the diagrams you've included have nothing to do with hashing power and its distribution, pools are intermediary entities and belong to another class of centralization threats and I've read something about Bitmain hiding its dominance behind fake names.

But it doesn't change the debate that Bitmain's technological advantage is not as dominant as what it was before. Making of ASICs is more distributed today, and hashing power is also more distributed.

A hard fork to change the mining algorithm would be an opportunity for a highly capitalized company to be a dominant force in making ASICs for the new algorithm. It would be back to the same problem again, and with one company having the technological advantage.

Your claim about the end of Bitmain dominance is not supported by facts. The most crucial problem with such companies is not the lack of competition tho, it is more about the lack of transparency and accountability and conflict of interests.

As of your argument regarding the algorithm change having bad impacts on competition, it is based on a false assumption, a hype about some magical rule that says every algorithm is breakable by an ASIC. It is discussed and refuted above thread already:

ProgPow for instance is provably safe against such an attack because it is designed for a commodity ASIC from the first place: a GPU. Attempting to break ProgPoW by a hypothetical non-gpu ASIC is highly de-incentivized because the feasible efficiency gains would be lower than 20% which comes with very high R&D and manufacturing costs that users should compensate for by paying very high prices. Nobody invests on such a project ever.