Topic: Is the blockchain's purpose being redefined by the forked Ethereum Community? (Read 3118 times)

A hardfork, more often than not, modifies more then just technicalities. A hardfork literally makes changes that
are currently violations and not allowable, and transforms that aspect to normal allowable protocol process.
Hardforks are not normally used to just upgrade the network. In fact, hardforks actually breaks the network.

What you have been arguing all along is that blockchains are not immutable if they can become mutable.
You are now articulating a distinction between a social contract violating hard fork and a "technicality upgrade" hardfork.
Either way you want to argue it, the reason that immutability is not guaranteed is because of the ability to hardfork.
So your real issue, stems around the fact that blockchains can be hardforked, when broken down to its simplest form.
If a immutable blockchain has the ability to be hardforked at all, it can be made mutable at any time.

Your "tautological" argument only applies when mutability occurs with the chain.
All blockchains, by their nature, are quasi-immutable systems, that is being upheld by their communities.
If the community no longer maintains the immutability aspect, then yes, their blockchain isn't serving the original intent.


No, they are specifically different.
One is secured by the blockchain's game theory and the other is only externally secured by the community.
ETH's immutability was mutated by this external hardfork consensus, and not by the blockchain's internal game theory's failure.



No, what you are saying is wrong. All nodes can either follow the rules or not. There can be different software from different devs,
but if that software has rules that violate the current protocol, then that violating node is rejected from the network.
The node consensus network makes sure everyone is playing by the same rules, software origin is irrelevant.
This is the internal blockchain consensus and maintains the immutability in the blockchain (game theory).

When majority of nodes agree to all collectively violate the current protocol, causing a split and new chain,
that is a hardfork and is an external blockchain consensus that may disregard immutability in the blockchain.
This is what Ethereum did and is why I argue the blockchain did not fail and the community did, since this is
external consensus, which does not use the game theory as envisioned like with internal consensus.

In theory, if a malicious actor started up 50,000 malicious nodes, they could fork the chain as a majority consensus.
This type of action is sperate from the internal "game theory" blockchain consensus which Satoshi envisioned.



No, if a miner creates a violating block, it is rejected automatically by majority of nodes and the miner gets no reward.
Game theory says he wasted his time and money, so he will not do such a thing under normal circumstances.
If you personally accept that miner's violating block, you are no longer part of the Bitcoin internal consensus.
You are now using a violating internal consensus that only you and the miner are now following.
So, you are no longer on the Bitcoin network and now on a separate chain & coin that is not
interchangeable with the majority node Bitcoin network. You essentially hardforked yourself.

Now, if you are conforming to the internal consensus, like all the other nodes, and then by agreement, a majority of all  
nodes agree to apply "their external consensus" to cause a change with the internal consensus rules, then the majority
basically hardforked themselves from the original Bitcoin blockchain and created a new chain & coin.
This new chain will have a new internal consensus that was founded and instituted on the majority external consensus.
So, back to the original argument, the blockchain never did, nor can stop the collusion of the community to hardfork.
The "external consensus" is not bound by the blockchain system nor game theory.

Internal consensus and external consensus are entirely separate and serve two separate purposes.
Internal is designed to maintain and secure. External is designed to fork and change.
They clearly exist for different purposes and thus are different consensus mechanisms.

These are, to me, two aspects of the SAME concept of immutability, indeed.  After all, a hard fork, or the "internal block chain consensus", are almost identical processes.

After all, the "internal consensus" is agreeing which blocks are valid blocks.  This agreement depends just as well on the rules one wants to apply to decide whether they are valid in the first place, as to the application of those rules to the chain and the blocks themselves.   If everybody were to write their own node software, and implement their own vision on their own interpretation of the rules they want to see, we would immediately recognize that the internal and external forking consensus is in fact the same thing.  It is only because we all just use the "dev version" of the software that it seems that the internal consensus is what happens when we run the dev version, and "external consensus" is when we decide to download and install or not, a new version.

Suppose that different miners publish different blocks with different transactions in them.  If I build (or accept) one block over another, I'm applying "internal consensus" according to certain rules ; but according to whether I change these rules or not, I'm applying "external consensus".  If you think about it, this is in fact the same principle.  The best way to see that is if I wouldn't use software, but if I would "check the blocks I prefer" by hand.    Then my preferences, and my way of selecting (and constructing) blocks is MY way, and if I find consensus over that, there is no clear distinction any more between "internal" and "external". (of course this is practically not possible because of the computational load it represents, but as gedanken experiment, you see the point I hope).

Eh, yes, of course, if you understand by "hard forks", hard forks that modify more than just technicalities, but modify the original intend.  That's indeed exactly what I mean. 

For instance, a bitcoin hard fork that changes all integer representations from little-endian to big endian, is a hard fork, but doesn't modify the original intend of the bitcoin protocol.  It is a technical hard fork, but not a modification of immutability.  So that's OK.  But, say, changing the emission rate of bitcoin, is a change in the intend of the bitcoin protocol, and yes, that is a failure of the immutable block chain system.

It is almost tautological: if there is mutability, then logically, the immutable system failed, no ?

Its been very interesting following the Ethereum fiasco. Eagerly watching how the ETC/ETH thing turns out.

Found an interesting quote from here:
http://nakamotoinstitute.org/mempool/ethereum-is-doomed/#selection-357.0-357.305

 

ETH's consensus mechanism was lowered, so that they could argue that majority reached a consensus for hardfork.
The ETH community argued that since ETHers who had a vote, did not vote, that non-vote is a vote for yes,
as opposed to a vote for no or neither. So in their new mechanism, an empty vote is a yes vote.
That is malicious consensus, since in normal systems, those empty votes fall under no or just not counted at all.
It was not a "play fair consensus", even if others would wish you to believe that. They changed the rules.
There is a real difference between real consensus and faked consensus.

The fact that you agree that ETH was able to implement it's hardfork in light of the game theory you claim exists
to prevent such change, only proves my argument that game theory does not apply to this form of consensus.

I think your real theory is, without articulating it, seems to be that:
hardforks are a failure (or a failure vector) for the immutable blockchain system.




No. You have mixed up apples and oranges. You have taken two separate systems with totally separate rules
and laws, devised as such intentionally, and have combined them incorrectly to form the basis of your argument.
You are incorrectly combining internal blockchain consensus, with external forking consensus.
Game theory consensus can not apply to malicious external forking consensus, such as what ETH did.



There may be double negatives in your statement, or not. I don't understand your statement.
Please rewrite your intention in a different way.
Game theory only works within the system that it was specifically designed for.



You are talking about internal blockchain participants, not the external ones.
External participants are not bound by the consensus systems of the blockchain.
In fact, they have separate consensus that is only controlled and monitored by
the "community" and not game theory or code function.

One of the jobs of the community is to monitor developer commits/pulls to each new version of Bitcoin.
That is one reason (there are many other reasons as well) why Satoshi created it as open-source software.
Satoshi placed the future of the system in the hands of the public gatekeepers, to prevent private control.
If the blockchain was intended to be what you are arguing, then it doesn't require open-source or observance.



Except your example lacks one important detail which causes it to become entirely invalid and not in context.

Normally, a company will go bankrupt because of a past incorrect action that led to the future overall financial failure.
So, bankruptcy is a 50/50 possibility in all forms of business where value and attaining more value is the goal.
What actually occurred with ETH is that "their stockholders" willingly and purposefully voted to become "bankrupt",
as per your example. So in effect, they voted to become a mutable chain through the external consensus process.
That is not a company (blockchain) failure, but a stockholder (community) failure. They willingly voted to lose future
value by removing immutability, to undo a failure with the product they were "selling".

How does a "game theory consensus", prevent an uneducated community from a majority destructive vote?



Yes, but Satoshi very well understood that game theory for immutability only exists within the blockchain,
and when it comes to outside the blockchain, such as things like hardforks, it does not apply.



No, it does not apply to all types of consensus within the system.
Game theory can only be applied to the internal blockchain consensus.

There is no difference.  Calling things malicious or not is a moral verdict, but doesn't change a thing.  I call the ETH fork malicious, but others will call it "fair play consensus".  It doesn't change the fact that it got implemented.


But that's the whole (sociological/game-theoretical) essence of the block chain invention: that it is SOCIOLOGICALLY and GAME-THEORETICALLY not possible to find a consensus, other than the immutable one.

That was, in my idea, the fundamental assumption of a block chain: that it is sociologically not feasible to find any majority collusion over anything else than the original rules, and the correct history recorded until then ; that the advantages that it can bring by breaking that, would never reach a majority in a very diverse and large set of actors, and that never ever, a majority would find an advantage in doing anything else but keeping with the original rules, and keeping with the correct history (in other words, that no majority can be found that finds an advantage in a *specific* deviation from immutability, although almost every actor would find an advantage in a *different* deviation from immutability).  In other words, that immutability is the least bad solution for the majority.


Game theory assumes of course that every actor optimizes its gains ; the idea is that there is no possibility that every *specific* deviation of immutability would never profit a majority of CPU power more than keeping immutability.


There is no "job" and there is no "community".  There is only a set of people looking for maximal gain.  The idea of a block chain is that those people are constrained, against their own wishes, to keep immutability.  Call immutability a Nash equilibrium of the game-theoretic analysis of a block chain setup.


You could put it that way.  But you could also say that during the period of immutability, people have had the opportunity to use it as an immutable system.  Those that crossed the "border of immutability" had bad luck, like people keeping stock of a bankrupt company.

It is somewhat similar.  The stock of a company reflects normally the diminished cash flow of future dividend.  As long as the company is live, the value of its stock is the market's best projection of its future cash flow.  So if you buy stock, and you sell that stock later, while the company was running all right, you have been using the "stock system" to your likings.  When the company goes broke, the stock drops to 0.  Those holding the stock at that moment, have bad luck.

Same for a block chain of which the immutability is broken.  As long as it wasn't broken, it was running like a company stock, and you could buy it, get dividend, and then sell it again.  Once immutability is broken, it is like a company that went bankrupt.  (with ETH, the funny thing is that people still trade the stock of a bankrupt company as a betting token...).

The idea is that the chain doesn't break, like the idea is that a company doesn't go bankrupt.  But things can fail.


As I told you already, this is my own analysis.  I don't consider Satoshi as the final prophet of truth, although he did bring a lot of good ideas in.  So I would think that in as much as Satoshi understood the system he was building, he had to assume immutability (in INTEND, not in technicalities !).  I'm somehow convinced that he intended that for instance, 21 million bitcoin was to be something immutable.  He talked about irreversible transactions.  For someone assuming mutability, that would be strange statements, no ?
But in as much as he was confused about that (I don't think so), that's no reason to remain confused about it.
Einstein invented General Relativity, but people have improved upon it even if Einstein didn't understand certain aspects of his own theory in the details later scientists understood it. 



Of course it does !  It is the essence !

Not only do I mean code part, but I also mean mechanically and functionally.
The game theoretically mechanisms of the blockchain, only apply and function within the blockchain.
I do not believe that game theory applies to consensus.
Game theory can not prevent malicious consensus, only a assumed play fair consensus.

Immutability can never be guaranteed within any human devised system.
The only aspect of game theory that would apply to immutability is that the gatekeepers
would keep the old immutable chain alive, since their is no action the gatekeepers can take to
prevent a malicious consensus hardfork.

Game theory does not prevent a violation, it only uses human assumptions to control the humans.
If I could bribe everyone with more value, they may destroy their own chain willing.
That is not a failure of the chain or the game theory aspect, but of their community and its own ignorance
of what their own job is. Users and other members of the community are not passive in this type of system.

Game theory in itself is a failed theory, since it assumes humans would not risk value over destruction.



I don't understand.
You are basically arguing that if a blockchain that is immutable now, is made mutable by a violating change later,
then the blockchain was never immutable to begin with? If yes, then you are arguing semantically and I would agree
semantically. I would then go on to advise you that there is no blockchain that is immutable and there will never be,
by your semantical reasoning. In fact, I would also advise that humans will never be able to create any perfect
immutable system, since all matter (in this universe) is mutable.

Semantical reasoning and argument is not beneficial toward true understanding of this system.



Satoshi did not create or intend perfect immutability and I'm not aware of where he made claims of such.



The game theoretical assumption(s) do not apply to consensus.
They only apply within the blockchain's functional dynamics.
Consensus exists outside that function, so game theory does not exist with consensus.



Humans can not create perfect systems, so, humans will never create perfect immutability.
Hardforking is not a mechanism to prevent the colluding majority, it is the result of their collusion.
When a hardfork occurs, it is because the majority has agreed (in theory) to that new version.
Maintaining the old chain is the only way to prevent a full "imposing of the collusion".



No, a 51% majority of the hash power is not the same as a majority of node consensus signaling, for a protocol change.
A 51% attack is a malicious action by a miner, who has attained 51% share of the hash, and now is
attempting to revert/edit/undo past work ON THE CHAIN that will benefit themselves (and destroying the Bitcoin system outright).
A 51% attack is not power to change the protocol and its intended function, but only power to reverse work on the chain.
The reason why miners do not reverse work is because game theory says it will cause loss in value of the chain and then token.
But a malicious government miner attacker doesn't value that, so they could perform this type of attack and
the game theory will not apply to them, since their intention is only the destruction of Bitcoin.



I don't understand.
Miners can not do what you are describing. Your example will be rejected by the rest of the network.
There are thousands of nodes and other operator systems that form the full consensus mechanism.
You can get all the miners on board, but that does not guarantee full consensus if the users and economies do not agree.
For Bitcoin consensus, you need around 95% of the whole community, is my understating.

Now if the whole community did allow the miners to do what was in your example, than yes,
the community failed itself by coming to consensus to allow a social contract violating action into the system.



Yes, Bitcoin is not entirely just a majority vote = change. But the community can vote on anything, technically.
But the fact that a vote is done and agreed upon does not mean that the result is a correct action.
Sometimes humans vote for their own destruction (ex: Ethereum hardfork that violates social contract).

Bitcoin's Immutability aspect is not governed nor maintained by game theory in any way.
In fact, I think there is no possible mechanism that can be created to stop humans from being human.
Immutability is only protected by gatekeepers who are informed and stay vigil.

Of course, what I write is my own understanding of block chain technology, not Satoshi's.  But I think I'm actually in agreement with the spirit of his paper.  If it were true that the idea was that block chain technology is there to implement whatever the majority decides, then there would not be any mention of a 51% attack, because that would not be an attack at all, but just an expression of the desire of the majority (majority in Satoshi's paper being measured by computing power).

You cannot at the same time claim that the system is to be the implementation of whatever the majority decides, and talk about "irreversible transactions", and whine about an "attack" if ever the majority as you defined it (namely, the majority of "CPU power" in Satoshi speak - he didn't realize ASICS were possible) decides to revert a transaction - which is his example of a 51% attack.  It would be just "the will of the majority" and everybody should be happy because the system "works".  (that's exactly what happened with ETH).

This, to me, is the clearest proof that "block chain technology" was not meant to be just the implementation of "consensus", defined as whatever the majority (of mining power) decides.

That said, that is what block chain CODE actually implements.  The block chain code principles allow a majority of computing power to change just anything, whatever they agree upon: doing away with past transactions, modifying entirely all the rules of coin creation and difficulty, .... but this was clearly NOT the idea.

To bring both in agreement, one has to make game-theoretic and hence sociological assumptions, namely that no majority ever will be found colluding over a deviation from immutability.  This is why I include them in my definition of "block chain tech".

I suppose that by "inside" you mean "code", the technical part, not the sociological part.
Well, it is obvious that that cannot work.  There is absolutely no "code/technical" way to establish consensus if there is no sociological/game theoretical part to the system.

You cannot just have code, without any game-theoretical assumptions, and suppose that any form of immutability will come out of that by itself.  It is sufficient that people change the software, and the block chain can be totally altered, or its interpretation can be totally altered.  If the software checking for the validity of PoW is altered, no amount of PoW will keep what so ever immutable. 

Immutability comes from game-theoretical (and hence also sociological) assumptions, not from code.  If people start propagating altered versions of the block chain, different software checking validity and so on, no piece of code can ever stop that and "provide immutability".  The whole immutability comes from the fact that people will be put in such a situation where it is very unfavourable for them, or essentially impossible to organize for them to do anything else but to STICK to the immutable consensus.


But it is essential.  If the devs could intervene at their will to change those things that are supposed to be immutable, your block chains would not keep any little bit of immutability, and be totally mutable and random.  The whole point is that even the devs cannot do this.

However, they CAN change the code, as long as this code remains COMPATIBLE with those aspects that are deemed needed to keep immutability, and that itself is the whole game-theoretic premise on which the whole of immutable block chains is based.

That is:
- the devs (or just anybody) can bring out another node software that is perfectly compatible with the former v0.1, but, say, with a fancier GUI, or with faster algorithms, *perfectly* respecting the protocol on the technical side.  You could have myriads of different but compatible node software.

- the devs (but not just anybody) can bring out node software that is technically not totally compatible with the former v0.1 and hence is technically a soft or hard fork, but which RESPECTS IN ALL ASPECTS the INTEND of the former protocol, and just improves a technicality.  The reason why people will upgrade is that the intend of immutability (all the economic implications) is perfectly respected, so there is no reason NOT to upgrade (like accepting a security update to your computer system), and not upgrading can put you with an incompatible node.  In other words, game-theoretically, there's no good reason not to upgrade if people agree that the technicality is improved now.

- and that's it.  The devs, or any other "leadership" shouldn't be able to modify whatever to the intend of the immutability.  If they can, then there is no immutability at all.


Well, I consider that as a failure of the block chain purpose.  If that happens, the block chain didn't do what it was made for, namely keeping immutability of intend.  Once you can do that, anything goes.


Call it as you like: the system failed.  The "code" didn't fail, but the block chain system failed, because the game-theoretical assumptions turned out not to hold.


That is a strange statement.  I will tell you why:


I would rather think that this was a shortcoming of the technology, rather than a feature.  That is, one couldn't think of a mechanism where one could even block a colluding majority from imposing its will (in fact there is one: hard forking).

Because when you think about it, if a colluding majority is the "normal way to alter block chains", then there is nothing wrong with a "51% attack".  It wouldn't be an attack at all.  It would just be the normal majority wanting to alter history, if you keep your viewpoint.

The idea of a block chain was not to have a "majority vote over what was the distant past", although that is what comes technically out of it.  The fact itself that you call a "community where a majority decides to DESTROY their block chain" indicates that you yourself consider that even if a majority decides something, that this is NOT what consensus is about.

Imagine that a majority of bitcoin miners decides that only they will now hold bitcoins, and that all former transactions that assign coins to addresses of which they do not have the keys, are invalid ; for that they have to write a lot of "exception rules' in the code accepting modified old blocks.  Hell, in your viewpoint, such a total alteration of the past would be "consensus" because decided by the voting majority of miners ?  Of course not, this is total breaking up of immutability and - as you actually call it - the destruction of the block chain system.  Nevertheless, the CODE is running correctly, has been altered with majority consensus and "everything works by consensus as wanted by Satoshi" ?

Of course not.

A block chain system is not supposed to be a majority vote system over just anything.  The only thing the majority is supposed to vote over, is the respect of immutability.  If anything else happens, the system fails.  So the game-theoretical part is an essential part of the block chain system, because without it, there's no such thing as immutability, and hence no purpose for a block chain (the code/technical part).

The blockchain was intended to maintain immutability ONLY within itself and not from outside intervention.
Outside intervention is left open so that the protocol can be upgraded by "consensus" when the time is appropriate.
Without this intervention by the devs, the code could never be "upgraded, fixed, or etc.", so version 0.1 would be the only one.
The consensus mechanism exists outside of the blockchain code and there is no sociological code within the blockchain.
The sociology aspect that you are relying upon, is actually based upon a community/miner/node gatekeeper premise.

If the gatekeepers (majority of consensus) wants to destroy the blockchain, it will happen, since Satoshi intended it to.
If Satoshi intended it to, and in fact designed it as so, then the failure rests upon the gatekeepers and not the blockchain tech.
The community failed to maintain their social contract and their "ideals".



The consensus mechanism does not exists within the blockchain and each altcoin's community can set out and
agree to their own consensus mechanism. The consensus mechanism was NEVER intended to prevent changes
to the original rules. The consensus mechanism is ACTUALLY intended to allow for a way to make changes, if
deemed appropriate by the majority of the community. If the community wants to destroy their original social
contract, and they have majority, they will destroy it. Consensus allows possibility for change, not prevent it.



Where in the whitepaper does Satoshi outline what you are saying?
When Satoshi talks about the "consensus mechanism" he is ONLY addressing a way to vote for a change, not to prevent a change.
If majority do not vote for a change, the correct course of action is that no change should occur, under his reasoning.
Like was, if the majority do vote for a change that violates the social contract or other important systems, they can by consensus.
Satoshi designed and intended the "community" to use free consensus to determine the future of a blockchain, not the other way.
A blockchain and it's "technology" can not determine it's own social contract or community consensus.



The current implementation of Bitcoin (I assume you mean the immutable aspect) is being enforced by the
devs, miners, and users, and not by the blockchain code itself. The current community consensus says not to
change the code to "reverse/roll back/edit" a past work on the blockchain. The current thinking is that it would
violate the intent and goal of the blockchain and the purpose of it's creation.
We can agree that there has been no purposeful change to the Bitcoin blockchain, that would violate
the original intent and goal of Bitcoin/bitcoin.

But it may be possible that around the years of the 2100s, the Bitcoin community could form a majority of consensus to
change the 21 million coin cap to a higher limit, in order to continue the mining process because it is determined that a
small subsidy (0.125btc per block, for example) is needed indefinitely because of factors and understandings that are not
knowable at this time. If that scenario becomes a reality, then the blockchain will change and conform to the wishes of the
new gatekeepers, since the blockchain social contract is not contingent on the blockchain technology, but the community itself.

So, if a blockchain performs a purposeful fork by consensus, and it violates the original social contract,
that community failed itself by going back on their word. In theory, when you participate within a coin and it's community,
you abide by those rules and ideal and would uphold them since you normally participate with things you agree with.
If you wish to fundamentally change the social contract, usually it is due to community ignorance or individual(s) malicious intent.

As I have to re-define them to address the points I'm making, I cannot "misunderstand" them.  It is true that I have to redefine them, because their standard definitions take implicit properties into account, which are not correct.  As such, I cannot use the standard definitions, but I try to find meanings that are as close as possible to these standard definitions.

I consider that "block chain technology" is a cryptographic system which *includes* a sociological part (as in fact, most cryptosystems do, but this one is more involved).  I consider that this technology is designed to achieve a specific property, which I call "immutability":
1) an agreed-upon history of events, such that once they are agreed-upon, this history is never going to change (no early additions, no early removals of events).
2) an agreed-upon set of rules, the "protocol intend", which gives meaning to the recorded events, which are also never going to change.
We call the realisation of these two properties "immutability".  The GOAL of block chain technology (including sociology) is to realize immutability (of history, and of protocol).

There is a technological part to block chain technology, which is all the thing with software, cryptography, hashes, keys, and all that stuff.  And there is a sociological part to it.  Both elements (the purely technological part, and the sociological part) are MEANS to reach the GOAL of immutability.

The sociological "technology" part of block chain technology is "consensus".  Consensus, in this frame, means: not being able to reach a majority between participants over anything else than the correct history, and the original protocol intend.

The cryptographic technology is such that a majority of miners can impose their will ; the sociological technology, called "consensus" is such that no such majority can be found, if it deviates from the correct history, and the original protocol intend.  Both together are supposed to guarantee "immutability", which was the GOAL of block chain tech.

So immutability can fail, if the crypto fails, if a minority can crack, say, the PoW scheme, can do double spending or whatever.  This is a failure of the technological part of block chain technology.  And immutability can also fail if a majority CAN be found that colludes over something else but the correct history of events and the original protocol intend.

If one of these things happens, the block chain technology has failed in its goal to maintain immutability.


I fail to see the difference.  The block chain was intended to maintain immutability by preventing the realisation of a majority over anything else but the original protocol and the correct history of events ; however, such a majority was nevertheless found.  So the consensus mechanism, the sociological pillar of block chains, broke down, as something happened that was supposed not to happen (finding a majority over something that was not supposed to find a majority).


You are perfectly right, but the consensus mechanism is a big part of block chain technology.  One pillar is code, the other is sociology, the famous consensus mechanism, that was supposed to prevent being able to find a majority over anything else but the original rules.


No.  Again, the block chain technology paradigm rests on two pillars: code and sociology.  You cannot put the sociology in the code, but it is nevertheless part of the fundamental hypothesis of block chain tech, in order to achieve its goal: immutability.


Uh, of course I'm talking about "an idealized version of block chain technology", namely its intend and goal.  However, to my knowledge, bitcoin still implements that ideal pretty well.  I have no knowledge of bitcoin ever finding a majority to deviate from the original intend of bitcoin, nor of any history that had been totally altered even though that history was generated perfectly according to the intend of the protocol (i'm not talking about buggy code that generated events that are not correct according to the intend of the protocol).

The only way I can reconcile what you have and are arguing is a misunderstanding of the words you are using.

For example, if you use the phrase of "a failure of blockchain technology" to mean that the blockchain was designed to prevent an
action, yet that action is currently allowable, then that is incorrect as to what has occurred. But if you use the phrase of "a failure of
blockchain technology" to mean that the blockchain lacks the ability to prevent an action, then that is correct and has occurred (in ETH).

The point is that you seems to be using the phrase as an actual failure in the system, which is wrong since the system was not intended
to fix those issues you address, and fall under consensus mechanisms and not code. In fact, those three issues you address above, do not
and have never existed in the Bitcoin code. If your understanding is that Satoshi did address those in the code, you are misinformed.

On the other hand, if you are just stating that blockchain technology can not programmatically address those three issue in code form,
and that alone is itself "a failure of the blockchain technology" than that is technically correct but using bad word phrasing.

Though, I am sure you are using the term of "failure" to mean an actual failure in the system, which I disagree with, I have attempted
to dissect your words and meanings so that they are made consistent with the actual history and understanding of Bitcoin and the
blockchain, and not an idealized version that the media and others have been perpetrating upon the ignorant masses. If you disagree
with me and intend to argue an idealized version of blockchain technology, then I have no choice but to assume you are purposefully
intending to obfuscate with semantics and word games OR don't know what you are talking about.

Otherwise I do not understand why your argument is: humans are bullshit artists, so thus blockchains can be airplanes, or anythings.

The problem with established definitions is that they take sometimes properties of theorems for granted.  By re-defining terms, it is true that conversation becomes more difficult, but on the other hand, that makes one think of implicitly assumed properties that weren't stated or obvious: by excluding them explicitly from the new definition, one immediately sees the difference.

We all agree on the purely engineering side of block chain: it is a software object that is distributed amongst different agents, with "code" (methods) and "block chain data" (attributes) that takes on a specific form, namely a linked list, in such a way that there are cryptographic techniques to add an element to the list ("block") and that there are testing methods to verify whether this happened "according to the rules".

The whole question is: what good is such a software object ?  What purpose does it serve ?  And this is where we make extra assumptions.  Given that this software object is distributed amongst different agents, they can all alter it at their guise, but there's one thing that all of them have to do: the "checking methods" have to agree that the linked list is "according to the rules".  If you distribute an object with code and a chain of which the code says that the chain isn't valid, you won't get anywhere.  But there are (many) ways to alter the chain, and/or alter the code, in such a way, that the new object is saying of itself that it is "OK".   So there must still be a SEMANTIC IDEA that this software object is implementing, because if not, there are miriad of ways to distribute different, and internally consistent block chain objects.

My claim is that the semantic idea of a block chain object - the reason why it was invented - was to implement
1) immutability of the past and of the rules (the *semantic* rules)
2) consensus about the newly added blocks (according to 1) ).

There are essentially 3 ways in which this can fail (in which this kind of software object is not going to implement said semantic idea:
1) If one changes the data of the past while respecting the original rules, this is called a 51% attack.  The software object is constructed in such a way that this is only possible if one can deliver more PoW than those keeping the past as it is.
2) by changing the rules in such a way that the rules are semantically different and have the other agents accept that: a hard fork
3) by not finding a consensus over newly added blocks (this ends up violating one of the rules such as "largest PoW chain wins" but one could pretend not having "seen" the other block chain fork: the code will work.  Imagine the great wall of china blocking bitcoin network traffic).

Now, if the question is, whether the block chain object has failed to deliver its semantic meaning if one of these three things happen, I would say "yes".  Once you consider that these 3 things happening are "normal" block chain behaviour, of course you can still use the engineering object "block chain", but one can wonder what it is semantically good for, if you can change the past, if you can change the semantic rules, or if you cannot find consensus on the next block.  Of course technically it still exists, but what does it mean now ?  

That was my whole point: if you do not consider 1) or 2) or 3) as a block chain FAILURE then the semantics of block chains is anything anybody wants to give to it.


On the contrary.  My point was that IF you consider that the block chain concept DIDN'T fail with ethereum, THEN one has to accept that block chain now has just any semantics any person gives to it.

A limited analogy: a screw driver was invented to drive screws.  You can of course also eat rice with it.  But if you ask whether people eating rice with screwdrivers have redefined the meaning of screw driver, you have to conclude: yes.  There's nothing wrong with buying screw drivers to eat rice with.  But obviously that was not the design purpose.  The analogy fails because the designer of the screw driver didn't have in mind an IMPOSSIBILITY of eating rice with his screw driver, he couldn't care less, while the inventor of block chain had an impossibility in mind.  When this impossibility materialized, you can only call this a failure of the original design, can't you ?

So in as much as we do consider ETH still a block chain, yes, this has re-defined the notion of block chain.  My point is rather that ETH illustrated the failure of the block chain semantics in their case (which means that block chains CAN fail as we just saw one failing).

It seems there can be no reconciliation of our two view points.
We fundamentally disagree on known definitions and established understandings in many different areas.
You are attempting to rationalize that the blockchain system has failed, by redefining common things and terms.
Interestingly, my original question was based around parties attempting to redefine established cryptocurrency terms.

Thank you for explaining your rational, but I think with your type of reasoning, you are saying that nothing should
be classified and thus things can not be defined, thus a blockchain is whatever the individual thinks or wants it to be
(even what it should have been), thus my original question is without merit, since in your vision of the world the current
human system of classification, categorization, and specialization is improper and possibly immoral.

Because of those reasons, this conversation isn't very interesting nor actually amounting to anything.

You may have to rethink that.  Every bank, and every credit money processor issues his own token, making the holder believe that it is another token.  This is the essence of fractional reserve banking: they can "peg" their tokens against another one, as long as they only need to actually do that with a fraction that is smaller than the fraction of their own tokens they hold for real.  VISA, the credit company, issues credit tokens, making the accepters of these tokens think they are dollars or Euros, themselves not "real" dollars or Euros but specific bank tokens of whatever bank is connected to the VISA account with which one paid.

It is the basis of the modern fiat system: pegged tokens with a fractional reserve of "real" tokens of the layer above it.
There used to be gold, there's nothing any more.
The next layer is the central bank with their own tokens, the actual dollars/Euros....
The next layer are the commercial banks, directly holding accounts at the central banks.  Their tokens are "bank accounts" which are pegged to the central bank tokens of which they hold a certain reserve.
The next layer are credit and other financial institutions with accounts coupled to commercial bank accounts.  Their tokens are pegged to the tokens in commercial bank accounts.  VISA is such an example, but a supermarket giving credit is exactly the same.


They are pegged to a few central bank tokens, but they AREN'T central bank tokens, but their own.


I don't think that there are different classes of users.  They are all attackers, including miners, developers, traders, commercial people.  That's what decentralisation is all about: no particular status, no privileges, no castes.  So there is not the caste of developer.  Everybody who wants is a bitcoin developer, trader, attacker, ... People are not classified into groups, or the system would be non-decentralized.  We are all the same, so we are all attackers.  We would all modify the block chain record to have more.  We would like to distribute code which forks, and assigns us an advantage.  We would all like to rewind the block chain record when we pay someone.  On the other hand, we wouldn't want to rewind when we receive coins and provide services.  But we would like to crack other people's keys so that we can assign them to us.

Only, WE CAN'T, and the most profitable way to "attack" the block chain individually, is, in the end, to defend its protocol and to play by the rules, no matter how we would all like to change them to our advantage.

THAT, to me, is the fundamental part of block chain technology.  In as much as Satoshi didn't say this (but this is how I understand it), then it is Dinofelis who says so


Ah ?


I take this one step higher on the scale of abstraction, and I consider that there are no classes of people called miners, users, and developers.  There are just AGENTS.  They all do what they think is most profitable for them.  Some think that by modifying the code they will make profit, others think that by mining they will make profit, others think that by "using" bitcoin as a means of exchange or speculation, they will make profit, still others think that by trying to cheat on the system they will make profit but these are simply STRATEGIES of AGENTS.  And to me, block chain technology is an invention that makes that out of all this antagonism and non-collusion, APPEARS an immutable system (protocol) with an immutable ledger.

I read Satoshi that way, but in as much that this is not what he meant (we'll never know), or in as much as people don't understand it that way, consider it my view on block chain technology.

I consider "block chain technology" somewhat similar as the invention of life: an emergent property of complex organic chemistry.  "Life technology" is then the fact that competing reproducing organic systems give rise to evolution and more involved forms of life.  If you say that "life technology is only DNA", then I tell you that you are missing the essence of life technology.

No, because for many different reasons.
The most obvious one is that VISA and other payment processors are not their own currency or token.
All those systems are directly pegged to a world recognized currency such as USD, CAD, EURO, YUAN, YEN, and so on.

A blockchain does not automatically equal crypto-currency, and a crypto-currency does not always need to use a blockchain.


...