DECENTRALIZED crypto currency (including Bitcoin) is a delusion (any solutions?)

[Fuserleer]

I want a solution. Period. Who ever provides it will get my respect. If I was an independent observer, I would say both you and I have failed for 3 years, so I would say we are both full of shit.

And thats why you won't succeed.

You see it as we're full of shit, I see it as we're evolving, maturing and improving.

You see failure as the end, I see it as the beginning of something better.

You see failure as a lack of knowledge, I see it as a learning experience

You see the glass half empty, I see it half full.

Get it yet?

What I get is that I need to reach a conclusion. Eventually I learn all the possibilities and realize the fundamental truths. I have a very abstract conceptual intellect and eventually it comes to the point where I understand what is possible and not possible. I might be wrong, but any way we are derailing the thread.

Let's talk specifics on technology. Talking about my personality and whether I am too negative is off topic. If I for example have some further insightful comments to make, they should be taken on the merits of the technological points I make.

I am not entirely pessimistic. I am just trying to be realistic. I have lost 3 years. Three fucking years to this shit. I want to drill down now to the conclusions. Going on and on isn't necessary for me, because I have now the holistic view of all this. Again you will say I am being egotistical if I claim that. So again let's just stay on topic of specific technical points. Until you reveal the details of  your design, I can't make specific technical comments on it.

Btw I don't dislike you. And I am not a person who is unfriendly. I just hate losing time. I am 51. I have many problems to solve. No time to lose...

I think we all just need to accept that there's a high possibility that there is going to have to be some compromise in the system.

We are all striving for perfection, which is commendable of course, but no one knows if there is a solution to achieve the perfection we all desire.

Just accept the compromise, but take solace in the fact that why it might not solve ALL problems, future generations of this technology will likely be an improvement over Bitcoin and clones by some degree.

Going back to my cancer statement, surely its better to provide a solution that prolongs the life of someone by 10 years, if an outright cure is not available.

The same is true here, Bitcoin is the first attempt to cure a cancer, and while it might not be an outright cure, its inspired and motivated many to also start looking for that cure.  Many will follow that are not outright cures, but does that make them entirely worthless?  Of course not.

[TPTB_need_war]

Okay let's get back on track of trying to determine what is the best design (in terms of best of the imperfect possibilities) and enumerating the tradeoffs and flaws.

[TPTB_need_war]

Getting this topic back on track, I have a question:

* If mining rewards cause centralisation, how do you handle distribution fairly in a system with no mining rewards?

Mining rewards can still be paid to PoW and it can still be unprofitable if the level of difficulty rises too high because every transaction is including a PoW share. Transactions just keep growing and growing while the debasement rate is a constant percentage. If the value of the currency grows as fast as the transaction rate, well then I guess that is a great problem to have.  

If mining farms are dumping their coins while the price is rising that fast, they are leaving all that money on the table for us.   Meaning the banksters are shrinking their relative net worth.

I assume you are referring to ongoing (a.k.a. tail) emission. For initial distribution, then profitability of mining appears to not be a flaw. I would argue that distribution to HODLers is a market failure plan though. I'd prefer to initially distribute to users than investors. Let the investors buy from the users in secondary markets such as your metaexchange. I would prefer if earliest investors have to work even harder to get coins early by creating businesses which accept the coins from the users. So the early investors add even more value to the ecosystem. As I had explained to you in my vaporcoin thread, I think users would have no idea what an exchange is and how to access it, thus earliest investors would have to work harder to source coins in volume. And that IMO is the way it should be. It shouldn't be so easy to invest as an early adopter, that all you need to do is buy an IPO. There is no capital created from such an activity, unless you can argue that these investors work on improving your coin (but they seem to mostly just clog up the forums with shrilling) or they are funding your coding (but that would normally come from angel investors not an IPO in a non-pump&dump). At least with Kickstarter, your customers who are your early investors actually help you design and refine your product.

Think out of the box.

[Gladimor]

You guys should take a look at SpreadCoin- the truly decentralized cryptocurrency (no pools, prevents centralization of hashpower, etc).

https://bitcointalk.org/index.php?topic=1045373.0

[minor-transgression]

"You must use PoW or PoS and thus the problems of centralization and government control are the end result of all of what we are doing here. And helping to force a world government cooperation of regulating the internet, encryption, and Bitcoin mining. Bitcoin is a Trojan Horse that weakens the nation-states and traditional banks, which must fight back by cooperating with a world governance regulation of the internet and Bitcoin mining."

Disagree - once inequality passes a critical limit revolution is inevitable. While some would prefer
a world government, it is not an optimal solution, indeed, there may not be a single optimal solution.

With regard to cryptocurrencies, I doubt that a single cryptocurrency ie bitcoin, can provide an optimal
solution to the paradox of thrift, and for that reason alone other currencies must exist.

Before taking this further, what is the CAP theorem referenced earlier?

[Peachy]

Before taking this further, what is the CAP theorem referenced earlier?

No offense, but that question is akin to walking into the Large Hadron Collider complex to ask "Where's the bathroom?"   

Relevant Info:
http://lmgtfy.com/?q=CAP+theorem

[minor-transgression]

My biggest mistakes have happened when the obvious question went unasked.

Thanks for the reply, though.

[Come-from-Beyond]

No offense, but that question is akin to walking into the Large Hadron Collider complex to ask "Where's the bathroom?"   

Relevant Info:
http://lmgtfy.com/?q=CAP+theorem

I used this and noticed that http://www.infoq.com/articles/cap-twelve-years-later-how-the-rules-have-changed appears on the 2nd page which is read by noone...

[Fuserleer]

No offense, but that question is akin to walking into the Large Hadron Collider complex to ask "Where's the bathroom?"   

Relevant Info:
http://lmgtfy.com/?q=CAP+theorem

I used this and noticed that http://www.infoq.com/articles/cap-twelve-years-later-how-the-rules-have-changed appears on the 2nd page which is read by noone...

Ive read that many times, and the first time I read it was AFTER I'd formed the basis of theory from which channels operate (believe it or not).

But you are right that is seems that many haven't.  Its a very important article of observation.

[TPTB_need_war]

No offense, but that question is akin to walking into the Large Hadron Collider complex to ask "Where's the bathroom?"  

Relevant Info:
http://lmgtfy.com/?q=CAP+theorem

I used this and noticed that http://www.infoq.com/articles/cap-twelve-years-later-how-the-rules-have-changed appears on the 2nd page which is read by noone...

That article about databases doesn't have to contend with Consistency of double-spends which infect everything irreversibly downstream. This added wrinkle for cypto currency is what makes CAP more strictly inviolable on us, i.e. we can't in general recover Consistency from multiple partitions.

[minor-transgression]

From my reading of the CAP Theorem, Bitcoin has consistency and availability but cannot be partitioned.
Hence cryptocurrencies, as a group, have inherent partitions, availability, but no consistency as a group.

[TPTB_need_war]

From my reading of the CAP Theorem, Bitcoin has consistency and availability but cannot be partitioned.
Hence crytpocurrencies, as a group, have inherent partitions, availability, but no consistency as a group.

The 1st sentence is correct. The 2nd sentence is incorrect because CAP is inapplicable to a relationship between two block chains. There is no double-spending (Consistency to maintain) between chains.

Actually I explained upthread that crypto currency can theoretically tolerate rare partitioning, but in the case where these are no longer Partitions but treated as orthogonal block chain forks. The Consistency is recovered by allowing everyone to double spend on both forks. However willy-nilly forking would be incredibly chaotic to manage. I suppose it is possible to design clients that could interopt with an unbounded number of forks. There are many issues, such as spending on multiple chains and securing multiple PoW chains.

[minor-transgression]

If you think of alternative currencies as providers of information on the state of fiat currencies,
then exchanges provide a relationship (information exchange) between partitions. You are
correct in that there is no "formal" transmission mechanism, the question is whether the
CAP theorem renders such a link impossible.

Edit: The question perhaps should be stated as how important could such a mechanism be?

[TPTB_need_war]

If you think of alternative currencies as providers of information on the state of fiat currencies,
then exchanges provide a relationship (information exchange) between partitions. You are
correct in that there is no "formal" transmission mechanism, the question is whether the
CAP theorem renders such a link impossible.

Edit: The question perhaps should be stated as how important could such a mechanism be?

Blockstream's Side Chains indeed would cause CAP to apply interchain. This is why smooth and others (including eventually myself) thought Side Chains are going to be intractable.

Exchanges don't link Consistency interchain because the risk of double-spend falls on the counter parties each independently w.r.t. each chain.

[Fuserleer]

No offense, but that question is akin to walking into the Large Hadron Collider complex to ask "Where's the bathroom?"  

Relevant Info:
http://lmgtfy.com/?q=CAP+theorem

I used this and noticed that http://www.infoq.com/articles/cap-twelve-years-later-how-the-rules-have-changed appears on the 2nd page which is read by noone...

That article about databases doesn't have to contend with double-spends which infect everything irreversibly downstream. This added wrinkle for cypto currency is what makes CAP more strictly inviolable on us, i.e. we can't in general recover Consistency from multiple partitions.

Really?  You want to stick with that train of thought?

Aside from the fact you seem to be conflating the issue of double-spending to a single problem, when it is in fact two issues, a double-spend is akin to many issues that a database has to deal with and is VERY relevant.

Assume we have 3 machines with a copy of a database, and this database contains the state of some object.   The state of some object is A in all 3.  Then, one of these machines sets the state of the object to B,  and another sets the state to Z.  The remaining machine A now has to decide which of the 3 states are correct, A, B or Z.  The database is in a partitioned state, which is exactly the same as what an internal double-spend results in, a conflict of state.

In crypto, A is the existing state, B is a spend to one place and Z is a spend to another.  The machines with state A then have to decide whether to ignore the new conflicting 2 spends entirely and stick with A, or pick between one of the spends B or Z.

In the case of a database, the choice of selecting A, B or Z effects everything irreversibly downstream also, just like it does here, they are the same thing.

[TPTB_need_war]

No offense, but that question is akin to walking into the Large Hadron Collider complex to ask "Where's the bathroom?"  

Relevant Info:
http://lmgtfy.com/?q=CAP+theorem

I used this and noticed that http://www.infoq.com/articles/cap-twelve-years-later-how-the-rules-have-changed appears on the 2nd page which is read by noone...

That article about databases doesn't have to contend with double-spends which infect everything irreversibly downstream. This added wrinkle for cypto currency is what makes CAP more strictly inviolable on us, i.e. we can't in general recover Consistency from multiple partitions.

Really?  You want to stick with that train of thought?

Aside from the fact you seem to be conflating the issue of double-spending to a single problem, when it is in fact two issues, a double-spend is akin to many issues that a database has to deal with and is VERY relevant.

Assume we have 3 machines with a copy of a database, and this database contains the state of some object.   The state of some object is A in all 3.  Then, one of these machines sets the state of an object to B,  and another sets the state to Z.  The remaining machine A now has to decide which of the 3 states are correct, A, B or Z.  The database is in a partitioned state, which is exactly the same as what an internal double-spend results in, a conflict of state.

In crypto, A is the existing state, B is a spend to one place and Z is a spend to another.  The machines with state A then have to decide whether to ignore the new conflicting 2 spends entirely and stick with A, or pick between one of the spends B or Z.

In the case of a database, the choice of selecting A, B or Z effects everything irreversibly downstream also, just like it does here, they are the same thing.

In the case of write-only database, the inconsistency can always be merged. I was stating that in general due to double-spends, paritions can't be merged. Indeed in a read/write database where clients of the database base their data writes on the state of the database, then inconsistency infects irreversibly same as for double-spends.

My point is that cases of relaxed CAP implications that apply to one type of network do not apply to our case which is strictly inviolable.

For example that article says:

Second, the choice between C and A can occur many times within the same system at very fine granularity; not only can subsystems make different choices, but the choice can change according to the operation or even the specific data or user involved. Finally, all three properties are more continuous than binary. Availability is obviously continuous from 0 to 100 percent, but there are also many levels of consistency, and even partitions have nuances, including disagreement within the system about whether a partition exists.

Which is inapplicable to the inviolable point the double-spends in multiple partitions can't be merged. This ends up infecting every possible design you can think of to temporarily enable partitions, unless you accept reversibility of transactions (tolerance for inconsistency) or give up Access during that temporal partitioning. CAP is strict on our situation.

[TPTB_need_war]

Fuserleer,

This is a design afair I hadn't contemplated until I wrote this. Perhaps this is what you mean by allow Partitions and then letting any user spend on any Partition? But this is essentially a partitioning of access to coins, because there is no cross-spending between forks below...

Actually I explained upthread that crypto currency can theoretically tolerate rare partitioning, but in the case where these are no longer Partitions but treated as orthogonal block chain forks. The Consistency is recovered by allowing everyone to double spend on both forks. However willy-nilly forking would be incredibly chaotic to manage. I suppose it is possible to design clients that could interopt with an unbounded number of forks. There are many issues, such as spending on multiple chains and securing multiple PoW chains.

[Come-from-Beyond]

That article about databases doesn't have to contend with Consistency of double-spends which infect everything irreversibly downstream. This added wrinkle for cypto currency is what makes CAP more strictly inviolable on us, i.e. we can't in general recover Consistency from multiple partitions.

Can't argue about this because we failed to come to a common denominator in definitions, particularly of Consistency.

[TPTB_need_war]

I agree with your statement "decentralized crypto currency (including Bitcoin) is a delusion".

This is a quote by Dan Larimer that has stuck with me for a while and was part of the reason I got into supporting alternatives to PoW:

All systems tend toward centralization whether planned or unplanned. DPOS gives the end user as much control over that centralization as possible, where as in the other systems the small player has less control.

This is part of the main ideology behind dPoS... all consensus algorithms eventually tend towards centralization, so it provides for a method for shareholders to decide who produces blocks (ideally whoever best benefits stakeholders.) A few articles you might find interesting:

http://bytemaster.github.io/article/2015/01/12/Decentralization-Scalability-and-Fault-Tolerance-of-Bitcoin/

Daniel constructs a strawman that claims we need Proof-of-Stake due to the claim that fault tolerance and throughput scaling is just sufficient hardware redundancy and hardware optimization, so decentralized control should be delegated so that delegates with more resources can provide better redundancy and optimization because any system will centralize as necessary to obtain that redundancy and optimization.

Conflations and logical errors in his reasoning.

Even if the claim is true, it is an assertion of his omniscience to conclude that PoW can't be reorganized in a design that also enables redundancy and optimization. I have a specific design variant in mind to accomplish that with PoW. Thus he constructed a strawman to fool readers into thinking DPOS is superior to PoW.

Also I think fault tolerance and throughput scaling is more than just  hardware redundancy and hardware optimization. For example, if the politics of DPOS causes de-optimization in some game theory such as the example I mentioned in this thread:

• even 0.1% stake can attack the coin because block solutions are exclusive to some stake holder so the stake holder can delay transactions[1]

[1] Another scenario is DDoS attack other stake holders when their turn to mine a block, then jack up your transaction fees sky high when its your turn to mine a block.

His egregious myopia is that fault tolerance and throughput scaling depend on decentralized control!

At the post I quoted above, I summarized why PoS is less secure for sustaining decentralized control. If we just punt and say centralized control is coming any way, so let's accelerate it by choosing to replace Satoshi's attempt at decentralized consensus with the same political mess we are trying to fix by inventing Bitcoin in the first place, then we have entirely defeated our reasoning.

My aim is to try to improve on Satoshi's design to add more power to the ability to sustain decentralization. Or give up and quit.

Decentralization is the normal mode of society. It is when the normal oscillating balance between decentralization and top-down control entirely fails in one extreme direction that society enters a Dark Age for 600 years. Without a balance of decentralization and top-down control, nothing functions and everything collapses.

So I am trying to figure out how we construct a crypto coin so that there remains a tension or competition between decentralization and centralization ongoing. We don't want entirely decentralization and no centralization as that is just as bad as entirely centralization, e.g. see the reply I made to ArticMine upthread and how 100% decentralized control over what goes in the block chain means a choice between unbounded spam or oligarchy control. Thus the problem is the lack of balance and Bitcoin flip flops either to too much decentralization forcing too much centralization (a Tragedy of the Commons). Credit CoinCube for making me aware of the applicable math on that point. Message him if you need the link to it (I don't have time to go find it).

https://web.archive.org/web/20150428162046/http://bytemaster.bitshares.org/bitshares/2015/01/04/Delegated-Proof-of-Stake-vs-Proof-of-Work/

I have agreed with the problem he outlines, and have proposed making mining unprofitable by submitting PoW share with each transaction as the way to get very high security at very low percentage of debasement (necessary to replace lost coins).

However he has constructed that strawman to lead into a non-sequitur concluding that PoS (even DPOS) can cause a great rise in price. An inferior consensus algorithm is not going to see wide adoption.

This is a similar issue along the same lines involving cryptocurrency trading, the centralization of liquidity/volume, where trading eventually centralizes around the exchange(s) with the most volume and liquidity: http://bytemaster.github.io/article/2015/01/05/The-Future-of-Crypto-Currency-Exchanges/

He assumes that all users in the world will trust BitShares block chain more than any other competing block chain offering similar features.

The point he makes about preferring lower spreads also applies to preferring not to exchange in a BitAsset that has to be converted again to the real asset desire, e.g. where the BitAsset is the proxy for another altcoin you are exchanging for.

The notion that we could use a proxy for the dollar without ever needing to cash out to actual dollars, also requires that everyone will use crypto currency, thus we could just use the crypto currency instead of the BitAsset proxy (if that crypto currency was ubiquitous which is inherent in his assumption).

I think Bytemaster has some of the most sound ideological blog posts on consensus algorithms, and is one of the reasons why I originally got into Bitshares.. sorry to plug Bitshares but I feel the things I brought up are applicable to the conversation.

I hope I have explained why when I used to debate Daniel in these forums back in 2013, I realized he couldn't effectively lead me. He is smart but comes up with the great strawmen and leaps in logic to support his flavor of politics and socialism. Is he still pushing that inane concept of paying an interest dividend to stake HODLers? We are supposed to be designing an End-to-End Principled crypto currency not a country club membership or Amway multi-level marketing scheme.

You, and the other posted in this thread, seem to be focused on the shortcomings of every facet of Bitshares without considering (or mentioning) the positives that come along with the different technologies that Bitshares is composed of.

BitShares is so diverse in what it purports to do, I haven't had time to compile a detailed analysis. I do know they liedhyped/mislead about the 100,000 TX/s rate of the 2.0 release, and I had to do some research that it really only is about 10 - 100 TX/s realistically as of now.

I will need to spend more time in the future doing a comprehensive analysis of all major coin technology. But for now that would slow me down too much on more urgent work.

I must admit I still have a philosophical bad after taste lingering from my debates/discussions with Daniel Latimer from 2013 in this forum, because he was proposing features which seemed to me to be the antithesis of decentralization, e.g. some thing about paying interest rate to stake holders in a namecoin variant which I explained to him was socialistic and opposite of our goals for End-to-End principled protocols. He dismissed me, as most good socialists do to the ideals of Libertarianism (the true anarchistic kind). So when I read him writing that decentralization is impossible therefor we should make a coin a corporation, I want to puke. I try to believe people can change, but I know rarely they do. I almost got desperate enough recently to consider testing whether I could implement some anonymity for BitShares and get paid for it. But then it only took 3 minutes at their forum to change my mind. Search my old username there "AnonyMint".

I am not accusing scam. Rather I think difference in core philosophy.

[Peachy]

As for all these details, bear in mind that when we are writing posts rapid fire like this, we may miss key details. I will go offline soon and think about all this in detail again (rehash in my mind).

That was 12 hours ago and you've posted every hour since then. 

Seriously.  Consider some downtime and get some much needed rest.