Quote from: ABISprotocol on January 22, 2016, 10:08:29 AM

Quote from: TPTB_need_war on January 08, 2016, 06:58:11 PM

Let's have a frank discussion about the technical realities of crypto-currency.

Apologies in advance to all those who have worked so hard on trying to advance crypto currency. I am not doing this to spite you. I don't want to waste more time. If we can convince ourselves we have a solution worth working on, then let's do it. Otherwise let's be honest with ourselves.

Edit: for those who want to jump straight to understanding how Iota's DAG works, click here and also here.

Quote from: TPTB_need_war on January 08, 2016, 03:54:14 PM

Upthread in my discussion with monsterer, I mentioned that I thought I could solve the problems Satoshi's Proof-of-Work by crediting all block solutions instead of the just the first one that arrives. When I went quiet on Jan 3, it was because I realized my design was faulty because there would still be an ambiguity around when the block solutions to be credited were propagated. I was trying to rush out a first iteration when while programming, I realized the detail that wasn't clear to me before.

I went off on several days of just thinking all day. I contemplated all the possible designs (including Iota's DAG, Lightning Networks, DPOS, Proof-of-Stake, Masternodes, Raiblocks/Blocklattice, etc), and I can't think of any design that uses a block chain or a DAG (or any other form of determining the longest chain of truth) which doesn't either centralize (factoring in society's ability to regulate the consistent partition) or diverge into inconsistent truths. Due to the CAP theorem it is fundamentally impossible for there to exist any block chain or consistent DAG design that won't centralize (even without regulation once you require scaling). Worse yet, it is impossible to attain any sort of end-to-end principled, decentralized scaling of transaction processing, because consistency is lost without centralization (even Proof-of-Work centralizes economically due to the Power Law distribution of capital).

The CAP theorem is fundamental. There will be no way to solve it. You all can spend the next 1000 years fooling yourself will all sorts of designs, but they will also end up either inconsistent or centralized or unable to scale. PERIOD. PERIOD.

I realized that Bitcoin and everything else so far is destined for failure. We are only mining each other here. We are not producing any fundamental breakthrough on the problem of decentralized electronic money. I do not like to work on things that I feel are misdirected and destined for failure in the end. I don't want to get rich by fooling other people (or fooling myself). All of you including the core Bitcoin developers are fooling themselves. I've been through all the designs. It is fundamental. There won't be any solution in any of the directions being pursued by any of the current and upcoming crypto projects. It is all delusional bullshit.

I felt rather hopeless about this, and so spent a few days thinking about other potential directions for my life, work, etc..

After all that, I decided the only way to get a breakthrough on electronic money is to admit the CAP theorem and decide which of the three, Consistency, Access, Partitioning to forsake in a design. Bitcoin can't tolerate any Partitioning, thus Access is and will be centrally controlled. Iota allows Partitioning and thus forsakes Consistency (watch it blow up).

Specifically Iota forsakes Consistency in a very chaotic way, where there can be multiple Partitions of truths and so no one will know which truth is valid. Or they will have to centralize to force a consistent truth.

[...]

Quote from: TPTB_need_war on January 08, 2016, 06:48:04 PM

Quote from: Fuserleer on January 08, 2016, 06:38:33 PM

Quote from: TPTB_need_war on January 08, 2016, 06:32:19 PM

Quote from: Fuserleer on January 08, 2016, 06:24:39 PM

Quote from: TPTB_need_war on January 08, 2016, 06:16:20 PM

Edit: the idea I proposed as a solution is also flawed.

I will be starting a new unmoderated thread to discuss in detail all the flaws in crypto currency.

So this can be explained well so that everyone can understand what they are investing in.

I'm confused.

One second you're stating you have the holy grails, the next you are starting a thread to point out there aren't any? :|

As you wrote, temporary chargebacks (inconsistent channels a.k.a. partitions) can be allowed and resolved with a proof-of-work scheme. Any other schemes you are contemplating won't work and I will tear them apart once you detail them.

They can not be resolved in a DAG scheme (e.g. Iota) without some centralized control.

Proof-of-work is centralized at 51% control. This encourages the nation-states to organize into cooperation on regulation of the internet in order to regain control over money. Governments and society will not give up this control and will instead decide to cooperate so the system described below can control to move us towards the 666 system which is rapidly taking form. Bitcoin is designed to drive us towards a world governance.

Centralized control is loss of permissionless principle. It means the government takes control (because the Power law distribution of capital always drives a collusion amongst government and big capital in a winner take all paradigm).

There is no POW required to secure a CL (channeled ledger), nor is it a block chain, or a DAG. Neither are there any charge backs in a CL design because the state of truth never gets to a point where you have to revert to them to resolve inconsistencies...which also means the risk of any centralization required to resolve said issues is next to nil, the network can always resolve itself.

If block chains or DAGs or Ripple style ledgers could do what a CL allows, well, I wouldn't of wasted 3 years reinventing the wheel and tearing it apart myself and starting over.

The truth of each channel still has to be a consensus. It doesn't change the fundamental issues of how to prove consensus about double-spends within the partition. Even there are chosen nodes who are signatories for determining the truth of the channel, this then not permissionless because the government can attack those specific nodes.

Detail your design and I will rip it to shreds.

Not intending to be unfriendly, but I am tired of bullshit (especially bullshitting myself because I don't want to waste any programming effort). I have done all these designs in my head.

If there is a design that can improve upon Bitcoin, then I want to work on programming it. If not, then I want to not waste effort. It would be better for me to make some fast transaction addon for Bitcoin than to waste effort on designs that won't improve upon what is.

You state that the CAP theorem is fundamental to all of what you are describing. The CAP theorem is essentially this:

a statement that it is impossible for a distributed computer system to simultaneously provide all three of the following guarantees:

   Consistency (all nodes see the same data at the same time)
   Availability (a guarantee that every request receives a response about whether it succeeded or failed)
   Partition tolerance (the system continues to operate despite arbitrary partitioning due to network failures)

Let's start with an assumption that I agree with the CAP theorem, and further assume that this notion is actually correct.

In that case, let's say all nodes don't see the same data at the same time. Seems reasonable, after all, how could they?
Availability - for the sake of argument, let's say every request receives a response about whether it succeeded or failed, or if it doesn't the first time, it is programmed to repeat the same request until it receives a reply, which it will receive within a reasonable period of time.
Partition tolerance - Let's say for the sake of argument the system continues to operate despite arbitrary partitioning. One example perhaps being a network failure leading to event like the 2013 Bitcoin fork - analyzed here by Arvind Narayanan, who interestingly poses arguments in favor of a certain degree of centralization within bitcoin development: https://freedom-to-tinker.com/blog/randomwalker/analyzing-the-2013-bitcoin-fork-centralized-decision-making-saved-the-day/

So under these conditions as described above, which I suggest are roughly representative of the actual condition of bitcoin (as an example) at most times, then it would seem to me that despite the presence of the CAP theorem, the system continues. Why? Because the system is dynamic, not static. Possibly for systems that are very fragile or inflexible then I think that changes in these C-A-P conditions could cause them problems, of varying degrees of seriousness. But even for highly centralized systems (e.g. corporation-states), which are highly resistant to change, they are curiously persistent. This may be because of people's desire to perpetuate an institutional memory and cultural history of organizational (and national or tribal) ideology. Belonging to a large group - identity which conveys a larger sense of belonging - has been, for good or for ill, branded into the human psyche. War, government failures, economic disasters, mass murder - nothing seems to stop the bulk of people in society from falling into the trap of the state, again and again. But I digress. What about those decentralized systems? Does the issue of the CAP theorem necessarily mean that they can't work?

Not necessarily. If a system is dynamic enough, and is managed well by its community (however that needs to happen with a distributed system in order for some degree of balance to be attained between a "centralized" development structure and a "decentralized" system, as no system is ever 100% "decentralized," then a well-cared for decentralized, distributed system can be continuously propagated (or re-propagated, much like a plant's seeds are used to regrow the fields).

Interesting discussion though!

Coming back for a bit (just a bit!) to add onto this discussion, I wanted to throw some flames onto the fire, for example, things like the bankruptcy or other forms of failure of certain firms (like for example KnC Miner, a recent bankruptcy example) which results in more centralization (and thus more China concerns) - stuff which I brought up before in 2015 on a dev list here:

https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-August/009861.html

As pointed out in my earlier comment in this thread,

"even for highly centralized systems (e.g. corporation-states), which are highly resistant to change, they are curiously persistent. This may be because of people's desire to perpetuate an institutional memory and cultural history of organizational (and national or tribal) ideology. Belonging to a large group - identity which conveys a larger sense of belonging - has been, for good or for ill, branded into the human psyche. War, government failures, economic disasters, mass murder - nothing seems to stop the bulk of people in society from falling into the trap of the state, again and again."

Anyway, as I suggested, regarding Partition tolerance - " Let's say for the sake of argument the system continues to operate despite arbitrary partitioning. One example perhaps being a network failure leading to event like the 2013 Bitcoin fork - analyzed here by Arvind Narayanan, who interestingly poses arguments in favor of a certain degree of centralization within bitcoin development: https://freedom-to-tinker.com/blog/randomwalker/analyzing-the-2013-bitcoin-fork-centralized-decision-making-saved-the-day/ "

It may be that the system thrives despite this condition not because of the presence of bitcoin but rather because of the emergence of new systems (some of which may be found amongst the at least 700+ decentralized and / or distributed virtual currencies commonly known or in use today) which are actually more distributed and / or decentralized than bitcoin is itself or because of exchange behaviors in the global economy which result in resilience of system(s) even where the distributed system does not meet C-A-P (here I will not enter into an argument over whether or not it is completely impossible for a distributed system to meet the C-A-P condition(s) / guarantees because it is possible that one might, so I won't completely rule it out even if it seems unlikely). But in contemplating these new systems (including ones that don't provide all three of the guarantees / conditions under C-A-P) then I have to consider things like Bitsquare and OpenBazaar, for without such systems I think that no system of decentralized virtual currency would be truly useful. Ultimately examining the declining profitability of mining in the context of PoW issues and so forth, for most currencies the issue then becomes not so much how will one continue on generating it but rather, how will it be exchanged and used, and can it be sustained.

Here is another thing I'm not sure has been contemplated in this post. A large number of coin systems have been developed -- see mapofcoins.com -- and not all of these will succeed, but it's also probable that over time more will be initiated that have not yet been seen. Over time these will consume energy, resources, and talent (development, programming, maintenance energies, etc). It is difficult to say how sustainable that is, but I think that overall as people orient themselves to a smaller set of currencies, at least some of which have privacy or strong anonymity, then this will be preferable to a larger number, and more sustainable as well.