The Ethereum Paradox

[r0ach]

Once again, we come full circle

Such is the case with everything in crypto.  The last resort is always shifting to some type of reputation based system with a completely unquantifiable security model - que Fuserleer...

I talked to Realsolid and ironically this Microcash thing he was working on was similar to your idea with centralized verification, but in current state, I analzyed it as being a federated chain.  If he was to try and convert it entirely to your system, it seemed to me like it would easily suffer from permanent preemption or constant service outages from government powers attacks.  I'm not really sure how you get past that issue.  Profitable PoW just seems harder to attack by irrational actors than unprofitable PoW on an unpartitioned system.

TLDR:  Bitcoin has problems but actually works somewhat.  Eth is doomed and not much progress has been made in any other areas.

TLDR2: I believe Fuserleer does use the Oxford comma, so that's one positive aspect of Emunie.

[1715439199]

1715439199

[1715439199]

1715439199

[1715439199]

1715439199

[smooth]

TL;DR: Bitcoin, like house dealers and cool bands, was great when it was obscure. Success just doesn't become it

I largely agree here. I don't know that any of these systems can truly scale, nor that any of them will ever be useful for large amounts of value.

The solution may be much the same as cool bands, or anything else (including anything "cool" that only works when it is small). Churn.

On this score, even Ethereum has its role, just don't buy in too high.

[r0ach]

I largely agree here. I don't know that any of these systems can truly scale, nor that any of them will ever be useful for large amounts of value.

It's not a big problem to me for Bitcoin.  Like I said before, you can fit all of the entire world's house and car transactions per day into 1MB blocks.  Around 8-10MB is where I see much of the glass ceiling on price coming off.  It's hard to imagine a future where Bitcoin is unable to scale to such small goals.  Even if Lightning Network is a flop and useless, there will be other transaction bundling services as well.

[monsterer]

Profitable PoW just seems harder to attack by irrational actors than unprofitable PoW on an unpartitioned system.

Rational actors in unprofitable PoW have no incentive not to attack the system. At least if it's profitable there is the mining incentive.

edit: some centralisation is necessary; this is why Iota has opted for checkpoints

[r0ach]

Profitable PoW just seems harder to attack by irrational actors than unprofitable PoW on an unpartitioned system.

Rational actors in unprofitable PoW have no incentive not to attack the system. At least if it's profitable there is the mining incentive.

edit: some centralisation is necessary; this is why Iota has opted for checkpoints

Yea, I meant to use an all encompassing term for attackers and used "irrational actors" instead.  I didn't know Iota was using checkpoints now.  How are these checkpoints derived?  Bitcoin uses them to avoid people initially downloading bad chains and black swan complete rollbacks, but if Iota requires them to work at all, is this an admission the system just dosn't work?

[monsterer]

but if Iota requires them to work at all, is this an admission the system just dosn't work?

Their problem is two-fold; at the inception of the currency, the network hashrate landscape will be sparse and patchy since there is no mining competition, only user's transactions - this means any actor with large hashrate relative to the early network hashrate can cause havoc by double spending. Ergo the need for them to use checkpoints early in the life of the currency.

Personally, I think they will be necessary throughout it's life, though, simply because there is no mining competition, therefore no bound on the network hashrate or anything to value the PoW such that transaction acceptability can be easily bounded.

I'm sure @Come-from-Beyond will clarify your question in more detail.

edit: I suggested that they pay some honest miners with their IPO fund instead of using checkpoints, but they didn't like that idea.

[r0ach]

edit: I suggested that they pay some honest miners with their IPO fund instead of using checkpoints, but they didn't like that idea.

Haha, I already knew that problem would exist in their system at 0 adoption, but I always thought Come from Beyond would try to keep the network up on his own shoulders at day one.  I guess he figures if no adoption ever comes, he doesn't want to be responsible to keep it up forever.

[monsterer]

Haha, I already knew that problem would exist in their system at 0 adoption, but I always thought Come from Beyond would try to keep the network up on his own shoulders at day one.  I guess he figures if no adoption ever comes, he doesn't want to be responsible to keep it up forever.

If they were to add fees collectable by miners, the problem might well simply fade with time, just as everyone hopes will happen to bitcoin when the block reward runs out. There still needs to be a competition at some level, though, IMO.

[hv_]

Haha, I already knew that problem would exist in their system at 0 adoption, but I always thought Come from Beyond would try to keep the network up on his own shoulders at day one.  I guess he figures if no adoption ever comes, he doesn't want to be responsible to keep it up forever.

If they were to add fees collectable by miners, the problem might well simply fade with time, just as everyone hopes will happen to bitcoin when the block reward runs out. There still needs to be a competition at some level, though, IMO.

.. when the block reward runs out, the fees come in (in BTC that happens already due to block size issue..)

[Come-from-Beyond]

Ergo the need for them to use checkpoints early in the life of the currency.

It works slightly different:

There is a node that doesn't roll back transaction that it considered finalized. Others who have "iri.coordinator = [that node IP]" in their configuration file will request the coordinator what transactions to approve. The coordinator returns the youngest finalized transactions on such requests.

As we see, if a node operator thinks that the network hashrate is strong enough he can remove that line from the configuration file and enjoy decentralized wild west.

[Come-from-Beyond]

If they were to add fees collectable by miners, the problem might well simply fade with time, just as everyone hopes will happen to bitcoin when the block reward runs out. There still needs to be a competition at some level, though, IMO.

Fee payment can be included as one of the transaction outputs should necessity arise. It's possible to do right now. So if the fee-less system fails to sustain it will fallback to "classical" scheme (or its hybrid).

[YarkoL]

There is a node that doesn't roll back transaction that it considered finalized.

Who runs that node?

[monsterer]

Fee payment can be included as one of the transaction outputs should necessity arise. It's possible to do right now. So if the fee-less system fails to sustain it will fallback to "classical" scheme (or its hybrid).

IMO you should make sure it is collectable by any miner, not a specific miner - this will create the mining competition necessary for equilibrium.

[Come-from-Beyond]

Who runs that node?

Iota team.

[YarkoL]

Who runs that node?

Iota team.

It's hard to see why anyone would ever let go of nanny's hand
and enter the wild west.

[TPTB_need_war]

Click & scroll down for the Casper progress...   don't miss the links to the hangout video!  

http://blog.synereo.com/2016/03/06/Synereo-Update/

At 1:11:45, Greg Meredith claims that the Nash equilibrium can be ignored because all game theories can be considered within the Pi calculus model.

Naive viewers might assume this implies the loss of Nash equilibrium (which we explained upthread) is irrelevant. Rather what he is saying is that any effects from the loss of Nash equilibrium will be observed in the Pi calculus model. He has not shown a proof that no negative effects exist within the Pi calculus model.

Worse yet, what he doesn't appear to realize is that his Pi calculus model is incomplete. It doesn't model the economic externalities that can't possibly be captured by a model of recursive (self-referential) validation, such as shorting the coin and the value of double-spends of sharded gas to the attacker. The Pi calculus is a process model, thus it doesn't capture these economic externalities.

In short, more technobabble bullshit.

[TPTB_need_war]

If he was to try and convert it entirely to your system, it seemed to me like it would easily suffer from permanent preemption or constant service outages from government powers attacks.  I'm not really sure how you get past that issue.  Profitable PoW just seems harder to attack by irrational actors than unprofitable PoW on an unpartitioned system.

If the government wants to take down a coin, they can always do it even for Bitcoin. But that is a nuclear option, because there is no direct profit to 51% attacking a coin.

Whether the mining is profit or not, is irrelevant to the profitability of attacking a coin. The profitability of mining enables economies-of-scale to centralize the coin, which helps enable attacks. What you are erroneously claiming is that by overpaying for mining, then it is more secure than by paying consumerately to the level of transactions for unprofitable mining. But the latter may be 1000X more popular, thus have a higher hashrate (and use much less electricity per transaction as well as being decentralized).

Society needs for digital currency to work (which it can't if it is centralized because centralized shit always fails), so that society doesn't collapse into a Dark Age. I have to believe society will limit government, else we can all kiss our current quality of life goodbye.

Really you guys are clueless about my design and why it will work. And I like it that way, so you can't copy it.

Apparently Satoshi agreed:

>[Lengthy exposition of vulnerability of a systm to use-of-force
>monopolies ellided.]
>
>You will not find a solution to political problems in cryptography.

Yes, but we can win a major battle in the arms race and gain a new territory of
freedom for several years.

Governments are good at cutting off the heads of a centrally controlled
networks like Napster, but pure P2P networks like Gnutella and Tor seem to be
holding their own.

Satoshi

[TPTB_need_war]

TL;DR: Bitcoin, like house dealers and cool bands, was great when it was obscure. Success just doesn't become it

I largely agree here. I don't know that any of these systems can truly scale, nor that any of them will ever be useful for large amounts of value.

The solution may be much the same as cool bands, or anything else (including anything "cool" that only works when it is small). Churn.

Indeed the powers-that-be are going to try find a way to control anything that is popular. Period. That is why you see everyone here has been reduced to doing P&D to steal from each other, while the whales (powers-that-be) walk away our money.

We are fucked. Society is fucked.

I'd rather try to do something about that and maybe fail, then to do nothing and certainly fail.

And I can earn some money on that attempt as well, and so can others.

[TPTB_need_war]

Profitable PoW just seems harder to attack by irrational actors than unprofitable PoW on an unpartitioned system.

Rational actors in unprofitable PoW have no incentive not to attack the system. At least if it's profitable there is the mining incentive.

This is utterly dumb confused crap, especially after I had already explained to you recently the profits from 51% attacking.

You conflate unrelated economic issues and you have so confused yourself.

I told you I am not going to repeat myself again. I have no incentive to give away my design. I already described the design to a large extent and I just smile as you butcher your understanding of it.

[TPTB_need_war]

Personally, I think they will be necessary throughout it's life, though, simply because there is no mining competition, therefore no bound on the network hashrate or anything to value the PoW such that transaction acceptability can be easily bounded.

That is not the generative essence issue that causes Iota to centralize. Rather it is that the choice of the mathematical algorithm to use when payers sign and payees accept, is a Prisoner's dilemma— i.e. they can't be sure that every other payer and payee is using the same algorithm and the algorithm is only provably convergent on consensus when most (a super majority?) participants choose the same algorithm.

There simply isn't enough mathematical analysis of Iota yet to fully characterize the game theory. It needs centralization to prevent edge cases.