Topic: The Ethereum Paradox - page 55. (Read 99876 times)

As far as I know that is not correct. What he said was:

The foundation currently has ~1.65 million ETH, plus ~$750k in non-ETH assets. 1650000 * 6.1 + 750000 = $10,815,000. Based on our current ~$200k/month burn rate, that will last us ~54 months ~= 4.5 years. That said, we are planning some substantial expansions which will increase our expenses but also get casper and other fun stuff out the door much faster, and we are also starting to get interest for corporate sponsorships coming in, which could secure us a more sustainable funding path in the long term.
https://www.reddit.com/r/ethereum/comments/45bhus/so_the_ethereum_foundation_can_now_fund_itself/czwpr04

They did not sell. And the price of $ 6.1 was correct two days ago but after the price drop it's currently at about $ 4.90. Makes less 2.7 Mio Dollar and I expect it to go down more.

The problem: They are highly dependent on the ETH-price.

That is an interesting tidbit the eth shills and bagholders would not want public.

I wonder how much VB brought in dumping eth? No one will ever know the true figure. Another 18 mil to piss away?

It all about the money. It always is. 

TPTB_need_war, I can't for the life of me figure out if you have completely lost the plot, or if you are some sort of wizard able to see what all these other "geniuses" can't..

I mean, for this to be true, that Ethereum is built in such a way that it will not work, at all. Then this is some elaborate scam from prominent and thus far honorable community members, or some incredible naiv "hive mentality" of pure wishful thinking from seemingly very intelligent people. The amount of hype Ethereum has atm is unreal, just to give an examples, my occupation atm consistent of answering questions from friends and colleagues about this new drug called "Ethereum", questions with perhaps even more wonder and entusiasme that I saw for bitcoin 2012/13.

I wish someone could chime in and just bury this, just so regular Joes like myself could rest assured that Ethereum is not fundamentally flawed and doomed as you are claiming..

I feel you have little credibility though as you seem to present arguments that are not factual as facts and you have your own competing project under development..

And those point you make that I understand fully such as:


Are blatantly wrong, as you know given the fact that VB said himself that the foundation have been selling Eths all the way up to $6 to ensure economical safety for the years ahead.

For those that aren't kids: https://bitcointalksearch.org/topic/seeking-partners-to-implement-automated-transactions-at-as-a-side-chain-1364594

(AT has been running "live" for over one year on two separate blockchains without any serious issue and supports parallelisation of smart contracts)

Let the others "play with their Lego".

(and don't forget that the Myth Busters did show that you can't make too big a ball of Lego and expect it to roll properly)

Partitions weaken the network overall, so there must be an incentive to merge partitions (whether that be by including unrelated partitions into a single block, or by actually merging branches). If you incentivise this merger, the miner's cost of creating a partition must be greater than the reward for merging otherwise the rational behaviour for a miner is to create and then merge partitions over and over. Moreover, this can only work when the path of largest cumulative difficulty is rewarded more than the other branches, otherwise the incentive is still to diverge instead of converge.

I've maintained for weeks now in my comments that the verification/validation will always become centralized for crypto currency (including smart contracts) and the cost of verification is more acute for long-running scripts.

I've also written that I think the problem can be solved by controlling centralized verification with decentralized UNprofitable proof-of-work.

Thus I believe smart contracts are still plausible on public decentralized block chains. Ethereum is pursing the wrong design though. And they are running out of funding.

The problem is that partitions destroy the Nash Equilibrium because validators can't be sure they can trust other validators and yet it is also impossible prevent state transitions from one partition from leaking into another.

Even exploiting the fact (explained our prior posts) that the scripts in a block can be computed in parallel if they only reference block chain state in prior block, or even extending this restriction further into the past history, will not fix the prior sentence in that the parallel computations can't be split up among different validators because once they can cheat each other on validation (another game theory with externalities) then the Nash equilibrium implodes (and PoS adds more Nash equilibrium failure modes).

Also ETH appears to me to be illegal:

There appears to be a conceptual solution for smart contracts which could enable partitions.

That is to lock down input to the block chain. In other words do not allow any data to be input into partition state from external source, i.e. not even allow a new script (which can contain new constants) to be added to run in the locked down partition. So this means no new user accounts in the locked down partition, no new contract instances, etc..

Not sure how useful that is though. Seems basically useless.

Correct. Agreed. See my edit. I realized that also while eating after posting my prior post. I was rushing because food was on the table waiting for me.

Although I am admittedly not familiar with the low-level details of their design the way that AT was designed does allow for parallelisation of AT processing as communication between ATs does not occur "whilst they are processed" (any token amounts or messages sent from one to another don't apply at the point that they are executed but effectively after all the ATs have been processed for that block).

It may be that the Ethereum design doesn't work the same way but for sure you can design "smart contracts" that allow for parallel execution (you simply cannot have the state of one affect any other within the bounds a block).

r0ach, Vitalik did not respond to my challenge.

I genuinely believe based on listening to their presentations which I linked upthread, that they sincerely didn't realize that partitions are unbounded due to externalities. Math nerds tend to see every nail with their math hammer and don't think of other perspectives such as the following bolded by one of the very smart mathematicians (Greg Meredith) working on Casper (he is also the author of Synereo's white paper):


We'll see how they react to this, whether they ignore it, offer a "solution", or capitulate honestly.

On throughput, smart contract block chains can't even exploit parallelism of the CPU! So without partitions you can only use one thread for validation. Basically it isn't scalable for anything. Completely useless.

I quoted Vitalik upthread and linked to that multichain.com page and indicated he had a had solution in mind for parallelization, but I suppose he is thinking along the lines of what I quoted upthread:


Edit: thinking about this more while eating, I think parallelization could be exploited within a block if the scripts can provably not depend on each other w.r.t. to the data in the prior blocks (e.g. employing partitioned data stores or 100% dependently typed scripts) because the prior blocks are static data so no cross-partition indeterminism can be introduced by externalities (all scripts submitted for the block are computed based on inputs from the historic static data which can't be impacted by running any script for the current block). So then the externalities wouldn't apply within the block. This still won't allow partitions to span block boundaries because externalities across block boundaries will apply in that case as I explained upthread.

So thus I guess you could scale this as a centralized service as r0ach wrote.

Since this coin has so many somewhat intelligent people working on it, they had to of known for a long time it was not comercially viable.  This brings up the questions:  Was it a scam from day one?  Is it a bunch of compartmentalized retard savants all working on different pieces that somehow can't put the big picture together and think that once each one finishes their module it's going to magically work?

Without partitioning (which it's obviously not going to have), the load bearing capacity will be so small, it will only serve a subset of one tiny industry maybe.  Anything involving finance is usually high volume, so their hopes of being Jamie Dimon's personal fluffer are not realistic at all.  I haven't determined just how much capacity Ethereum should have without partitioning, but maybe someone else has.  Let's say users could justify the fees (the fees would likely go way beyond what they would think is reasonable for the market), could Ethereum replace the backbone of only Kickstarter with nobody else even using it?

The real Kickstarter would be able to offer a cheaper service due to operating in a centralized fashion, which brings back the question, why or would anyone bother using Ethereum Kickstarter?  You don't use Kickstarter as a store of value, so it's not like Bitcoin where decentralization is some kind of great benefit.  The risk of having your money stolen by the real Kickstarter isn't really high enough to force people into paying more to do it in a different manner, and it's not like Ethereum removes the counterparty risk of a Kickstarter delivering on it's actual product anyway.

As I mentiond upthread, at the end of the day, there will likely be only one specific, tiny sector that can justify the fees for it's use case scenario.  At that point, the coin would be more like a corporation instead of a general purpose platform.  This is assuming the use case scenario actually exists.  Nobody knows if it does.  The likelihood of it being economically unviable for all sectors is high, which would make the coin value zero.  Bitcoin can scale on-chain, it can scale off-chain, it can scale through bundling, there are numerous tricks to scale Bitcoin.  Ethereum has few if any means of scaling, while relying on scalability even more than Bitcoin does.

Beautiful.

suppose we have shards, or islands, or whatever domains with something resembling
a partial order, can we have a possibility of an asynchronous stable protocol that can keep them afloat?

The UI and wallet availability has nothing to do with the "Paradox" we are writing about in this thread.

I suppose you could award subsidy based on the number of partitions you include in a block - this is actually similar to how the ethereum GHOST variant works atm.

But do they have an incentive to? Is the Nash equilibrium lost?

This was one of the issues I had to work through in my design.

As long as the block producers are aware of all partitions, this is valid.

Not necessarily. If the partitions are provably self-contained (e.g. no spending between partitions thus no possible double-spend conflicts between partitions) then all partitions can be added to the same block without having partitions validate each other.

The problem for scripting is it can't be proven to be self-contained in the partition due to externalities (which even convert Bitcoin scripting to Turing complete).

Partitions in PoW are bad because the hashrate of the entire network gets split by the number of partitions, meaning each individual partition is weaker than the whole unpartitioned network would have been. In fact, the strength of the network tends towards 0 as the number of partitions tends towards infinity.

Okay let's replace PoS consensus-by-betting with Satoshi's PoW. So the problem remains every full node needs to validate every script (which can't scale). If we instead split validators into partitions, then they can't all agree on adding the other partitions to a block unless they've all validated all the scripts thus destroying the scaling advantage of having partitions. If we could assume the partitions can't impact each other in terms of incorrect validation impacting the output a script which impacts the input of another script in another partition, then the partitions wouldn't care to validate each other. But I showed that with scripting, it is impossible to know whether scripts in different partitions impact each other. The externalities that the block chain can't know means it can't guarantee the partitions are bounded (i.e. self-contained).

In short, if the partitions can impact each other, then all full nodes must validate/verify all scripts. Thus partitions are useless for scaling.

Whereas with the directed acyclic graphs that are crypto asset transfer, we can keep these self-contained in a partition, thus we can use partitions to do scaling.