Topic: Ethereum flawed? (Read 2935 times)

I was assuming that the attacker need only pay for completed machines (that reach a halted state).

If everyone pays for cycles regardless of whether or not the machines halt, could a malicious miner DOS the network by operating machines up until their final state minus one cycle?

(I'm asking dumb questions that are all answered in the referenced paper, just like ppl do when they don't read the original Satoshi paper, aren't I? Be honest....)

The cycles are not free. The success of such an attack is limited by the amount of money the attacker has.

I don't know anything about Ethereum, so this is probably a stupid question, but what prevents the contract writer from offering a huge per-cycle fee and then providing an algorithm which (non-trivially) doesn't halt?

The contract states the amount the sender will pay the miner per cycle and a limit on the number of cycles.

Interesting. I'm on the fence.

How does Ethereum prevent DoS attacks by rogue contractors (contracts taking advantage of Turing-completeness) ?

Government meddling would be the least of it' s problems. Looking at the panic around the malleability bug, which is not even a bug (afaik the bug means that 05, 5 and 5.0 etcetera are all equally valid) the chaos and panic around ethereum and unforseen consequences will probably make it unsustainable. The idea of using ether as a currency AND a fuel for your entire system does not sound like a good idea.

Inmagine a flash crash in ether because of a bug. Not only will the price drop, but also it's value as fuel (incentive for miners) will drop as well, unleashing a chain reaction. Contracts grinding to halt because they need more ether to run, causing more failures, causing more panic and pricedrops, spiralling down.

I'm a bit irritated about the whole Turing-completeness is better argument for a base transaction protocol. Can anybody explain why anything ethereum hopes to be able to do is not possible on top of bitcoin? Or to continue the hype analogy, why would you want to have javascript inside tcp/ip instead of making tcp/ip good for it's task and running javascript on top? I'd much rather they seperate the two of them and let each camp concentrate on their own separate issues, instead of mixing them.

Please note I'm not saying Turing-complete applicability of bitcoin is useless, but I am asking why anyone would want them merged.

Why about Ethereum makes it more susceptible to government meddling than bitcoin? Couldn't political clout also be used to force bitcoin exchanges to do business only with approved people?

That's the point.
Ethereum is a centralized and controlled by contracts and people are not aware about it.

Interesting.

Lol, yeah I certainly used a lot of words there.

I'm spending so much time trying to explain bitcoin and others to people unfamiliar and hostile to the idea, that I'm having trouble switching back. I should find more people "who get it" in my nick of the woods.

Yes Ethereum is centralisation with all the related problems, the obvious ones :

 - over-bloated blockchain : one blockchain supporting infinitely growing applications ? obviously impossible

 - huge incentive for hackers to break the system : one breach and all applications are doomed

quote author=JoeyD link=topic=433986.msg5259059#msg5259059 date=1392905318]
I don't get the impression of Vitalik being greedy or trying to scam people, but people should indeed use their own judgment. He is of course only human and might be susceptible to human weaknesses after all. He is also only one person and there might be wolves in sheep clothing close to him, trying to bend his talents to their own objectives. Something I've witnessed happening up close more than once to several very bright people with very admirable projects.

I've got a few reservations about the project myself, which keep me from sharing the same enthusiasm and bitcoin2.0 scanting. I might be a bit overly critical here, but I think the current hype will more than compensate for my points.

- The project does not have a real goal/purpose itself imo. Taking a very critical view it looks a bit like the battery-manufacturer analogy, who gave away free radios, but those radios only took one type of battery (guess who made them). In this case it would be proposing to manufacture power-tools that will only work on ether.

- Expanding on the previous point, why would you want to run "everything" including your own monetary/fuel system on the same blockchain? Apart from that being enormously more inefficient no matter what tricks you use. Running every conceivable transaction network including other complete coin-ecosystems on the same blockchain, poses a lot more bandwidth and scalability issues than bitcoin is facing. Isn't this going backwards more towards centralization? Also having everything be interdependent and share the same point of failure plus the added security risks of Turing-complete scripting, doesn't immediately fill me with confidence. I don't agree with the argument that Satoshi was too scared to attempt a Turing-complete scripting system and therefore created bitcoin as a flawed system. From my perspective limiting the scripting-options in bitcoins to things it should be able to do, might not only be the only way to secure the system, while still giving you all the options you need to do everything imaginable.

- Many essential features have solutions that might be solved in the future, by offering bounties for people coming up with the solution. To me this does not sound the same as investing in a working concept that can be expanded upon.

- Last but not least (for me) what bothers me is the notion that bitcoin (and the others) exclude the option of Turing-completeness or enabling javascript-like options. Vitalik wrote an article in which he mentions the possibility of contracts being able to use encrypted private keys for bitcoin, without the private keys getting exposed. If that really is possible, then doesn't that make Ethereum a solution to a problem that doesn't exist? That would mean that any program,contract,DAO, purpose built blockchained network or you name it could be created completely independent of the base currency fuel (or fuels) using any Turing-complete language anyone would want, as long as it can manage and react to it's own wallet.

Completely separate from the above points, I'm not a very technical person and I lack the skills to be able to read the source code and then make decisions based on that. So instead I have to rely on my observations of the persons involved and trying to figure out their motivations by their publications and interviews. There where a number of remarks that rubbed me the wrong way and killed my enthusiasm, although most of these came from people other than Vitalik.
- All the references at what Satoshi would have wanted and would have done and that's why Ethereum is better than bitcoin and everything else. Also suggesting that Ethereum is an improvement compared to all the other "competitors" while currently the project is not even guaranteed in working order let alone comparable or better.
- References to Ethereum being the new google or android and investors being shareholders similar to Microsoft, don't inspire me with confidence at all. I'm not really interested in helping create new walled gardens, I was under the impression that were trying to get away from the androids and microsofts.
- Mentioning the many different approaches and experimentations in the development of crypto-currencies as being signs of a disease, was one of the most shocking to me and suggests a complete lack of understanding of how bottom up Bazaar-style decentralized development works. Without which even bitcoin would have been impossible in my opinion. It sounds more like the top down central corporate view afraid of losing control and their bottom line.
- High focus on ROI instead of the projects own merits. To me this sounds like a lack of confidence in your own project. The project should garner peoples attention and investments in time, money and effort without having to be promised short term riches. If you have to pay people to see merit in the project, than there is something not quite right.

Just to be sure these are my personal opinions and in no way fact or expert opinion. Nevertheless they represent some of the concerns keeping me personally from investing in the project.
[/quote]

Tell us more.

I don't get the impression of Vitalik being greedy or trying to scam people, but people should indeed use their own judgment. He is of course only human and might be susceptible to human weaknesses after all. He is also only one person and there might be wolves in sheep clothing close to him, trying to bend his talents to their own objectives. Something I've witnessed happening up close more than once to several very bright people with very admirable projects.

I've got a few reservations about the project myself, which keep me from sharing the same enthusiasm and bitcoin2.0 scanting. I might be a bit overly critical here, but I think the current hype will more than compensate for my points.

- The project does not have a real goal/purpose itself imo. Taking a very critical view it looks a bit like the battery-manufacturer analogy, who gave away free radios, but those radios only took one type of battery (guess who made them). In this case it would be proposing to manufacture power-tools that will only work on ether.

- Expanding on the previous point, why would you want to run "everything" including your own monetary/fuel system on the same blockchain? Apart from that being enormously more inefficient no matter what tricks you use. Running every conceivable transaction network including other complete coin-ecosystems on the same blockchain, poses a lot more bandwidth and scalability issues than bitcoin is facing. Isn't this going backwards more towards centralization? Also having everything be interdependent and share the same point of failure plus the added security risks of Turing-complete scripting, doesn't immediately fill me with confidence. I don't agree with the argument that Satoshi was too scared to attempt a Turing-complete scripting system and therefore created bitcoin as a flawed system. From my perspective limiting the scripting-options in bitcoins to things it should be able to do, might not only be the only way to secure the system, while still giving you all the options you need to do everything imaginable.

- Many essential features have solutions that might be solved in the future, by offering bounties for people coming up with the solution. To me this does not sound the same as investing in a working concept that can be expanded upon.

- Last but not least (for me) what bothers me is the notion that bitcoin (and the others) exclude the option of Turing-completeness or enabling javascript-like options. Vitalik wrote an article in which he mentions the possibility of contracts being able to use encrypted private keys for bitcoin, without the private keys getting exposed. If that really is possible, then doesn't that make Ethereum a solution to a problem that doesn't exist? That would mean that any program,contract,DAO, purpose built blockchained network or you name it could be created completely independent of the base currency fuel (or fuels) using any Turing-complete language anyone would want, as long as it can manage and react to it's own wallet.

Completely separate from the above points, I'm not a very technical person and I lack the skills to be able to read the source code and then make decisions based on that. So instead I have to rely on my observations of the persons involved and trying to figure out their motivations by their publications and interviews. There where a number of remarks that rubbed me the wrong way and killed my enthusiasm, although most of these came from people other than Vitalik.
- All the references at what Satoshi would have wanted and would have done and that's why Ethereum is better than bitcoin and everything else. Also suggesting that Ethereum is an improvement compared to all the other "competitors" while currently the project is not even guaranteed in working order let alone comparable or better.
- References to Ethereum being the new google or android and investors being shareholders similar to Microsoft, don't inspire me with confidence at all. I'm not really interested in helping create new walled gardens, I was under the impression that we were trying to get away from the androids and microsofts.
- Mentioning the many different approaches and experimentations in the development of crypto-currencies as being signs of a disease, was one of the most shocking to me and suggests a complete lack of understanding of how bottom up Bazaar-style decentralized development works. Without which even bitcoin would have been impossible in my opinion. It sounds more like the top down central corporate view afraid of losing control and their bottom line.
- High focus on ROI instead of the projects own merits. To me this sounds like a lack of confidence in your own project. The project should garner peoples attention and investments in time, money and effort without having to be promised short term riches. If you have to pay people to see merit in the project, than there is something not quite right.

Just to be sure these are my personal opinions and in no way fact or expert opinion. Nevertheless they represent some of the concerns keeping me personally from investing in the project.

Thanks for the list.
And to think that I was even considering "participating" at one point...
Hopefully, people will stay away.

Ethereum is flawed because :

- the creator is too greedy (36 millions Dollars IPO + 50% pre-mine + people will have their coin confiscated for 1 year)

- the creator is the first in the world of cryptocurrency to be a public figure and want fame

- people from the banking system are behind it

- the devs are not going to develop any application, so they are not going to add any value

- the development toolkit is a simplification of C++ which mean that : no application will ever be optimised, no application will ever be as effective a an application directly written in C++

- the bloated blockchain problem is going to be infinitely worse in Ethereum than Bitcoin

- the Proof-of-Work system (makes it a 1st gen currency not a 2nd) is not viable in long term because of the growing energy and hardware requirement

- the Ethereum “Dagger” PoW function is flawed

- Ethereum push people to do copycoin and scam IPO, soon there will be MILLIONS of this shit everyday

Yes, that is exactly what I'm worried about.
If they could design a system of interlocking scripted contracts, they could create an "approved" system of wallets that they control.

Then they could use existing laws, or have new ones enacted that force exchanges (or anybody else) to trade only within the group of "trusted" accounts.

Like I said, I'm not too much of a programmer or software engineer. But I have dabbled with relational databases that perform scripts and calculations (Filemaker). It is almost scary what you can construct with a few scripts and calculations. You can really make data jump through hoops.

Don't get me wrong, I like the potential here. It's just that something like this could really be misused  and I don't know how it could be brought back in check.

Edit:
As I re-read this, I feel that I may come off as too strident about this.
I think that I should stress that I am in no way an expert, and honestly don't know if what I described is possible. It just seems to me that it might be.
I look forward to hearing from others who know more about it.

Here's a link to the whitepaper
http://ethereum.org/ethereum.html

And to their website
http://ethereum.org/

The blog portion of the site also has some interesting articles about the "turing complete" aspect of their built in script language.

Several ways at coin distribution. A btc++ if you will.

So you afraid that ethereum will become a decentralized platform for centralized operations?