Digipl's battle. Read the assault and help provide clarity. Call to arms!

I made small grammatical corrections. Google did the rest. :relaxed:

1 Like

Scary stuff. Looks like people don’t believe the double spend problem is taken care of in SAFE? I thought we already addressed this.

(Looking ahead) please @moderators DON’T delete this.

Wow, actually just finished reading all of it.

It seems like the major concern here is that the main achievements of the network are not properly documented yet, which worries people because it should be easier to write out a clear explanation than to actually code out the product, as the team opts to do.

Am I understanding the raised points correctly?

2 Likes

Can we get a TL;DR on this please?

Yes, I think that’s the main point that is arousing suspicion that the “Byzantine general” problem is not solved is an apparent lack of a clear explanation of the “quorum forming” mechanism used to solve this problem.
The other main point being contended by Vitalik and others is that there is an attack possible by way of vaults downloading a “patch” to misbehave. D Irvine’s position is that misbehaving nodes will be rejected by the Network - I think the retort to this has been that taken further, if “enough” nodes download the patch, then it becomes like a 51% attack?
The other raised issue I can see is the claim that latency will be a huge issue using “Kadmelia”(?) - the accusation is that a dodgy “workaround” or “patch” has been coddled together I think.
I’m probably one of the least technical people on here, but believe I have a reasonable grasp of where we’re up to and how these things are spposed to work (some in theory).
The question of “timing” and “consensus” keep coming up, which leads me to think that there is some misunderstanding with some detractors about this and the “atomic” nature of Safecoin.
As to those with the “killer arguments” to “humble Irvine on his own forum” …he already seems pretty humble and self effacing to me - the opposite of those who appear to want to belittle to aggrandize themselves.
Anyway…[quote=“javi618, post:3, topic:7999”]
Poor Irvine, which awaits you …
[/quote]

We’ll see…love the tone btw… can’t wait to get to know you better: :smile:

4 Likes

There’s no explanation because the way that part works was never denied in the first place. I stated my thoughts on this several times, e.g. this topic:

The risk of “micro-attack” focused on your group only is independent of the size of the network and must be handled properly. It can’t be minimized or avoided by having a large network.

1 Like

Random XOR address allocation makes “micro attacks” very difficult in a large network. So I can’t agree. Other reduction or mitigation ideas are welcome. Most systems come with their caveats. I believe this one is tolerable. Even if it doesn’t immediately or consistently provide the anonymity and safecoin protection you’d prefer. For greater anonymity use Tor (to connect to SAFE). When it comes to safecoin, I’d see it as no more than a narrow use token in the short term. You win some and you lose some. Relying on them for financial security is no beueno before a perfect implementation arrives. Until then let’s enjoy a network far superior to the current incumbents. :wink:

4 Likes

I followed your links, but I’m not really convinced there is any real problem - attacks seem to be pretty far fetched to me and appear to be most likely cost prohibitive. Your comments were helpful as always though:

gohan00760:
“How long can you maintain this attack? even less than the first idea cos its costing you even more than before.”

Janitor
“I’ve considered those problems and I have some ideas how they could be mitigated or worked around, but I’m not going to detail them.”

This actually leads on to the other criticism that “Game Theory” hasn’t been used to “incentivise” good behaviour. I believe that it actually has really (though not recognised as such) by dis-incentivising it financially. This attack vector also seems to be premised on nodes “knowing” their close groups which they don’t - so it would cost loads and become exponentially less effective/feasible as the Network grows…wouldn’t it?
I’m getting that the double spend is not possible due to the atomic nature of Safecoin, (which also seems to negate a lot of the “incentive” “consensus” “game theory” claims too).
I’m getting that the “Sybil” type attack is highly unlikely due to prohibitive costs and diminishing effectiveness over time.
I’m also getting that I don’t know/understand the technical details of how the “quorum” is established and maintained (and prob never will… ):smile:

1 Like

I think these people will be really surprised when the MVP goes live in a little while. It’ll be like “You mean it actually works as described? He pulled it off?”

First they ignore you.
Then they laugh at you.
Then they fight you.
Then you win.

7 Likes

Seriously, I can’t ****** wait! :laughing: What kills me is that these people can attack this project so strongly without first thoroughly absorbing the details. You know they’ll find some way of justifying their derisive commentary to themselves. I’m hoping humility is the state they accept or at least consider when the time comes. But hey, few adapt quickly. We march on… :expressionless:

5 Likes

Lol try debating the dangers of vaccines or GMOs with someone from the mainstream sometime. You’ll get the same kind of response. Total brainlock and conditioned denial. Of course they aren’t going to do their research. That would require them to challenge their most deeply held beliefs about the world. Whether it’s health, politics, the environment, cryptocurency, people are people and they operate pretty much the same.

Maidsafe is challenging the very foundation of what people have come to believe about the internet: servers, the blockchain, it’s breaking almost every rule imaginable and changing the whole paradigm. Of course people are going to have cognitive dissonance issues.

9 Likes

My brotha blindsite, please say it again… A paradigm shift on so many levels is ahead. The wait is excruciating!! :confounded::tired_face: !!!

4 Likes

Yes, however consider the idea from the text you pasted:

  • Create 28 patched bots. Make each join the network and broadcast their group info to the other 27.
  • Unleash them and wait.

Sooner or later, they’ll happen to for a group of 28 members of the same SAFE group

2 Likes

Try it and see if you can win the hackathon.

Seriously there’s all this talk about how SAFE has all these security issues and whatever. We’re in bloody testing right now. And even after this we’ll be in BETA testing. So if you think there’s some kind of security hole then hack the bloody network! The Maidsafe team has even stated this explicitly and encouraged people to do it. There’s even talk of putting money up to ENCOURAGE people to try it so they can patch security holes. Being a whitehat hacker pays good money. You crack the system, expose the hole and help the team patch the leak. If people really believe SAFE is going to be falling apart at the seams why don’t they start lining up for the hackathon and try their hand at the winnings?

4 Likes

The wait will be long and very unpredictable. Churn and random XOR assignment make this difficult. Don’t forget that XOR addresses are regularly changed for vaults. “Them” would have to be hosted by the attacker. Zombies will be harder to obtain. Direct hacks and JavaScript exploits being a narrowing window. Without cheap node acquisition, this attack becomes impractical.

2 Likes

I can answer this one, if a node is unavailable then the next closest one takes it place. This decreases the “density” of the address space of the quorum. Before others accept a message as valid from another group, they check this density and compare it with the density of the part of the network that they know. Since address asignment is random this should be in roughly the same range. If it’s not dense enough it’s not considered valid. They also AFAIK send messages to surrounding nodes of the close group in question to check if the addresses are real.

I think vitalik’s second concern is the most valid one. I’d love to see better answers on that front, especially with SafeCoin implementation coming up.

His first concern about incentives is I think not a problem, because a vault will have to invest resources to become and stay a full member of a close group. Getting disconnected due to strange behaviour makes you lose that investment (just like in BTC mining, you lose your mining investment when you create a block that is invalid instead of a valid one).

8 Likes

WTF!!! My biggie boy supa seneca is Baaaaaack!! :astonished: Give em the noise and hit em with the brains my dude!! :sunglasses:

1 Like

I’m sorry, but this whole thread and attitude displayed here is only alienating people from outside the community.

The way I see the translated thread is that there are apparently some folks who’re eager to understand these technical details of some of things that are proposed as a solution. But right now, they can’t verify it because in their opinion
a) There is not enough details in the whitepapers to really verify how it’ll work. A high level overview is good to lay down the requirements that should be met, but what is missing is the actual documentation of how the implementation will look like.
b) There is not enough working code at the moment that could be used to verify it there.

Either of those things are necessary to verify that it actually works and so it’s quite understandable that there still people out there who’re sceptical that it’ll work when under attack and apart from the personal attacks, I don’t see anything wrong with saying: I don’t believe it until I see it.

Either side could be wrong, it could be that there are problems that can’t be overcome with the proposed approach and until that happens it is rather pointless to retort to dismissing claims that it won’t work just because we’re more faithful that it’ll actually work.

It is always important to listen to critics and what their arguments are, countering with facts and being rational about it. Being emotional and dismissing them out of hand just because we want to be right, doesn’t help and only sheds a bad light on this project.

Remember, the things that unite us are far greater than the things that divide us.

18 Likes

I agree, we should ignore the turd throwing and just try to answer the actual issues.

11 Likes

I’m fine with listening to critics. I also acknowledge the validity of their concerns. They went beyond this and began insulting because desired information wasn’t/isn’t available. Not due to clearly defined flaws. Speculation should IMO be less derisive and never so conclusive. They were being ridiculous. It’s okay if they find it reasonable… Narrow minds inherit this.

2 Likes