Topic: PPCoin is NOT a decentralized cryptocurrency - page 3. (Read 10995 times)

Now that I truly understand where you were going with that I agree :-)

Yes, that is the point. Proof-of-stake allows stakeholders to send a signal that one fork is correct. However, conditional on you having the necessary stake, each signal is costless and you can simultaneously signal for multiple forks (i.e. you can take both sides of the bet without any additional cost). Each signal will earn a reward if the fork turns out to win. Therefore you might as well signal for as many forks as possible.

I've been agitating for costly signals to avoid this problem. If each signal is costly, then you would only send it out for chains that are likely to succeed. The cost would be paid in work.

For some reason, only my proof-of-stake proposal incorporates costly signaling. I don't understand why.  I'm trying to provoke an argument about whether costless signaling is a major problem.
I think it is. Therefore, I think PPC coin should be modified to make signaling costly.

Thanks.

Would that not nullify any perceived advantage of PoS though and then you might as well go back to PoW?  PoS in this case would really only be serving the purpose of transaction processing between PoW which would again contain all the security responsibility no?

How hard would it be to modify the PPC coin client so that it attempts to generate a stake block for not just the main chain, but also for any forks from the main branch?

Adoption of such a modified client would seem to slightly increase stake income.
(i.e. in the event of an orphaned block, you get your stake earlier. This means you accumulate coin age more quickly. This means your next block comes slightly earlier too.)

Complete adoption of this modified client would seem to make all forks equivalent in terms of stake content. Therefore only proof-of-work would determine chain selection, correct?

Also, as more people adopt this client forks become more frequent. Therefore, the rewards from adopting the updated client increase. Right?

I do not want to run anything, I'm merely suggesting a simulation.

For example, you can simulate proof-of-stake mining where one malicious miner tries to mine many blocks in one go.

Setup is described here: https://bitcointalksearch.org/topic/m.1139483

Suppose we have N equal 'stakes'. (N-n) stakes belong to honest miners, n belong to a greedy one who wants to do double-spend.

You can simulate PoS mining in this way: each block should have some random number X associated with it. Then miner rolls a dice to get random number Xi. If |X-Xi| < coin-days miner has then he signs PoS block.

So you go through all miners trying to find one who have solved PoS block. If nobody solves it then next block is PoW block, so you just roll another X.

For a malicious miner things are a bit different, though. He tries all his n shares one by one. If one or more shares solve the block, he immediately tries to build another blocks on top of it using his other stakes. If chain of k blocks can be built malicious miner wins a double spend. If his best chain is less than k blocks he doesn't submit his blocks so coin-days of his stake isn't reset.

For everybody else, solving the block resets coin-days, not solving it in this round increases it by one.

This is a very basic simulation. Ideally you should use parts of PPCoin code to simulate real mining. (Of course, only PoS blocks, you don't need to make valid PoW blocks.)

E.g. CTransaction::CheckProofOfStake https://github.com/ppcoin/ppcoin/blob/master/src/main.cpp

It isn't easy, it might take a couple of days, or maybe a week of work. But it's not a rocket science either...

No the question is whether it is by the minute or the hour in PPC is he getting paid.

LOL 

How is it not a paraphrase?

 

Edit: Detract...change subject...oh back on topic....."it's difficult/complex....that's why"

Honestly Andy, your knob slobbing is almost as bad as smoothie's trolling. Are you getting paid?

How is that a paraphrase?

http://youtu.be/C_c7cmeDAOw

kind of like a taco, within a taco, within a Taco Bell within YOUR MIND!  Its super cool and complex.

It is more of a paraphrase of what Sunny has said recently on why he has no formula and analysis of PPC.

"It's difficult/complex....that's why"

Who are you quoting?

Sounds like lots of people have questions that aren't being answered and are responded to, mostly by Sunny, as "It is the way it is because it is complex"...

lol

I would run the simulation too. However, Sonny needs to provide a detailed explanation of the random process that generates proof-of-stake blocks.

I cannot read the code. I cannot write a simulation if I don't know what I'm simulating.

So basically its another SolidCoin, and isn't even willing to actually claim he is not simply solidCoin back again under another username?

-MarkM-

OK, so let's put it straight. The purpose of proof-of-stake, and of blockchain protocol in general, is to prevent double-spends.

You claim that you've invented a crypto currency which uses proof-of-stake to prevent double-spends. But you don't know whether this protection is actually strong.

Too much cognitive dissonance... I'll give you an analogy:

"Hey guys, I've invented a new kind of tank armor, it's very lightweight, thus it's much more efficient and will likely replace other kinds of armor in future. You ask what damage it can withstand? Well, I don't know really, I neither can do modeling nor I can test it in practice. But how about this: you'll make some tanks with it and use it in battlefield, so we'll see what explosions and shells it can deal with. By the way, my recipe for coating is all open, so any material expert can analyze it."


It's not a "final say", it's a basis, a very first block for a proof-of-stake design. Proof-of-stake's purpose is to prevent double-spends, so first and foremost you need to check how good is your scheme against double-spends. Once you know it's good enough you can consider other "attacking vectors".

And I don't think it's hard, you probably can come up with a good estimate in a couple of days if you have skills. If you don't want a theoretic solution, you can do a simulation. It isn't hard to implement, again just a couple of days worth of work.

(Control of the blockchain is a different thing, by the way. It's obvious that one who controls blockchain can do double-spends, but small scale reorgs -- and thus double-spends -- are possible with much less effort, and this is what matters for security of payments first and foremost.)


Logic escapes you here. Again, perhaps analogy can help: suppose you build houses and I build bridges. Suppose I've noticed house you're building when I was walking in my spare time, and as I know a bit about structural integrity I've noticed that house you're building might collapse at any time. So I ask you to check your calculations (I might be missing something) and I also tell other people that they shouldn't go inside that house since it can collapse and kill them.

So now you say that you don't have calculations, but that's OK because you are the only one who builds houses. And apparently I shouldn't comment because I build bridges, not houses.

Does it make sense?

I do not work on proof-of-concept schemes, but I still can comment on them. It really isn't convenient because I don't get anything out of it, I do it in my spare time to entertain myself.


That's simply not what I do, I'm not an expert on that matter. You see, your problem is that real experts do not even want to touch your stuff because they don't want to dig into your code to guess the details. They just have better things to do. So you only get attention from guys like me which do it just for entertainment.

If you publish a better paper MAYBE you'll get more attention from real experts. But I cannot guarantee you anything, of course.

BTW I've already mentioned a couple of times what you can do to improve security.


It isn't a crusade, I'm just warning people about situation about ppcoin because a lot of people do not get it, they cannot do analysis on their own and they trust you.

But I didn't post anything about ppcoin at least for a month... Smoothie resurrected old topic which was relevant only before you released v0.2 (?). I just posted an update here, and that's all.


ppcoin is vulnerable by design, you can't fix it unless you implement significantly different algorithm. Do you want me to do your work for you or something?


Yeah, that's ok. I'm already quite bored with this ppcoin stuff.

@Sunny and @AndyRossy,

How about that EGO claim of Sunny's eh? Someone has a huge ego claiming their development time into PPC which they have no formula, have no analysis, and have an incomplete white paper is worth $100k-$200k and that bitcoin development is ONLY worth about $500k, don't you think?

So funny that many of the claims that are made are so true of yourselves in your attempts to continually detract attention from PPC and the important questions that people have asked and will continue to ask.

Oh nice try to change the subject on an important point. The intelligence insults don't work very well on me. Keep trying though, it is entertaining.

Like I asked in the other thread, does Sunny Drag Queen pay you in PPC by the hour or the minute to keep detracting from the SHADINESS of all things that are PPC?

lol