This a fundamental flaw with PoS in general. A decentralized currency is based on game theory. In PoW we have miners competing with hash power to win blocks. In PoS we have the game owner making up the rules as they go along. That's not much of a game. A hybrid money game would work, expecially with competing verification systems, but the rules have to clear, up-front, and fair.
No, it's not. Read the wiki page on proof-of-stake. The existing proposals have been designed to not hand the keys to the network over to anyone. Proof-of-stake must be implemented as PoW *AND* PoS. Proof-of-work is meant to add security, not take it away. But that's exactly what happens in PPCoin--PPCoin greatly lessens the security of the network. |
|
|
|
|
PPCoin is not proof-of-stake, at least not as that term has come to be used. It has nothing in common with either existing PoS proposals--on the surface it resembles Cunicula's, but it's totally different in the details. It fails #3. (Given that it's centralized with a checkpointing super-node, I'd say it fails #1 and #4 too.)
IMHO, keep the bounty going, or at least wait. This proof-of-stake "solution" has not been community vetted. If anything, it's going to end up making proof-of-stake look bad when the coin collapses.
|
|
|
|
We're working on a beta, to be released within a week or two. Discussion is happening on the freicoin forums: http://www.freicoin.org/ |
|
|
|
You are surely aware that with growth incoming transactions can start to number into the hundreds and the thousands per second.
Can it really? Current limitations on block sizes limit the number of transactions to no more than a thousand per block, or a few per second. Changing those limitations would result in a hard-fork. Or am I missing something? |
|
|
|
|
Interesting. That is the first time I have ever encountered an openssl without built-in sha256 support...
Anyway, you can open contrib/vagrant/Makefile and do a find-replace, changing 'openssl sha256' into 'sha256sum -b'. That *should* work. I'm not near my development machine or else I would update the pull-request and test it myself.
|
|
|
|
Wonderful. Looks like a ruby bug. But thankfully that shouldn't concern you as there is an alternative: download the binary RPM from here: http://downloads.vagrantup.com/tags/v1.0.3and install: rpm -i vagrant_1.0.3_x86_64.rpm
|
|
|
|
Wicked. Can you post the results of: # gem install -V vagrant
|
|
|
|
Install VirtualBox. Then: Download and install any requirements it complains about: sudo yum install ...
Build: make download
time make all
It'll take a while the first time, after than builds will be considerably faster. You'll find the zipped binaries in the 'output' directory. can I has bounty too? 17SRxATG3LZrD7WWTCr5EfCapprShVEtP |
|
|
|
Great work @jgarzik, this is critically important stuff. Even if nobody seems to notice  |
|
|
|
|
Remember that when that was done transactions were effectively free and instant. Thanks to Satoshi Dice that is no longer the case, and the cost of doing your own ASCII Bernanke will only go up with time.
|
|
|
|
I assumed PoS would be like mining where you need % of coins to fork the chain.
It'd be a hybrid of both. You'd need X% of coins (51% of the weighted signatures) to fork the chain, and 51% hash power to maintain that fork. |
|
|
|
|
That would be a natural extension of proof-of-stake. Signatures in proof-of-stake are weighted by the number of coins they represent, but there's no reason another reputation-based weighting function can't be added on top of that.
That said, you'd have to find a way to deal with when the reputation evaluations don't agree.
|
|
|
|
It needs some kind of incentive to get people to sign the checkpoints.
PoS is better than relying on one developer to lock the chain.
The incentive is: it keeps their (bit|lite)coins from becoming worthless. When you have or control a significant portion of the total supply of coins, that is no small incentive. |
|
|
|
The more stakecoin you have the higher the reward the more you accumulate the more stakecoin you have, etc...
Stakeholders are not rewarded with coins, just the ability to select checkpoints. I recommend reading the wiki page on the subject: https://en.bitcoin.it/wiki/Proof_of_Stake. |
|
|
|
It doesn't really matter where the income is coming from, if it is anything which can be expressed as compound interest it is bad IMO.
What compound interest? |
|
|
|
Some principal questions:
If the amount of the coins I hoard increases my income, why would I ever spend them?
If I need to acquire coins in order to get income why would I bother with the currency in the first place?
While there may or may not be valid reason to do this for security it is absolutely senseless from an economical perspective.
I can't see how that is related to the issues I brought up. I recognize some people want a mechanism which enables them the charging of interest. But I consider interest a harmful concept in a limited base. Worse the concept does amplify the early adopter advantage issue. Those with many coins will get even more, to the point where joining isn't simply worth it anymore for latecomers.
I would even consider the idea itself harmful btw, but I don't wanna derail the thread.
I'm not sure what relation, if any, that has to do with proof-of-stake. Could you elaborate? Proof-of-stake is a mechanism for negotiating checkpoints weighted by bitcoin balances. Checkpoints are way of mitigating attacks on the network and have nothing to do with income. |
|
|
|
@coblee, thanks for fighting the good fight  |
|
|
|
Yes. In a sense this is less disruptive than P2SH, since it doesn't involve a backwards-incompatible change to the interpretation of the block chain structure. It's essentially an overlay network on top of bitcoin for negotiating checkpoints (I'm assuming we're talking about Mini's proposal). There will be a hard-fork as soon as there is disagreement over a block, however. You'd better have near-unanimous mining pool support before you turn it on... EDIT: Your SE question is a misquote of me. PoS does require a hard-fork no matter how you slice it, but that's simply due to PoS working as advertised, not because of a change to the blockchain structure. I simply maintain that it would be a diversion to create an alt coin specifically for this purpose, when it could be developed as a patch against the mainline clients of Bitcoin or Litecoin (or Freicoin), and tried out on their test nets. |
|
|
|
|
Why a new coin? PoS requires no block chain or network protocol changes. Or rather, it is a proper superset of the current protocol rules that includes new rules for choosing checkpoint blocks.
Why not implement PoS directly on top of Bitcoin (or Litecoin) with a modified client?
|
|
|
|
|
Show Posts
