Bitcoin Forum

A well-written article for what i wanted to say.


https://ossbox.com/bitcoin-pool-ddos.htm

I agree, lets not have mining pools, lets just have p2p mining integrated into the spec. Nobody ever gets 50BTC, everyone always gets a share of the mined block according to their contribution.

*waves hand mystically*

make it so.

On a serious note, has anyone actually proposed a mechanism for implementing this? Any threads? Seems kinda difficult.

I think we really need a way to decentralize mining pool. It seems btcguild is DDos at the moment and deepbit has again a too big part of the Network power, anyone who could hack their system could use their user power and put all in a mess. We should take it really seriously, we are quite lucky because nothing happened yet.

As the article says, it is really easy to develop a centralized system for everything but it is so vulnerable (to our level) that it shouldn't be a long-term solution and I think it is already late, but we can do something before anything really harmful happens. We need to think a way to make a P2P mining pool system which shouldn't depends on a such insufficient number of "master" peer.

We can make a centralized-decentralized system like Freenode which have super-nodes (>50 pools and >15 super-nodes/pool) and lots of other peers. I think the problem is that 50 BTC is huge and should be decreased. But I hope we will find a way to avoid cheating among those super-nodes (one has to collect the shares, make the block and distribute their reward). Maybe we should modify the bitcoin system to include the meaning of mining share or something like this.

The author is wrong when he says that Deepbit has all the power. The power belong to thousends of miners that use Deepbit. If Deepbit tried to abuse, it might get somethign temporarely (and would damage Bitcoin image), but the miners would pull out quickly rendering Deepbit useless. Also, Deepbit would be out of the game, so it has a strong incentive to keep being honest.

As for the DOS attacks, pcolbm has now option for backup servers, so when one mining pool server goes down the miners automatically switch to the next one in the list and the network keeps working.


The problem with a p2p mining pool is how do you keep the miners honest. It would be very easy to cheat. I think that just taking some care that no mining pool has 50+% and now that the miners allow for backup servers it will be enough.

I believe this situation will work itself out.  Pools will work for now, but there will be another, larger 'pool' that will just once again count basically for individuals mining.  Then it will be later tricked into more pools within pools within pools, but it will take time for all this to happen.   At this point, everyone will be into CoinCoin, which is of course what follows NickNameCoin, which obviously follows...

It would be pretty awesome if that was possible.

BiTCoin is itself a central point of (I'm gonna say success... ;))

As you might be aware, MI-5 and MI-6 are fictional, which need not necessarily imply that Martial Intelligence is an oxymoron.

The Hacker, Martian, European and United Nations civilisations (among others) are, in mainline Freeciv software, listed as fictional.

Just so you know. ;)

So. In the spirit of, uh, well lets say "hypotheticals"...

Let us imagine someone... we might as well call them Martians for now... had learned of the interesting mathematical cryptological computation network based currency used by an even more advanced civilisation whom we might as well refer to as Hackers.

For a number of reasons, perhaps not least among them some unfortunate effects contact with relatively advanced cultures sometimes has upon cultures, the Martians imagined there might be room in the universe for other currencies using basically the same technology as that used by the Hackers' currency. Accordingly, though by what precise means history might not entirely make clear, the Martians obtained sufficient knowledge of the relevant Hacker technology. Thus were born Martian BotCoins.

You don't have to believe this - not everyone does it seems, but Martian propaganda seems to like to at least pretend that it does - but the Martians are not as brutally aggressive and warlike as various "Mars is the god of war, Martial means relating to war" etc etc etc myths and legends might have led some to suspect or imagine. Thus it is that United Nations Scrip, referred to by some as United Nations Shares came into existence. Yes, the propaganda arm of the Martian Intelligence 5ervices  would have you believe the Martians would like lo the many nations of sentient beings to co-exist peacefully.

Of course the very concept of a United Nations might itself be seen as yet another threat of centralisation. If it is true that Hackers and Martians are fictional, even in the eyes of a United Nations itself purported or thought by some to be fictional, might it be that the United Nations invented the myths or legends of the existence of Hackers and Martians in order to appeal to some notion that the United Nations might not itself be the most advanced and powerful civilisation? (What is in a name? Oh wait, are we supposed to consult the NaMeCoin system about such questions?)

General Mining Corp (GMC) conducts mining operations of various kinds and purports thereby to provide some notion of value to its own offering, General Mining Corp scrip (or shares, as some sometimes call them). General Retirement Funds (GRF) uses its own scrip or shares in various joint ventures with GMC and, in general, in pursuit of advancing the cause of excellent retirement plans to fit its customers' various needs and desires.

A prototype Graphical User Interface (GUI) derived from bitcoin-qt and multicoin is is now on github at https://github.com/knotwork/bitcoin-qt

Github has it as a fork of bitcoin-qt but mostly because I did not know how to rename it something more related to GRouPcoin, because it is an initial test of the cloning or forking of variants and this first variant is a variant headed toward use with GRouPcoins.

I would like to fork more variants directly off of bitcoin-qt or possibly off of this fork of bitcoin-qt, however I have not yet learned how to do that. I had in mind bitNicKeLs for my next fork...

-MarkM-

Yes, there was a proposal for a purely peer-to-peer mining pool. The basic idea is that I try to mine some blocks that include a payoff to you in them and I submit proof-of-work to you. You, in exchange, try to mine some blocks that include a payoff to me in them and submit proof-of-work to me.

The client would run promiscuously for a bit, adding small payoffs to various people who claim to offer the reciprocal service. After it has generated a few proofs of work, it would then switch to "tit for tat" mode, providing a tiny bit more proof of work in return for any proofs of work it got.

Are these threads germane?
http://forum.bitcoin.org/index.php?topic=18313.0
http://forum.bitcoin.org/index.php?topic=9137.0

Well I would expect a lot more mining pools to pop-up now that all the software for doing so is open source.

Well the original Bitcoin client mining is a pure p2p with no centralized implementations. and look what the community took it so far by creating centralized exchanges and mining pools.


I think Satoshi need to step back again to correct this situation, maybe fixing the original protocol to make the Bitcoin client mining work as cooperative task across teh network. Instead of making each node mine against the other one by default design.


Maybe then we get red of centralized pools that get daily-5-dollars DDoS.

as for exchanges I am out of ideas, but code a sample code be written for a decentralized Bitcoin exchange ?


Maybe this way everyone will be back to the original client and will be actually using the BTC instead of running the mining software! or using the exchangers.

Nice, thanks.

Anyone reading ?
anyone responding ?


my suggestion is to update the network protocol model to make the default network mining "cooperative mining". and issue a new client with these updates.

This way no more centralized pools are needed. and no more 200-dollar-DDoS attack can bring the network to its knees.

There is already a decentralized bitcoin exchange in the works: http://forum.bitcoin.org/index.php?topic=26063.0

It's still in development and having some serious issues right now though.

Wasn't there some fairly well-developed plan to include support for corporative mining in the system itself, bypassing the single point of failure of the current pools? I can't find the topic right now, but it sounded good.

The idea is good, but wouldn't this result in such a huuuge amount of traffic, that
some miners can't participate and thus destroying the concept of bitcoin?

regards, talpan

I do not think mining traffic is that big.