The threat of centralization through pools?

[Mahkul]

The recent rapid increase of generation difficulty got me thinking: soon for most of us pools will be the only option to generate any coins. The problem with that is that we are leaning towards centralization - if someone wanted to attack the network they would nearly had to attack the pools (e.g. slush's pool with 60GHash/s is a huge chunk of the overall power of the Bitcoin network).

Any thoughts on that?

[no to the gold cult]

"Combination is always better than competition".

-Jack London, The Iron Heel.

Can the bitcoin network revert to a more distributed form, can difficulty go down as well as up?

[grondilu]

I don't mind if there is a bit of centralisation, as long as it is not coerced.

There can be many pools, with different rules.  Plus, nothing prevents you from investing in a lot of CGU and do your own mining.

[Mike Hearn]

Right now the software Slush uses to run his pool is proprietary, so it's a bit harder for people to set up competing pools. You could try asking him to open source it.

Alternatively sipa posted a decent proposal for p2p pooled mining in the developer forum recently. It'd be good to see somebody make a serious attempt at implementing that.

[caveden]

Can the bitcoin network revert to a more distributed form, can difficulty go down as well as up?

Yes for both questions.

[Mahkul]

I don't mind if there is a bit of centralisation, as long as it is not coerced.

There can be many pools, with different rules.  Plus, nothing prevents you from investing in a lot of CGU and do your own mining.

Yes. But what I am trying to say is that it would be much easier for some organization hostile to Bitcoin to turn off (or attack) a few pool servers than to disconnect thousands of single users.

[Nefario]

By the way, how much is slush making from running his pool?

[myrkul]

Yes. But what I am trying to say is that it would be much easier for some organization hostile to Bitcoin to turn off (or attack) a few pool servers than to disconnect thousands of single users.

I don't think that would really work... even if every pool were systematically DDoS'd, others would pop up like mushrooms, it would provide incentive for that P2P network to get truly active, or the network would revert to a decentralized, low-difficulty stage. Ever try to nail Gelatin to a wall?

[grondilu]

Yes. But what I am trying to say is that it would be much easier for some organization hostile to Bitcoin to turn off (or attack) a few pool servers than to disconnect thousands of single users.

Well, even if they do, then the difficulty will decrease and the mining will be done by CPUs...

I really think there isn't much to worry about.

[no to the gold cult]

"thanks Ben" lol

[Ricochet]

By the way, how much is slush making from running his pool?

Probably not much, given that he recently had to enforce a mandatory donation in order to keep things running.  I highly doubt he's making a profit after all the bandwidth and server costs, and is providing us the service more or less out of goodwill.

[caveden]

So people are really paying just to get a more frequent supply of minted coins?

[kwukduck]

By the way, how much is slush making from running his pool?

Probably not much, given that he recently had to enforce a mandatory donation in order to keep things running.  I highly doubt he's making a profit after all the bandwidth and server costs, and is providing us the service more or less out of goodwill.

assuming an average rate of 57 GHash/s

which gives an average of 31 minutes per block

daily thats

24*0.5*50*0.02=12 BTC

[myrkul]

So people are really paying just to get a more frequent supply of minted coins?

 I donno about anyone else, but I'm in the pool because I'd rather have a slow trickle than nothing until I hit the lottery. And I'll gladly fork over a small percentage of that income to ensure that trickle keeps coming.

[Ricochet]

By the way, how much is slush making from running his pool?

Probably not much, given that he recently had to enforce a mandatory donation in order to keep things running.  I highly doubt he's making a profit after all the bandwidth and server costs, and is providing us the service more or less out of goodwill.

assuming an average rate of 57 GHash/s

which gives an average of 31 minutes per block

daily thats

24*0.5*50*0.02=12 BTC

We know nothing about the daily costs associated with hosting the pool server.  It could very well be more than 12 BTC worth, just as likely as it could be less.

[TheHoboHarvester]

There's still a significant amount of solo miners with 5970 and/or 5870 clusters.

Though it is true that the "get paid no matter what" pool that recently popped up will draw some people in due to giving out coins solely based on expected block generation rate.

[FooDSt4mP]

assuming an average rate of 57 GHash/s

which gives an average of 31 minutes per block

daily thats

24*0.5*50*0.02=12 BTC

Your math is lacking.

40 minutes = 48 blocks per day.

* 50  = 2400

* minimum 2% = 48 BTC per day

Just to avoid typo confusion, 30 minutes = 48 blocks per day is the correct conversion foo, but the result is correct.

[myrkul]

Isn't this assuming that the pool finds every block, something we can see on the result page is false?

[dissipate]

Isn't this assuming that the pool finds every block, something we can see on the result page is false?

If the pool had over 50% of the total block output, in theory the pool operator could manipulate the block chain.

[TiagoTiago]

Even thought as a whole the system won't be brought down, there is still the risk lots of money of lots of people will sudenlly disapear, suddenlly increasing the value of the bitcoins still circulating, possibly by several orders of magnitude depending on how big the hole is.