Namecoin Forum

indolering: Jeremy, I wanted to talk about your price discovery proposal.
Jeremy: What's up?
indolering: Who would actually control the final vote, the mining pool or the individual miner?
Jeremy: It probably depends on which mining API is used... I suspect that if getblocktemplate were implemented, we could make it so that individual miners would be voting... but I'm not sufficiently well-versed in the mining process to know for sure... maybe ask Daniel about that?
Jeremy: Keep in mind that GHash.IO's miners are mostly in house.

Domob, care to weigh in?

Also, if we could implement this via getblocktemplate, can I somehow estimate the number of independent miners within a given pool?

I have to admit that I'm not really expert in mining, either. I think that this is a very good point - with current mining methods, GHash.io would be able to dictate prices (basically). Something like getblocktemplate could be implemented, but I'm not sure how it would interact with merge-mining. I. e., if getblocktemplate is used, would the individual miners be responsible for creating the merge-mined block, constructing the auxpow and all that? Or would this be done by GHash.io and getblocktemplate only allow them to choose Bitcoin transactions they want to include? I think this definitely needs more research / external input from people who know about it.

domob wrote:I have to admit that I'm not really expert in mining, either. I think that this is a very good point - with current mining methods, GHash.io would be able to dictate prices (basically). Something like getblocktemplate could be implemented, but I'm not sure how it would interact with merge-mining. I. e., if getblocktemplate is used, would the individual miners be responsible for creating the merge-mined block, constructing the auxpow and all that? Or would this be done by GHash.io and getblocktemplate only allow them to choose Bitcoin transactions they want to include? I think this definitely needs more research / external input from people who know about it.

Luke-Jr might be a good person to ask about that. I don't think he's familiar with the price discovery proposal, but he knows a lot more about the mining process than most of us do.

Since Namecoin is merged-mined, it would actually be a bit of work to get GBT going for it.

Basically, it'd take these steps:

• Extend GBT to be merged-mining aware at the Bitcoin level; miners would need to tell the pool which merged mining chain(s) they wish to participate in, and the pool would respond which are allowed and what additional chain(s) might be required.
• Implement this extension in mining software: the software now needs to keep track of an independent Bitcoin pool list and an independent Namecoin pool list (which the Bitcoin pool may have restricted - eg, to its own Namecoin pool). The mining software then builds first the Namecoin block, then the merged mining merkle tree, then finally the Bitcoin block.
• Implement GBT's decentralisation in mining software: this is the same for both Bitcoin and Namecoin; it isn't done yet

Luke-Jr wrote:Since Namecoin is merged-mined, it would actually be a bit of work to get GBT going for it.

Basically, it'd take these steps:

• Extend GBT to be merged-mining aware at the Bitcoin level; miners would need to tell the pool which merged mining chain(s) they wish to participate in, and the pool would respond which are allowed and what additional chain(s) might be required.
• Implement this extension in mining software: the software now needs to keep track of an independent Bitcoin pool list and an independent Namecoin pool list (which the Bitcoin pool may have restricted - eg, to its own Namecoin pool). The mining software then builds first the Namecoin block, then the merged mining merkle tree, then finally the Bitcoin block.
• Implement GBT's decentralisation in mining software: this is the same for both Bitcoin and Namecoin; it isn't done yet

Well, it's probably safer to just assume that each mining pool controls 100% of their voting power anyway.

Thanks for your responses!

Luke-Jr wrote:Since Namecoin is merged-mined, it would actually be a bit of work to get GBT going for it.

Basically, it'd take these steps:

• Extend GBT to be merged-mining aware at the Bitcoin level; miners would need to tell the pool which merged mining chain(s) they wish to participate in, and the pool would respond which are allowed and what additional chain(s) might be required.
• Implement this extension in mining software: the software now needs to keep track of an independent Bitcoin pool list and an independent Namecoin pool list (which the Bitcoin pool may have restricted - eg, to its own Namecoin pool). The mining software then builds first the Namecoin block, then the merged mining merkle tree, then finally the Bitcoin block.
• Implement GBT's decentralisation in mining software: this is the same for both Bitcoin and Namecoin; it isn't done yet

Hi Luke,

I think this would be useful work; if bounties were available for these three things, would that be likely to speed up the process? Any idea what the price of a bounty would be for those 3 steps? There's not much funding at the moment, but I'm hoping that'll change at some point.

biolizard89 wrote:

Luke-Jr wrote:Since Namecoin is merged-mined, it would actually be a bit of work to get GBT going for it.

Basically, it'd take these steps:

• Extend GBT to be merged-mining aware at the Bitcoin level; miners would need to tell the pool which merged mining chain(s) they wish to participate in, and the pool would respond which are allowed and what additional chain(s) might be required.
• Implement this extension in mining software: the software now needs to keep track of an independent Bitcoin pool list and an independent Namecoin pool list (which the Bitcoin pool may have restricted - eg, to its own Namecoin pool). The mining software then builds first the Namecoin block, then the merged mining merkle tree, then finally the Bitcoin block.
• Implement GBT's decentralisation in mining software: this is the same for both Bitcoin and Namecoin; it isn't done yet

Hi Luke,

I think this would be useful work; if bounties were available for these three things, would that be likely to speed up the process? Any idea what the price of a bounty would be for those 3 steps? There's not much funding at the moment, but I'm hoping that'll change at some point.

Not sure. My time is already spread thin, so it would probably be most effective to find someone willing to do this aside from me.
Maybe things could/should be split down to more than just those 3 items? Then there might be some more reasonably-short goals people can work toward without having to worry about completing it entirely.

biolizard89 wrote: I think this would be useful work; if bounties were available for these three things, would that be likely to speed up the process? Any idea what the price of a bounty would be for those 3 steps? There's not much funding at the moment, but I'm hoping that'll change at some point.

Although this directly impacts your proposal, I don't think that these enhancements are necessary nor even moderately helpful. Assuming that the mining pools control the miners themselves is probably a correct assumption for ~1/2 of the market anyway. Consolidation to professional mining operations in which investors purchase mining power, but without direct control over the software, will only continue. Any system we emulate should make this assumption anyway.

We have many, many other things we could use Luke's help on, such as making merged mining easier. Just a handful of pools made up some 17% of Bitcoin hash power told me that they don't merge-mine because it's a PITA.