A case AGAINST merging Namecoin and Bitcoin mining [Converted to SUPPORT!]

[JohnDoe]

Auxilliary blockchains are endless in number, the parent blockchain they look to will have more and more and more such blockchains's data in its coinbase transactions.

So it would probably be better to make a blockchain specifically for the purpose of being the parent of a potentially quite large number of auxilliary blockchains. Then bitcoin can decide whether it wants to be an auxilliary to such a thing itself or continue on its own even when possibly this metachain might become even larger in hash power than bitcoin (theoretically, as more and more chains, some possibly of some actual usefulness or value, are added to it).

-MarkM-

This assume that there'll be an endless amount of auxiliary chains that will be worthy enough of being mined. Personally I don't even see a third chain rising to prominence in the foreseeable future. But even if so, it's ~33 bytes per block according to Mike Hearn. I could send BTC to myself once every block and I'll be adding more information than 7 auxiliary chains combined. If 33 bytes per block is a scalability problem then we are already doomed.

That's because you don't understand the concept of merged mining that's being discussed.   

I don't know if I'm getting trolled or what. Read first, speak later: http://dot-bit.org/Merged_Mining

[1714060657]

1714060657

[1714060657]

1714060657

[1714060657]

1714060657

[1714060657]

1714060657

[kiba]

If you guys have any more objections to voice, please do so.

I am quite sure that they are probably misunderstanding.

[casascius]

No, you can mine NMC by itself if you desire. Plus there's no way for Tor to support Namecoin, Namecoin will support Tor instead. The Tor project doesn't have to do anything, unless by "full support" you mean buying a Namecoin domain for every hidden service that asks for one.

Perhaps the real direction of "support" is on the Polipo proxy that is bundled with Tor to make browsing work.

By "full support" I mean that the client is able to cryptographically confirm the integrity of the Namecoin block chain, which it does by verifying the hashes from the genesis block to the current block.  If the Namecoin block chain is dependent on something that's being tacked onto the Bitcoin block chain, the only way it can confirm the integrity of the chain is to download both blockchains.  So, someone who wants to run a Namecoin DNS server suddenly has to download this Bitcoin block chain that will be terabytes soon and growing exponentially each time someone eats at Meze Grill or scores some weed.  All so miners can get an extra payout?

[JohnDoe]

By "full support" I mean that the client is able to cryptographically confirm the integrity of the Namecoin block chain, which it does by verifying the hashes from the genesis block to the current block.  If the Namecoin block chain is dependent on something that's being tacked onto the Bitcoin block chain, the only way it can confirm the integrity of the chain is to download both blockchains.  So, someone who wants to run a Namecoin DNS server suddenly has to download this Bitcoin block chain that will be terabytes soon and growing exponentially each time someone eats at Meze Grill or scores some weed.  All so miners can get an extra payout?

No, both chains will be independent of one another. You don't need to download the Bitcoin blockchain to "cryptographically confirm the integrity of the Namecoin block chain". You can be a regular Namecoin miner or node and still asses if a block/transaction is valid or not.

[casascius]

No, both chains will be independent of one another. You don't need to download the Bitcoin blockchain to "cryptographically confirm the integrity of the Namecoin block chain". You can be a regular Namecoin miner or node and still asses if a block/transaction is valid or not.

Then what's the point of sticking proof of anything in the Bitcoin block chain in the first place if nobody ever needs to read it?

[phillipsjk]

No, both chains will be independent of one another. You don't need to download the Bitcoin blockchain to "cryptographically confirm the integrity of the Namecoin block chain". You can be a regular Namecoin miner or node and still asses if a block/transaction is valid or not.

If that is true, then why?

For namecoin users this means a major change, as the blockains won't be compatible between the version below block 24k and above 24k.

Some kind of blockchain corruption is happening. Bitcoin is desinged not to take up too much memory. Why not let the people who want to run two bitcoin clients: one for each blockchain?

I understand namecoin is in a lull right now, but that does not mean you should force bitcoin users to deal with it.

Edit: I also don't like how it is assumed that all miners are using pools.

Mining becomes more profitable (mining two currencies at the same hash rate for the same kWh. If you don't care about Namecoin you can just sell the mined NMC for BTC).

So.. by halving how much work you do securing both block chains, you increase miner profits? If you get a successful transaction block, are both currencies included? If so, you are either leaving something insecure or bloating both blockchains.

[hashcoin]

lol @ people not understanding how things work making arguments that don't make any sense.
Protip:
1) read the alternative chains wiki page
2) understand what a merkle branch is
3) understand (2) allows securing an arbitrarily large namecoin/blahcoin block by adding ~30 bytes to a bitcoin block. 
4) understand a single bitcoin tx is more than 80bytes
5) stop whining

[phillipsjk]

lol @ people not understanding how things work making arguments that don't make any sense.
Protip:
1) read the alternative chains wiki page
2) understand what a merkle branch is
3) understand (2) allows securing an arbitrarily large namecoin/blahcoin block by adding ~30 bytes to a bitcoin block.  
4) understand a single bitcoin tx is more than 80bytes
5) stop whining

But then you can't use namecoin without downloading the entire bitcoin blockchain. You can't verify a specific merkle leaf without examining the whole block. Are you trying to claim that nobody will want to run a namecoin miner without the bitcoin blockchain?

Edit: what happens when bitcoin gets replaced by bictoin2, but we think namecoin is still good?

[hashcoin]

You can't verify a specific merkle leaf without examining the whole block.

Not only can you do that, but that is in fact the entire purpose of a merkle branch.  (Note I say branch not leaf; a markle branch is a path from root to leaf, along with the siblings, so it a string of length O(log n))).
http://en.wikipedia.org/wiki/Hash_tree

So someone interested in namecoin mining/verifiction would need:

1) The bitcoin block headers
2) The namecoin blocks
3) The merkle branch from the bitcoin block header to the hash of the namecoin block header in the bitcoin block

So the added overhead to bitcoin is adding the hash of the namecoin block header to the bitcoin block, or 30 bytes.
The added overhead to namecoin is a merkle branch from the bitcoin block header to the namecoin block header, or roughly an O(log n) additive amount of space, where n is the size of the bitcoin block.

So if anyone would complain, it would be namecoin fans, absolutely not bitcoin fans.  However IMO a small additive space overhead is worth the security guarantees of getting all bitcoin miners verifying namecoin too.

EDIT: Re your edit.  We jump with joy and take the opportunity to do things the correct way: put both bitcoin2 and namecoin as auxilliary blocks of a more general "proof-of-work-for-hash-tree" system, whose "blocks" will be nothing more than pointers to auxilliary block chain headers.  In particular, such blocks for the primary chain will always be very tiny, and so the auxilliary chains only incur O(log n) overhead where n is the number of auxilliary chains.

[kjj]

lol @ people not understanding how things work making arguments that don't make any sense.
Protip:
1) read the alternative chains wiki page
2) understand what a merkle branch is
3) understand (2) allows securing an arbitrarily large namecoin/blahcoin block by adding ~30 bytes to a bitcoin block. 
4) understand a single bitcoin tx is more than 80bytes
5) stop whining

The wiki page isn't very good.  Actually, were I in a less charitable mood, I'd say it sucks donkey balls. Took me several reads to "get" it, and that was only because I've read the forums extensively and was able to put the pieces together.

Summary:

Basically, they hide the namecoin block hash in the coinbase transaction of the bitcoin block, and if a hash is found for the bitcoin header that meets the namecoin difficulty standard, the namecoin chain can use the entire bitcoin header to extend the namecoin chain, even if the bitcoin difficulty target was not met.  It does not extend the header size, but it does extend the coinbase script field, which isn't a very big deal because it wasn't a fixed size to begin with.

[casascius]

Summary:

Basically, they hide the namecoin block hash in the coinbase transaction of the bitcoin block, and if a hash is found for the bitcoin header that meets the namecoin difficulty standard, the namecoin chain can use the entire bitcoin header to extend the namecoin chain, even if the bitcoin difficulty target was not met.  It does not extend the header size, but it does extend the coinbase script field, which isn't a very big deal because it wasn't a fixed size to begin with.

OK, I suppose I will consider myself sold on it.  This summary belongs in the Wiki, because I couldn't gather this from what was there.

Essentially what you are saying is that reference to the BTC block chain isn't ever needed, and that there is no dependency, but rather, that NMC will accept a "proof of work" formatted for Bitcoin as a Namecoin proof of work - and tack it onto the Namecoin block chain - so long as that Bitcoin proof of work contains a reference to the most recent Namecoin block.  And that NMC will look for the merkle root not in the merkle root field (since that's reserved for Bitcoin compatibility), but in the coinbase.

As a purely abstract argument, I assume this would mean one doesn't actually need to be mining real Bitcoins, the Bitcoin block chain will never be consulted by any non-mining Namecoin client, and that a block that ends up becoming stale or rejected off the Bitcoin block chain can still remain valid on Namecoins, and in fact never need to become part of the Bitcoin block chain, if they meet the Namecoin difficulty but not Bitcoin.

If I have understood this correctly, then I hereby do a 180 and endorse the idea and apologizing for not seeing the greatness (though the documentation could have been better).

In this case, why wait till block 24000?  That could be a while, especially at the low end of this manic-depressive mining cycle.  If the Namecoin exchange, the Namecoin block explorer, the Name DNS servers, and other critical infrastructure convert, the conversion can happen pretty much any time, because there will be no functional tools or any reason for anyone to stay on the old system.

[marcus_of_augustus]

So if anyone would complain, it would be namecoin fans, absolutely not bitcoin fans.

This.

But the namecoin "fans" don't know what they are wishing for ... all they see is money in their eyes from "more hashing power" ...

It is a very pernicious mod. to namecoin that has been rammed thru in a somewhat unseemly fashion imho.

The ruination of namecoin would serve as a very stern warning to any other aspirations for starting competing blockchain currencies.

[casascius]

So if anyone would complain, it would be namecoin fans, absolutely not bitcoin fans.

This.

But the namecoin "fans" don't know what they are wishing for ... all they see is money in their eyes from "more hashing power" ...

It is a very pernicious mod. to namecoin that has been rammed thru in a somewhat unseemly fashion imho.

The ruination of namecoin would serve as a very stern warning to any other aspirations for starting competing blockchain currencies.

Nah I don't see it quite the same. There will be 7200 nmc on average per day regardless of how it's mined. Right now namecoin is suffering from a manic depressive negative feedback loop where people rush to mine it when the difficulty is low, spiking it so no one wants to mine it anymore, making it lowregain for another cycle.  This proposal would completely solve that.

[Seraphim401]

I am in favor of merging.
Why?Let's have a look:http://dot-bit.org/tools/nextDifficulty.php.
If or when the majority of miners leave the network NMC is screwed.
 

[smoothie]

Summary:

Basically, they hide the namecoin block hash in the coinbase transaction of the bitcoin block, and if a hash is found for the bitcoin header that meets the namecoin difficulty standard, the namecoin chain can use the entire bitcoin header to extend the namecoin chain, even if the bitcoin difficulty target was not met.  It does not extend the header size, but it does extend the coinbase script field, which isn't a very big deal because it wasn't a fixed size to begin with.

OK, I suppose I will consider myself sold on it.  This summary belongs in the Wiki, because I couldn't gather this from what was there.

Essentially what you are saying is that reference to the BTC block chain isn't ever needed, and that there is no dependency, but rather, that NMC will accept a "proof of work" formatted for Bitcoin as a Namecoin proof of work - and tack it onto the Namecoin block chain - so long as that Bitcoin proof of work contains a reference to the most recent Namecoin block.  And that NMC will look for the merkle root not in the merkle root field (since that's reserved for Bitcoin compatibility), but in the coinbase.

As a purely abstract argument, I assume this would mean one doesn't actually need to be mining real Bitcoins, the Bitcoin block chain will never be consulted by any non-mining Namecoin client, and that a block that ends up becoming stale or rejected off the Bitcoin block chain can still remain valid on Namecoins, and in fact never need to become part of the Bitcoin block chain, if they meet the Namecoin difficulty but not Bitcoin.

If I have understood this correctly, then I hereby do a 180 and endorse the idea and apologizing for not seeing the greatness (though the documentation could have been better).

In this case, why wait till block 24000?  That could be a while, especially at the low end of this manic-depressive mining cycle.  If the Namecoin exchange, the Namecoin block explorer, the Name DNS servers, and other critical infrastructure convert, the conversion can happen pretty much any time, because there will be no functional tools or any reason for anyone to stay on the old system.

I guess the next important question is who gets the 50 namecoins for the proof of work that is generated on the bitcoin network? Where do those coins end up residing?

[casascius]

I guess the next important question is who gets the 50 namecoins for the proof of work that is generated on the bitcoin network? Where do those coins end up residing?

The person who mined them gets them, of course.

The coins are not generated "on the bitcoin network" nor do they reside there.  They reside completely in the Namecoin block chain.  The only thing is that the block is in a different format that keeps it compatible with being submitted to Bitcoin if it also happens to satisfy the Bitcoin difficulty.  Think of it another way - instead of thinking of "dual mining", think of mining for Namecoins with a special block format that happens to be compatible with Bitcoin.

This is what I originally did not understand.  I believed that Namecoin proof of work would be stored in the Bitcoin block chain and that the namecoin client would need to consult both chains just to follow all Namecoin proof of work.  I was wrong.

The point of this change is to come up with a new format of a block that is 100% backwards compatible with Bitcoin but which also benefits the Namecoin network, the same way when they invented color analog TV, they did it with a signal that was 100% compatible with black-and-white TV's which simply ignored the color portion of the signal.

[marcus_of_augustus]

tl;dr summary? namecoin is color TV, bitcoin is black 'n white??

[ttk2]

tl;dr summary? namecoin is color TV, bitcoin is black 'n white??

Merged mining works like this, you have two totally separate block chains, they are not related in any way nor does either contain any data from the other. When you mine you generate hashes that may be the solution to the current block, this is very very improbable per hash, its like a lottery where everyone generates tickets until someone finds the winning one. Normally you make tickets and check them against the Bitcoin block chain to see if they are the solution. With merged mining you create a ticket and check it against both the Bitcoin block chain and the Namecoin block chain, Bitcoin and Namecoin know nothing about each other, they are two totally different lotteries with different winning numbers, you just sent a copy of your ticket to both. Since you are sending the same ticket to two lotteries you increase your chances of winning one or the other. No Bitcoin data goes into Namecoin no Namecoin data into Bitcoin they remain totally separate, you simply run both the Namecoin and Bitcoin clients on the same machine and submit hashes to both networks, if your hash is the solution to the Namecoin block you get Namecoins if you hash is the solution to the Bitcoin block you get Bitcoins, its exactly like if you where mining on just one network, except you submit the same work twice.


Not quite tl;dr but i think its gets the point across. 

[JoelKatz]

Essentially what you are saying is that reference to the BTC block chain isn't ever needed, and that there is no dependency, but rather, that NMC will accept a "proof of work" formatted for Bitcoin as a Namecoin proof of work - and tack it onto the Namecoin block chain - so long as that Bitcoin proof of work contains a reference to the most recent Namecoin block.  And that NMC will look for the merkle root not in the merkle root field (since that's reserved for Bitcoin compatibility), but in the coinbase.

Correct. And the benefit is that both chains will tend to contain more hashes if this is done, thus making them both more secure.

In this case, why wait till block 24000?  That could be a while, especially at the low end of this manic-depressive mining cycle.  If the Namecoin exchange, the Namecoin block explorer, the Name DNS servers, and other critical infrastructure convert, the conversion can happen pretty much any time, because there will be no functional tools or any reason for anyone to stay on the old system.

Making a change in what is or is not considered a valid block can split the network. So you have to plan it way in advance, wait for the vast majority of the mining power to understand the new rules, and only then introduce blocks whose validity differs in the two schemes. And the vast majority of nodes must agree on precisely when the rules change. Otherwise, those with the old software will track a different hash chain than those with the new software.

Long-term I think the bitcoin protocol should be adapted to specifically include foreign merkle roots in a known format. This will allow people to pile bitcoin proof of work on top of anything they need without having to 'disguise' it as a dummy transaction or the like. (This could consist simply of an agreement that a particular type of transaction will be used. No actual code/format changes are likely needed.) An accepted fee of, say, .005 BTC to include one 256-bit chain root, one 256-bit merkle hash, and one 256-bit "accessory" field would do.

[smoothie]

I guess the next important question is who gets the 50 namecoins for the proof of work that is generated on the bitcoin network? Where do those coins end up residing?

The person who mined them gets them, of course.

The coins are not generated "on the bitcoin network" nor do they reside there.  They reside completely in the Namecoin block chain.  The only thing is that the block is in a different format that keeps it compatible with being submitted to Bitcoin if it also happens to satisfy the Bitcoin difficulty.  Think of it another way - instead of thinking of "dual mining", think of mining for Namecoins with a special block format that happens to be compatible with Bitcoin.

This is what I originally did not understand.  I believed that Namecoin proof of work would be stored in the Bitcoin block chain and that the namecoin client would need to consult both chains just to follow all Namecoin proof of work.  I was wrong.

The point of this change is to come up with a new format of a block that is 100% backwards compatible with Bitcoin but which also benefits the Namecoin network, the same way when they invented color analog TV, they did it with a signal that was 100% compatible with black-and-white TV's which simply ignored the color portion of the signal.

Okay so then that would imply that both the bitcoin and namecoin clients would have to updated the code to submit the proof of work to both block chains or no?

I guess for bitcoin since 95% generally pool mine, only the pools need to update their software to submit proof of work to the namecoin network in order to capture the majority for merged mining.

So then what will likely happen when this does take place is that since the namecoin difficulty is like 1/17th of that of the bitcoin network the difficulty will likely catch up to the bitcoin network and likely come to parity at some point in the near future.

And lastly if someone creates a proof of work that satisfies both bitcoin and namecoin at that moment they get 50 BTC and 50 NMC?

Does this sound right?