Why I believe Mastercoin should be encoded on the bitcoin blockchain

[ripper234]

There have been several well known bitcoiners that tried to convince J.R.Willet to move Mastercoin to an altchain, instead of using the bitcoin blockchain. Willet's opinion is that Mastercoin should stay on the blockchain, and I fully concur. Here is my reasoning:

It is easier to implement Mastercoin on the blockchain. The entire concept of mining "vanishes", or actually is satisfied by Bitcoin miners. This means the security of Mastercoin is on par with Bitcoin (up to potential bugs) ... there is no way to 51% attack it.

The same could be implemented using merged mining, but no matter how you look at it, merged mining is technically a more complicated solution, because it requires an entirely new blockchain one needs to reason about. You need to think about the hashrate of this separate blockchain and to encourage miners to merge-mine it, things that are just not needed when you directly encode things into the Bitcoin blockchain. As far as software development goes, I'm super lazy - I do not like to do something complicated if I can do something simple that serves the same purpose.

Now, it is true that this "burdens" the blockchain ... in the same way that Satoshi Dice does.
However, this burdening is something that Bitcoin users and developers will just have to live with. Nobody is the boss of Bitcoin, and any standard Bitcoin transaction (like Mastercoin transactions are) are legitimate transactions that Bitcoin simply has to learn to deal with. There is just no feasible way that I see that Bitcoin developers could ban such transactions (please enlighten me if I'm wrong).

Bitcoin is the future of finance, resistant to hacking, government censorship and regulation. Do you really want to base that future on asking other people to pretty please "not abuse it for you own purposes"? This is not the answer. Bitcoin has to be made scalable enough to meet its role as the one major, global, world currency.

Mastercoin developers have no intention of disrupting Bitcoin, and will gladly adapt to any friendly encoding that the brilliant minds of these forums will propose, as long as it doesn't harm Mastercoin (using non-standard transactions that are relayed more slowly is not an option at the moment). The Bitcoin blockchain must and will remain neutral, and if Willet hadn't come up with Mastercoin encoding, someone else would have created something similar sooner or later.

[btcdrak]

It is easier to implement Mastercoin on the blockchain. The entire concept of mining "vanishes", or actually is satisfied by Bitcoin miners. This means the security of Mastercoin is on par with Bitcoin (up to potential bugs) ... there is no way to 51% attack it.

The same could be implemented using merged mining, but no matter how you look at it, merged mining is technically a more complicated solution, because it requires an entirely new blockchain one needs to reason about. You need to think about the hashrate of this separate blockchain and to encourage miners to merge-mine it, things that are just not needed when you directly encode things into the Bitcoin blockchain. As far as software development goes, I'm super lazy - I do not like to do something complicated if I can do something simple that serves the same purpose.

So your entire reasoning is that, while merged mining is in fact a solution, you are too lazy to do it? You need a specialized client already, so why not mastercoin-qt. Sheesh. It takes almost zero effort to start an altchain, evidenced by the rash of them already.

Your argument is weak. Let's face it, Willet just wants to do it this way, there is no other reasoning other than, "I want to, I can, so I will".

[killerstorm]

The same could be implemented using merged mining, but no matter how you look at it, merged mining is technically a more complicated solution, because it requires an entirely new blockchain one needs to reason about. You need to think about the hashrate of this separate blockchain and to encourage miners to merge-mine it, things that are just not needed when you directly encode things into the Bitcoin blockchain.

Mastercoin is different from your typical merged-mined alt-coin: it does not rely on validation by miners, it only requires timestamping.

And so you can just use Bitcoin blockchain for timestamping and enjoy security of full Bitcoin hashrate.

As far as software development goes, I'm super lazy - I do not like to do something complicated if I can do something simple that serves the same purpose.

Well, the difference is that fees will likely be order(s) of magnitude lower on the alt-chain. So there is an incentive to use it.

As for laziness, if Namecoin devs were able to implement merged-mining without any funding, and you cannot do that having $500k in the bank, why should people trust you?

[Come-from-Beyond]

However, this burdening is something that Bitcoin users and developers will just have to live with. Nobody is the boss of Bitcoin, and any standard Bitcoin transaction (like Mastercoin transactions are) are legitimate transactions that Bitcoin simply has to learn to deal with. There is just no feasible way that I see that Bitcoin developers could ban such transactions (please enlighten me if I'm wrong).

If the devs were able to do this - http://bitcoinmagazine.com/4465/bitcoin-developers-adding-0-007-minimum-transaction-output-size/ - they can ban Mastercoin transactions. Bitcoin's decentralization is just a myth. All big pool owners will do what Gavin and Co. will ask them to do.

[ripper234]

As far as software development goes, I'm super lazy - I do not like to do something complicated if I can do something simple that serves the same purpose.

Well, the difference is that fees will likely be order(s) of magnitude lower on the alt-chain. So there is an incentive to use it.

As for laziness, if Namecoin devs were able to implement merged-mining without any funding, and you cannot do that having $500k in the bank, why should people trust you?

If Mastercoin fees become a problem, we could always migrate away to a separate blockchain (and it's not 100% that they will be).
Right now this is the fastest route to push development forward.

We'd rather spend the $500k to implement features, not infrastructure that may-or-may-not be needed.

Still, when the time is right, various optimization features will warrant bounties of them own.
If a separate blockchain is deemed necessary because the Bitcoin blockchain is drowning in Mastercoin transactions, and this is really the way to keep Mastercoin fees at decent levels, then I'm sure a part of the 1Exodus funds could be allocated to investigating this option.

However, this burdening is something that Bitcoin users and developers will just have to live with. Nobody is the boss of Bitcoin, and any standard Bitcoin transaction (like Mastercoin transactions are) are legitimate transactions that Bitcoin simply has to learn to deal with. There is just no feasible way that I see that Bitcoin developers could ban such transactions (please enlighten me if I'm wrong).

If the devs were able to do this - http://bitcoinmagazine.com/4465/bitcoin-developers-adding-0-007-minimum-transaction-output-size/ - they can ban Mastercoin transactions. Bitcoin's decentralization is just a myth. All big pool owners will do what Gavin and Co. will ask them to do.

I hardly see Gavin or the core devs, or pool operators for that matter, trying to outright ban or slow down Mastercoin transactions, which are always above the dust level and pay all required miner fees.
Mastercoin can adapat if needed, but I don't think it will come to that. Gavin and the core devs will not want to play a game of cat-and-mouse with the Mastercoin protocol - they have far more important things to do. And I'll add that Mastercoin developers aren't just ignoring everything Bitcoin devs are saying - rather they are rather cooperative in finding better, more optimal ways to encode Mastercoin transactions. It is possible that a sort of merged-mined alt-chain will be deemed the optimal solution one far day in the future ... but for now it's certainly not the case.

The entire merged mined / timestamping service approach to Mastercoin should just be viewed as a possible future optimization, and as we know optimizations are best implemented at the correct time, and not before.

[btcdrak]

The altchain part is not a bottleneck. It takes a few hours to make the alterations and launch. And yes, Gavin *has* said there is a strong possibility of making changes that will prevent embedding of arbitrary data. It's just not his focus yet, but it is very much on his radar.

[ripper234]

The altchain part is not a bottleneck. It takes a few hours to make the alterations and launch. And yes, Gavin *has* said there is a strong possibility of making changes that will embedding of arbitrary data. It's just not his focus yet, but it is very much on his radar.

It's not a matter of a spending the few hours - the design as an altchain is very different, and will require miners, which the current design does not. Ergo, the current design is simpler and was chosen as the implementation.

We do not want another altchain, mining, block rewards ... all these needlessly complicate the design.
Mastercoin just doesn't require these elements to function.

In the future, if there is a pressing need (remains to be seen), then we might adapt to it by introducing a sort of mining, but it will definitely not be mining in the traditional sense (e.g. it can't create additional mastercoins or bitcoins).

[dacoinminster]

One important reason I'm not moving to an alt-chain is the distributed exchange between bitcoin and MasterCoin, which isn't possible if we are on an alt-chain, since it requires using data from standard bitcoin transactions AND MasterCoin transactions to work.

I don't think I've mentioned that reason anywhere else, but it is probably the primary reason I'm not even willing to consider it. There are other reasons, like total cost of development, and even the dirty stink of starting yet another alt chain (For instance, bitshares is doing it "right", but they were banished to the alt section with all those scam coins, which they of course bitterly resent).

I am totally convinced that building on top of bitcoin is the right way to go, for all the reasons here and above in Ron's posts.

[ASICSRUS]

MST  = hijacks the blockchain ? kewL weeeeeeee

[Patel]

Even if mastercoin stays on the bitcoin blockchain, bitcoin needs to evolve to counter others from bloating the blockchain

mastercoin was the first (not against it, i am for the idea of colored coins), there will be others and they may not have the best interest of bitcoin in mind. and want to do more harm to bitcoin than good

[ASICSRUS]

Even if mastercoin stays on the bitcoin blockchain, bitcoin needs to evolve to counter other parasites from bloating the blockchain

mastercoin was the first (not calling it a parasite, im all for colored coins), there will be others and they may not have the best interest of bitcoin in mind. technically, you can even consider this an attack on the blockchain

i'm having difficulty understanding how any mechanisim that will increase use of the bitcoin system and builds faith in the bitcoin blockchain as it is now "bad biz".
Think: if Gavin/btcGeekzzz are not even able to figure out how to handle this minute amount of extra data what make's you think they will be able to handle the real crazy gangsta wall street chit that's coming? LOL

[btcdrak]

The whole thing is so foolish it's not funny.

[ASICSRUS]

The whole thing is so foolish it's not funny.

= da next satoshi martial arts center!

[thezerg]

One important reason I'm not moving to an alt-chain is the distributed exchange between bitcoin and MasterCoin, which isn't possible if we are on an alt-chain, since it requires using data from standard bitcoin transactions AND MasterCoin transactions to work.

I don't think I've mentioned that reason anywhere else, but it is probably the primary reason I'm not even willing to consider it. There are other reasons, like total cost of development, and even the dirty stink of starting yet another alt chain (For instance, bitshares is doing it "right", but they were banished to the alt section with all those scam coins, which they of course bitterly resent).

I am totally convinced that building on top of bitcoin is the right way to go, for all the reasons here and above in Ron's posts.

It would be easy enough to build a client that could do atomic cross-blockchain txns... essentially the txn in mastercoin-chain would reference a bitcoin txn and maximum block and be invalid if that bitcoin txn "pair" is not in the bitcoin chain before that block.  A subsequent mastercoin-chain "txn" could set a marker that explicitly validates or invalidates the mastercoin-txn so mastercoin-only clients could exist.

None of your mining, minting etc. arguments hold water because its just a fork of the existing OSS bitcoin code.


But really none of these arguments matter.  Its great that MC txn fees encourage bitcoin use/velocity.   And if it becomes an issue for bitcoin, Bitcoin txn fees will rise to the point where you will move.

I think you are really feeling a backlash because you advertised MC as a service on top of Bitcoin rather than a competing alt-coin.  But in truth it is a competing alt-coin that uses the bitcoin infrastructure.  Its like you said you are going "sailing" but are actually just sitting in your car on a ferry.  MC is technically a service on top of Bitcoin but in essence it is a separate coin.

In contrast, Casascius coins are technically not a service on top of Bitcoin (to transfer them you don't use the blockchain).  But they ARE in their essence.


But the reality is that an alt-coin is an easy way to raise money without going to VCs.  And projects (especially those about money) lose steam without some funding as we are seeing with colored coins.

[ASICSRUS]

One important reason I'm not moving to an alt-chain is the distributed exchange between bitcoin and MasterCoin, which isn't possible if we are on an alt-chain, since it requires using data from standard bitcoin transactions AND MasterCoin transactions to work.

I don't think I've mentioned that reason anywhere else, but it is probably the primary reason I'm not even willing to consider it. There are other reasons, like total cost of development, and even the dirty stink of starting yet another alt chain (For instance, bitshares is doing it "right", but they were banished to the alt section with all those scam coins, which they of course bitterly resent).

I am totally convinced that building on top of bitcoin is the right way to go, for all the reasons here and above in Ron's posts.

It would be easy enough to build a client that could do atomic cross-blockchain txns... essentially the txn in mastercoin-chain would reference a bitcoin txn and maximum block and be invalid if that bitcoin txn "pair" is not in the bitcoin chain before that block.  A subsequent mastercoin-chain "txn" could set a marker that explicitly validates or invalidates the mastercoin-txn so mastercoin-only clients could exist.

None of your mining, minting etc. arguments hold water because its just a fork of the existing OSS bitcoin code.


But really none of these arguments matter.  Its great that MC txn fees encourage bitcoin use/velocity.   And if it becomes an issue for bitcoin, Bitcoin txn fees will rise to the point where you will move.

I think you are really feeling a backlash because you advertised MC as a service on top of Bitcoin rather than a competing alt-coin.  But in truth it is a competing alt-coin that uses the bitcoin infrastructure.  Its like you said you are going "sailing" but are actually just sitting in your car on a ferry.  MC is technically a service on top of Bitcoin but in essence it is a separate coin.

In contrast, Casascius coins are technically not a service on top of Bitcoin (to transfer them you don't use the blockchain).  But they ARE in their essence.


But the reality is that an alt-coin is an easy way to raise money without going to VCs.  And projects (especially those about money) lose steam without some funding as we are seeing with colored coins.

no worries, if someone creates a mechanism that operates on top of the existing Bitcoin blockchain just expect to be at the mercy of Gavin and crew pulling the plug on you!
 
LOL

[dacoinminster]

But the reality is that an alt-coin is an easy way to raise money without going to VCs.  And projects (especially those about money) lose steam without some funding as we are seeing with colored coins.

So true. I remember when I first read about colored coins, I got really excited. Then I realized that there was no way to invest in their success (other than just owning bitcoins). It's a fantastic project (and killerstorm is doing a great job with it), but it has struggled to get momentum, and I think it will continue to struggle for exactly that reason.

The first rev of my spec didn't have any functionality like colored coins, but I added "smart property" in the latest rev, so hopefully we'll get IPOs and user-currencies with floating values as a result, even if the colored coins project dies on the vine.

Also, I like your analogy about going sailing. Colored coins is more of a "pure" sailing ship. As you say, mine is more of a hybrid.

[adpinbr]

for people who want to revolutionize the current financial system due to its inefficiency and unfairness, it seems like you are going down the exact same path. It is a clear moral hazard to have an "I can so I will" attitude with mastercoin integration. the blockchain is more than a public ledger, its an economic asset of itself, hogging it for personal profit is unfair towards the rest of us that want to/will want to have an unbloated block chain in the future. The fact that master coin and satoshi dice can piggy back their software for personal profit is a parasite that can't survive without their host, if in fact master coin is viable and therefore profitable (i believe it is) then like any other business it can build its own infrastructure and block chain, it can used merged mining and set up its own reward system even if it has been pre mined (such as transaction fees). If it can't be self sustaining on its own merit it is unfair to leech off the block chain. If it is such a great idea it should be able to succeed just like bitcoin was able to. but its not about long term viability its short term greed, and then after you are a billionaire maybe it will serve the better good of human kind. Satoshi Dice needed the block chain cause it was a vanity business, master coin has enough of its own merit to do its own thing. we do need distributed data systems so that we can trade stocks without brokers, make bets without bookies and create currencies without governments. But greed and short term profit has blinded the developers/supporters of this project and I am afraid you are taking a great idea and doing a bad implementation

[killerstorm]

One important reason I'm not moving to an alt-chain is the distributed exchange between bitcoin and MasterCoin, which isn't possible if we are on an alt-chain, since it requires using data from standard bitcoin transactions AND MasterCoin transactions to work.

It is possible.

Mastercoin client can follow both Bitcoin chain and "alt-chain" at the same time.

Mastercoin works by inserting messages into Bitcoin blockchain. It is done for two reasons: to distributed these messages, and to make sure that the order they appear in blockchain is fixed and can be agreed upon.

But what happens if we would do it in other way: assemble a list of Mastercoin messages, hash it, and publish this hash in the blockchain?

This will definitely make the order of messages certain, just like in the previous case. But you'll have to distribute the list by other means.

Of course, Mastercoin client still has to follow Bitcoin blockchain to know the order of messages, but the difference is that you only need to insert much less data into the Bitcoin blockchain: only 32 bytes per block. The rest can be stored elsewhere.

And as Mastercoin client has to follow Bitcoin blockchain anyway, you can easily implement cross-chain trade, it won't be in any way different from implementation you have now, I think.

Of course, it isn't as easy as I described: it isn't clear who assembles Mastercoin message list and how it is inserted into Bitcoin blockchain.

A design similar to merged mining would be somewhat inconvenient: mastercoin message list hash won't be included into each Bitcoin block, and so  it would take more time to confirm mastercoin transactions.

But possibly there is some clever way to deal with it. For example, differentiation between soft and hard confirmations.

In any case, there is no strong need to put data into Bitcoin blockchain.

It might be an easier thing to do, but it isn't a fundamental need.

[Mike Hearn]

Not only is it possible, but I documented how to do it years ago:

https://en.bitcoin.it/wiki/Alternative_chain

[ASICSRUS]

Not only is it possible, but I documented how to do it years ago:

https://en.bitcoin.it/wiki/Alternative_chain

/\hit mm with the merkle branch !!!