BitcoinX vs. Alternative Chains

[BlueMeanie]

 Hello,

   Why not use an Alternative Block Chain to trade alternative digital currencies and assets?

   Different mining rules and parameters could be set depending on desired outcomes.  Existing Bitcoin technology could be used(for the most part).

   Most importantly this avoids the foreseen block chain bloat.  This problem could become enormous.  By attributing new value to small amount of BTC, you increase the probable number of transactions in ratio to BTC.  This might change the economics of mining in a potentially destructive way.

   By having a separate block chain fine tuning the economic parameters of mining is possible on a per-currency basis.  Also Bitcoin Addresses are capable of holding only ONE asset type.  This simplifies things immensely.  Also latency problems only emerge when a currency becomes widely used.

   Has this option been fully explored?  I realize it has been raised before elsewhere, but have the tradeoffs been considered sufficiently?

 thanks,  -bm

[Tamas Blummer]

What would be the incentive to commit sufficient mining power to protect the assets? 

[BlueMeanie]

 Hi Tamas,

  This concept has been discussed before on the Bitcoin forums: https://bitcointalk.org/index.php?topic=1790.msg28696#msg28696

  In this message Satoshi himself suggests that it is possible for separate block chains to share computing resources.  He suggests that the rewards come in terms of the generated asset.  So if I were generating a digital gold currency then the miners would receive(a presumably small amount of) gold. This implies a whole new milieu of mining entrepreneurship.  Here though they are not discussing alternative chains for financial currencies, but rather for other purposes(domain name ownership).  I assume they just dont recognize the value in multiple currencies, or it wasn't relevant here.  The Wiki does seem to suggest this idea has been presented before.

  In addition you can use existing software as is.  Clients would work exactly the same, you would just point them to a different block chain.  Bitcoin addresses could hold many asset types if they were used on many block chains.  Of course eventually people would demand better features for managing multiple currencies and the clients would require customizations.

  The forums also discuss what is called BitX which is a generalization of Bitcoin.  I think what we're looking at is something like Bitcoin 2.0, a generalization of Bitcoin technology.  It would require something like Ricardian Contract(discussed before) that define the rules for block validation(for miners) and the Issuer's contract with the public.  https://bitcointalk.org/index.php?topic=1790.0

  ex.  I'm Meanie Gold Services and I make a new currency called MeanieBucks.

       I purchase 10000 Krugerrands and write in my Ricardian Contact the following:

          1 MeanieBuck shall be redeemable for 1 Krugerrand at any time.

          100000 MeanieBucks shall be created on this day, and no more.

          Every miner who solves a transaction block shall receive exactly 0.000001 meaniebucks (worth 0.05 USD) from address xHASHw32fwe3fss (which is the Meanie Gold Services Reserve Bank)

       This contract is hashed and put into the blockchain for all posterity.  It cannot be removed, and it cannot be withdrawn or modified.  If the Issuer does not follow through with their claims then they face public disgrace and scandal.  They cannot deny that they signed it and agreed to it- the signature is in the contract and it is tied to their domain name.  It is possible to automate these obligations as well.

   -bm

[Alex Mizrahi]

   Why not use an Alternative Block Chain to trade alternative digital currencies and assets?

This is a frequently asked question... 

We want to use specifically Bitcoin for a couple of reasons...

The main reason is that this way people will be able to buy colored coins (e.g. shares, bonds, tickets) with Bitcoins in a secure way, via so-called atomic coin swap transactions. This means that p2p exchange is possible.

Another reason is that Bitcoin offers highest degree of security against double-spends.

Also it is easier to set up this way since we only need modified clients and can re-use same infrastructure.  Existing investment into Bitcoin might be a reason too: believe it or not, some people want Bitcoin to be more and more useful.

Additionally, if you are talking about one-chain-per-asset model, it has extra disadvantages, particularly maintenance overhead.

 

   Different mining rules and parameters could be set depending on desired outcomes.

It only makes sense to create another mined currency if you want to compete with Bitcoin.

 

   Most importantly this avoids the foreseen block chain bloat.  This problem could become enormous.  By attributing new value to small amount of BTC, you increase the probable number of transactions in ratio to BTC.  This might change the economics of mining in a potentially destructive way.

When people buy and sell colored coins for Bitcoins via atomic coin swap transactions, it isn't really different from normal trade transactions, like buying cheeseburgers.

Really, such transactions are just twice larger on average, so impact from a transaction which buys, for example, shares is same as of two transactions which buy cheeseburgers.

If people trade colored coins for other colored coins, it indeed creates extra strain... I'm not sure whether it is actually a problem, but in this case it is possible to migrate such colored coins to alt-chain.

Finally, one might say that colored coins increase UTXO set size. This is true, in theory, you might get some extra UTXOs for each (user, color) pair.

But then again, we can say that UTXO set size is proportional to number of users, does that mean we should discourage new users from joining?

That's ridiculous. Colored coins, if they will be widely used, will make Bitcoin more attractive and useful, so more users will join. Isn't that what we want?

Also it's worth mentioning that there are some proposals to enhance transaction privacy by avoiding merging coins... It can have bigger impact on UTXO set size, why don't people complain about such proposals?

In the end, I think 'bloat' is paid for by transaction fees. If this mechanism doesn't work correctly then it is a flaw in Bitcoin design.

 

   By having a separate block chain fine tuning the economic parameters of mining is possible on a per-currency basis.

WTF is "mining on per-currency basis"? How is that useful?

Do you want to create a bajillion of Bitcoin copycats?

 

  Also Bitcoin Addresses are capable of holding only ONE asset type.  This simplifies things immensely.

Well, we can create addresses for colored coins, that was a plan...

But I'd say a concept of addresses is too low-level, people need to use payment protocol which would offer end-to-end security.

 

   Has this option been fully explored?  I realize it has been raised before elsewhere, but have the tradeoffs been considered sufficiently?

 

Frankly, I don't think we really need to consider trade-offs.

I'm working on colored coins.

You might start working on these alt-chains if you think they are better.

Market will decide which approach is optimal.

Ripple can be seen as a competitor of colored coins too, and I don't think it is a problem:

they might co-exist, occupy different niches or one of this projects will eventually die...

Nobody can predict what will happen, so we need to try both, I think.

[Alex Mizrahi]

  In this message Satoshi himself suggests that it is possible for separate block chains to share computing resources.  He suggests that the rewards come in terms of the generated asset.  So if I were generating a digital gold currency then the miners would receive(a presumably small amount of) gold. This implies a whole new milieu of mining entrepreneurship.  Here though they are not discussing alternative chains for financial currencies, but rather for other purposes(domain name ownership).  I assume they just dont recognize the value in multiple currencies, or it wasn't relevant here.  The Wiki does seem to suggest this idea has been presented before.

It already works this way with merged-mined alt-chains like Namecoin, Devcoin etc. 

  In addition you can use existing software as is.

 No, you can't.

[BlueMeanie]

 Hi Alex,

   thanks for your response.

On Monday, April 22, 2013 3:25:05 AM UTC-4, Alex Mizrahi wrote:

   Why not use an Alternative Block Chain to trade alternative digital currencies and assets?

This is a frequently asked question... 

We want to use specifically Bitcoin for a couple of reasons...

The main reason is that this way people will be able to buy colored coins (e.g. shares, bonds, tickets) with Bitcoins in a secure way, via so-called atomic coin swap transactions. This means that p2p exchange is possible.

   This is an interesting value proposition.  It's also possible to use another block chain to create a decentralized exchange for all asset-types.  It seems that working this into the basic technology is 'feature bloat', IMHO.
 

Another reason is that Bitcoin offers highest degree of security against double-spends.

   Not sure how that works.  As I mentioned, it is possible to use the computing power from other block chains simultaneously(ipse dixit Satoshi Nakamoto).
 

Also it is easier to set up this way since we only need modified clients and can re-use same infrastructure.  Existing investment into Bitcoin might be a reason too: believe it or not, some people want Bitcoin to be more and more useful.

   Granted there may be some pluses there, I'm not altogether clear on how everything works with color coins.
 

Additionally, if you are talking about one-chain-per-asset model, it has extra disadvantages, particularly maintenance overhead.

   I still contend that color coins are going to seriously exacerbate the block chain bloat problem.  Think about it, you are tacking on a new value to bitcoins.  It could be anything.  Suddenly one specific BTC is now worth a million dollars.  Naturally the transactions for that particular BTC(and its sub-denominations) is going to be VERY HIGH, however the incentives for miners remain the same.  We don't really know how this will effect the economics of mining.  Just seems that adding some kind of value to specific bitcoins is just plain irregular- sorry if I'm being critical here.  You still have a lot of proving and validation to do before this is ready for something like a digital gold currency.

 

   Different mining rules and parameters could be set depending on desired outcomes.

It only makes sense to create another mined currency if you want to compete with Bitcoin.

 

  This is an interesting point.  What are your ideological goals?  Many believe that competing currencies are the ideal. 

     http://www.forbes.com/sites/realspin/2012/08/23/private-currency-competition-is-the-monetary-answer/

     http://www.forbes.com/sites/johntamny/2012/08/06/are-ron-pauls-competing-currencies-the-answer-to-monetary-mischief/

     http://www.maxmore.com/hayek.htm

     J. Orlin Grabbe was really the trailblazer in this field:  http://www.orlingrabbe.com/smartcards.htm

  Bitcoin is a curious phenomenon as far as economists are concerned.  It began with no backing at all, and somewhat defied the laws of monetary theory.  In my view, it became valuable because it was the first in its kind: a p2p cryptocurrency.  It's basic utility was a value in itself.  This however doesn't mean that alternatives built with similar technoloogy are not going to be valuable, perhaps even more valuable.

 

 

   Most importantly this avoids the foreseen block chain bloat.  This problem could become enormous.  By attributing new value to small amount of BTC, you increase the probable number of transactions in ratio to BTC.  This might change the economics of mining in a potentially destructive way.

When people buy and sell colored coins for Bitcoins via atomic coin swap transactions, it isn't really different from normal trade transactions, like buying cheeseburgers.

 
   True, but you are introducing new factors that will change the ratio of BTC to Number of Transactions.  It is not difficult to implement a p2p exchange using a modified block chain.  Seems there is resistance to the idea of introducing new features such as this into the main BTC chain.  I think your argument that there is more maintenance overhead with the alt-chain proposal is not very solid.

Really, such transactions are just twice larger on average, so impact from a transaction which buys, for example, shares is same as of two transactions which buy cheeseburgers.

If people trade colored coins for other colored coins, it indeed creates extra strain... I'm not sure whether it is actually a problem, but in this case it is possible to migrate such colored coins to alt-chain.

Finally, one might say that colored coins increase UTXO set size. This is true, in theory, you might get some extra UTXOs for each (user, color) pair.

But then again, we can say that UTXO set size is proportional to number of users, does that mean we should discourage new users from joining?

That's ridiculous. Colored coins, if they will be widely used, will make Bitcoin more attractive and useful, so more users will join. Isn't that what we want?

Also it's worth mentioning that there are some proposals to enhance transaction privacy by avoiding merging coins... It can have bigger impact on UTXO set size, why don't people complain about such proposals?

In the end, I think 'bloat' is paid for by transaction fees. If this mechanism doesn't work correctly then it is a flaw in Bitcoin design.

 

   By having a separate block chain fine tuning the economic parameters of mining is possible on a per-currency basis.

WTF is "mining on per-currency basis"? How is that useful?

Do you want to create a bajillion of Bitcoin copycats?

 
   This means that the that the terms and conventions for mining can be set on a per-currency basis.  It may be possible to achieve similar results with your system.  See my example below[1].
 

 

  Also Bitcoin Addresses are capable of holding only ONE asset type.  This simplifies things immensely.

Well, we can create addresses for colored coins, that was a plan...

But I'd say a concept of addresses is too low-level, people need to use payment protocol which would offer end-to-end security.

 

   Has this option been fully explored?  I realize it has been raised before elsewhere, but have the tradeoffs been considered sufficiently?

 

Frankly, I don't think we really need to consider trade-offs.

I'm working on colored coins.

 

  Not trying to dissuade you from that, just trying to get some questions answered about your technology.  :)

 

You might start working on these alt-chains if you think they are better.

 

  I appreciate the spirit of competition here!

  Equal Rights for All Block Chains!!!!

Market will decide which approach is optimal.

  indeed!

  [1]  so consider the example I raised before, except this time, I will construct a scenario that is actually profitable.

  We decide to create MeanieBucks2.0, it has slightly different terms this time.  We again purchase 10k Krugerrands and put it in our reserve safe deep in the mountains of Switzerland.

    1) 1 MEA may be redeemable for 1 Krugerrand at MeanieBank locations worldwide.

    2) It is suggested to include a fee in your transactions, but not required.

    3) Exactly 10k MEA are created, miners do not create new MEA at any time, the issuance remains constant.

    4) MEA are sold to the public at slightly above market cost for a Krugerrand.  This creates a profit margin X(based on this differential X 10k).  (similar to 'spot price' nomenclature in the Gold Bug world).

  Now we have something that the Digital Gold Currency world has sought after for decades.  A reliable digital gold currency with no central bank at all!  The overhead of transfers are managed by 3rd party miners whose incentive are transaction fees.  The cost to start such a scheme is incredibly low compared to similar schemes in the past(that profit margin over 'spot price' would be historically extremely low providing excellent value to users).  Such an idea raises many legal questions, and certainly the question of the practicality of enforcement.  Most importantly, it is possible, the idea is financially solvent, and it is technically possible and proven.

  Now we also get something else in this scenario: the opportunity to pilot and adopt new innovations to the core Bitcoin technology.  For instance it is the prerogative of the Issuer to incorporate various add-ons such as Zerocoin.  Of course they must manage the problems associated with client compatability, etc. but this is a business decision that is related to the scale of profit and cost for support.  As far as the Open Source world goes- it offers far more opportunity to innovate new features to the core technology.

  thank you so much for your response!

  -bm

[Joshua Josh]

  Just a quick point, I don't think we're being altogether clear here.  If I make an alt-chain with identical technology, then the client is the identical and does not require modification.  Yes it is true that for something like Namecoin, the standard client cannot be used.

  It does suggest something significant:  that Bitcoin technology should be generalized to accommodate this new class of use cases.  This has been raised on the bitcoin forums already(see:BitX).  thus the 'Bitcoin 2.0' vision...

  -bm

[Alex Mizrahi]

The main reason is that this way people will be able to buy colored coins (e.g. shares, bonds, tickets) with Bitcoins in a secure way, via so-called atomic coin swap transactions. This means that p2p exchange is possible.

   This is an interesting value proposition.  It's also possible to use another block chain to create a decentralized exchange for all asset-types.

Bitcoin is the most established of cryptocurrencies. It has "market cap" above $1B USD, daily trading volume above $10M USD, a lot of media attention etc.

Thus we securities to be tradeable for Bitcoin specifically.

If you start a brand new blockchain you have to bootstrap it all from zero. 

Technically speaking, it is possible to do cross-chain trading, but from technical perspective it is way less attractive than atomic coin swapping.

Thus we want colored coins on Bitcoin blockchain.

 

  It seems that working this into the basic technology is 'feature bloat', IMHO.

Why do you think so?

Let's start with the fact that ArmoryX implements p2p exchange in form of p2ptrade component.

It implements fairly simple protocol and it is just 500 lines of Python code. (To be fair, it's less than perfect, but you got the idea.) So:

1. It is optional. You can implement client without p2ptrade support, but I don't see why  you wouldn't want to add it.

2. It doesn't constitute 'feature bloat' because it is fairly simple to implement.

3. I believe that secure exchange is raison d'être of colored coins. Without it colored coins offer very small advantages compared to other technologies. For example, Open Transactions. If you cannot buy shares for Bitcoins atomically, you can as well use centralized Open Transactions server. Why not?

4. I personally believe that decentralized exchange would be a killer both for colored coins and for Bitcoin in general. It is the only reason why I am interested in colored coins, without it I'd rather do something else.

 

Another reason is that Bitcoin offers highest degree of security against double-spends.

   Not sure how that works.  As I mentioned, it is possible to use the computing power from other block chains simultaneously(ipse dixit Satoshi Nakamoto).

Well, I assume that not all of Bitcoin miners will participate in merged mining of alt-chain.

It is already the case, see here;

http://dustcoin.com/mining

Namecoin has 20% of Bitcoin's hashing power

Devcoin has 6%.

 

  I still contend that color coins are going to seriously exacerbate the block chain bloat problem.

Let's deal with this problem when it appears. It is possible to migrate colored coins to different chain.

It won't be 100% disruption-free, but there is fairly simple way to do it in a semi-automatized way.

I believe that blockchain bloat problem can be solved. So either it will be solved in Bitcoin, or we all will convert to a different cryptocurrency which solves it.

 

  Think about it, you are tacking on a new value to bitcoins.  It could be anything.  Suddenly one specific BTC is now worth a million dollars.  Naturally the transactions for that particular BTC(and its sub-denominations) is going to be VERY HIGH,

however the incentives for miners remain the same. 

How so? Miners receive a fee for each transaction.

We don't really know how this will effect the economics of mining.

Nobody understands economics of mining anyway, with or without colored coins.

I'd say that the only possible problem is that Bitcoin won't get as valuable as it could otherwise, but it depends on many factors, so nobody knows for sure.

 

[BlueMeanie]

On Monday, April 22, 2013 12:02:03 PM UTC-4, Alex Mizrahi wrote:

 

  Think about it, you are tacking on a new value to bitcoins.  It could be anything.  Suddenly one specific BTC is now worth a million dollars.  Naturally the transactions for that particular BTC(and its sub-denominations) is going to be VERY HIGH,

however the incentives for miners remain the same. 

How so? Miners receive a fee for each transaction.

   Let's concentrate on this particular point.

   Miners receive a fee for each transaction in terms of BTC, however what you're doing here is tacking on an additional value to that coin.  This changes the likely behavior of those who use this particular coin which you have colored.  It's already been noted that Microtransactions pose problems.  In a BTC context many, if not all color coin transactions will appear as microtransactions.  The transactions will be very small BTC amounts, although the actual value of those coins would be much more(depends on Issuance contract).  While I'm not offering proof here of this problem, I think it certainly warrants consideration.

  -bm

[Cameron Stewart]

Transactions fees aren't based on the amount of BTC transferred. There's no inherent problems with microtransactions it's just its not economical to send a 0.0001btc transaction with a 0.0005btc, if that 0.0001btc has a much higher value this no longer becomes a problem. 

The only problem you could construct is coloured coins don't contribute to the fixed block reward as they aren't effectively devalued like normal bitcoins are in response to the inflation, but that block reward was only really temporary (at least until 2140 :)). All coloured coins are doing is speeding up the transition from block rewards to transaction fees.

>
>   -bm
>
> --
> You received this message because you are subscribed to the Google Groups "Colored bitcoin - BitcoinX" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to bitcoinX+u...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>  
>  

[Alex Mizrahi]

Transactions fees aren't based on the amount of BTC transferred. There's no inherent problems with microtransactions it's just its not economical to send a 0.0001btc transaction with a 0.0005btc, if that 0.0001btc has a much higher value this no longer becomes a problem. 

True.

People were concerned about uses which send 1 satoshi as a signal, and that 1 satoshi isn't supposed to be spent. It isn't the case with colored coins...

The only problem you could construct is coloured coins don't contribute to the fixed block reward as they aren't effectively devalued like normal bitcoins are in response to the inflation, but that block reward was only really temporary (at least until 2140 :)). All coloured coins are doing is speeding up the transition from block rewards to transaction fees.

It is hard to say what will be the net effect:

1. If some AwesomeCorp Inc. will issue their shares in form of colored coins, people will have to buy such shares for Bitcoins, which increases demand for Bitcoins.

2. When Digital Gold Coins will be issued, some people will buy them, but they would have bought Bitcoins instead if Digital Gold Coins did not exist. This reduces demand for Bitcoins.

3. People need to buy Bitcoins on market to issue colored coins, which increases demand for Bitcoins.

4. People who transact in colored coins need to buy Bitcoins to pay fees, which increases demand for Bitcoins.

 

[BlueMeanie]

  This is highly significant.

  What you state here is that your design priorities include price maintenance or appreciation of the BTC currency.

  I, OTOH, see a far off land where all block chains are created equal.  This land is filled with wonder and opportunity.

  There are some key considerations that I don't see being discussed here.  The colored coin concept is irregular at the minimum.  It is unproven.  It has no precedence.  A technology with these characteristics is going to be a very hard sell to people who are serious about building digital currencies.  If I were optimistic, I might say it's ambitious.  Think of how difficult it was to get the public to adopt Bitcoin in the first place, and the technology is based in proven Computer Science concepts.  Unfortunately I don't see that with colored coins.  Thus this is going to be a tough sell to the finance world, and that would be a shame.

  We DO have an option here though to meet the Use Cases being raised here, and that option is Alternative Chains.  It's already happening.  Alex, if your primary need/want here is a distributed exchange mechanism, it's possible to build a very powerful exchange by customizing the block chain.  I would suggest perhaps reevaluating this possibility.

  Regarding Cameron's points- these seem to be mysterious and obscure matters.  Its seems that there is some quality that requires that the transactions scale properly to the transaction fees.  Without this, there will be abuse eg. using the block chains for messaging.  It's somewhat analogous to the credit card industry, payment risk is best managed on a per domain basis eg. pornography sites have much different requirements than a clothing store.  The moral of the story, this abuse is best managed on a per-currency basis and the only way to implement this is in an alt-chain.

  What I see as ideal is generalizing the bitcoin technology and generating standards to make it easy for miners, clients, issuers, etc. to manage, customize and optimize the technology.

  Seems the BTC loyalists don't like this idea because they feel it will cause depreciation in BTC- and it will, but overall the community is far better served.

  Bitcoin 2.0: Where All Blockchains Are Created Equal.

  cheers!

  -bm

 

[BlueMeanie]

  There is also another key case that I didn't mention below- strong Crypto.

  If people want to trade valuable assets internationally, they will likely want to do so with Crypto algorithms that are illegal in the US.  It is impossible(to my knowledge) to incorporate stronger crypto into the existing Bitcoin Block Chain(not to mention potentially illegal).

  here is a great article by J. Orlin Grabbe.

  -bm

[Cameron Stewart]

"The rest of this page is of historic value only. It may be I that the RSA sig played some small part in the eventual relaxation of the US crypto export laws."

I suspect you might start to piss people off by spamming this list with FUD.

--

[BlueMeanie]

  http://en.wikipedia.org/wiki/Export_of_cryptography_in_the_United_States

Current status

As of 2009, non-military cryptography exports from the U.S. are controlled by the Department of Commerce's Bureau of Industry and Security.^[9] Some restrictions still exist, even for mass market products, particularly with regard to export to "rogue states" and terrorist organizations. Militarized encryption equipment, TEMPEST-approved electronics, custom cryptographic software, and even cryptographic consulting services still require an export license

  It's a well accepted fact that some crypto is illegal to use and export in the US.  ASMOF, the Java encryption APIs are built around this fact.

  here's some additional reading.  Also see: Cypherpunks.

 -bm

[Cameron Stewart]

Well...

 - bitcoin uses algorithms that would be considered "strong crypto"

 - export regulations only cover exporting not use in the united states

 - plenty of open source software already uses such algorithms and is widely available

 - it may be illegal to use bitcoin nonetheless in the us due to patents on the ecdsa but thats another story (http://en.wikipedia.org/wiki/ECC_patents)

Anyway that's it from me.

[BlueMeanie]

 so your claim is that this factor is 1) not a risk 2) already being done x) at first you claimed it was fraud('FUD') - I guess youre retracting that claim?

 -bm

[Alex Mizrahi]

value to specific bitcoins is just plain irregular- sorry if I'm being critical here.  You still have a lot of proving and validation to do before this is ready for something like a digital gold currency.

Don't forget that Satoshi did not wait for 'proof' or 'validation', he just released his software and let market to decide.

I bet if you ask economists whether Bitcoin is a good idea they would tell you it is bullshit and can affect world's economy in an unpredictable way.

 

 

   True, but you are introducing new factors that will change the ratio of BTC to Number of Transactions. 

So what? Wider adoption changes this ratio too.

 

It is not difficult to implement a p2p exchange using a modified block chain.

  Seems there is resistance to the idea of introducing new features such as this into the main BTC chain.

What do you mean?

 

   This means that the that the terms and conventions for mining can be set on a per-currency basis.  It may be possible to achieve similar results with your system.  See my example below[1].

Technology to create Bitcoin copycats already exists, so I don't see why it is worth discussing.

 

  Now we have something that the Digital Gold Currency world has sought after for decades. 

I don't see principal difference between this and e-gold.

 

A reliable digital gold currency with no central bank at all!

It is still centralized at issuer.

 

  The overhead of transfers are managed by 3rd party miners whose incentive are transaction fees.  

The cost to start such a scheme is incredibly low compared to similar schemes in the past

You can easily replace those miners with some centralized service, basically a web site.

Users have to trust you to keep gold coins and offer a way to redeem digital coins for physical ones.

[BlueMeanie]

 Hi Alex!

On Monday, April 22, 2013 4:17:27 PM UTC-4, Alex Mizrahi wrote:

 

value to specific bitcoins is just plain irregular- sorry if I'm being critical here.  You still have a lot of proving and validation to do before this is ready for something like a digital gold currency.

Don't forget that Satoshi did not wait for 'proof' or 'validation', he just released his software and let market to decide.

I bet if you ask economists whether Bitcoin is a good idea they would tell you it is bullshit and can affect world's economy in an unpredictable way.
 

 

 

   True, but you are introducing new factors that will change the ratio of BTC to Number of Transactions. 

So what? Wider adoption changes this ratio too.

 

It is not difficult to implement a p2p exchange using a modified block chain.

  Seems there is resistance to the idea of introducing new features such as this into the main BTC chain.

What do you mean?

   Am I correct in believing that Colored Coins requires NO special customizations to the existing bitcoin block chain?
  

 

   This means that the that the terms and conventions for mining can be set on a per-currency basis.  It may be possible to achieve similar results with your system.  See my example below[1].

Technology to create Bitcoin copycats already exists, so I don't see why it is worth discussing.

 

  Now we have something that the Digital Gold Currency world has sought after for decades. 

I don't see principal difference between this and e-gold.

  E-gold requires connection to a specific server(and access to that server) to perform transactions.  There are many many legal, economic, and technological implications there.
 

 

A reliable digital gold currency with no central bank at all!

It is still centralized at issuer.

   Yes you need to have some relationship with the issuer, but that relationship can be very abstract and indirect.
 

 

  The overhead of transfers are managed by 3rd party miners whose incentive are transaction fees.  

The cost to start such a scheme is incredibly low compared to similar schemes in the past

You can easily replace those miners with some centralized service, basically a web site.

Users have to trust you to keep gold coins and offer a way to redeem digital coins for physical ones.

  yes, but the majority of the holders of the currency need never interact with the Issuer at all, this does not hold true for e-Gold and similar currencies.

 -bm
 

[Alex Mizrahi]

  Just a quick point, I don't think we're being altogether clear here.  If I make an alt-chain with identical technology, then the client is the identical and does not require modification.

Well, if you want to rely on merged mining, you have to apply merged mining patch.

 

  This is highly significant.

  What you state here is that your design priorities include price maintenance or appreciation of the BTC currency. 

No. I just said that it might be a nice side effect.

I simply work on software, I do not care that much about consequences...

  I, OTOH, see a far off land where all block chains are created equal.  This land is filled with wonder and opportunity.

From what I know, it is just impossible. They will be equal w.r.t. security only if you force all miners to mine all chains. But that means that instead of solving a problem with bloat you will create a bigger one.

I strongly recommend analyzing how blockchain works and how merged mining works... 

Longest chain wins. Not one which was timestamped...

However, this is an off-topic here, I think, as merged mining is totally unrelated to colored coins.

 

  There are some key considerations that I don't see being discussed here.  The colored coin concept is irregular at the minimum.  It is unproven.  It has no precedence.

Any new technology is "unproven" and "has no precedence".

It is very easy to see why colored coins work, so I don't see any problem here.

Its interactions with Bitcoin in economic plane might be complex, but I don't think that people care much about purely theoretic economic studies. After all, Bitcoin itself is more than weird economically.

 

  A technology with these characteristics is going to be a very hard sell to people who are serious about building digital currencies.

Currently there is no shortage of people who want to use colored coins for digital currencies and what not. There is only a shortage of software.

So I am working on software and I do not care much about business aspects.

 

Bitcoin in the first place, and the technology is based in proven Computer Science concepts. 

Colored coins are based on exactly same concept: ownership of a colored coin is confirmed through a chain of digital signatures originating from the genesis transaction.

 

  We DO have an option here though to meet the Use Cases being raised here, and that option is Alternative Chains.  It's already happening.  Alex, if your primary need/want here is a distributed exchange mechanism, it's possible to build a very powerful exchange by customizing the block chain. 

I have no idea what you're talking about...

To make it significantly different from Bitcoin you'll have to introduce significant changes, and at that point it will no longer follow the security model of Bitcoin.

 

  Regarding Cameron's points- these seem to be mysterious and obscure matters.  Its seems that there is some quality that requires that the transactions scale properly to the transaction fees.

  Without this, there will be abuse eg. using the block chains for messaging.

People already do that. It is one of  Bitcoin's flaws.

 

  What I see as ideal is generalizing the bitcoin technology and generating standards to make it easy for miners, clients, issuers, etc. to manage, customize and optimize the technology.

Well, good luck with that...

[Alex Mizrahi]

   Am I correct in believing that Colored Coins requires NO special customizations to the existing bitcoin block chain?

Yes, that's the point.

You can download ArmoryX software right now and use it to issue colored coins, send them to others, trade them on p2p exchange etc.

Obviously, colored coins already exist, and we did not change anything about Bitcoin...

The only problem is that Armory requires 1+ GB of RAM in addition to running Bitcoin-Qt, and there is a couple of bugs and gotchas here and there.

  yes, but the majority of the holders of the currency need never interact with the Issuer at all, this does not hold true for e-Gold and similar currencies. 

I see, but if that's the problem you can try to solve it without using Bitcoin-style blockchain and mining...

Say, via Open Transactions server and something like Tor.

[BlueMeanie]

 Well, I think I made my point for now.  I think others might also be interested to know what color coins have over alt block chains, so at least we now have some public reading material.  Thanks for spending the time defending your case.

 re. OpenTransactions, the proposal does not offer anything resembling p2p(even when using Chaum e-Cash).  Some of the claims of this project can be quite misleading.  It's a system that attempts to tie together Digital Vouchers, Chaumian E-Cash, Triple-Signed Receipts and some other features.  There is already a competing Digital Voucher system called VoucherSafe.  It's kind of like the author just threw a lot of ingredients together and made a salad.  Also there are some questions as to proprietary technology, as there is a company formed attempting to commercialize it.  Seems we have a wave of amateur developers overstating their contributions these days.

 Your software isn't going to be used for business?  this is a money system right?  what is money used for exactly?  ;)

  -bm

[Jorge Timón]

Hi, BlueMeanie.
I think the only two reasons that make and alternative chain necessary
are these:

1) Bitcoin modifies its protocol to prevent spam and as a side effect
it disables colored coins. I don't think this is likely to happen.

2) We want to support more features than those that can be supported
by colored coins as they are and the needed modifications in the
bitcoin protocol are not accepted on the community.

Here's some features the current colored coins cannot support:

1) Issuance and trade assets that have in-chain interests (positive or
negative).
2) Open binding orders to buy a given color without defining where the
coins must come from.
3) Use the binding orders at 2 to build and broadcast a transitive
transaction (ripple transaction).

As said in another thread, if Freicoin wants to support colored coins
at all, it must allow custom coinbases in its chain, because "freicoin
satoshis" are lost to demurrage and can't be used to persistently
represent assets.

So we're working on a design for Freicoin (starting from the ideas I
had from ripplecoin) and given that we need a hard-fork anyway, we
want to support those additional features. Once implemented and tested
(that's not a priority for us right now), it could be used for bitcoin
if accepted.

We plan to merge mine with bitcoin once Freicoin has enough mining
support on its own.

Here's the previous thread I'm talking about:
https://groups.google.com/d/topic/bitcoinx/otDtyIgHR80/discussion

I don't buy the "bloated chain" argument. And I think colored coins
should continue as they are since many use cases can be supported and
we don't need to wait for complex modifications that require more
design, implementation and testing.
Colored coins in the unmodified bitcoin chain is right now. Supporting
more use cases is for the future.

By the way, chaumian "cash" (you must trust an issuer so doesn't fit
my definition of cash well) is not atomically tradeable for anything,
not even other chaumian cash tokens. So in my opinion is worthless.
Other OT features are better, and the system is reasonably trust-less
though not p2p. And it doesn't even have centralized ripple yet.
I still like to follow the project though, I think they've probably
advanced more in legally binding contracts than any other similar (in
the sense that you can issue arbitrary tokens) system.

> --
> You received this message because you are subscribed to the Google Groups
> "Colored bitcoin - BitcoinX" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to bitcoinX+u...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>


--

Jorge Timón

http://freico.in/

[Cameron Stewart]

On Tue, Apr 23, 2013 at 6:32 PM, Jorge Timón <jti...@monetize.io> wrote:

Hi, BlueMeanie.
I think the only two reasons that make and alternative chain necessary
are these:

1) Bitcoin modifies its protocol to prevent spam and as a side effect
it disables colored coins. I don't think this is likely to happen.

2) We want to support more features than those that can be supported
by colored coins as they are and the needed modifications in the
bitcoin protocol are not accepted on the community.

Here's some features the current colored coins cannot support:

1) Issuance and trade assets that have in-chain interests (positive or
negative).

What's the advantage of this over say exchanging for yearly bonds?

 

2) Open binding orders to buy a given color without defining where the
coins must come from.
3) Use the binding orders at 2 to build and broadcast a transitive
transaction (ripple transaction).

Another advantage is you wouldn't have to have the full chain of transactions to the coinbase to verify the currency greatly simplifying things and allowing pruning of the blockchain 

As said in another thread, if Freicoin wants to support colored coins
at all, it must allow custom coinbases in its chain, because "freicoin
satoshis" are lost to demurrage and can't be used to persistently
represent assets.

So we're working on a design for Freicoin (starting from the ideas I
had from ripplecoin) and given that we need a hard-fork anyway, we
want to support those additional features. Once implemented and tested
(that's not a priority for us right now), it could be used for bitcoin
if accepted.

Probably not really relevant to this list but I don't see the appeal in Freicoin. The demurrage fee just seems an incentive not to use the currency. There doesn't seem any real cost to holding money in the currency as only processing transactions costs money. Further if you did implemented coloured coins or some other system it becomes a benefit to use coloured coins rather than Freicoin's as coloured coins would not get inflated. 

Also you claim that traditional currencies are inherently inflationary but this isn't entirely true because while some of it is kept as cash most people keep a large portion of in banks where you get interest usually according to the cash rate. So you really have to consider the inflation / cash rate to get the real inflation. After you do this not all currencies are inflationary and if they were can could be predicted to do such people would be moving money out of them very quickly. 

Further the reason countries generally have inflation is because the economy is not doing well and the country is using this effective tax on capital to fund public services or alternatively to fund wars not just to fund the money system. Such a tax I think could be implemented in bitcoin anyway by forcing reasonably large businesses to pay a tax on any funds have received where the tax had not been paid. The government could publish a list of txouts that had paid. Therefore txouts which the tax had not been paid would be effectively worth less as any one would know as soon as it hit some complying organisation they would have to pay the tax.

Also "80% of the total will be distributed through the Foundation.", yeah good luck with that...

Having said that I support the idea of integrating coloured coins in a blockchain as I think in the future it might be possible to introduce it in the bitcoin chain via a hard fork.

[Jorge Timón]

On 4/23/13, Cameron Stewart <cam...@shptech.com> wrote:
>> 1) Issuance and trade assets that have in-chain interests (positive or
>> negative).
>>
> What's the advantage of this over say exchanging for yearly bonds?

More use cases. For example, having a gateway (in ripple's
terminology) of a currency with demurrage.
You could owe me a debt with interest and I could re-sell it.
Or just to repay the debt and the system automatically account for
interest, it would make things much easier for p2p lending.

> Another advantage is you wouldn't have to have the full chain of
> transactions to the coinbase to verify the currency greatly simplifying
> things and allowing pruning of the blockchain

Yes, that's another important one I forgot to mention.

> Probably not really relevant to this list but I don't see the appeal in
> Freicoin. The demurrage fee just seems an incentive not to use the
> currency. There doesn't seem any real cost to holding money in the currency
> as only processing transactions costs money. Further if you did implemented
> coloured coins or some other system it becomes a benefit to use coloured
> coins rather than Freicoin's as coloured coins would not get inflated.

If we're going to go deeper, probably freicoin.org is a better place.
But to answer fast.
The demurrage fee is an incentive not to hoard the currency. It does
incentive you to spend it, invest it or loan it when you have it. The
incentive to accept it, like with every other cash money is the
expectation that you will be able to trade it for something else
later.
And Freicoin is not inflationary, it converges at a fixed supply of
100 Millions.

> Also you claim that traditional currencies are inherently inflationary but
> this isn't entirely true because while some of it is kept as cash most
> people keep a large portion of in banks where you get interest usually
> according to the cash rate. So you really have to consider the inflation /
> cash rate to get the real inflation. After you do this not all currencies
> are inflationary and if they were can could be predicted to do such people
> would be moving money out of them very quickly.

I don't usually use the term "traditional currencies", what do you
mean gold or keynesian money?
Also when the money is "kept at interest at the bank", it is loaned to
another person and therefore circulating. That's what banks used to do
anyway.

The portion that really matters for prices rapid moves in prices is
the part that is hoarded, because that's what can change more rapidly
if people "increase hoarding" or decrease it all at the same time.

The core gesellian theory circulates around interest and capital
yields as rents and not around general prices, although cycles in
interest rates are what cause cycles in prices (or what caused them
when we were on gold).

> Also "80% of the total will be distributed through the Foundation.", yeah
> good luck with that...

Thank you, we expect it to be a better issuance method in terms of
adoption than mining.
We were also concerned about the wastefulness of brute-hashing issuance.
That's good for the security of the system, but didn't really had to
be reused for issuance.
It won't be an easy task though and if freicoin had been the first
cryptocoin I guess wasteful but fully p2p issuance would have been
better, as the whole task of proving the concept would have been
enough of a task.

> Having said that I support the idea of integrating coloured coins in a
> blockchain as I think in the future it might be possible to introduce it in
> the bitcoin chain via a hard fork.

Yes, it would be completely possible. We try to stay close the
bitcoin's code for things that are unrelated to demurrage. And we
would commit this changes so that bitcoin could take them easily.
Feel free to make any suggestions you want about our design draft at
the other thread.

[BlueMeanie]

  Just caught this...

color leaking: when a sum of colored inputs is matched to sum of color outputs which is lesser. in that case we will assumed a color leaking has occurred and the total number of colored bitcoin in circulation has decreased by the diff of the two sums

 Why not create the convention that any colored coins 'leaked' in this way, go as a transaction fee to the miner?  How transaction fees work exactly for Colored Coins?

 Does this break anything?  This somewhat addresses the issue, where you can create color specific awards to the miners.

 -bm

[Jorge Timón]

On 4/23/13, BlueMeanie <josh.j...@googlemail.com> wrote:
>

> Just caught this...
>
> color leaking: when a sum of colored inputs is matched to sum of color
> outputs which is lesser. in that case we will assumed a color leaking has
> occurred and the total number of colored bitcoin in circulation has
> decreased by the diff of the two sums
>
> Why not create the convention that any colored coins 'leaked' in this way,
>
> go as a transaction fee to the miner? How transaction fees work exactly
> for Colored Coins?
>
> Does this break anything? This somewhat addresses the issue, where you
> can create color specific awards to the miners.

Currently there's only BTC fees, but there's nothing impeding a rule
that makes some situations be interpreted as colored fees. In fact I
think it is the way to go.
We allow colored fees on freicoinx but didn't thought that they could
work similarly for regular colored coins.

Some miners will ignore transactions paying fees in colors that they
can't or don't want to see, but highly liquid assets shouldn't present
much problem to be accepted by miners. Maybe even sell the colored
fees for BTC in the same block, although not likely without binding
orders.

[Alex Mizrahi]

  Just caught this...

color leaking: when a sum of colored inputs is matched to sum of color outputs which is lesser. in that case we will assumed a color leaking has occurred and the total number of colored bitcoin in circulation has decreased by the diff of the two sums

 Why not create the convention that any colored coins 'leaked' in this way, go as a transaction fee to the miner?

Basically, colored coin protocol exists only between parties interested in colored coins.

Miner will simply see them as regular Bitcoins, so he will get fee in regular Bitcoins.

Note that color-aware wallet will prevent leaking and mixing, so such things happen only if there is a bug or  somebody creates misconstructed transaction intentionally.

Loss of colored coins is actually the right thing in such case: if it was an accident, issuer can compensate coins to the last legitimate owner... But it is up to issuer, of course.

 

  How transaction fees work exactly for Colored Coins?

Both colored coins and uncolored Bitcoin are included in same transaction, and fee is paid using uncolored Bitcoins.

Alternatively, you can use p2ptrade protocol to pay fee using colored coins via exchanger.

Basically you pay exchanger with colored coins, and exchanger pays fee with uncolored coins.

 

 Does this break anything?

No.

Design is robust: if there is a problem, it is isolated.

 

  This somewhat addresses the issue, where you can create color specific awards to the miners.

Well, Bitcoin is supposed to be THE currency to pay tx fees with... It just makes sense.

If you do not want to hold any Bitcoins a construction with 3rd party exchanger solves the issue.

Any other approach will be messy and hard to implement.

[BlueMeanie]

On Tuesday, April 23, 2013 4:52:08 PM UTC-4, Alex Mizrahi wrote:

 

  Just caught this...

color leaking: when a sum of colored inputs is matched to sum of color outputs which is lesser. in that case we will assumed a color leaking has occurred and the total number of colored bitcoin in circulation has decreased by the diff of the two sums

 Why not create the convention that any colored coins 'leaked' in this way, go as a transaction fee to the miner?

Basically, colored coin protocol exists only between parties interested in colored coins.

Miner will simply see them as regular Bitcoins, so he will get fee in regular Bitcoins.

Note that color-aware wallet will prevent leaking and mixing, so such things happen only if there is a bug or  somebody creates misconstructed transaction intentionally.

  Will using a regular wallet with colored coins have some kind of destructive effect?  Will I lose the colors?  Is using a regular wallet with an address that contains colored coins a bad idea?
 

Loss of colored coins is actually the right thing in such case: if it was an accident, issuer can compensate coins to the last legitimate owner... But it is up to issuer, of course.

  Is it possible to define a way to offer transaction fees that are colored?  eg. red input 1 is greater than red output 1, leaving the difference as a RED transaction fee.
 

 

  How transaction fees work exactly for Colored Coins?

Both colored coins and uncolored Bitcoin are included in same transaction, and fee is paid using uncolored Bitcoins.

Alternatively, you can use p2ptrade protocol to pay fee using colored coins via exchanger.

Basically you pay exchanger with colored coins, and exchanger pays fee with uncolored coins.

 

 Does this break anything?

No.

Design is robust: if there is a problem, it is isolated.

 

  This somewhat addresses the issue, where you can create color specific awards to the miners.

Well, Bitcoin is supposed to be THE currency to pay tx fees with... It just makes sense.

   This does make sense to make the basic transaction processing to be paid in BTC.  This does keep the core block chain running and preserves BTC economy.

  -bm

 

[BlueMeanie]

On Tuesday, April 23, 2013 4:28:37 PM UTC-4, Jorge Timón wrote:

On 4/23/13, BlueMeanie <josh.j...@googlemail.com> wrote:
>
>   Just caught this...
>
> color leaking: when a sum of colored inputs is matched to sum of color
> outputs which is lesser. in that case we will assumed a color leaking has
> occurred and the total number of colored bitcoin in circulation has
> decreased by the diff of the two sums
>
>  Why not create the convention that any colored coins 'leaked' in this way,
>
> go as a transaction fee to the miner?  How transaction fees work exactly
> for Colored Coins?
>
>  Does this break anything?  This somewhat addresses the issue, where you
> can create color specific awards to the miners.

Currently there's only BTC fees,

 

   there is an important usability consideration here.

   Users of a colored currency might be required to obtain BTC in order to make transactions.  Should this be a limitation?  Should we allow users to offer a colored transaction fee?

   the net effect here is also increased demand for BTC as it becomes the gold standard for transaction fees.  Seems this would be favorable in Alex's view.

  -bm
 

[Alex Mizrahi]

  Will using a regular wallet with colored coins have some kind of destructive effect?  Will I lose the colors?  Is using a regular wallet with an address that contains colored coins a bad idea?

Yes, regular wallet does not recognize colors of coins, so it can send those colored coins when you want to pay with uncolored, it can mix coins together etc.

We consider making special colored addresses to avoid confusion.

But it would be better to develop invoicing protocol which would eliminate both this and other problems.

 

  Is it possible to define a way to offer transaction fees that are colored?  eg. red input 1 is greater than red output 1, leaving the difference as a RED transaction fee.

Yes. But that would make things much more complicated and I see no benefits.

[Alex Mizrahi]

   there is an important usability consideration here.

   Users of a colored currency might be required to obtain BTC in order to make transactions. 

Should this be a limitation?  Should we allow users to offer a colored transaction fee?

I already mentioned it is possible to pay fees with colored coins: via p2ptrade protocol they will be auto-exchanged as needed.

So if user has no  uncolored Bitcoins wallet will ask whether he agrees to pay fee with colored coins.

Thus it is not a usability problem if such auto-exchange is implemented in wallet software.

[Jorge Timón]

@Alex

I don't really see the complications.
The worse that can happen, is that a miner gets some colored coins
fees thinking they're uncolored. So he will treat them like that and
eventually uncolor them by "accident".
What's the big deal?

@BlueMeanie

Even if we enable colored fees, BTC would still have a privileged
position for fees because one can assume that bitcoin miners will
always accept btc as fee.
I don't see anything wrong with that though. After all, btc is the
only cash in the system. Since they have an issuer, all colored coins
are credit.
It is similar to XRP in Ripple, but there you cannot pay fees with
other currencies because there's no miners to receive them. Remember
xrp fees are destroyed in Rippple. There you can't have colored fees,
period.

On 4/24/13, Alex Mizrahi <alex.m...@gmail.com> wrote:
>> there is an important usability consideration here.
>>

>> *Users of a colored currency might be required to obtain BTC in order
>> to make transactions.*

>>
>
> Should this be a limitation? Should we allow users to offer a colored
>> transaction fee?
>>
>
> I already mentioned it is possible to pay fees with colored coins: via
> p2ptrade protocol they will be auto-exchanged as needed.
>
> So if user has no uncolored Bitcoins wallet will ask whether he agrees to
> pay fee with colored coins.
>
> Thus it is not a usability problem if such auto-exchange is implemented in
> wallet software.
>

> --
> You received this message because you are subscribed to the Google Groups
> "Colored bitcoin - BitcoinX" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to bitcoinX+u...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>

[Alex Mizrahi]

I don't really see the complications.

It would require a different coloring scheme. 

We currently only need to consider a chain of transactions.

But if we add this "colored fee" thing, we also need to consider coinbases, and implement a special coloring rules for coinbases.

Perhaps it is easier to switch to satoshi-level tracking at that point and reimplement all coloring algos.

 

The worse that can happen, is that a miner gets some colored coins
fees thinking they're uncolored. So he will treat them like that and
eventually uncolor them by "accident".
What's the big deal?

It isn't a big deal, it just happens to be useless unless all miners recognize colored fees.

On the other hand, there are much better ways to handle colored fees:

1.  Pay a small amount of colored coins to output  OP_TRUE script. Anybody can spend it, but usually that would be miner who mines next block and can recognize OP_TRUE script. Thus miner can construct transaction which pays colored coins to himself and include both into block he mines.

2. Once miners will be smart enough to compute fees hierarchically (i.e. total amount of fees in a set of transactions is considered when it is added to the block), anybody who recognizes value of colored coins will be able to pay a fee. E.g. suppose transaction pays $0.10 worth of USDcoins to OP_TRUE. An agent who believes that USDcoins have that value will create a transaction which spends those USDcoins to his address, and he will include a fee which pays for both transactions. Say, that would be 0.0005 BTC. Now miner will include both transactions to grab 0.0005 BTC fee. And agent pockets a difference between value of $0.10 USDcoins and 0.0005 BTC. (However, he needs to take into account that he needs to pay again to spend this small output.)

3. I already  mentioned p2ptrade which tries to find agent who will pay the fee interactively.

Important difference is that you do not depend on a particular miner to recognize value of colored coins (and you do not force him to accept your colored coins): ANYBODY who recognizes their value can assist you.

Practically it means that you do not need to wait 10 blocks until you find a miner who recognizes your fee, if there is a liquid market, quite likely your coins will get into the very next block.

So I believe it is better to keep coloring scheme simple, and implement "paying fees with colored coins"  in separate layers.

[Alex Mizrahi]

(However, he needs to take into account that he needs to pay again to spend this small output.)

Correction: He can combine input with OP_TRUE scriptPubKey with input with his existing funds in one transaction. Since OP_TRUE input is small than a regular one, there is little overhead.

Additionally it is possible to claim many such inputs in one transaction.

And there is no problem with growth of UTXO set.

[Jorge Timón]

On 4/24/13, Alex Mizrahi <alex.m...@gmail.com> wrote:

> It isn't a big deal, it just happens to be useless unless all miners
> recognize colored fees.

You don't need all miners to care about colored fees. Miners that
don't just won't include your transactions, but you may be happy to
wait a little bit more if you know there's enough colored aware miners
that care.

> On the other hand, there are much better ways to handle colored fees:
>
> 1. Pay a small amount of colored coins to output OP_TRUE script. Anybody
> can spend it, but usually that would be miner who mines next block and can
> recognize OP_TRUE script. Thus miner can construct transaction which pays
> colored coins to himself and include both into block he mines.

Certainly this is simpler and better for colored fees than changing
the coloring rules. I hadn't thought about this.
I'm all for this implementation of colored coins, it seems this use
case doesn't need any modifications to the current rules.

Thanks for explaining it.

[Jorge Timón]

Correction: ...I'm all for this implementation of colored FEES...

[BlueMeanie]

On Wednesday, April 24, 2013 2:38:49 AM UTC-4, Alex Mizrahi wrote:

  Will using a regular wallet with colored coins have some kind of destructive effect?  Will I lose the colors?  Is using a regular wallet with an address that contains colored coins a bad idea?

Yes, regular wallet does not recognize colors of coins, so it can send those colored coins when you want to pay with uncolored, it can mix coins together etc.
 

We consider making special colored addresses to avoid confusion.

But it would be better to develop invoicing protocol which would eliminate both this and other problems.

  This is a salient characteristic.

  There is an additional onus on the owner of the assets that they use the correct Bitcoin client with their account.

  While certainly not a showstopper, it is by all estimation a negative.  Without color support in the client, they may lose wealth in their accounts without any warning from the UI.

  -bm