[Crowdfund] User-issued assets, p2p exchange, off-chain transactions, and more

[maaku]

Yes, I'm in favor of replace-by-fee. I think we mean something different by binding, however. In this case we mean that the orders are pre-signed, even though the counter-party is not known at the time the offer is created: the signature on the offer is all that's required to get the final matched exchange transaction on the chain. You could, for example, sign an offer, broadcast it, and then disconnect from the network and never come back. The offer remains in force until some other condition invalidates it, such as a double-spend, passing the nExpireTime, scriptValid ceasing to validate, etc.

The important distinction is not whether you can back out or what steps are required, but the fact that the participants need not be online at the same time. This is in contrast to some other proposals, which clear the market by generating regular bitcoin transactions, which then must be signed by each of the participants before being included in a block*. In these proposals the offer isn't really a binding offer, because the participation of both counter-parties is required up to the very last moment. Having offers be binding (sufficient for inclusion) allows the protocol to operate over a decentralized, p2p network with lossy message transmission/retention and occasionally-connected nodes.

* It is possible to do some limited pre-signing with SIGHASH_SINGLE, but sighash modes really are not general enough to handle all the use cases we desire. It was through investigating the problems with SIGHASH_SINGLE that we were eventually lead down the path of adding isolated sub-transactions.

[killerstorm]

The foundation was NOT created to fund Freicoin development but because although proof of work is a great security solution, we think is a wasteful issuance mechanism.

This is funny... Suppose Freicoin is wildly successful, and total monetary base is equivalent to 10 trillion 2013 US dollars.

Then miners get 500 billlion USD worth of Freicoins each year, and, due to the way mining economics works, they'll burn equivalent amount of energy (and pay for amortization of equipment). This is wasteful on epic scales.

On the other hand, Freicoins which Foundation gets now are worth pennies...

Do you see what I'm talking about? IMO it would make much more sense to give all created money to miners, but send 80% of demurrage money to Foundation.

This way you could redirect money miners are going to spend on electricity to charity... Potentially that might be a large sum each year.

But you did exactly the opposite...

Well, Freicoin economics in general looks completely alien to me. I guess it works in mysterious ways... Good luck with that.

Isn't private donations the funding model for colored coins as well?

Yes, but nobody have offered anything on scale of $100k so far... If they did, we could actually hire people and get it done. Instead, we have people who work in their free time... As you can guess, progress is kinda slow.

However, colored coins require much less work than Freimarkets.

[killerstorm]

Yes, I'm in favor of replace-by-fee. I think we mean something different by binding, however. In this case we mean that the orders are pre-signed, even though the counter-party is not known at the time the offer is created: the signature on the offer is all that's required to get the final matched exchange transaction on the chain. You could, for example, sign an offer, broadcast it, and then disconnect from the network and never come back. The offer remains in force until some other condition invalidates it, such as a double-spend, passing the nExpireTime, scriptValid ceasing to validate, etc.

The important distinction is not whether you can back out or what steps are required, but the fact that the participants need not be online at the same time. This is in contrast to some other proposals, which clear the market by generating regular bitcoin transactions, which then must be signed by each of the participants before being included in a block*. In these proposals the offer isn't really a binding offer, because the participation of both counter-parties is required up to the very last moment. Having offers be binding (sufficient for inclusion) allows the protocol to operate over a decentralized, p2p network with lossy message transmission/retention and occasionally-connected nodes.

* It is possible to do some limited pre-signing with SIGHASH_SINGLE, but sighash modes really are not general enough to handle all the use cases we desire. It was through investigating the problems with SIGHASH_SINGLE that we were eventually lead down the path of adding isolated sub-transactions.

OK, this makes sense.

But I don't think that "binding" is the right word here. Perhaps you could describe such orders as 'autonomously executed' as they can be executed even if one who submitted them is offline. But "binding" implies that somebody is forced to stick to his order.

[killerstorm]

killerstorm, if you're willing to invest in return for 6% of your investment amount as freicoins (current market prices), would you not also be willing to donate 94% of that original amount, and buy up your own coins on the exchange of your choice?

I guess you misunderstood me, it's not about buying Freicoins below market price, it is about bringing market price up to that level... One person cannot do that, but you bring investors' attention to it and explain how Freimarkets can make Freicoin much more useful than an average alt-coin, investors will be eager to invest into the next big thing...

Jorge mentioned that alt-coins can simply copy Freimarkets, but it's not so simple... It really matters which coin is endorsed by developers, as that one will, likely, get all the necessary maintenance, latest features, security fixes, etc.

Technically speaking, it could work without funding from Foundation, but we get a prisoners' dilemma-like situation. For each individual, it isn't rational to donate anything, even though if everybody donates a bit all would benefit from it. So you gotta depend on altruism... Or use a mechanism like Foundation, which you cannot use.

[maaku]

But I don't think that "binding" is the right word here. Perhaps you could describe such orders as 'autonomously executed' as they can be executed even if one who submitted them is offline. But "binding" implies that somebody is forced to stick to his order.

Hrm. "Autonomously executed" is a mouthful, but I see your point. I'll see if I can find a better way to phrase it...

killerstorm, if you're willing to invest in return for 6% of your investment amount as freicoins (current market prices), would you not also be willing to donate 94% of that original amount, and buy up your own coins on the exchange of your choice?

I guess you misunderstood me, it's not about buying Freicoins below market price, it is about bringing market price up to that level... One person cannot do that, but you bring investors' attention to it and explain how Freimarkets can make Freicoin much more useful than an average alt-coin, investors will be eager to invest into the next big thing...

Jorge mentioned that alt-coins can simply copy Freimarkets, but it's not so simple... It really matters which coin is endorsed by developers, as that one will, likely, get all the necessary maintenance, latest features, security fixes, etc.

Technically speaking, it could work without funding from Foundation, but we get a prisoners' dilemma-like situation. For each individual, it isn't rational to donate anything, even though if everybody donates a bit all would benefit from it. So you gotta depend on altruism... Or use a mechanism like Foundation, which you cannot use.

Our hope with this crowd-fund is that through public discussion of the proposal we could educate people on why it's valuable, and get people excited about seeing it implemented. User assets, distributed p2p exchange, off-chain transactions... these are all things that people on this forum are clamoring for, and this proposal delivers them all simultaneously, as well as other more esoteric features like ripple-esque credit lines, smart property tools, and AML/KYC compliance. Seeing that Freicoin developers are serious about making this happen would drive investment in freicoins, and getting donors to provide the financial resources to make this happen would protect that investment. But as you say, it is a form of the prisoner's delima, resolution of which will hopefully come if/when we get the momentum which follows our first big donor...

Then miners get 500 billlion USD worth of Freicoins each year, and, due to the way mining economics works, they'll burn equivalent amount of energy (and pay for amortization of equipment). This is wasteful on epic scales.

On the other hand, Freicoins which Foundation gets now are worth pennies...

We have ideas for how to distribute demurrage other than to the miners, if/when that is ever a problem. My personal idea is a proof-of-stake budgetary voting system I've called 'republicoin', which is designed to resemble a combination of participatory budgeting processes and parliamentarian democracy. But that's an issue we can address if/when it ever becomes a problem.

The initial distribution is a separate issue. It's given us significant street cred outside of the bitcoin community that we've setup a non-profit to handle most of the initial distribution. It's opened some doors for collaboration as well, although we have yet to see how that pans out. In either case, giving 20% vs 100% to the miners would have resulted in the same amount of burnt cycles and excess heat in either case during these first few years, so at least this way some charitable good comes of it.

[maaku]

I've updated the crowdfund goals in the OP to be denominated in fiat, since ultimately that is what we require (living expenses, etc.) and the wildly fluctuating price of bitcoin makes a bitcoin-denominated goal a moving target.

We are seeking $125k for the development of Freimarkets (including salaries of full-time developers, operational costs, and conference travel expenses), and an additional $50k for the development of private accounting services.

So far we have received about $4.5k in donations, which is enough to get started on the first components, but on its own not enough to make Freimarkets a priority over other our other projects we are currently working on.

[FrictionlessCoin]

reserved

[markm]

It is true that the kinds of stuff you are hoping to develop would be nice to have, but since well-funded teams are already working on such stuff does it realyl make sense to blow antoher huge wad of money on funding yet another team to diverge from all the already funded projects that are already working on what to and end user presumably is much the same thing?

If the already funded teams fail somehow then maybe looking around for a different team might make sense, and if the specific implementation-details of how to build such a thing fails then again considering funding yet another different way of accomplishing the same goals might make sense.

Maybe if I had invested in one or more of the others and it was truning out to be massively lucrative i might consider investing sme of the loot I gained from the success of bitshares or Mastercoin or whatever into yet another similar project but as far as I can tell at that point I would already have these functionalities.

Meanwhile if it is really the case that this team needs money, maybe we can hope that they will take a job with one of the already-funded teams whereas if we throw money at them now that probably just makes them even less available to work on the projects we the people or we the community or whatever have already thrown huge amounts of money at toward the same goals...

In a way it seems not all that different from the "yet another altcoin" craze. How many more alternate distributed exchanges are we supposed to fund all at once fragmenting the resulting markets, or is some way of having them all work together on globally integrated orderbooks also part of the plan?

-MarkM-

[jtimon]

I've been thinking about this long before Ripple.com even existed.
Ripple.com itself has several important design flaws, but much of the p2p exchange noise has only come after them.
So I don't feel we're diverting resources as you @markm suggest. Maybe we should have tried to get funded before having a design we were fully comfortable with and should have rushed to promise people huge returns like the rest did, instead of asking for donations to implement free software anybody can benefit from, donors or not. But I believe the best design will prevail independently of the funding model.

Mastercoin (in its current form) is even less scalable than regular colored coins.
Bitshares document looked like a random mix of the latest fuzzy altcoin proposals coupled with a great deal of ideology and economic dogmatism. They got funded and then they create protoshares as another funding mechanism, also spreading myths like "anti-ASIC pow algorithms can exist and are better than SHA256".
Honestly, I'm not following that very closely because it's hard for me to take it very seriously, despite the funding they can have.
Mastercoin has also a great deal of funding, but as said their current design is not scalable.

Our design is there for any of those projects to implement if they want, we're happy to clarify and discuss any detail of the design.  
If they fund this campaign they will also be able to use the resulting code on their own projects, but that's for them to decide since is their funding.
I don't see how our design can obstruct these other projects in any way.
We believe that everything our design offers is necessary: off-chain assets, ripple-like transactions, interest-bearing assets, indivisible tokens, options...
And we also believe our design is the best for the task.
That's why we want to implement this design and not another one. If other projects are interested in helping us, that would be great, but we don't chose that part.

[maaku]

So the main contendors I've seen are Ripple, Mastercoin, BitShares, and the general category of non-forking coin coloring schemes. None of these provide the complete feature set of Freimarkets, even in combination.

Ripple is not as decentralized and trust-free as OpenCoin / Ripple Labs would have you believe. It also suffers from technical and arbitrary implementation limitations which prevent it from being used in higher level applications outside of the forex and payment applications its creators are pushing. It is hindered by using a ledger, a heavyweight account system, and a host currency which is awkwardly injected in such a way as to making using the system expensive, a phenominon I call "awkward monetization."

Mastercoin is a completely unscalable solution. It couldn't even meet bitcoin's current level of activity, let alone a future where a million user-ussed assets are traded daily. Its problems are fundamental, and unfixable without completely reworking the design from the ground up. It has many, many other issues ranging from trivially easy to pull off of double-spends to complicated self-balancing mechanisms which have not been peer reviewed and not proven to work as advertised. But the scaling issues are an unfixable non-starter.

BitShares suffers from numerous technical problems which artificially limit scalability, functionality, and security. In too many places it hops on the "me too!" alt coin bandwagon without adequatly considering the associated costs of these poorly thought out "improvements." It's proof-of-work system is broken. It's order book system is bloated and inflexible. Assets are very heavyweight and a limited number per-chain. When I last looked at it I was unable to figure out how to do transitive transactions.

Colored coins are the only alternative who's proponents have spent considerable time working out things like SPV modes of operation, a critical aspect of scalability. The first Freimarkets prototype designs were in fact colored coin proposals as we tried to find a way to implement distributed ripple protocol over vanila bitcoin. However we quickly discovered that this was/is an impossible task, and found ourselves forced to implement new validation rules to get that and other funcationality and security we desired.

Colored coins using ordinary non-forking bitcoin suffers from some serious limitations. The scripting language and transaction format is not powerful enough to do things like binding offers or granular redemption. These limitations plus the lack of new opcodes makes it difficult if not impossible to do higher level applications like auctions, cross-server trade, or derivatives/options. (Since the validation rules of bitcoin are not extended to be color aware, you can't be sure that an offer will be matched with the color you actually desire. Nodes have to stay online to participate and can drop out at any time, so now you start worrying about DoS costs and reputation systems...)

Now there are also things which Freimarkets does which no other proposal here does, even inadequately. We provide generalized mechanisms for KYC compliance & authority delegation. We provide an integrated off-chain solution and tooling for mutual coordination of multiple private servers. We provide a mechanism for binding partial transactions which allow participants to construct and sign just the portions of a transaction that concern them. And we do all of this in a carefully designed way that enables access by lightweight nodes and the same scalability properties as bitcoin.

So no, I don't think of this as "yet another altcoin" or something substitutable by any of the above mentioned proposals. I have been approached with employment offers by both BitShares and the Mastercoin foundation and declined, because what Jorge and I are doing here can't be replicated in those frameworks without starting over from scratch. Jorge and I both firmly believe that in the long term people will use the framework judged to be the best on technical grounds, even if it is not the first developed. Scaling issues, security failures, and limited functionality will drive people to the more advanced protocols, in time.

Am I frustrated by the lack of interest? Yes, of course. But the things that are really worth doing are rarely recognized in that way at first.

[FrictionlessCoin]

A big fan of your previous work (i.e. Freicoin).

What specifically is in the Mastercoin specification that makes it unscalable?

I agree that the Bitshares specifications are an incoherent mess.

Regarding Colored Coins,  why can't arbitrary data be placed in the scrypt block instead of refactoring the transaction format?

[killerstorm]

What specifically is in the Mastercoin specification that makes it unscalable?

You need to scan almost the whole blockchain to validate Mastercoin transactions. Obviously, thin clients can't do that...

[FrictionlessCoin]

What specifically is in the Mastercoin specification that makes it unscalable?

You need to scan almost the whole blockchain to validate Mastercoin transactions. Obviously, thin clients can't do that...

Are you serious?  I know that miners can't check the double spending that happens with Mastercoin, so that responsibility goes with the receiver of the coins?

With color coins, you have to traverse the address chain to see if it ends up at the colored coin genesis transaction.

With mastercoin,  which transactions does a client look at to verify this?  Is there a transaction chain anywhere?

With the Freimarkets proposal,  is there a need to traverse the the chain to the origin transaction?

[killerstorm]

Are you serious?  I know that miners can't check the double spending that happens with Mastercoin, so that responsibility goes with the receiver of the coins?

Yes. Isn't it obvious?

With color coins, you have to traverse the address chain to see if it ends up at the colored coin genesis transaction.

Yes, one needs to scan transaction history for a particular color. Of course, it is possible that he needs only a fraction of the whole history, but in the worst case he needs all of it...

Which is why in the long term we either needs Freimarkets, or some alternative solution, possibly SCIP.

With mastercoin,  which transactions does a client look at to verify this?  Is there a transaction chain anywhere?

With Mastercoin, one needs a list of all Mastercoin transactions up to the transaction in question. The only way to obtain it is to scan blockchain from the first block which has transaction sent to exodus address to the block which includes the transaction in question.

Some people think that it's possible to obtain a list of Mastercoin transactions without scanning the blockchain, but, actually, it is impossible to make sure that it is the correct list: attacker can simple withhold information about some transactions to perform a double-spend. There is no way to check whether list is complete.

(Well, aside from magic SCIP.)

With the Freimarkets proposal,  is there a need to traverse the the chain to the origin transaction?

As far as I know, SPV clients do not need to check transaction history.

[FrictionlessCoin]

SCIP ?

[killerstorm]

SCIP ?

http://www.scipr-lab.org/

[maaku]

With color coins, you have to traverse the address chain to see if it ends up at the colored coin genesis transaction.

For SPV proofs you need the history tracing colored coins back to their definition transaction. You do not need to store the history of every input back to genesis, so there isn't a combinatorial explosion of history, but it's still a large concern. When you first create a colored coin definition, proofs are problably about a kilobyte or two in size. Each time they change hands the proof grows by the size of the transaciton plus a few hundred bytes of metadata. So coins can change hands a few dozen or maybe up to a hundred times before the proofs start to get unweildy, although one's tolerance probably depends on many factors.

You could have an online issuer that every so often creates new colored coin definitions pegged 1:1 with the old coins, and operates an exchange to trade in old heavyweight coins for fresh new ones.  This works, barely. It's very much non-ideal, but it's also as far as I know the very best you can do without forking bitcoin. Tons of credit goes to Alex Mizrahi, Meni Rosenfeld, and others who have pushed this concept as far as it can go.

With Mastercoin, one needs a list of all Mastercoin transactions up to the transaction in question. The only way to obtain it is to scan blockchain from the first block which has transaction sent to exodus address to the block which includes the transaction in question.

Some people think that it's possible to obtain a list of Mastercoin transactions without scanning the blockchain, but, actually, it is impossible to make sure that it is the correct list: attacker can simple withhold information about some transactions to perform a double-spend. There is no way to check whether list is complete.

(Well, aside from magic SCIP.)

Alas, there is no real magic in this world. SNARK systems hold out the promise for a compact proof that doesn't require the entire block chain to validate. But the downside is that creating SNARK proofs is a very, very computaitonally intensive process. I think it's an open question in what way SNARKs will end up being useful in the bitcoin ecosystem, for this reason.

With the Freimarkets proposal,  is there a need to traverse the the chain to the origin transaction?

No, by incorporating coloring into the validation rules, Freimarkets is the same as bitcoin in this regard. If a transaction has made it into the chain, you can assume it and its attached coloring is valid with SPV level of security. If/when commitments of the utxo state are added to the coinbase, you could use those indices to check validity and coloring as well. So proof size remains small and relatively constant over time.

[FrictionlessCoin]

With color coins, you have to traverse the address chain to see if it ends up at the colored coin genesis transaction.

For SPV proofs you need the history tracing colored coins back to their definition transaction. You do not need to store the history of every input back to genesis, so there isn't a combinatorial explosion of history, but it's still a large concern. When you first create a colored coin definition, proofs are problably about a kilobyte or two in size. Each time they change hands the proof grows by the size of the transaciton plus a few hundred bytes of metadata. So coins can change hands a few dozen or maybe up to a hundred times before the proofs start to get unweildy, although one's tolerance probably depends on many factors.

You could have an online issuer that every so often creates new colored coin definitions pegged 1:1 with the old coins, and operates an exchange to trade in old heavyweight coins for fresh new ones.  This works, barely. It's very much non-ideal, but it's also as far as I know the very best you can do without forking bitcoin. Tons of credit goes to Alex Mizrahi, Meni Rosenfeld, and others who have pushed this concept as far as it can go.

With Mastercoin, one needs a list of all Mastercoin transactions up to the transaction in question. The only way to obtain it is to scan blockchain from the first block which has transaction sent to exodus address to the block which includes the transaction in question.

Some people think that it's possible to obtain a list of Mastercoin transactions without scanning the blockchain, but, actually, it is impossible to make sure that it is the correct list: attacker can simple withhold information about some transactions to perform a double-spend. There is no way to check whether list is complete.

(Well, aside from magic SCIP.)

Alas, there is no real magic in this world. SNARK systems hold out the promise for a compact proof that doesn't require the entire block chain to validate. But the downside is that creating SNARK proofs is a very, very computaitonally intensive process. I think it's an open question in what way SNARKs will end up being useful in the bitcoin ecosystem, for this reason.

With the Freimarkets proposal,  is there a need to traverse the the chain to the origin transaction?

No, by incorporating coloring into the validation rules, Freimarkets is the same as bitcoin in this regard. If a transaction has made it into the chain, you can assume it and its attached coloring is valid with SPV level of security. If/when commitments of the utxo state are added to the coinbase, you could use those indices to check validity and coloring as well. So proof size remains small and relatively constant over time.

Curious,  what would be the minimum amount of changes to Bitcion (based on the Freimarkets proposal) to implement the functionality of Colored Coins?

[maaku]

Colored coins doesn't require any changes to bitcoin. That us rather the point.

[FrictionlessCoin]

Colored coins doesn't require any changes to bitcoin. That us rather the point.

What I meant was, what would be needed to incorporated into the bitcoin rules to ensure that colored coins are valid without the need to traverse back to the root transaction.