[color=#ff0000]UPDATE: [/color]I’ve recently taken a formal role on the Counterparty team, so it seems unlikely that I’ll have time to develop DEXR myself. To make sure that this project gets built I’m putting together a team to share the risks and rewards. If you’re interested in being a part of this project, please PM me.
[hr]
Over the past few weeks, I’ve been giving some thought to the challenge of reducing counterparty / issuer risk within the distributed exchange, and the outcome of all of this thinking is an idea for a new service. I ran the concept by some key members of the community (including the founders) and now I’d like to share the idea with all of you, and hopefully get your feedback.
A bit of context -
In a few years we’ll hopefully have a large and open marketplace with hundreds of issuers and thousands of assets being traded within the distributed exchange. This is likely to include all manner of cryptocurrencies, commodities, bonds, dividend paying stocks, and alternative investments like fine wine and art.
Say you want to buy some Litecoin via the distributed exchange; what you’ll end up holding are asset tokens that are essentially Litecoin IOUs. You’re hoping that the issuer will make good on these IOUs if / when you come to redeem them. With each asset token you’re holding, you’re taking on some issuer risk, because the person holding the property that backs these tokens might default, go rogue or be a charlatan.
In the future, there might be a dozen different issuers for any given asset, so how do you choose between them?
I think that what the Counterparty ecosystem will need in the future is a quick way of understanding which issuers are the most reputable, otherwise, using the DEX might end up as a bit of a minefield and tricky for new users to navigate.
I’m proposing a website (called DEXR.CO, meaning “the Distributed Exchange Ratings Company”) which will contain rich profiles for issuers (with detailed descriptions of the assets they offer) plus, a publicly available “trust score” for each asset that indicates the current level of trust that the community has that the issuer will meet their obligations going forward.
I want users to be able to visit the site, filter the results for say “Litecoin” or “Gold” and then see a list of issuers sorted by reputation score.
The idea is to build a kind of Standard and Poor’s of distributed finance.
It’ll be a free service that anyone can sign-up for, and it’ll enable anyone to start acting as an independent ratings agent, such that each issuer’s overall trust rating will be calculated by blending the ratings of individual users, and then weighting them by a number of factors that’ll make it difficult to game / manipulate the rankings.
As a user, you’ll be able to rate issuers / assets (on a scale between 0 and 10) based on your practical experience of dealing with them (deposits, withdrawals, customer service, etc) and also the amount of transparency you feel an issuer is offering into their business.
After posting an initial rating, you can post “upgrades” and “downgrades” to your score along with justifications for the change.
You’ll also be able to "follow” (e.g. subscribe) to the ratings of other users so you can get email updates when assets you hold are upgraded or downgraded by people you trust.
If you’re an issuer, you’ll be able to subscribe to be informed of upgrades or downgrades to your own assets, and you can use the service to monitor, build and maintain your reputation. You’ll also have the opportunity to issue rebuttals to ratings.
Here’s the cool bit.
Trust scores will be broadcast into the blockchain in order to help bootstrap an insurance market such that asset holders and insurers could take opposite sides of a bet on the future trust score of an issuer.
The end game is this: when you buy an asset via the distributed exchange I want you to be able to pay a small fee to an “insurer” that would give you a payout if the issuer defaults. In practice this would take the form of a bet on the trust score of an issuer falling below a certain threshold within a certain period of time.
So the poll question is this - how useful would you find this service?
I think it would be useful, and I’d love to be an advisor on the project. Also, I grabbed RATING and RATINGS on the asset exchange, so we have some good vehicles.
[quote author=AdamBLevine link=topic=220.msg1764#msg1764 date=1396887691]
I think it would be useful, and I’d love to be an advisor on the project. Also, I grabbed RATING and RATINGS on the asset exchange, so we have some good vehicles.
[/quote]
Hi Adam, I’d love to have you as an adviser on the project! If you could PM me some preferred contact details, I’ll drop you a message and introduce myself.
adam@letstalkbitcoin.com, not a secret 
Good idea, such a rating system is essential to the functioning of the distributed exchange. Without this application layer, users have no way of weather to trust the asset or not.
ok read this post for three times, this is by far the best idea I have seen for a while,if this is integrated properly it will help the counterparty ecosystem massively.
I do not know how complex the coding is, so if it is in any way possible, especially this part, it will be a boost for the whole sphere of decentralised trading:
[quote][font=verdana][size=11px]Here’s the cool bit. [/size][/font][font=verdana][size=11px]Trust scores will be broadcast into the blockchain in order to help bootstrap an insurance market such that asset holders and insurers could take opposite sides of a bet on the future trust score of an issuer.
The end game is this: when you buy an asset via the distributed exchange I want you to be able to pay a small fee to an “insurer” that would give you a payout if the issuer defaults. In practice this would take the form of a bet on the trust score of an issuer falling below a certain threshold within a certain period of time.[/size][/font][/quote][font=verdana][size=11px]
The beauty of this is idea is, that, in case you plan on running a thrustworthy asset/ excahnge trading IOUs, you can get an additional monetary reward by giving an insurance for your created IOU, if you do not behave properly or you miss some standards which will fuck up your service, you’ll get a monetary punishment, assuming you are going to insure your assets. Please correct me if I am wrong.
I would really love to see this innovation implemented
[/size][/font]
I think in addition to this, almost certainly, investors/traders will want leverage, margin accounts, and other advanced methods of trading shorts/longs etc., Now I’m not entirely sure, but it seems to me that, that is not baked in at a protocol level. I believe that a system that will be used by high profile traders will need to be secure, quick(est), and include features like seen on bitfinex.com…
Any thoughts ?
Something about a good idea, that I’ve noticed over the years, is that you (at least I) only have to read it like maybe a few sentences in before it just clicks because its almost obvious. And this idea is obvious (the one you have put forth) I think mine too.
[quote author=hodl link=topic=220.msg1771#msg1771 date=1396905303]
[font=verdana][size=11px]The beauty of this is idea is, that, in case you plan on running a thrustworthy asset/ excahnge trading IOUs, you can get an additional monetary reward by giving an insurance for your created IOU, if you do not behave properly or you miss some standards which will fuck up your service, you’ll get a monetary punishment, assuming you are going to insure your assets. Please correct me if I am wrong.[/size][/font]
[/quote]
That’s right, issuers could insure their own reputation score and win a small amount of XCP if their trust score remains above a certain threshold, and pay out a large amount if it doesn’t. The only downside is that issuers might need large amounts of spare XCP to do this at scale.
If an issuer doesn’t have the capital to back large numbers of bets in this way, professional insurers could step in who’d be willing to lock up their XCP to make a return backing issuers they’ve come to know and trust over time. The more open the relationship between an issuer and their insurer(s) and the more competitive the market, the lower the “insurance premium” side of the bets would be. I think people might be willing to do this for similar reasons that stock holders are willing to write covered calls in options trading.
To make the system work effectively, I think we’d probably need threshold style bets in Counterparty (a little like one touch binary options), but even without these, there are ways of making it work. If I go ahead with this idea, I’ll be chatting to the developers about extending the equals / not equals bets to cope with >= or <= numerics.
My hope is that if a service like this became widely used, it would push the industry in the direction of more transparency. Insurers would have an incentive to really get to know issuers and understand their businesses, and issuers would have an incentive to be more transparent (being less anonymous, submitting to periodic audits, using voting pools, etc).
Essentially, the insurance market would signal risk to buyers; if no insurer is willing to back an issuer, or if they’re charging very high prices, then this tells a story.
[quote author=halfcab123 link=topic=220.msg1773#msg1773 date=1396912059]
I think in addition to this, almost certainly, investors/traders will want leverage, margin accounts, and other advanced methods of trading shorts/longs etc., Now I’m not entirely sure, but it seems to me that, that is not baked in at a protocol level. I believe that a system that will be used by high profile traders will need to be secure, quick(est), and include features like seen on bitfinex.com…
Any thoughts ?
[/quote]
Yeah, to my knowledge, in Counterparty, the notion of leverage is unconnected with margin and just refers to how sensitive a bet is to movements in the value of a feed.
Bets in Counterparty use full escrow and are trustless, and I think that the idea of trustlessness and margin might be incompatible unless you also have the concept of debt in the system. I guess if you have debt, concepts like reputation might also be useful.
It could be interesting to think this through.
Great work on the How To videos BTW
This is great.
Make it so.
We will use it. We currently reference Blockscan in our Asset listing. We’d gladly reference any other site that aggregated reputation data of our services.
It sounds like everything you propose can be done on line. Will there ever be a need for an actual person to visit another person to authenticate the asset? For example, bullion banks get regular site audits by professional auditing agencies. Corporations get regular site visits to go over the records of large corporations. Such services are not cheap, but might be necessary for new asset holding businesses or individuals to acquire trust ratings quickly. For such speedy service, many will want to pay.
Individuals and small businesses might rely on someone like a notary to verify the quality and possession of the asset. Larger corporations with more and more expensive assets to post might require a more extensive audit from professionals.
You don’t have to hire and train all of these people; the experts and authenticators are already out there doing this kind of work. The question is whether you want to utilize them and their already established trust ratings (professional auditors and notaries, in this example) in order to give your initiative a head start. This could be a special service that some clients will want to use so that they don’t have to wait for their trust rating to gradually grow over time. And for such speedy, custom service, they pay an appropriate fee (in XCP).
No matter how sophisticated any distributed exchange, no matter how “trustless” any blockchain-based transaction and ledger protocol, there will always be the interface between the electronic data and the real world thing. This interface is the place where no amount of “trust in math” will work, because you have to trust people - not just their intentions but their ability to judge the quality and authenticity of their assets. You need people working in many parts of this interface each doing their bit to verify all aspects of this data-to-world connection, and what you’re proposing can still do it in a mostly decentralized way.
Or am I totally missing the point of what you are proposing?
Yeah, I’d definitely see real world audits taking place (esp. for physical assets) - I think they’d either be conducted by “insurers”, trusted members of the public, or by an audit company of some kind (paid for by the asset issuer themselves). I like your idea, I could maybe act as a matchmaking service between asset issuers and people in the business of auditing.
My goal is to create a system of incentives to give asset issuers every reason to be as open as possible in their dealings.
Asset issuers would want to increase their trust rating, because doing so would win them more business and also make them more “insurable”.
They could do this by putting useful evidence on the public record. E.g. if it’s a digital asset, this could be signed proof of reserves on a blockchain. Or it could be an audit document, perhaps with a reference number that it’s possible for someone outside of the company to validate.
Alternatively, they could work with “insurers” or trusted members of the public directly, and allow them to scrutinize their business. When the results are published, this would then be reflected in the trust ratings and in the “premiums” charged by insurers.
In my model, I’m using bets in Counterparty such that, on the one hand you have asset holders, who’d like to pay a smallish “premium” in order to get a lump sum if an issuer goes rogue (e.g. if their trust score falls below a certain threshold in a defined period of time).
On the other hand you have “insurers” who’d be willing to lock up XCP to get a predictable return, taking the other side of the bet, and effectively banking on an issuers score not dropping below an agreed threshold.
If an asset issuer can convince an insurer that they’re low risk, they’ll demand less of a return (because they’ll be a safer bet) and the insurance would be cheaper. My thinking is that this would create a natural pressure toward greater transparency.
Does this sound workable to you?
So the insurer is different from the auditor, in one of your scenarios. This makes sense. Consider this and tell me if it is possible: If the auditor and the insurer were the same entity, it would be possible to game the system by giving high ratings to worthless assets and then taking the “losing” side of the bet.
Instead, the auditors could be free-lance individual experts or for-hire companies who specialize in particular industries. They would be trustworthy because their own reputation for honesty and competence would be on the line, and subject to a ratings ledger you keep. I see this auditor niche being a huge industry in its own right, and would work not just for Counterparty but for any interface between decentralized blockchain marketplaces and real world stuff. (Centralized operations allow chargebacks, and bank in a certain level of default, so they have a different model.)
In your original model, you were focusing on the trust ratings of the individual asset issuers and buyers. In the enhancement I propose, the focus is on verifying the assets themselves. Eventually, the reputation of the asset issuer will be comprised of the verification work of the independent auditors in addition to the ratings of clients who have done business with the asset issuer. That is, good assets and good transactions will both contribute to good asset issuer trust ratings. Eventually, these asset issuers may not need to have their assets audited as long as they continue to have a very high individual trust rating. But let’s say they screw up. They can begin to repair their trust rating by hiring these independent auditors to audit their assets again, and give them an immediate boost in reputation.
Thanks for these really useful thoughts, sparta_cuss.
Up until now, I’d seen every user in the system as being a ratings agent, and some subset of these users choosing to take on the extra (somewhat informal) role of auditor. In my model, you absolutely could have scams where, say, a malicious auditor had built up a great reputation and then gives a high rating to a worthless asset.
Following this through - this may have the effect of making insurance cheaper for that asset, and perhaps also, pushing up the price. They could buy insurance in the hope of the issuer going under and collecting a large payout. They could also buy the asset before they publish their ratings in the hope of selling at a higher price.
In my model, this is mitigated by users in the system punishing the auditor’s reputation / influence by “unfollowing” them. In the ratings algorithm I’m proposing, the number of users who are “following” (subscribing to) your ratings, the larger the weighting that your individual score plays in the overall score for that asset / issuer.
In your model, an auditor does not issue ratings, they simply verify facts.
Maybe something like this could work -
In the future, there could be a roster of professional auditors, who’s job it is to verify simple propositions that are published by an issuer.
For example, an issuer could publish claims like -
[list]
[li][font=verdana][size=1em]“I have a safety deposit box that contains 100 silver coins”.[/size][/font][/li]
[li][size=1em][font=verdana]“I have a small workshop in Sausalito, California”.[/font][/size][/li]
[li][size=1em][font=verdana]“I employ an artisan silversmith named Bob Peterson”.[/font][/size][/li]
[li][size=1em][font=verdana]“I use a Peddinghaus Forming Hammer”.[/font][/size][/li]
[li][size=1em][font=verdana]etc…[/font][/size][/li]
[/list]… and then they’d offer an XCP bounty to have someone come and verify the claim(s).
Maybe customers could even issue public challenges to issuers that take a similar form, e.g. “Prove that you have a safety deposit box that contains 100 silver coins”.
Then (qualified) auditors could bid against each other for the work, and get paid in XCP.
Also, for any given job, there could be a small probability that a second audit is triggered and assigned to the auditor with the second best bid (the price an auditor pays could have a premium to factor this in).
Issuers can’t collude with auditors because they can’t guarantee who’ll be assigned the job, and auditors are kept honest (and must be accurate in their work) because there’s always a small chance that their work will be repeated by a second auditor.
What do you think?
I think I might play around with splitting out the roles further, it seems to help reduce opportunities for collusion -
1) Customers
2) Auditors
3) Insurers
4) Issuers
Thanks again for all your thoughts.
I like the way your project is developing, and I am glad that my questions are helpful. I have a few more followups.
I should let you know, though, that I have no professional expertise in any of these arenas except as a customer. I’m trying to ask challenging questions, but I don’t speak from a position of industry experience or expertise. Just so you know.
"[font=verdana][size=11px]In your model, an auditor does not issue ratings, they simply verify facts.[/size][/font][font=Verdana][/size][size=78%]" [/size][/font]
I agree with how you put this.
Preventing people from gaming the system by having specific, non-overlapping roles seems like the right strategy. Here is an analogy for the auditor’s role. When you buy a house, you need to have it inspected. What is the incentive for the inspector to do an honest and competent job?
As you say, “[font=verdana][/size][size=11px]Issuers can’t collude with auditors because they can’t guarantee who’ll be assigned the job, and auditors are kept honest (and must be accurate in their work) because there’s always a small chance that their work will be repeated by a second auditor.[/size][/font][font=Verdana][/size][size=78%]”[/size][/font]
In addition, the auditor wants to get a good rating in the eyes of the buyer and seller. If a string of buyers becomes critical of his work and feels like they have been taken advantage of (the inspector failed to note critical, expensive-to-repair flaws that each buyer is now stuck with), then his reputation will suffer. If a string of sellers becomes critical of his work because they think he is obstructing sales (he is far too picky about issues or exaggerates the seriousness of some issues), then they will give him a bad rating. In both cases, there is a dynamic tension produced by the interests of the buyers and sellers that causes the inspector to find the “right balance.” This is an art and not a science. Your rating system promotes that dynamic equilibrium and allows that balance to be sought, I think.
By what mechanism will auditors be selected? Will there be a place for issuers or buyers to select auditors? Certainly price for service will be relevant, and auditors would compete on price. But they will also want the ability to compete based on experience and their reputation.
In your system so far, every interested party has a reputation score and history, right?
The proposal is for
1) A listing service
2) an auditing repository - professional audits
3) a rating service - Web of Trust style
4) insurance…what… broker…coordinator?
5) something else I missed or don’t understand
Each of these could be viable, so you might want to design the whole, and prioritize development to be able to release the modules as they are ready.
[quote author=sparta_cuss link=topic=220.msg1862#msg1862 date=1397180588]
I’m trying to ask challenging questions, but I don’t speak from a position of industry experience or expertise. Just so you know.
[/quote]
Challenging questions are exactly what I’m looking for right now, so thanks. I’m not an expert in the area of financial ratings or auditing either, so that makes two of us!
[quote author=sparta_cuss link=topic=220.msg1862#msg1862 date=1397180588]
Here is an analogy for the auditor’s role. When you buy a house, you need to have it inspected.
The auditor wants to get a good rating in the eyes of the buyer and seller.
[/quote]
I like the analogy, although with house buying, the buyer, the seller and the inspector all get to see the house. In financial audits, only the seller and (to a lesser extent) the auditor get to see the reality behind the tokens that buyers hold.
In our financial system, we have a situation where ratings agencies are employed by sellers of financial instruments to rate their products. This is a bit of a recipe for corruption, because the seller has a free choice of auditor, who then ends up on the seller’s payroll.
In house buying (at least in the UK) the surveyor is paid by the buyer and is supposed to act in their interests. If the surveyor was employed by the seller, I’d be less inclined to trust their findings.
[quote author=sparta_cuss link=topic=220.msg1862#msg1862 date=1397180588]
By what mechanism will auditors be selected? Will there be a place for issuers or buyers to select auditors? Certainly price for service will be relevant, and auditors would compete on price. But they will also want the ability to compete based on experience and their reputation.
[/quote]
Let’s see, I can imagine -
[list]
[li][font=verdana][size=1em]Buyers would need to have some evidence that auditors are [/size][/font]competent, even before their first audit[size=1em]. Maybe there’s an application process, or some kind of voting. [/size][/li]
[li][font=verdana][size=1em]Issuers should pay for audits, because otherwise, I think we’re left with buyers needing to crowdfund audits. [/size][/font][/li]
[li][font=verdana][size=1em]I think that issuers should be denied a free choice of auditor. If they have a free choice, they can collude. For example, I could pick my wife’s auditing company to audit my assets. [/size][/font][/li]
[li][font=verdana][size=1em]If we’re going to deny issuers a free choice, but make them pay, I think they’d need to have some assurance that the auditor chosen will be fair and competent. [/size][/font][/li]
[li][font=verdana][size=1em]Lastly, the issuer needs to feel that they’re paying a fair / market price for the auditing service. [/size][/font][/li]
[/list]I get the sense that the only people who should be judging auditors are fellow auditors; the buyers never actually get to meet the issuer, or see the issuer’s business from the inside, and we already know the issuer’s view, as they’re the ones making the claims.
Maybe “right of appeal” is the way to go.
So if an auditor refuses to validate a claim the issuer has made about an asset, the issuer could appeal.
In this case, both the issuer and the auditor could collect and submit evidence to a panel of auditors, who would then have a vote (perhaps, in some cases, this could trigger a second in person visit from another auditor).
If the decision goes in favor of the issuer, the auditor gets scored down, if it goes in favor of the auditor, the issuer is scored down.
How do we choose which auditors get selected for any given job?
How about something like this -
[list]
[li][size=1em][font=verdana]Start with all the auditors and collect bids.[/font][/size][/li]
[li][font=verdana]Filter out all auditors from the list who’re bidding above the maximum threshold that the issuer is willing to pay.[/font][/li]
[li][font=verdana][size=1em]Use Bayesian statistics to rank auditors by a score derived from 1) the total [/size][/font]monetary[size=1em] value of the work they’ve completed, vs 2) the [/size]monetary [size=1em]value of the work completed that did not result in a successful appeal by an issuer. This could be a proxy for the competence of the auditor, weighted by experience.[/size][/li]
[li][font=verdana][size=1em]Rank the results by some blend of the price that an auditor is quoting for the work and the “competence score”.[/size][/font][/li]
[li][font=verdana][size=1em]Pick the top result and assign them the work.[/size][/font][/li]
[/list][quote author=sparta_cuss link=topic=220.msg1862#msg1862 date=1397180588]
In your system so far, every interested party has a reputation score and history, right?
[/quote]
Here’s a version that tries to separate out roles a little more (I have another model where the Auditor and User roles are combined) -
Users
[list]
[li][font=verdana][size=1em]Have the power to rate issuers, on scale from 0 to 10, based on 1) their experience of deposits, withdrawals and general customer service 2) the claims validated (and left un-validated) by auditors and 3) their general feelings about the issuer.[/size][/font][/li]
[li][size=small][font=verdana]In the case of withdrawals, it’s possible for users to prove that they’re real customers (rather than trolls) by using the blockchain and key signing.[/font][/size][/li]
[li][font=verdana][size=1em]Have the ability to challenge issuers, and request that they prove claims about their business phrased as simple statements, e.g. “Prove you have 100 silver coins in a safety deposit box”, or “Prove you really are the person in that photo.”[/size][/font][/li]
[li][font=verdana][size=1em]Have the ability to “follow” other users, and get notifications when other users they trust upgrade or downgrade assets or issuers they’re interested in.[/size][/font][/li]
[/list]Issuers
[list]
[li][font=verdana][size=1em]Have the ability to issue rebuttals to the ratings that users submit, and can use the service as a kind of customer service channel, like Twitter.[/size][/font][/li]
[li][font=verdana][size=1em]Can make claims about their business and pay to have an auditor verify those claims.[/size][/font][/li]
[li][font=verdana][size=1em]Can challenge the results of an audit and trigger an appeals process.[/size][/font][/li]
[li][font=verdana][size=1em]Get a mark on their record when these appeals go against them.[/size][/font][/li]
[/list]Auditors
[list]
[li][size=small][font=verdana]Get paid to validate claims made by issuers.[/font][/size][/li]
[li][font=verdana][size=1em]Can bid on any work they feel qualified to do, but take a reputation hit if a finding is appealed and the vote goes against them.[/size][/font][/li]
[/list]Insurers
[list]
[li][font=verdana][size=1em]Put XCP to work, in the hope of earning a return, by offering to take the “insurance premium” side of a bet with asset holders.[/size][/font][/li]
[/list][font=verdana]I think what I’ll probably end up doing is to make a very simple version of this service to begin with, and then make it more granular later. [/font]
[font=verdana]But this is all really useful thinking for the future. [/font]
[font=arial][size=small]I stayed away from this thread because I did not want to hog the microphone. But I’ve thought a bit more about it and wanted to check in. [/size][/font][font=arial][size=small]If I may, here are a few more thoughts, and I hope this is helpful as you continue to develop this project.[/size][/font]
[font=arial][size=small][/size][/font]
[font=arial][size=small]Basically what you are designing is a mechanism which uses cryptocurrencies/blockchain technology to allow people to build trust in other people, yes? The reputation scores are attached to people (or business entities) primarily and only derivatively to their assets, right? [/size][/font]
[font=verdana]
[font=arial]It’s better than what many in this space seem to be doing: [size=small]using people[/size][/font][/font][size=1em] (developers, in particular) to build mechanisms that build trust in cryptocurrency technology and their particular coin.[/size]
[font=verdana]
[font=arial]One oft-repeated Bitcoin slogan is “trust in math.” But that is only an instrumental goal, isn’t it? Trust is built when people agree on the facts and the facts are proven to be true. So it is [/font][/font][font=arial][size=small]possible (and, in my opinion, desirable) to build trust in other human beings by [/size][/font][font=arial][size=small]using[/size][/font][font=arial][size=small] math. I think this is more in line with what Satoshi had in mind.[/size][/font]
[font=verdana]
[font=arial][/font][/font]
[font=arial]Your original plan of building out from the trust rating system to eventually include auditors is the best plan. That way you don’t have to worry about creating an interface for auditors until the system that they are interfacing with is built. [/font]
[font=arial][size=small]In your original model, trust of the issuance is a matter of the accumulated trust rating of the issuer. This is still, I think, the primary model to work with. An auditor might not always be needed, say, once the issuer develops a high reputation score. The work/role of the auditors would support, not replace, the reputation system. Here are some scenarios:[/size][/font]
[font=arial][size=small]1. The asset issuer is new to the system and no one knows who this entity is. This asset issuer is strongly incentivized to pay for an audit of asset in order to jump-start their reputation score. (You might even say that it’s the “price of admission.”)[/size][/font]
[font=arial][size=small]2. The asset issuer’s reputation has been damaged somehow. Perhaps the issuers’s internal quality control process was poor, or they had an incompetent or dishonest individual on board. Somehow, the once good reputation of this asset issuer has been justifiably hurt. But now the asset issuer has fired the incompetent person and put controls in place. They have their house in order and want an independent auditor to verify their assets and processes. [/size][/font]
[font=arial][size=small]In both of these scenarios, the asset issuer is has an incentive to use the auditor. I think this incentive is important. It means that the asset issuer can be required to pay for the audit, and the issuer will understand that to ensure that the audit is legitimate, the asset issuer cannot select the auditor (as you already noted). [/size][/font]
[font=arial][size=small]Making sure that each of the roles that you have identified has incentives to use your service/protocol to connect with the other roles will help this system be interconnected and collectively supported. [/size][/font]
[quote author=sparta_cuss link=topic=220.msg1981#msg1981 date=1397841710]
[font=arial][size=small]Basically what you are designing is a mechanism which uses cryptocurrencies/blockchain technology to allow people to build trust in other people, yes? The reputation scores are attached to people (or business entities) primarily and only derivatively to their assets, right? [/size][/font]
[/quote]
[font=verdana]Yeah, the system should make apparent the level of confidence (the community has) that the issuer of a particular asset will meet their obligations in the future. If an issuer issues two assets, they’re very likely to have different trust scores (although one would imagine that, in general, they’d be fairly similar). [/font]
[font=verdana]I guess we could end up with a situation such that an issuer is a company with lots of staff, and some of those staff offer poor customer service or are untrustworthy. [/font]
[quote author=sparta_cuss link=topic=220.msg1981#msg1981 date=1397841710]
[font=arial]Your original plan of building out from the trust rating system to eventually include auditors is the best plan. That way you don’t have to worry about creating an interface for auditors until the system that they are interfacing with is built. [/font]
[/quote]
[font=verdana]
Agreed, I think this is the way forward[/font]
[quote author=sparta_cuss link=topic=220.msg1981#msg1981 date=1397841710]
[font=arial][size=small]Making sure that each of the roles that you have identified has incentives to use your service/protocol to connect with the other roles will help this system be interconnected and collectively supported. [/size][/font]
[/quote]
Thanks for those scenarios.
If you wouldn’t mind, maybe I could show you some mock-ups of a very basic version of the system, when I’ve found time to Photoshop them up?
[quote author=Niki Wiles link=topic=220.msg2044#msg2044 date=1398181611]
[quote author=sparta_cuss link=topic=220.msg1981#msg1981 date=1397841710]
[font=arial][size=small]Basically what you are designing is a mechanism which uses cryptocurrencies/blockchain technology to allow people to build trust in other people, yes? The reputation scores are attached to people (or business entities) primarily and only derivatively to their assets, right? [/size][/font]
[/quote]
[font=verdana]Yeah, the system should make apparent the level of confidence (the community has) that the issuer of a particular asset will meet their obligations in the future. [color=#ff0000]If an issuer issues two assets, they’re very likely to have different trust scores[/color] (although one would imagine that, in general, they’d be fairly similar). [/font]
[font=verdana]I guess we could end up with a situation such that an issuer is a company with lots of staff, and some of those staff offer poor customer service or are untrustworthy. [/font]
Thanks for those scenarios.
[color=#0000ff]If you wouldn’t mind, maybe I could show you some mock-ups of a very basic version of the system, when I’ve found time to Photoshop them up?[/color]
[/quote]
[color=#ff0000]Red:[/color]
This is confusing. For comparison I am thinking of how eBay or Amazon work. Any particular vendor is treated like a “store,” and their reputation is part of their identity. They can sell whatever assets they want, and their reputation remains separate from the “product ratings” on individual products. Is this the kind of thing you had in mind? This only makes sense if it’s possible for two separate asset issuers to offer the exact same asset type. For example, Vendor ABC on eBay and Vendor XYZ on eBay both sell a particular kind of Levi’s Jeans. They don’t produce the product; they merely act as a purveyor of a product made by someone else. So it’s possible that they are each offering the “same” product. That product has a “rating” associated with it that is logically (and programmatically) distinct from the vendor and the vendor’s reputation.
I am not sure whether this is the right model for what you are proposing. Is it the case that asset issuers would only ever sell assets that they are creating? We need some use cases to explore this further.
[color=#0000ff]Blue:[/color]
Absolutely. Let’s keep this conversation going.