BitShares Forum

Main => General Discussion => Topic started by: Agent86 on September 19, 2014, 04:09:27 pm

Title: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 19, 2014, 04:09:27 pm
I'm not a big fan of using proof of burn for reputation as planned.  I want to propose an alternative.  I think although these ideas are more complex they are more useful.

Every username would have a birthday and three reputation metrics or scores:
   1) Trust
   2) Devotion
   3) Lifespan
   
Everything builds off the devotion metric which is accomplished by mining your user name via coin-day mining "CDM".  Essentially you can point a balance to a user name to increase (or decrease) it's devotion score proportional to the coin days of the stake directed toward the user name.  This metric is scarce because a given balance cannot mine multiple usernames at the same time.  Multiple balances can however mine a single username concurrently.

The process starts when someone pays a reasonable fee to register a username.  This initiates a 30 day "auction" that is decided by competitive coin-day mining.  If no one else competes for the username or mines against it, than the person who registered it/mined it the most gets it after 30days.  This date is the "birthday" of the user name.

The devotion score for a user name is generally given by the formula: supportive_CDM - opposing_CDM = devotion.

Any username may choose to trust, not trust, or remain neutral toward another username.  The user name's total trust score is determined by the sum of the devotion scores of all usernames who trust it minus the sum of the devotion scores of all usernames who distrust it.

The lifespan for most accounts will be equivalent to the age of the account since the birthdate and goes up with time.  However if a username's total devotion score is brought negative by opposing CDM, the lifespan instead decreases with time for as long as the devotion score remains negative.  If a username's lifespan goes to zero, the user "dies" (username is revoked and it may be re-registered by someone else.)

The robo-hash images can hash both the username and the birthdate so that if a username ever changes hands the robo-hash image also changes.

Most names will have positive metrics for all three and the more positive the better. 

If there is a user you don't trust you can simply distrust the user and encourage others to do the same.  Any name with a negative trust score should typically not be trusted.

If the user is particularly offensive or the username is an attempt to spoof, you can opposition mine the name to destroy it's devotion score and eventually kill it/get it revoked. Any name with a negative devotion score is a huge red flag.  This account is quite likely a spoof/fake name and is on the road to revocation unless they can mine it to bring it positive.  There would be BIG warnings before sending any funds to a username with negative devotion.

The longer you have maintained your username in good standing the longer your lifespan and the longer time you have to react if someone starts opposition mining your username.

During the initial auction if there is more opposing CDM than the highest user competing to own the name then the name is never created.  So if someone tries to register an offensive name we can neg-mine it and it is killed in utero.

If you want to take a real close look at a username you can see specifically who trusts them, who doesn't, how recently was their devotion mined etc.

There is a natural opposition between trust & reputation and anonymity/privacy.  My feeling is that anonymity has generally been overemphasized at the expense of more important things.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: emski on September 19, 2014, 04:49:41 pm
I generally like your idea! I think it is better than burning.

However there are some things about it I find as problematic:

I don't like the idea of automatically revoking account name if its trust is negative.
How will you prove it is not registered again by the same person?

Also I do not like the robohash avatars - here is my proposal about this: https://bitsharestalk.org/index.php?topic=9109.0 .
Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 19, 2014, 05:05:50 pm
I generally like your idea! I think it is better than burning.

However there are some things about it I find as problematic:

I don't like the idea of automatically revoking account name if its trust is negative.
How will you prove it is not registered again by the same person?

Also I do not like the robohash avatars - here is my proposal about this: https://bitsharestalk.org/index.php?topic=9109.0 .

Names are not automatically revoked if trust is negative.  Trust has no bearing on revocation; only devotion & lifespan impact revocation.  Lifespan must go to zero for a name to be revoked.

I'll give an example:
If you have a username that you have had for 2 years and you have 10 shares that you have been mining the username with over that timeā€¦.  People can distrust you all they want and your trust is negative but it doesn't otherwise impact your name.  If someone with the same amount of stake as you (10 shares) starts opposition mining your username and you continue to also mine your username then your devotion would no longer increase (because the mining would even out) but you would never lose your user name.  If someone with twice the stake as you (20 shares) starts "neg-mining" your user name than your devotion score would start to decline.  It would take 2 years of this before your devotion score declined to zero and became negative.  At that date your lifespan would be 4 years (for 4 years you have owned the user name with a positive devotion score).  It would take 4 more years with your devotion score staying negative and your lifespan now declining before your lifespan would finally reach zero and your username is revoked.  So if you had your name for 2 years it took someone with twice the stake 6 years to revoke it from you (for the last 4 years your devotion was negative so probably no one trusts you but you haven't lost your name.)  Does that make sense?
Title: Re: Reputation & username system - alternative to proof of burn
Post by: emski on September 19, 2014, 05:14:01 pm
Got it.

However I still don't like the idea that someone could revoke your name at any cost.
It could go away due to inactivity or unpaid fees but not due to someone with larger stake mining/minting against you.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 19, 2014, 05:21:42 pm
Got it.

However I still don't like the idea that someone could revoke your name at any cost.
It could go away due to inactivity or unpaid fees but not due to someone with larger stake mining/minting against you.

I disagree, I think this allows the shareholders as a whole to decide that this particular name is a spoof or inappropriate and revoke it before it ever gets a chance to be used.  It also allows the shareholders to decide that the person squatting "toast" can't do that and toast can get his name back.  Generally one other person with a little more stake than you can't afford to waste the effort trying to fight you about your username, it's mostly a community driven thing.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: Brent.Allsop on September 19, 2014, 05:23:17 pm
Help me get up to speed with what you guys are talking about.  Can you provide more info, or references to where I can get more info about what you mean by "mining", "planned proof of burn", "opposition mining"

It seems to me that any type of reputation systems needs the ability to know, concisely and quantitatively, what everyone thinks of the experience they've had with whatever it is right?  In other words, for a generic reputation system you need to have a concise description of the experience large numbers of people have had with it, and a measure of how many people have had that experience vs competing camps/descriptions.  Or, the kind of stuff you can do at canonizer.com?

Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 19, 2014, 05:28:54 pm
Help me get up to speed with what you guys are talking about.  Can you provide more info, or references to where I can get more info about what you mean by "mining", "planned proof of burn", "opposition mining"

It seems to me that any type of reputation systems needs the ability to know, concisely and quantitatively, what everyone thinks of the experience they've had with whatever it is right?  In other words, for a generic reputation system you need to have a concise description of the experience large numbers of people have had with it, and a measure of how many people have had that experience vs competing camps/descriptions.  Or, the kind of stuff you can do at canonizer.com?

I'm using "mining" figuratively, it's just a way of using "coin-days" to demonstrate commitment to something.  Coin-days are a scarce resource that you can use rather than "proof of burn" where the scarce resource is the stake/coins themselves.  "opposition mining" is just using the same method to indicate commitment in opposition to something (a username).  Proof of burn is basically indicating that you have invested in your username by burning money (shares), I think people will generally be reluctant to do this.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: emski on September 19, 2014, 05:31:15 pm
Got it.

However I still don't like the idea that someone could revoke your name at any cost.
It could go away due to inactivity or unpaid fees but not due to someone with larger stake mining/minting against you.

I disagree, I think this allows the shareholders as a whole to decide that this particular name is a spoof or inappropriate and revoke it before it ever gets a chance to be used.  It also allows the shareholders to decide that the person squatting "toast" can't do that and toast can get his name back.  Generally one other person with a little more stake than you can't afford to waste the effort trying to fight you about your username, it's mostly a community driven thing.

Do you remember the case with Mike Rowe ?
http://en.wikipedia.org/wiki/Microsoft_vs._MikeRoweSoft (http://en.wikipedia.org/wiki/Microsoft_vs._MikeRoweSoft)
How can you fight that?
My opinion is that once mined/registered names should be personal and only in extreme cases they should be revoked.
Something at least 100000x your stake.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: oldman on September 19, 2014, 06:23:30 pm
I think the proposal has merit... in concept. The trick is implementation and BM/devs will need to chime in.

Not completely sold on whether the advantages outweigh the increased complexity; open Bazaar has some pretty convincing material on proof-of-burn for rep.

Could you perhaps elaborate on the aspects of proof-of-burn that you feel are inadequate or undesirable?

Title: Re: Reputation & username system - alternative to proof of burn
Post by: gamey on September 19, 2014, 06:32:05 pm

Waiting 30 days to get your username is horrible for adoption.  I want to dislike the idea of large stakes buying trust systems, but I can't think up a good argument against it at the moment.  I would need examples etc.  Can this be abused to shake people down?  I get a large stake behind me, we find legitimate business and bad rep them unless they pay up.  What prevents this ?
Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 19, 2014, 06:52:57 pm
I think the proposal has merit... in concept. The trick is implementation and BM/devs will need to chime in.

Not completely sold on whether the advantages outweigh the increased complexity; open Bazaar has some pretty convincing material on proof-of-burn for rep.

Could you perhaps elaborate on the aspects of proof-of-burn that you feel are inadequate or undesirable?

Well I don't have to burn my stake; that's a good enough reason for me.  Really it just creates the wrong incentives because you don't really want to burn stake and makes people reluctant to participate to get the rep so the rep isn't very representative of anything IMO.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 19, 2014, 06:58:54 pm
Waiting 30 days to get your username is horrible for adoption.

I'm thinking of the username more as a luxury.  I would want it so it is not necessary to get started and use the system.  I think the current "chicken and egg" situation where you need to ask someone to register you is a little ridiculous.

I want to dislike the idea of large stakes buying trust systems, but I can't think up a good argument against it at the moment.  I would need examples etc.  Can this be abused to shake people down?  I get a large stake behind me, we find legitimate business and bad rep them unless they pay up.  What prevents this ?
I don't think these concerns have merit.  As soon as you try to blackmail someone that person will distrust you and makes the situation publically known.  Then if you are trying to go around doing this to everyone, they all distrust you and you seem to distrust everyone and people talk so it's not hard to figure out what's going on here.

Title: Re: Reputation & username system - alternative to proof of burn
Post by: emski on September 19, 2014, 07:02:33 pm
  I think the current "chicken and egg" situation where you need to ask someone to register you is a little ridiculous.

Again check the first proposal here:
https://bitsharestalk.org/index.php?topic=9075.0 (https://bitsharestalk.org/index.php?topic=9075.0)

It should allow us to get rid of "Post here if you want 1 BTSX" threads.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: bytemaster on September 19, 2014, 07:42:15 pm
The burn isn't really a means to define reputation, but can be done for vanity purposes. 
Title: Re: Reputation & username system - alternative to proof of burn
Post by: gamey on September 19, 2014, 08:44:26 pm
Waiting 30 days to get your username is horrible for adoption.

I'm thinking of the username more as a luxury.  I would want it so it is not necessary to get started and use the system.  I think the current "chicken and egg" situation where you need to ask someone to register you is a little ridiculous.

I want to dislike the idea of large stakes buying trust systems, but I can't think up a good argument against it at the moment.  I would need examples etc.  Can this be abused to shake people down?  I get a large stake behind me, we find legitimate business and bad rep them unless they pay up.  What prevents this ?
I don't think these concerns have merit.  As soon as you try to blackmail someone that person will distrust you and makes the situation publically known.  Then if you are trying to go around doing this to everyone, they all distrust you and you seem to distrust everyone and people talk so it's not hard to figure out what's going on here.

I agree having to register first is ridiculous, but that is an issue with the exchanges forcing that requirement.  I suppose that for BTSX waiting 30 days is somewhat reasonable, where for KeyID DAC it would be a killer.  It just seems strange to me.  There will be a certain subset of guys who are actually involved in this process.  The naming thing just isn't important enough to justify this auction type system and all the weird market that'll occur in the background.  It just makes BTSX all that more complicated in a way that doesn't really bring enough value.  The names are meant to aid ease of use, we're going backwards trying to create yet another market within the system.

The person blackmailing doesn't have to be known and can move their stake around at will.  I would need to study the issue more if this isn't true.

This also doesn't solve spoof scammers.  If I understand correctly, those spoof bter type accounts that get random deposits can be negative mined at which point they show up on the market again.  What prevents a scammer from doing the whole thing again?  It is up to bter to try and vote these guys down  ?  I am not sure there is a big incentive from others.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 19, 2014, 09:37:57 pm
Waiting 30 days to get your username is horrible for adoption.

I'm thinking of the username more as a luxury.  I would want it so it is not necessary to get started and use the system.  I think the current "chicken and egg" situation where you need to ask someone to register you is a little ridiculous.

I want to dislike the idea of large stakes buying trust systems, but I can't think up a good argument against it at the moment.  I would need examples etc.  Can this be abused to shake people down?  I get a large stake behind me, we find legitimate business and bad rep them unless they pay up.  What prevents this ?
I don't think these concerns have merit.  As soon as you try to blackmail someone that person will distrust you and makes the situation publically known.  Then if you are trying to go around doing this to everyone, they all distrust you and you seem to distrust everyone and people talk so it's not hard to figure out what's going on here.

I agree having to register first is ridiculous, but that is an issue with the exchanges forcing that requirement.  I suppose that for BTSX waiting 30 days is somewhat reasonable, where for KeyID DAC it would be a killer.  It just seems strange to me.  There will be a certain subset of guys who are actually involved in this process.  The naming thing just isn't important enough to justify this auction type system and all the weird market that'll occur in the background.  It just makes BTSX all that more complicated in a way that doesn't really bring enough value.  The names are meant to aid ease of use, we're going backwards trying to create yet another market within the system.

The person blackmailing doesn't have to be known and can move their stake around at will.  I would need to study the issue more if this isn't true.

This also doesn't solve spoof scammers.  If I understand correctly, those spoof bter type accounts that get random deposits can be negative mined at which point they show up on the market again.  What prevents a scammer from doing the whole thing again?  It is up to bter to try and vote these guys down  ?  I am not sure there is a big incentive from others.
Gamey, you are not understanding these issues at all.

I does solve spoof scammers... The user names never even become active because when someone tries to register them (must pay for this) they have no chance to out-mine the community/exchanges.  Mining has an opportunity cost in that they can't build a reputation on a name they can actually keep and their misbehavior is not rewarded...  Blackmailers can't "hide"; all these transactions are publically auditable... in the worst case for bad actors they risk having their staked forked/removed from the network.

Reputation is not some "needless distraction."  It is important, useful, and a problem that a lot of people have been trying to solve.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: arhag on September 19, 2014, 10:15:48 pm
First, I want to say that I don't think we need trust/reputation and even account names on every DAC. In my view we should keep the account names and the trust/reputation of an account on a single DAC: BitShares DNS. Other DACs should simply refer to the DNS DAC when determining the Account Keys and reputation of an account from a human-readable account name, just like they would have to refer to the DNS DAC to get the IP address and public keys of a .p2p domain.

Okay, that still leaves some trust/reputation mechanisms needed for KeyID accounts. Agent86, I think your proposal is pretty decent, but I have some concerns. First, I don't like that someone needs to wait 30 days to get a novel account name set up. That is putting up unnecessary barriers for new users in my opinion. And, there is already a great name system with a proper auction for names that we want proper discovery on: the .p2p domains. As for KeyID names, I think if someone picks a new account name and people don't think it should be registered, it will be killed by negative devotion fairly quickly anyway (since lifetime would be very small by the time people realize the name was registered). If an account name has been killed, then we can require waiting 30 days before it can be re-registered. During that time you can have people competing with the devotion mining as in your proposal.

By the way, I want to make the quick side remark that I am not very bothered by the idea of account names dying and being reclaimed. Your robohash patch is clever, but I don't even think that is necessary. I personally never liked the robohash idea to begin with (it looks too unprofessional). I would rather rely on adding accounts I communicate with or send money to in my favorites and use that to warn me against typos. And if an account died and was re-registered, the client could automatically warn me when that happens to any of the accounts in my favorites. In the case of adding new contacts to favorites or sending money to a contact not in my favorites, since this is an unusual event, I think relying on registration date and trust level should be enough to prevent mistakes (if the client GUI is designed properly that is). Also with our own version of BIP 70, payments to merchants would be secured with their .p2p name not their KeyID name.

The second concern I have is regarding privacy. While people can use their coin-age to add or subtract devotion of any user. I bet that it is going to be far more likely that any positive devotion is going to be to their own accounts. This links the balances of the user together to their account. And even that wouldn't be such a problem if it wasn't for the fact that you really should use your coin-age to boost your devotion to protect your account from the trolls. I wonder if there can be a technical solution to this using linkable ring signatures (or whatever BitShares Vote is using). You could use signatures to prove ownership of balances that have already been transferred but not yet claimed as coin-age. The sum of the claimed coin-age is rounded down to the nearest N*c coin-age (where N is an integer and c is some fixed coin-age constant). This entitles the recipient to N coin-age votes. Then using linkable ring signatures, the user can disassociate the N coin-age votes from the original signatures granting access to the coin-age votes, and use each of the disassociated coin-age votes to either add to or subtract from any account's devotion. A lot of complication for the sake of privacy. Not sure if it is worth it.


Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 19, 2014, 11:05:58 pm
First, I want to say that I don't think we need trust/reputation and even account names on every DAC. In my view we should keep the account names and the trust/reputation of an account on a single DAC: BitShares DNS. Other DACs should simply refer to the DNS DAC when determining the Account Keys and reputation of an account from a human-readable account name, just like they would have to refer to the DNS DAC to get the IP address and public keys of a .p2p domain.

Okay, that still leaves some trust/reputation mechanisms needed for KeyID accounts. Agent86, I think your proposal is pretty decent, but I have some concerns. First, I don't like that someone needs to wait 30 days to get a novel account name set up. That is putting up unnecessary barriers for new users in my opinion. And, there is already a great name system with a proper auction for names that we want proper discovery on: the .p2p domains. As for KeyID names, I think if someone picks a new account name and people don't think it should be registered, it will be killed by negative devotion fairly quickly anyway (since lifetime would be very small by the time people realize the name was registered). If an account name has been killed, then we can require waiting 30 days before it can be re-registered. During that time you can have people competing with the devotion mining as in your proposal.

By the way, I want to make the quick side remark that I am not very bothered by the idea of account names dying and being reclaimed. Your robohash patch is clever, but I don't even think that is necessary. I personally never liked the robohash idea to begin with (it looks too unprofessional). I would rather rely on adding accounts I communicate with or send money to in my favorites and use that to warn me against typos. And if an account died and was re-registered, the client could automatically warn me when that happens to any of the accounts in my favorites. In the case of adding new contacts to favorites or sending money to a contact not in my favorites, since this is an unusual event, I think relying on registration date and trust level should be enough to prevent mistakes (if the client GUI is designed properly that is). Also with our own version of BIP 70, payments to merchants would be secured with their .p2p name not their KeyID name.

The second concern I have is regarding privacy. While people can use their coin-age to add or subtract devotion of any user. I bet that it is going to be far more likely that any positive devotion is going to be to their own accounts. This links the balances of the user together to their account. And even that wouldn't be such a problem if it wasn't for the fact that you really should use your coin-age to boost your devotion to protect your account from the trolls. I wonder if there can be a technical solution to this using linkable ring signatures (or whatever BitShares Vote is using). You could use signatures to prove ownership of balances that have already been transferred but not yet claimed as coin-age. The sum of the claimed coin-age is rounded down to the nearest N*c coin-age (where N is an integer and c is some fixed coin-age constant). This entitles the recipient to N coin-age votes. Then using linkable ring signatures, the user can disassociate the N coin-age votes from the original signatures granting access to the coin-age votes, and use each of the disassociated coin-age votes to either add to or subtract from any account's devotion. A lot of complication for the sake of privacy. Not sure if it is worth it.
Thanks for taking the time to understand the proposal and thanks for the feedback  :)  I think the comments are reasonable.  I think there is time to iron things out.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: Brent.Allsop on September 19, 2014, 11:07:41 pm

In my opinion it is just stupid to think Alibaba has anything to do with the crash in Total Crypto Currency Market Cap.

In my opinion it is all about uncertainty,  everyone knows capital flees uncertainty.  The more 2.0 currencies there are, the more people are becoming aware of the weakness in Bitcoin.

My prediction is that in 2015, a new currency will overtake Bitcoin with the leading market cap, and that Bitcoin will crash very fast when this happens.  It will only take a few months for the entire industry to switch to the new leader, sending it to the moon, and continuing the "Canonized Law of the Crypto Currency"  http://canonizer.com/topic.asp/154/2 (http://canonizer.com/topic.asp/154/2)

If more experts get involved in these kinds of amplification of the wisdom of the crowd surveys, they become exponentially more intelligent.  And you can see how well people are at predicting by checking their history of what camp they were in.  I know I'm bragging, but you can see from my history that I've been in the right on the money camp, way before anyone else, many times already.  It'd be very cool if anyone could beet me in their predictions, and such would help amplify everyones knowledge about the revolution that is about to happen, so everyone can be expecting, and profiting from it, instead of getting turned off, and loosing money from it.







Title: Re: Reputation & username system - alternative to proof of burn
Post by: gamey on September 19, 2014, 11:55:12 pm
Waiting 30 days to get your username is horrible for adoption.

I'm thinking of the username more as a luxury.  I would want it so it is not necessary to get started and use the system.  I think the current "chicken and egg" situation where you need to ask someone to register you is a little ridiculous.

I want to dislike the idea of large stakes buying trust systems, but I can't think up a good argument against it at the moment.  I would need examples etc.  Can this be abused to shake people down?  I get a large stake behind me, we find legitimate business and bad rep them unless they pay up.  What prevents this ?
I don't think these concerns have merit.  As soon as you try to blackmail someone that person will distrust you and makes the situation publically known.  Then if you are trying to go around doing this to everyone, they all distrust you and you seem to distrust everyone and people talk so it's not hard to figure out what's going on here.

I agree having to register first is ridiculous, but that is an issue with the exchanges forcing that requirement.  I suppose that for BTSX waiting 30 days is somewhat reasonable, where for KeyID DAC it would be a killer.  It just seems strange to me.  There will be a certain subset of guys who are actually involved in this process.  The naming thing just isn't important enough to justify this auction type system and all the weird market that'll occur in the background.  It just makes BTSX all that more complicated in a way that doesn't really bring enough value.  The names are meant to aid ease of use, we're going backwards trying to create yet another market within the system.

The person blackmailing doesn't have to be known and can move their stake around at will.  I would need to study the issue more if this isn't true.

This also doesn't solve spoof scammers.  If I understand correctly, those spoof bter type accounts that get random deposits can be negative mined at which point they show up on the market again.  What prevents a scammer from doing the whole thing again?  It is up to bter to try and vote these guys down  ?  I am not sure there is a big incentive from others.
Gamey, you are not understanding these issues at all.

I does solve spoof scammers... The user names never even become active because when someone tries to register them (must pay for this) they have no chance to out-mine the community/exchanges.  Mining has an opportunity cost in that they can't build a reputation on a name they can actually keep and their misbehavior is not rewarded...  Blackmailers can't "hide"; all these transactions are publically auditable... in the worst case for bad actors they risk having their staked forked/removed from the network.

Reputation is not some "needless distraction."  It is important, useful, and a problem that a lot of people have been trying to solve.

There are going to be multiple spoof scam names for any name.  Some spoof addresses have more value others won't.  Seems to me this is another way to allow a person with a large stake to remove the name of a smaller person.  These assumptions that the community will behave in a certain way need to be examined.  Much like approval voting is great, but realistically the average stake holder can't and will not be aware enough of 101 delegates.

As far as blackmailers, I don't see why people won't be able to mix up who and where they come from.  Nor do I think this would ever become severe enough for anyone to be forked out.

Blackmailers can hide.  Do you think you'd know who a blackmailer is ?  Why would you assume this?  To me it seems quite easy to hide your id in these situations. What is going to keep them from mixing their coins in at some point.  You would seriously consider forking the network over this as a solution ? 

And yes I think making another market for names inside bitshares x will make the product needlessly complicated and convoluted.  Lets see..  we had a system meant to make the whole thing easier to use.  It has the spoof problem albeit minor.  Dan proposes an ok system.  Then you come along and want there to be a 30 day wait to obtain the name while others bid on it in yet another market embedded inside bitsharesx.  How do these 2 objectives fit together ?  People aren't going to want to screw with all that.  Yes it is needless.  We need a somewhat simpler solution and no account name revocation by large stakes.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: luckybit on September 20, 2014, 03:02:10 am
I'm not a big fan of using proof of burn for reputation as planned.  I want to propose an alternative.  I think although these ideas are more complex they are more useful.

Every username would have a birthday and three reputation metrics or scores:
   1) Trust
   2) Devotion
   3) Lifespan
   
Everything builds off the devotion metric which is accomplished by mining your user name via coin-day mining "CDM".  Essentially you can point a balance to a user name to increase (or decrease) it's devotion score proportional to the coin days of the stake directed toward the user name.  This metric is scarce because a given balance cannot mine multiple usernames at the same time.  Multiple balances can however mine a single username concurrently.

The process starts when someone pays a reasonable fee to register a username.  This initiates a 30 day "auction" that is decided by competitive coin-day mining.  If no one else competes for the username or mines against it, than the person who registered it/mined it the most gets it after 30days.  This date is the "birthday" of the user name.

The devotion score for a user name is generally given by the formula: supportive_CDM - opposing_CDM = devotion.

Any username may choose to trust, not trust, or remain neutral toward another username.  The user name's total trust score is determined by the sum of the devotion scores of all usernames who trust it minus the sum of the devotion scores of all usernames who distrust it.

The lifespan for most accounts will be equivalent to the age of the account since the birthdate and goes up with time.  However if a username's total devotion score is brought negative by opposing CDM, the lifespan instead decreases with time for as long as the devotion score remains negative.  If a username's lifespan goes to zero, the user "dies" (username is revoked and it may be re-registered by someone else.)

The robo-hash images can hash both the username and the birthdate so that if a username ever changes hands the robo-hash image also changes.

Most names will have positive metrics for all three and the more positive the better. 

If there is a user you don't trust you can simply distrust the user and encourage others to do the same.  Any name with a negative trust score should typically not be trusted.

If the user is particularly offensive or the username is an attempt to spoof, you can opposition mine the name to destroy it's devotion score and eventually kill it/get it revoked. Any name with a negative devotion score is a huge red flag.  This account is quite likely a spoof/fake name and is on the road to revocation unless they can mine it to bring it positive.  There would be BIG warnings before sending any funds to a username with negative devotion.

The longer you have maintained your username in good standing the longer your lifespan and the longer time you have to react if someone starts opposition mining your username.

During the initial auction if there is more opposing CDM than the highest user competing to own the name then the name is never created.  So if someone tries to register an offensive name we can neg-mine it and it is killed in utero.

If you want to take a real close look at a username you can see specifically who trusts them, who doesn't, how recently was their devotion mined etc.

There is a natural opposition between trust & reputation and anonymity/privacy.  My feeling is that anonymity has generally been overemphasized at the expense of more important things.

It might be better if people can't see who trusts who but instead just have a point and badge system. Connections between people is not necessary to track for this purpose.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: bitder on September 20, 2014, 03:56:50 am
  I think the current "chicken and egg" situation where you need to ask someone to register you is a little ridiculous.

Again check the first proposal here:
https://bitsharestalk.org/index.php?topic=9075.0 (https://bitsharestalk.org/index.php?topic=9075.0)

It should allow us to get rid of "Post here if you want 1 BTSX" threads.

Wouldn't it be simpler to just allow transfers to any address as well as TITAN transfers to a name?
A problem with only supporting TITAN is that the exchanges have a record of all fund transfers by name (which arguably is less private than just allowing transfers to any address where the exchanges can't definitively associate the participant on the other side of the transaction).
i.e. the exchange knows you transferred 50K BTSX to account Fred instead of just knowing that you transferred 50K BTSX to some address that may or may not be yours.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: gamey on September 20, 2014, 04:06:17 am
  I think the current "chicken and egg" situation where you need to ask someone to register you is a little ridiculous.

Again check the first proposal here:
https://bitsharestalk.org/index.php?topic=9075.0 (https://bitsharestalk.org/index.php?topic=9075.0)

It should allow us to get rid of "Post here if you want 1 BTSX" threads.

Wouldn't it be simpler to just allow transfers to any address as well as TITAN transfers to a name?
A problem with only supporting TITAN is that the exchanges have a record of all fund transfers by name (which arguably is less private than just allowing transfers to any address where the exchanges can't definitively associate the participant on the other side of the transaction).
i.e. the exchange knows you transferred 50K BTSX to account Fred instead of just knowing that you transferred 50K BTSX to some address that may or may not be yours.

This is a problem with the way exchanges implement their withdrawals.  It is not a technical problem that needs to be solved on Bitshares X end.  The threads with the small giveaways from registration are just a temporary solution until exchanges start to allow transfer to addresses that are not registered names.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 20, 2014, 04:07:40 am

There are going to be multiple spoof scam names for any name.  Some spoof addresses have more value others won't.  Seems to me this is another way to allow a person with a large stake to remove the name of a smaller person.  These assumptions that the community will behave in a certain way need to be examined.  Much like approval voting is great, but realistically the average stake holder can't and will not be aware enough of 101 delegates.

As far as blackmailers, I don't see why people won't be able to mix up who and where they come from.  Nor do I think this would ever become severe enough for anyone to be forked out.

Blackmailers can hide.  Do you think you'd know who a blackmailer is ?  Why would you assume this?  To me it seems quite easy to hide your id in these situations. What is going to keep them from mixing their coins in at some point.  You would seriously consider forking the network over this as a solution ? 

And yes I think making another market for names inside bitshares x will make the product needlessly complicated and convoluted.  Lets see..  we had a system meant to make the whole thing easier to use.  It has the spoof problem albeit minor.  Dan proposes an ok system.  Then you come along and want there to be a 30 day wait to obtain the name while others bid on it in yet another market embedded inside bitsharesx.  How do these 2 objectives fit together ?  People aren't going to want to screw with all that.  Yes it is needless.  We need a somewhat simpler solution and no account name revocation by large stakes.
For a blackmailer to threaten to distrust you he/she must have a username(s) with a lot of devotion that took time for the blackmailer to build.  So you of course don't send any money and then you are distrusted by the blackmailer's username(s).  You go to the forum to alert people to the behavior.  If the blackmailer can't publicly defend their action, they may quickly have a problem.  If they pull the same stunt on multiple community members, now they are basically screwed.  Everyone will neg mine the blackmailer's username until it has no devotion left.  Now no one cares if he/she trusts them or not because they don't have any devotion and it doesn't affect anything.  In any case the blackmailers gain nothing and this behavior doesn't make any sense.

The spoof-scam names will die quickly and the spoofers will soon give up.

As far as approval voting 101 delegates I already have a method in mind for dynamic delegate number.  You may not like the idea anyway because it might seem too complicated but not everything that seems complicated is pointlessly complicated.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: tonyk on September 20, 2014, 04:21:05 am

There are going to be multiple spoof scam names for any name.  Some spoof addresses have more value others won't.  Seems to me this is another way to allow a person with a large stake to remove the name of a smaller person.  These assumptions that the community will behave in a certain way need to be examined.  Much like approval voting is great, but realistically the average stake holder can't and will not be aware enough of 101 delegates.

As far as blackmailers, I don't see why people won't be able to mix up who and where they come from.  Nor do I think this would ever become severe enough for anyone to be forked out.

Blackmailers can hide.  Do you think you'd know who a blackmailer is ?  Why would you assume this?  To me it seems quite easy to hide your id in these situations. What is going to keep them from mixing their coins in at some point.  You would seriously consider forking the network over this as a solution ? 

And yes I think making another market for names inside bitshares x will make the product needlessly complicated and convoluted.  Lets see..  we had a system meant to make the whole thing easier to use.  It has the spoof problem albeit minor.  Dan proposes an ok system.  Then you come along and want there to be a 30 day wait to obtain the name while others bid on it in yet another market embedded inside bitsharesx.  How do these 2 objectives fit together ?  People aren't going to want to screw with all that.  Yes it is needless.  We need a somewhat simpler solution and no account name revocation by large stakes.
For a blackmailer to threaten to distrust you he/she must have a username(s) with a lot of devotion that took time for the blackmailer to build.  So you of course don't send any money and then you are distrusted by the blackmailer's username(s).  You go to the forum to alert people to the behavior.  If the blackmailer can't publicly defend their action, they may quickly have a problem.  If they pull the same stunt on multiple community members, now they are basically screwed.  Everyone will neg mine the blackmailer's username until it has no devotion left.  Now no one cares if he/she trusts them or not because they don't have any devotion and it doesn't affect anything.  In any case the blackmailers gain nothing and this behavior doesn't make any sense.

The spoof-scam names will die quickly and the spoofers will soon give up.

As far as approval voting 101 delegates I already have a method in mind for dynamic delegate number.  You may not like the idea anyway because it might seem too complicated but not everything that seems complicated is pointlessly complicated.

I like your ideas in this thread A86. They are just missing an approval voting, to decide on wheatear the voting on each account's devotion is indeed done with enough devotion and or honesty... then and there  I believe we will have truly solid system.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: gamey on September 20, 2014, 04:32:13 am

There are going to be multiple spoof scam names for any name.  Some spoof addresses have more value others won't.  Seems to me this is another way to allow a person with a large stake to remove the name of a smaller person.  These assumptions that the community will behave in a certain way need to be examined.  Much like approval voting is great, but realistically the average stake holder can't and will not be aware enough of 101 delegates.

As far as blackmailers, I don't see why people won't be able to mix up who and where they come from.  Nor do I think this would ever become severe enough for anyone to be forked out.

Blackmailers can hide.  Do you think you'd know who a blackmailer is ?  Why would you assume this?  To me it seems quite easy to hide your id in these situations. What is going to keep them from mixing their coins in at some point.  You would seriously consider forking the network over this as a solution ? 

And yes I think making another market for names inside bitshares x will make the product needlessly complicated and convoluted.  Lets see..  we had a system meant to make the whole thing easier to use.  It has the spoof problem albeit minor.  Dan proposes an ok system.  Then you come along and want there to be a 30 day wait to obtain the name while others bid on it in yet another market embedded inside bitsharesx.  How do these 2 objectives fit together ?  People aren't going to want to screw with all that.  Yes it is needless.  We need a somewhat simpler solution and no account name revocation by large stakes.
For a blackmailer to threaten to distrust you he/she must have a username(s) with a lot of devotion that took time for the blackmailer to build.  So you of course don't send any money and then you are distrusted by the blackmailer's username(s).  You go to the forum to alert people to the behavior.  If the blackmailer can't publicly defend their action, they may quickly have a problem.  If they pull the same stunt on multiple community members, now they are basically screwed.  Everyone will neg mine the blackmailer's username until it has no devotion left.  Now no one cares if he/she trusts them or not because they don't have any devotion and it doesn't affect anything.  In any case the blackmailers gain nothing and this behavior doesn't make any sense.

The spoof-scam names will die quickly and the spoofers will soon give up.

As far as approval voting 101 delegates I already have a method in mind for dynamic delegate number.  You may not like the idea anyway because it might seem too complicated but not everything that seems complicated is pointlessly complicated.

It isn't that any of this is "pointlessly complicated".  It is a question of whether there is enough value in the added complexity. I think there is a bit too much expecting people to network and go to forums to discuss all these issues when your system is abused.

I will consider it more in terms of devotion as I didn't take 100% of my time to understand your proposal fully. It does seem more reasonable in that regard.  So perhaps blackmailing can't work.

What are the problems with BM's proposal ?  Is it listed what problems your proposal solves?  His proposal seemed fairly clean to me.  Yes, legit people can be red repped, but at least they keep their account.  Someone can red rep the real bter account, but it just means people will need to double check.  It is far from perfect, but at least it doesn't muddy up creating a registered account and losing the value that was supposed to bring. (ease of use)  - That is my main issue here.  I also do not like accounts being able to be revoked.  Maybe I just want to keep my name and not have to go check that it hasn't been revoked everytime someone sends me something.  What happens if i make an automated website and my account has no devotion ?  All at once it is revoked because I wasn't paying attention and some competitor doesn't like me.  Meh.

Instead of revoked, have the network not allow deposits into the account.  Let them spend but receive no more.  Thats at least far better than revoking and putting the account back in circulation.

When you start talking about revoking names and then relying on 'the community' to protect the bullied, the idea loses interest for me.  These are just account names.  People just work around it with new names when theirs is squatted.  Just like google emails. 

Don't have high expectations from users and the effort they are willing to put forth.  You will consistently be let down with reality.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 20, 2014, 04:33:56 am
I like your ideas in this thread A86. They are just missing an approval voting, to decide on wheatear the voting on each account's devotion is indeed done with enough devotion and or honesty... then and there  I believe we will have truly solid system.
I assume this is sarcastic  ;)   So you don't like the idea?  Do you have some specific thoughts/suggestions?
Title: Re: Reputation & username system - alternative to proof of burn
Post by: tonyk on September 20, 2014, 04:43:41 am
I like your ideas in this thread A86. They are just missing an approval voting, to decide on wheatear the voting on each account's devotion is indeed done with enough devotion and or honesty... then and there  I believe we will have truly solid system.
I assume this is sarcastic  ;)   So you don't like the idea?  Do you have some specific thoughts/suggestions?

No, I actually LOVE it.
If it does not work, we will appoint a comity (or 5) selected to only take decisions on account creation/dilation and last but not lest rank/devotion. The comity members will be selected by combination of delegate votes/and users with high devotion accounts. After selection new comity members will have to wait 3 to 782 days (depending on the exact votes from any specific group), but will be replaceable in the waiting period from any account providing big collateral/and or devotion....


{EDIT} seriously though... dare to share how your new market model truly increases the collateralization? In any way other then in vague terms, and - because it does/you will see, it is significant improvement.
{EDIT2} The above or at your own will - what is the big need, after you uselessly change the market engine in one day, to go and suggest highly complicated and not necessitated by any problem account value system?
Title: Re: Reputation & username system - alternative to proof of burn
Post by: gamey on September 20, 2014, 04:43:58 am

Ok, I've reread it for the third time.  I don't think I'm missing anything.  Maybe blackmailing isn't a viable business model.

This whole system basically means that you have to proactively protect your account by CDM-mining it from the start.  And you better have a large stake.  So you have to mine your account when you create it (which was meant for ease of use initially)   Then you need enough stake to proactively defend it.  And if by some chance you do not have the means to fend of an attacker, you can go to the forums and ask for CDM-defensive backup.

Dan's system just has you register an account and be able to use it.  If you want to use the rep system, then you can.  You are not forced to understand it though to defensively protect your account name.  (The one which you happen to receive funds on.. very important)

So now we are back to another system that adds this whole new layer of effort to keep track of a simple address to send and receive from.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: toast on September 20, 2014, 04:53:14 am
Proof of burn is specifically not called "points" or "reputation" because its not very good for the job. I call it "vanity"...

I think a proper reputation system can't have "global reputation" values (reputation is a function of two individuals). I need to process agent's model more to respond.

Sent from my SCH-I535 using Tapatalk

Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 20, 2014, 06:29:42 pm
Ok, I've reread it for the third time.  I don't think I'm missing anything.  Maybe blackmailing isn't a viable business model.

This whole system basically means that you have to proactively protect your account by CDM-mining it from the start.  And you better have a large stake.  So you have to mine your account when you create it (which was meant for ease of use initially)   Then you need enough stake to proactively defend it.  And if by some chance you do not have the means to fend of an attacker, you can go to the forums and ask for CDM-defensive backup.

Dan's system just has you register an account and be able to use it.  If you want to use the rep system, then you can.  You are not forced to understand it though to defensively protect your account name.  (The one which you happen to receive funds on.. very important)
I think saying you have to "proactively defend" your account is an exaggeration.  You just have to make a transaction that devotes your stake to your account name.  You can then have everything in cold storage while your name gains devotion.  I think of it more as a reputation metric than as "defending" your account.  You are proving that you have invested in the username over time because stake can't be devoted to more than one name.

I don't think anyone is going to attack your username for no reason when you aren't bothering anyone.  I don't think it will be an issue that people will be having their name attacked for no reason.

So now we are back to another system that adds this whole new layer of effort to keep track of a simple address to send and receive from.
Again, I think a username is not needed for simple send and receive functions.  Even if you lost a username, you don't lose your public key.  I think a rep system for user names and getting rid of spoofers is helpful

what is the big need, after you uselessly change the market engine in one day, to go and suggest highly complicated and not necessitated by any problem account value system?
What are the problems with BM's proposal ?  Is it listed what problems your proposal solves?  His proposal seemed fairly clean to me. 
I don't see the current TITAN username system as a raging success:
-There have been a lot of spoof accounts created
-A number of people have been hurt by mistakenly sending funds to an incorrect name
-New user confusion with the "chicken and egg" name registration deal
-Issues with scanning the blockchain, missing info etc.
-Greater barriers and less focus on: Cold storage, offline transaction signing and watch only wallets.
   (I think these things are very important and there should be more effort in this regard)
Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 20, 2014, 06:31:31 pm
I want to say something about privacy vs. reputation tradeoff.

I think one of the most important attributes of these systems is they codify what is ultimately a social consensusā€¦

Let's say tonyk has 10mil BTSX and there is this social consensus that he owns this 0.5% of the network.  Let's say tonyk runs afoul of the law for not renewing his "bitlicense" or some such nonsense and the government demands that tony forfeit his BTSX.  The network/community simply invalidates that stake so the gov't gets nothing, the network happily pays for tony's legal defense and whenever he is out of harms way he freely claims back his remaining balance.

I think this type of thing is way more powerful than this obsession with anonymity and untraceable transactions.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: gamey on September 20, 2014, 06:58:26 pm
Ok, I've reread it for the third time.  I don't think I'm missing anything.  Maybe blackmailing isn't a viable business model.

This whole system basically means that you have to proactively protect your account by CDM-mining it from the start.  And you better have a large stake.  So you have to mine your account when you create it (which was meant for ease of use initially)   Then you need enough stake to proactively defend it.  And if by some chance you do not have the means to fend of an attacker, you can go to the forums and ask for CDM-defensive backup.

Dan's system just has you register an account and be able to use it.  If you want to use the rep system, then you can.  You are not forced to understand it though to defensively protect your account name.  (The one which you happen to receive funds on.. very important)
I think saying you have to "proactively defend" your account is an exaggeration.  You just have to make a transaction that devotes your stake to your account name.  You can then have everything in cold storage while your name gains devotion.  I think of it more as a reputation metric than as "defending" your account.  You are proving that you have invested in the username over time because stake can't be devoted to more than one name.

I don't think anyone is going to attack your username for no reason when you aren't bothering anyone.  I don't think it will be an issue that people will be having their name attacked for no reason.

So now we are back to another system that adds this whole new layer of effort to keep track of a simple address to send and receive from.
Again, I think a username is not needed for simple send and receive functions.  Even if you lost a username, you don't lose your public key.  I think a rep system for user names and getting rid of spoofers is helpful

what is the big need, after you uselessly change the market engine in one day, to go and suggest highly complicated and not necessitated by any problem account value system?
What are the problems with BM's proposal ?  Is it listed what problems your proposal solves?  His proposal seemed fairly clean to me. 
I don't see the current TITAN username system as a raging success:
-There have been a lot of spoof accounts created
-A number of people have been hurt by mistakenly sending funds to an incorrect name
-New user confusion with the "chicken and egg" name registration deal
-Issues with scanning the blockchain, missing info etc.
-Greater barriers and less focus on: Cold storage, offline transaction signing and watch only wallets.
   (I think these things are very important and there should be more effort in this regard)

People run businesses and are screwed with _ALL_ the time on the internet.  It is amazing how many people get DDOS'ed.  I've had the thought that it is the DDOS mitigation companies doing this (possibly indirectly) just to drum up business, because there are times when it doesn't make sense.

So if I start a business that takes deposits at a certain name that has competitors, why wouldn't they attack it ?  Especially if they don't need their CDM elsewhere ?   So now I have to be capitalized just to defend that name.  Weird. Previously I might get negative rep, but my accounting system wouldn't stop working over it.

With your list, I agree with 1 & 2.  I think the rest are just part of development and have largely been fixed.  I might even agree that there is too much effort put into privacy with Titan, but all that work is done.

People do send funds to wrong address in btc, but btsx just makes it easier.  Yet it also makes it possible for me to remember my name.  You want to send me funds?  just send them to 'gamey'.  You might be right that we'd be better off without names, but I think the whole name system should be axed before we go down the rabbit hole of more complexity.  You want the devs to signficantly increase complexity, but then you want them to focus on other things.  Issues with scanning the blockchain etc are due to the dev process and time required to track down all the bugs.  This proposal will just destabilize the code base again as it is being implemented.

It isn't like I am totally against what you are trying to accomplish, but things like this need to be really really thought out on the ways they can be abused before they should even be considered technically.  I'd love to see the exchange spoof names revoked, but it would be very hard to find a way that is agreeable to me to do this.

Also - the far easier solution here is for exhanges to not use named accounts to begin with.  If spoofing is too much of a problem, they should just switch to the regular BTSX address with the checksums.  Problem solved ! :)
Title: Re: Reputation & username system - alternative to proof of burn
Post by: arhag on September 20, 2014, 07:44:32 pm
I don't see the current TITAN username system as a raging success:
-There have been a lot of spoof accounts created
-A number of people have been hurt by mistakenly sending funds to an incorrect name
-New user confusion with the "chicken and egg" name registration deal
-Issues with scanning the blockchain, missing info etc.
-Greater barriers and less focus on: Cold storage, offline transaction signing and watch only wallets.
   (I think these things are very important and there should be more effort in this regard)

I don't view spoof accounts and people sending their funds to the incorrect name as such huge problems that require the community to kill off "bad" account names (I am not saying I am against your lifetime proposal necessarily, I just don't think that is the proper way of handling these issues). In my view people shouldn't be sending money often to people not already on their contacts list. If they are, then it is because they initiated the transfer from either clicking on a link on a website or having their mobile client directly communicate with a point-of-sale device of a merchant. In both cases, the validity of the to address information should in the future be protected by the security of .p2p domain names. Other people they communicate with, send money to, work with, etc. should be added to their contact list. Global unique names are useful because it means that you can get back in touch online with a person you met in real life and were only able to exchange a human-memorizable name (and one that is easily written down). Even this becomes less important as more people have smartphones that can automatically transfer their contact information with near field communication (and a confirmation code on each party's phone to protect against any man-in-the-middle attacks as rare as they might be).

Regarding the chicken egg problem, I see that as the fault of the exchanges that still have not implemented sending funds to a public key for some reason. Anyone know why they haven't done this yet? Anyway your proposal doesn't solve this issue. We need to at least have a transaction fee for registering a name, otherwise people could just spam the network for free.

Regarding your last two points, again this has nothing to do with your proposal, but I do agree that they are important issues to solve. The way I understand missing transaction info (other than ones causes by bugs) is that if you don't back up your exported JSON, you would not be able to recover the recent outgoing transaction information because you wouldn't know which user the outgoing transaction was sent to in order to re-derive the shared secret (someone please correct me if I understood incorrectly). I think it would be nice if the client allowed you to supply a list of contacts that you remember sending money to and it would go through that list to brute force your outgoing transactions to try to recover that info. Cold storage, multisig, and offline transaction signing are all very important security features that are badly needed, and I'm sure the devs are either working on it or will get to it soon enough. Another critical improvement necessary before this can go mainstream are lightweight clients and the ability to get notifications of new received funds through a mail server rather than require the user's client to scan the entire blockchain.


People run businesses and are screwed with _ALL_ the time on the internet. 
...
So if I start a business that takes deposits at a certain name that has competitors, why wouldn't they attack it ? 

In my opinion, if you are running a business you should not use account names as your public facing identity but rather your .p2p name. A BIP 70-like payment protocol can allow customers to know they are sending their money to the right place (they will be able to verify the .p2p domain name in their payment confirmation).
Title: Re: Reputation & username system - alternative to proof of burn
Post by: gamey on September 20, 2014, 08:12:39 pm

In my opinion, if you are running a business you should not use account names as your public facing identity but rather your .p2p name. A BIP 70-like payment protocol can allow customers to know they are sending their money to the right place (they will be able to verify the .p2p domain name in their payment confirmation).

I am not familiar with bip70 enough to comment, but I think there definitely is a point to be made to not use a registered account.  It seems so useful at first, but then you realize the added effort of registering the spoof names or potential losses.  Not even considering this thread's proposal, maybe Bitshares dev's should change around best practices for businesses to not use named accounts.  Names help with offline transactions I suppose, but usually addresses are just a cutnpaste away.  It is people trying to rely on their memory that screws things up and allows spoofers to gain from misdirected transactions.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: tonyk on September 20, 2014, 08:34:09 pm
I don't see the current TITAN username system as a raging success:
-There have been a lot of spoof accounts created
-A number of people have been hurt by mistakenly sending funds to an incorrect name
-New user confusion with the "chicken and egg" name registration deal
-Issues with scanning the blockchain, missing info etc.
-Greater barriers and less focus on: Cold storage, offline transaction signing and watch only wallets.
   (I think these things are very important and there should be more effort in this regard)

So this whole deal is solving 'problems' creating by TITAN ?????????????????????????????????????????????
How about reversing back to not using named accounts.

And NO I do not think 'scanning the blockchain' issues have anything to do with naming accounts, directly.
Title: Re: Reputation & username system - alternative to proof of burn
Post by: Agent86 on September 20, 2014, 08:59:43 pm
I don't see the current TITAN username system as a raging success:
-There have been a lot of spoof accounts created
-A number of people have been hurt by mistakenly sending funds to an incorrect name
-New user confusion with the "chicken and egg" name registration deal
-Issues with scanning the blockchain, missing info etc.
-Greater barriers and less focus on: Cold storage, offline transaction signing and watch only wallets.
   (I think these things are very important and there should be more effort in this regard)
...Regarding your last two points, again this has nothing to do with your proposal, but I do agree that they are important issues to solve...
Yes, of the issues I mentioned, I'm really only trying to solve #1 & #2 with my proposal.

The other issues I'm pointing out wouldn't be such issues if we considered TITAN an advanced feature instead of the only way to use the system.