[ SNAPSHOT: 8/21 ] VOTE

[vikram]

A few comments on the whitepaper:

• I think it should be made clear in the whitepaper that account verifiers not only need to verify the demographic credentials of a user's account, but they also need to make sure that the account belongs to a user that has not already been counted. The paper explicitly mentions how the DAC does not allow a single user account to request multiple ballots for a particular election and how this prevents voting more than once in an election. But that assumes that the user cannot create another account for which he gets its demographic credentials verified again. Thus, to actual prevent multiple votes in an election, each account verifier needs to keep track of all the human beings they already verified and make sure before verifying another account that an account does not belong to a person that they already verified.
  
  
• It would be nice if the whitepaper went into a little bit of detail on how the zerocoin technology provides anonymity so that user's known accounts are not linked with their votes. I'm thinking metaphors not the actual cryptography/mathematics.
  
  
• I think another point that can be added to the list of benefits is flexibility in voting method. As far as I see it the VOTE DAC blockchain and network should be agnostic to the actual contents of the ballot. Anyone going through the public vote record on the blockchain who understands the rules of the election can independently determine whether a particular ballot is malformed, voting for a nonexistent candidate, or otherwise not appropriate for inclusion in the election. This allows the clients to be updated to create and interpret any future format of a ballot desired. We could have plugins added to client for all kinds of voting methods (first-past-the-post, approval, instant runoff, reweighted range voting, etc.). In fact, in order to reduce blockchain bloat, I think it would make sense to just include the hash of the ballot on the blockchain and allow anonymous uploading of the contents of a ballot on some other server.
  

These are good ideas and particularly for #3 you might want to make a separate thread for discussion.

[tonyk]

Thanks.

[toast]

The majority of the work is in validating users... as the network grows the number of validated users grows and *THIS* is where the value is.  A new chain would have no validated users.

Am I the only one not seeing how I, as a share holder, profit from that? Help please.

When the government wants to do an election, it pays some contractors to set up the voting booths and collect votes. Instead, it will buy enough VOTEs for all their citizens to vote for free. Same goes for any org that wants to vote.

[tonyk]

The majority of the work is in validating users... as the network grows the number of validated users grows and *THIS* is where the value is.  A new chain would have no validated users.

Am I the only one not seeing how I, as a share holder, profit from that? Help please.

[arhag]

I am still failing to see investment value here.

What is stopping someone from copying the source code, then launching their own chain? Voting is based on districts/geographical areas.  The fact that Bitshares users are spread across the globe gives no geographic monopoly. There are no global elections. So this doesn't seem practical unless you are going to do surveys.

This is how I see blockchain tech combining with voting:

Every election will have to be a new chainlike btsx. Whether it's your local city or county, up to countrywide elections.

This new chain will have opposing candidates listed in place of an assets. Instead of bitGold, you have Romney/Obama token as an "asset".  Then another would be prop. XXX as an "asset".

Each registered voter in that jurisdiction or district will be assigned an address in the genesis block with 1token of each "asset"

They will then send each token to the candidates specified address. i.e. you send your Romney/Obama token to Obama's public address. You send your propXXX to the Yes address or vice versa and so on.


Not sure TITAN would work for this. It would have to be open and visible.

The majority of the work is in validating users... as the network grows the number of validated users grows and *THIS* is where the value is.  A new chain would have no validated users.

I'm not sure if that is true. I mean, I agree that the majority of the work is in validating users, but a cloned Vote DAC could always piggyback off the hard work of the account validators, assuming they trust the validators. I don't think there is any lock-in network effect in the DAC. The network effect is in the validator, which is a centralized firm.

And frankly, it may not be so horrible if there isn't any lock-in effect in the DAC. I rather have the fixed cost of validation be made useful on as many Vote DAC clones as possible. Cloning the DAC will allow for maximum scalability. If transaction costs get too high on one DAC, just clone it and reduce the transaction burden. The transaction fees will obviously need to be high enough to motivate delegates to operate the network in the first place (pay their computing costs and give a decent enough profit to make it worth their time), but I think the lack of network effect would reduce the profit margin of this DAC to near zero.

The only issues are funding development and marketing. Hopefully the inflation model can sustain that enough until it succeeds. If we can manage that, then it doesn't matter if this doesn't become a high market cap DAC. It would still be sustainable. It would have low transaction fees so that voting isn't expensive. Users would likely have to cover the costs of their own validation, which then becomes public information. And it would still be an incredibly useful tool for society.

[santaclause102]

Voting from at home would be especially nice for direct democracy decision making. There are a ton of things (plebiscite) to vote on each year here and I often am too lazy to go to the next voting office to vote, which is just possible on sundays.

Follow my vote probably knows sites like this http://www.avaaz.org/de/ (petitions verified by email addresses and Names) already. If not they should definitely take a look!

[bytemaster]

I am still failing to see investment value here.

What is stopping someone from copying the source code, then launching their own chain? Voting is based on districts/geographical areas.  The fact that Bitshares users are spread across the globe gives no geographic monopoly. There are no global elections. So this doesn't seem practical unless you are going to do surveys.

This is how I see blockchain tech combining with voting:

Every election will have to be a new chainlike btsx. Whether it's your local city or county, up to countrywide elections.

This new chain will have opposing candidates listed in place of an assets. Instead of bitGold, you have Romney/Obama token as an "asset".  Then another would be prop. XXX as an "asset".

Each registered voter in that jurisdiction or district will be assigned an address in the genesis block with 1token of each "asset"

They will then send each token to the candidates specified address. i.e. you send your Romney/Obama token to Obama's public address. You send your propXXX to the Yes address or vice versa and so on.


Not sure TITAN would work for this. It would have to be open and visible.

The majority of the work is in validating users... as the network grows the number of validated users grows and *THIS* is where the value is.  A new chain would have no validated users.

[donkeypong]

Elections are extremely costly. If VOTE can get a few small takers and demonstrate through experiments that this can work, then it can make a positive impact while also appealing to larger entities. Initially, this DAC will not be a billion dollar industry, but over time it could grow to fill an increasing demand.

[Ggozzo]

I am still failing to see investment value here.

What is stopping someone from copying the source code, then launching their own chain? Voting is based on districts/geographical areas.  The fact that Bitshares users are spread across the globe gives no geographic monopoly. There are no global elections. So this doesn't seem practical unless you are going to do surveys.

This is how I see blockchain tech combining with voting:

Every election will have to be a new chainlike btsx. Whether it's your local city or county, up to countrywide elections.

This new chain will have opposing candidates listed in place of an assets. Instead of bitGold, you have Romney/Obama token as an "asset".  Then another would be prop. XXX as an "asset".

Each registered voter in that jurisdiction or district will be assigned an address in the genesis block with 1token of each "asset"

They will then send each token to the candidates specified address. i.e. you send your Romney/Obama token to Obama's public address. You send your propXXX to the Yes address or vice versa and so on.


Not sure TITAN would work for this. It would have to be open and visible.

[Ggozzo]

So someone emails a copy of his ID and tells the "verifier authority" that he wants the ID to be associated with a certain public key? That would be a lot of manual work? Any way (proposed) to automate/scale it?

have you ever participated in counting out the ballots? THAT is hell of a lot of work.
Getting your pubkey/i.e. account name (ONCE) shouldn't be too difficult. But sure the system needs a lot of 'keep-it-simple' user interfaces .. I'd too it much like a post-ident in germenay.

1.) create your pubkey
2.) print out a specialized form (includes your name, id numer and the pubkey) -- all in one QR-code
3.) go to your authority (local "Einwohnermeldeamt") and let them scan/check and sign your request

pretty much what is well kown from SSL's Certificate Signing Request (CSR)

Voters don't do the work of counting ballots. Volunteers do that.

[xeroc]

So someone emails a copy of his ID and tells the "verifier authority" that he wants the ID to be associated with a certain public key? That would be a lot of manual work? Any way (proposed) to automate/scale it?

have you ever participated in counting out the ballots? THAT is hell of a lot of work.
Getting your pubkey/i.e. account name (ONCE) shouldn't be too difficult. But sure the system needs a lot of 'keep-it-simple' user interfaces .. I'd too it much like a post-ident in germenay.

1.) create your pubkey
2.) print out a specialized form (includes your name, id numer and the pubkey) -- all in one QR-code
3.) go to your authority (local "Einwohnermeldeamt") and let them scan/check and sign your request

pretty much what is well kown from SSL's Certificate Signing Request (CSR)

[santaclause102]

So someone emails a copy of his ID and tells the "verifier authority" that he wants the ID to be associated with a certain public key? That would be a lot of manual work? Any way (proposed) to automate/scale it?

[xeroc]

How do you make sure you tie one private key to one real world identity? Or is that not part of the solution? I cross read the whitepaper and couldn't figure out how the problem is solved you described initially...

My understanding of that issue on the white paper is that your private key gets verified (i.e. signed) by an authority .. seeing the second image on page 9 in the white paper it states 'Identity Verifyer' and is connected to the privkey and the account(pubkey)

I guess it'll be pretty much what SSL vertificates make them trustworthy (a signature from a CA)

//edit: if my understanding is correct .. then they need a way to revoke the verification once s.o. looses its privatekeys (i.e. by theft).

//edit2:
from the whitepaper page 10

these   account   holders   are   users   that   wish   to   serve   as   
account   verifiers   to   validate   certain   user   credentials   (i.e.   the   U.S.   government,   
Republication   Party,   Democratic   Party,   etc.).   Once   verified,   these   account   verifiers   
will   confirm   the   user’s   credentials   inside   of   the   system,   at   which   point   these   
credentials   will   officially   be   associated   to   the   user’s   account.   

ps: quoting pdf sucks

[santaclause102]

How do you make sure you tie one private key to one real world identity? Or is that not part of the solution? I cross read the whitepaper and couldn't figure out how the problem is solved you described initially...

[bytemaster]

I don't get the business model, why should we except people will want to pay to vote?

Not only that, but how do you control the jurisdictions?

Account identities can be validated by 2 or more competing parties that each certify global uniqueness, location, and any other qualifications necessary to filter.