Bitcoin's first industrial purpose

[the founder]

This is a breakoff from this thread:

you want the feds off your back? 

Treat bitcoins like Silver and not gold.

Silver is used as currency AND x-ray machines AND parts for your electronics AND solar panels... etc...

Gold is just a currency for the most part.

If we had bitcoins used for 4000 differing things,  like DRM or whatever.. it would completely diffuse why people use bitcoins into a much wider pool... it also would solidify the fact that bitcoins are a commodity not a currency...  and hence completely sidesteps many of the issues that we are running into (and will run into).

https://bitcointalk.org/index.php?topic=58616.msg696154#msg696154



Volia,  2 days later.... 

Clark and his colleague Aleksander Essex at the University of Waterloo, also in Ontario, realised they could convert a message - for example, a list of codes that securely link voters to their votes - into a Bitcoin address. Sending a tiny fraction of a bitcoin - a small transaction - to that address would allow the holder of that list to store it in the public record without revealing its contents. When they later publish the message for verification, anyone can repeat the conversion to a Bitcoin address and confirm its age by checking the public record.

Faking Bitcoin's public record would be very difficult as you'd need more computing power than the rest of the Bitcoin network combined - a feature that ensures the currency's security.

The pair have used their method, known as CommitCoin, to close a loophole in a voting system they helped develop. In the Scantegrity system, voters receive a confirmation code from the list that is cryptographically linked to their selected candidate and can be used to check on the election website that their vote is counted.

Now, if an unscrupulous election official tries to change votes they would be outed, because the code used to record the vote would change, and would not match up with the BitCoin network entry. "CommitCoin allows you to not trust anyone," says Clark.


http://www.newscientist.com/article/mg21328476.500-bitcoin-online-currency-gets-new-job-in-web-security.html

[1714187088]

1714187088

[Steve]

That's very cool.  I just read about CommitCoin as well…also very cool.  Your point about alternative uses of the block chain being important for bitcoin is a good one as well.

In reading about CommitCoin, a thought occurred to me…sending a small amount of coins to one of these hash addresses loses those coins…however, I wonder if it's possible to include an output of a transaction with 0 coins going to it…all the other coins can just be sent back to a change address.  With a transaction fee, the transaction itself shouldn't be considered spammy (and a transaction fee for the miners would be quite appropriate considering you're using their services to put a timestamp on something).

Excellent stuff.

[Fizzgig]

There is a bitcoin raffle that relies on the blockchain. This adds backing/value to the blockchain even though the application is 100% bitcoin related. Any dependence on the blockchain adds value to the blockchain I think!

[bb113]

Electronic voting is a huge problem. I haven't verified yet but on the radio today a lady was saying how most US votes get sent to some company in Florida, and then from there overseas to Italy. Scary stuff. Who would be held accountable if some kind of fraud occurred using CommitCoin though? I haven't read the technical details, but neither will most people. So they still need to trust at a minimum, the CommitCoin developers and some other programmers.

[Hunterbunter]

How awesome.

Does anyone know in more detail what they're actually doing? I'd like to understand this idea.

[Mark Oates]

Who would be held accountable if some kind of fraud occurred using CommitCoin though?

Assuming CommitCoin is decentralized (haven't looked at it,) I would imagine that the band of super nerds would notice the fraud and expose it, much the same way we could expect a rogue blockchain to be exposed by a similar process.

[bb113]

Most people will want someone to blame if the new idea goes wrong. I think the adoption of it also needs to be decentralized which would require a hybrid transition period. Voters are given a choice as to whether or not to encode their vote, the traditional methods of counting are used as well. Honestly I don't see why paper and electronic ballots (and CommitCoin) can't be used at the same time, then the results compared to each other. Counting votes is important.

[the founder]

in my humble opinion,   every single use that depends on the blockchain WILL increase value to bitcoin,   need an effective timestamp,  use the blockchain,  need a way to prove that a vote went to this guy and not that one,  use the blockchain,  need to transfer money and not have to say "check is in the mail" use the blockchain.

You get my point... 

Eventually bitcoins would be used all over the place, each of them placing more and more value on bitcoin.

 

[farfiman]

in my humble opinion,   every single use that depends on the blockchain WILL increase value to bitcoin,   need an effective timestamp,  use the blockchain,  need a way to prove that a vote went to this guy and not that one,  use the blockchain,  need to transfer money and not have to say "check is in the mail" use the blockchain.

You get my point... 

Eventually bitcoins would be used all over the place, each of them placing more and more value on bitcoin.

 

How will using the blockchain but not bitcoins themselves change the value of the coins?
Or do you mean bitcoin as a technology ?

[Elwar]

While I love the idea of being able to verify your vote...it has not been verifiable for a reason.

The ability to verify your vote opens the door to being able to prove that you voted one way or another.

Being able to prove that you voted one way or another opens the door to someone paying you to vote one way or another and then being able to show proof that you did so.

Or even worse, a government with enough power to have a vote and then "check" to make sure you voted "correctly".


The ideal voting machine is an open source software that allows you to vote in the voting booth, then get an easily read receipt showing clearly how you voted. You take that receipt over to a box and you put that receipt into the box.

You have an electronic vote that is cast with a back up ballot box that can verify that vote.


Sure, you may say "paying someone to vote is illegal" and try to fix the problem there. But tampering with votes is already illegal, so why would you need to come up with a better system if you trust the law to take care of such things?

[Elwar]

That's very cool.  I just read about CommitCoin as well…also very cool.  Your point about alternative uses of the block chain being important for bitcoin is a good one as well.

In reading about CommitCoin, a thought occurred to me…sending a small amount of coins to one of these hash addresses loses those coins…however, I wonder if it's possible to include an output of a transaction with 0 coins going to it…all the other coins can just be sent back to a change address.  With a transaction fee, the transaction itself shouldn't be considered spammy (and a transaction fee for the miners would be quite appropriate considering you're using their services to put a timestamp on something).

Excellent stuff.

Sending .00000001 BTC would not be much of a loss. You could have a full nationwide election and only lose 1 BTC. This adds 1BTC worth of value to the rest of the Bitcoins in use.

And the divisibility can always be extended down the line as has been brought up numerous times.

[DeathAndTaxes]

While I love the idea of being able to verify your vote...it has not been verifiable for a reason.
The ability to verify your vote opens the door to being able to prove that you voted one way or another.
Being able to prove that you voted one way or another opens the door to someone paying you to vote one way or another and then being able to show proof that you did so.

No it doesn't.  Hashes are one way functions.

It would be like saying if people know public address they can recreate the private key and take the funds.

[Elwar]

While I love the idea of being able to verify your vote...it has not been verifiable for a reason.
The ability to verify your vote opens the door to being able to prove that you voted one way or another.
Being able to prove that you voted one way or another opens the door to someone paying you to vote one way or another and then being able to show proof that you did so.

No it doesn't.  Hashes are one way functions.

It would be like saying if people know public address they can recreate the private key and take the funds.

In the Scantegrity system, voters receive a confirmation code from the list that is cryptographically linked to their selected candidate and can be used to check on the election website that their vote is counted.

So, is this saying that you can verify that your vote was counted but not what your vote was?

[westkybitcoins]

I think any "industrial" use of bitcoins is a good idea, provided fees are included, and there's not too much bloat. (Otherwise, an alternate chain could handle it, and I still think it would just strengthen Bitcoin by confirming the validity of the design.)

This particular use of bitcoins sounds quite promising.

[the founder]

you don't want,  nor could you really have an alternate chain do it...  you need the robust computing power that bitcoin already has...  start forking it and you'll have 3 guys mining (most likely owned by the voting booth company) and controlling everything.

You really have to use bitcoin.

[casascius]

Someone could already offer a CommitCoin service without having to burn any coins, the same way Merged Mining works for namecoin.  Merged Mining adds one extra field to the bitcoin block - which is the hash of the commit used for namecoin.  No reason it has to be just namecoin.

Miners can bury extra hashes in their blocks, and those hashes could refer to anything.  They could be hashes of hashes of hashes.  Luke and his mining pool could offer "commitment" services, and sell an unlimited number of "commits" all day long, without needing to send them to the block chain.  He merely sends the hash of all the commits to the block chain, so one hash in a block could stand for one commit, or a million.

[westkybitcoins]

you don't want,  nor could you really have an alternate chain do it...  you need the robust computing power that bitcoin already has...  start forking it and you'll have 3 guys mining (most likely owned by the voting booth company) and controlling everything.

You really have to use bitcoin.

Good point.

Few things would ruin the introduction of a new voting method faster than finding it's been taken over power-wise within days of operation.

[DeathAndTaxes]

Someone could already offer a CommitCoin service without having to burn any coins, the same way Merged Mining works for namecoin.  Merged Mining adds one extra field to the bitcoin block - which is the hash of the commit used for namecoin.  No reason it has to be just namecoin.

Miners can bury extra hashes in their blocks, and those hashes could refer to anything.  They could be hashes of hashes of hashes.  Luke and his mining pool could offer "commitment" services, and sell an unlimited number of "commits" all day long, without needing to send them to the block chain.  He merely sends the hash of all the commits to the block chain, so one hash in a block could stand for one commit, or a million.

Except now where is the list of commits stored to verify against the single hash?

[casascius]

Except now where is the list of commits stored to verify against the single hash?

Anywhere else on the internet.

Such a scheme uses the block chain as a means to certify a list published somewhere else, a function not co-dependent on what amounts to using the block chain as a cloud hard drive.

The hashes themselves aren't necessarily secret, as they are useless without knowing what they are of.  One could publish the list, and include only the tree root in the block chain.  Unlike Bitcoin, where all transactions are needed to verify the complete history, if the list of commits were lost, the results would not be catastrophic system-wide - it would merely deprive the users of the commitment function the ability to prove their commitment.  A risk they could remedy simply by keeping a copy of the list so it doesn't get lost.

[phillipsjk]

I am skeptical this would work. I fail to see how they can gaurantee each code corresponds to exactly one voter.

Since the voters are unlikely to compare codes, the is little risk in assinging duplicates if you vote for the "wrong" candidate, and extras for the "right" candidate.

Computers are too insecure for e-voting, and likely will be for generations.