[2014-01-26] 4 New Bitcoin Features Revealed by Core Developer Mike Hearn

[LiteCoinGuy]

4 New Bitcoin Features Revealed by Core Developer Mike Hearn

http://www.cryptonews.biz/4-new-bitcoin-features-revealed-by-core-developer-mike-hearn/

[jinni]

This is great.

Is there a timeline?

[ArticMine]

So if one has dual citizenship and can consequently have more that one passport one can run more than one node. 

[dave111223]

So if one has dual citizenship and can consequently have more that one passport one can run more than one node.  

I have two passports, my kids have 3 passports each, so I can run 8 nodes on a wifi network attack and you'd think you were connected to the network, so this passport concept is flawed from the start.

Not to mention the fact that if you go to any factory around here you can guarantee the manager has like 100 Cambodian passports in a stack (so the employees can't run away).  There are probably some factories will access to literally thousands of passports.

Not to mention if I traveled to somewhere like Nigeria with a pocket full of cash...how many passports could I scan in 10 minutes?

[Peter Todd]

So if one has dual citizenship and can consequently have more that one passport one can run more than one node.  

I have two passports, my kids have 3 passports each, so I can run 8 nodes on a wifi network attack and you'd think you were connected to the network, so this passport concept is flawed from the start.

Not to mention the fact that if you go to any factory around here you can guarantee the manager has like 100 Cambodian passports in a stack (so the employees can't run away).  There are probably some factories will access to literally thousands of passports.

Not to mention if I traveled to somewhere like Nigeria with a pocket full of cash...how many passports could I scan in 10 minutes?

It's one of those ideas that rather than keeping the bad guys out, discourages the good guys from contributing - "I'm not trusting this random program with my passport. How do I know this crypto moon-magic actually works?" - while giving a way for the bad guys to easily use their resources to buy fake legitimacy.

Much better to design systems where the bad guys can't do any harm in the first place.

[Paladin69]

Are people upset at Mike Hearn over these features?  Why?

[ArticMine]

Are people upset at Mike Hearn over these features?  Why?

This proof of passport idea is a non starter. It is also the responsibility of members of the community to constructively point out possible flaws with these kind of ideas. In this particular case this those of us who have multiple citizenships, and are as a consequence are eligible for multiple passports, have a particular responsibility to bring this issue to the attention of the community.

[420]

article 2 days old from
http://www.cryptocoinsnews.com/2014/01/24/4-new-bitcoin-features-revealed-mike-hearn/

[Mike Hearn]

The article doesn't actually mention ZKPOPs anywhere, by the way!

[dave111223]

The article doesn't actually mention ZKPOPs anywhere, by the way!

It's in the video (at the bottom).

RE: Passports I'm not trying to cramp your style; I'm just saying they are certainly not as sacrosanct as you may think.

There is really nothing in this world that everyone has one;  even if you based it on fingerprints or something there would still be people willing to sell the usage of their fingerprints.

[crazyalfy]

Surely "Proof of Stake" would be optimal instead of "Proof of Burn" or "Proof of Passport"? It costs nothing and is intrinsic to the network itself.

Proof of Stake would allow for verifiable proof of coin controlled and is impossible to duplicate without being noticed. Stake size for anonymous nodes would be a good measure of a nodes trust-ability.

[Mike Hearn]

I don't see why there is any correlation between owning lots of Bitcoin and running a node. Especially, proving stake requires signing and you don't want private keys to a large wallet on a remote server.

[Mike Hearn]

Oh, to answer the first question - timelines for the features discussed in the article.

HD Wallets are being implemented in bitcoinj and Electrum right now. As bitcoinj is a library, once done that will flow through into the Android Bitcoin Wallet app (by Andreas), MultiBit, Hive, MacWallet and Bitcoin Touch. I am hoping HDW will be ready for the bitcoinj 0.12 release which optimistically might come out end of March or April. I expect wallet apps to catch up soon after, probably within a 4-6 weeks. I don't know what the timeline is for Electrum, ask Thomas.

The payment protocol reference implementation in Bitcoin Core (the new name for Bitcoin-Qt) is already done and will ship with v0.9, which is likely to come out in February. Merchant adoption is a bit of an unknown, BitPay definitely want to do it and have some code done internally but right now they're drowning in customer demand for their existing service and that's taking up all their development time. Other payment processing companies also committed but I guess it'll take a while after wallets start supporting it for them to catch up.

Tor in bitcoinj - Miron said he'd work on it in February when I am on vacation. I am hopeful it will ship with bitcoinj 0.12 like HD wallets will.

Proof of sacrifice/passport, no plans to implement these and thus no timeline.

[crazyalfy]

As the network stands now, all nodes are treated equally. Why should this always be the case?

There is nothing currently stopping anyone from running multiple nodes, which is fine, provided no significantly large number of nodes falls under any particular entities control (colluding nodes are viewed as an entity).

My point is that if I (or any entity) wishes to support the network with multiple nodes (even over tor) I should be allowed to. What is required is a disincentive to be dishonest. Requiring proof of passport doesn't remove the incentive for dishonesty. It merely creates an artificial and arbitrary limit on node creation.

Proof of stake requires economic commitment to the Bitcoin ecosystem, much like proof of work does. This makes it prohibitively expensive to build nodes to create a Sybil attack, exactly like proof of burn would, but without the cost. It's impossible to fake and can (if required) have a strictly enforceable 1-1 relationship with a node.

It also provides a powerful disincentive to generate dishonest nodes. Coins in addresses used in a proof of stake by cheating nodes may be voluntarily neglected by miners. The opposite is also true, and co-operative nodes with honest and reliable reputations may receive donations or prioritised transactions.

Agreed, having private keys on a remote node with a large balance isn't a good idea. But there are solutions to this. For example: A large balance offline wallet could be used to publicly authorise an empty address as a proxy for the offline wallet.

We all know Bitcoin is supposed to be decentralised, but just as importantly: It's supposed to be a free market solution. It's the reason Satoshi ultimately wanted artificial and arbitrary limits removed from the system.