[Off-topic part] [ANN] Trezor: Bitcoin hardware wallet

[allten]

. . . Mike now goes on to reinvent representative democracy.

Haha, thanks Actually that's an old interest. Here's a paper I wrote about 6 years ago:

https://docs.google.com/document/d/1jidmNJHWAtsPLCUD7EPPm8jOEV93kSXbZOMycqCWOyA/edit?authkey=CN7BnLUG&authkey=CN7BnLUG

It covers how to use secure hardware to build a new kind of democracy where votes can be delegated by topic up a tree of representatives. I think these days people call it "liquid democracy". I never did anything with the idea. The paper talks about smart cards but Trezor style devices are a better fit.

Mike. That was a really neat article!
Truly among great minds. I feel humbled.
This was one of my motivations that got me started on the Bitsafe HW Wallet project.
It's so nice to see it on paper.

[1715026358]

1715026358

[1715026358]

1715026358

[1715026358]

1715026358

[Mike Hearn]

One future direction that would be nice to explore is if Trezor gets more secure storage, it could store enough data to ensure it's not creating double spends.

Combined with a device-specific hardware key (signed by the manufacturer) this would allow a Trezor to create transactions that carry extra information, proving they were created by a genuine device. This can help people using lightweight clients. Currently, to obtain confidence that an unconfirmed transaction is valid, an SPV client must wait to hear it announced by multiple peers on the network.

This works OK unless (1) there's no internet connection or (2) the connection is controlled by an attacker. For instance, a rogue Bitcoin trader who agrees to meet you in a coffee shop, and is in cahoots with the coffee shop owner. You turn up, use the coffee shop wifi and end up connected to fake nodes.

If the transaction you receive was signed by a Trezor that could detect attempts to sign invalid or double-spend transactions, that would be a way to gain confidence in the transaction even though you weren't able to reach the internet.

It's even a way for people on regular full nodes to gain confidence. Note that the tx broadcast message can contain arbitrary data after it. Nodes will relay all the data even though they only understand some of it. Only the actually understood tx data will be included in the block chain, but if you broadcast a transaction that contains a Trezor signature after it, the receive can have more confidence you won't double spend, because that would require you to defeat the hardware security on the device.

Of course, this all assumes the devices can be made sufficiently tamperproof. Perhaps for version 1 this is not a priority, it's sufficient to protect from software based attacks that might be triggered by a virus. But making tamperproof chips is quite an old art and it's definitely applicable to the Bitcoin world.

[2112]

. . . Mike now goes on to reinvent representative democracy.

Haha, thanks Actually that's an old interest. Here's a paper I wrote about 6 years ago:

https://docs.google.com/document/d/1jidmNJHWAtsPLCUD7EPPm8jOEV93kSXbZOMycqCWOyA/edit?authkey=CN7BnLUG&authkey=CN7BnLUG

It covers how to use secure hardware to build a new kind of democracy where votes can be delegated by topic up a tree of representatives. I think these days people call it "liquid democracy". I never did anything with the idea. The paper talks about smart cards but Trezor style devices are a better fit.

Mike. That was a really neat article!
Truly among great minds. I feel humbled.
This was one of my motivations that got me started on the Bitsafe HW Wallet project.
It's so nice to see it on paper.

The whole paper is an expose on how the ultra-naive think about democracy. Here's a short quote:

They can then walk away with their token, secure in the knowledge that it cannot be tied to their real identity.

Then they walk to the party headquarters where they exchange their voting token for one issue of election sausage.

In addition to the above the paper has a whole section entitled "Vote buying" that completely omits the discussion of buying the voting smartcards.

In summary: the whole paper meets the definition of ultra-naivette. May God bless the heart of its author.

[cbeast]

This central-bank-killing-liquid-democracy device may become the killer gadget for Occupiers world-wide.

[2112]

<off-topic>

In addition to the above the paper has a whole section entitled "Vote buying" that completely omits the discussion of buying the voting smartcards.

It does, it clearly states that such risk is irrelevant. Buying individual votes like this is much less efficient than buying politicians today. Plus, the keys could expire once each 5 years or something.

Also, selling your vote is not much different than blindly delegating it, or blindly voting. Do you really expect somebody that easily sells his vote rights like that to even mind about voting at all? What you seem to be criticizing is democracy per se, not his particular instance.

OBS: I'm not sure I support his idea, although I'm inclined to think it might be a good improvement on status quo. A more direct democracy might considerably decrease the effect of "dispersed costs, concentrated benefits", which are pretty much the norm in democracies. But it's still far away from full decentralization of power, which is the "nirvana", let's say.

In summary: the whole paper meets the definition of ultra-naivette. May God bless the heart of its author.

Do you honestly believe that the current widely adopted representative democracy models are so superior? Buying politicians is pretty much the norm on them. I guess you're the naive one there if you think otherwise.

</off-topic>

I made a full quote to protect against further editing.

Buying politicians became an expensive requirement when the previous strategy of buying votes (one vote -> one sausage and/or beer) became hard to enforce by combination of voting in public but casting the actual vote in secret.

Now we are dscussing a supposed improvement where for the cost of one sausage buyer gets all future votes cast with particular token.

I'm just going to quote some select parts from the Wikipedia's article on idiots:

An idiot in Athenian democracy was someone who was characterized by self-centeredness and concerned almost exclusively with private—as opposed to public—affairs. [...] Declining to take part in public life, such as democratic government of the polis (city state), was considered dishonorable. "Idiots" were seen as having bad judgment in public and political matters. Over time, the term "idiot" shifted away from its original connotation of selfishness and came to refer to individuals with overall bad judgment–individuals who are "stupid".

I can muse a little about how IQ is not an one-dimensional quantity. An individual can be a competent software (Mike Hearn) or hardware (allten) designer; and yet completely fail the basic civic class material where grade-school children are taught why everyone is asked to vote in public, but the actual vote marking process is obscured and why nobody is allowed to accompany the voter while marking the vote; not even a husband can assist his voting wife (or wives).

This central-bank-killing-liquid-democracy device may become the killer gadget for Occupiers world-wide.

Again: may God bless your little hearts. The brown-shirted functionaries of our party will allow you your continuous use of your knees just for the cost of letting them watch while you press the buttons on your trezor in the privacy of your house. If you drop your trezor in our party headquarters you will be paid one sausage for your kindness and cooperation in furthering the efficient democracy.

[cbeast]

Again: may God bless your little hearts. The brown-shirted functionaries of our party will allow you your continuous use of your knees just for the cost of letting them watch while you press the buttons on your trezor in the privacy of your house. If you drop your trezor in our party headquarters you will be paid one sausage for your kindness and cooperation in furthering the efficient democracy.

Geez, you think this is some sort of conspiracy or something? The Trezor represents an idea, not just a single device. Bitcoin is an experiment just like democracy was an experiment. The latter failed. Let's just give these little wide-eyed optimists the benefit of a doubt to play their own experiment out. You can sit by and make insults, but it doesn't really add anything except to the ignore files. Besides, there is nothing wrong with buying votes directly if you are not allowed to just print your money though monopolistic powers.

[2112]

This thread has gone a bit off topic (Yes - I know I am guilty too).

Could we get it back to Trezor stuff please.

I support Jim in the request for thread-splitting.

If possible please start with Mike Hearn's post:

https://bitcointalk.org/index.php?topic=122438.msg1466565#msg1466565

but exclude the post by commonancestor:

https://bitcointalk.org/index.php?topic=122438.msg1468915#msg1468915

and change the title to something like: Bitcoiners in support of voter itimidation and vote buying via crypto hardware.

Geez, you think this is some sort of conspiracy or something? The Trezor represents an idea, not just a single device. Bitcoin is an experiment just like democracy was an experiment. The latter failed. Let's just give these little wide-eyed optimists the benefit of a doubt to play their own experiment out. You can sit by and make insults, but it doesn't really add anything except to the ignore files. Besides, there is nothing wrong with buying votes directly if you are not allowed to just print your money though monopolistic powers.

[hazek]

. . . Mike now goes on to reinvent representative democracy.

Haha, thanks Actually that's an old interest. Here's a paper I wrote about 6 years ago:

https://docs.google.com/document/d/1jidmNJHWAtsPLCUD7EPPm8jOEV93kSXbZOMycqCWOyA/edit?authkey=CN7BnLUG&authkey=CN7BnLUG

It covers how to use secure hardware to build a new kind of democracy where votes can be delegated by topic up a tree of representatives. I think these days people call it "liquid democracy". I never did anything with the idea. The paper talks about smart cards but Trezor style devices are a better fit.

Mike. That was a really neat article!
Truly among great minds. I feel humbled.
This was one of my motivations that got me started on the Bitsafe HW Wallet project.
It's so nice to see it on paper.

The whole paper is an expose on how the ultra-naive think about democracy. Here's a short quote:

They can then walk away with their token, secure in the knowledge that it cannot be tied to their real identity.

Then they walk to the party headquarters where they exchange their voting token for one issue of election sausage.

In addition to the above the paper has a whole section entitled "Vote buying" that completely omits the discussion of buying the voting smartcards.

In summary: the whole paper meets the definition of ultra-naivette. May God bless the heart of its author.

I didn't even read the paper but I already know I'd mostly likely agree with the above poster. (well except with the very last sentence  )

Mike Hearn I appreciate your coding work but man do I hope you never end up in any position of influence or decision making because your convoluted worldviews actually scare me. I can't even understand how a coder, working with logic and math everyday, can fall victim to such an obvious and blatant fallacy as is the belief in a democratic system as the system we should strive to improve and implement for our benefit. That just baffles me.  


I apologize for my off topic post but I was very concerned with Mike for a long time now and I just couldn't hold it back anymore and had to get it off my chest. I just can't help it but psychopathic statists scare me.

[hazek]

I didn't even read the paper but I already know I'd mostly likely agree with the above poster. (well except with the very last sentence  )

Mike Hearn I appreciate your coding work but man do I hope you never end up in any position of influence or decision making because your convoluted worldviews actually scare me. I can't even understand how a coder, working with logic and math everyday, can fall victim to such an obvious and blatant fallacy as is the belief in a democratic system as the system we should strive to improve and implement for our benefit. That just baffles me.  

I've seen many of your posts and I'm aware you're a libertarian. So, I wonder (honestly, this is not an irony), why do you think representative democracy is less worse than a more direct one?

The reason is exceptionally simple: I do not agree to being governed by anyone without my contractually arranged explicit consent and so I do not and ever will agree to voting. The whole concept to me is repugnant and I could never understand how anyone could think they may force me or anyone else to do something or follow some rules just because they took a vote and some majority of some people somewhere voted so.

I apologize for my off topic post

Why don't you use your admin privileges to create a new thread with the posts talking about this "delegated democracy" thing, and remove them from here?
The topic is interesting, but it indeed has no place here.

My moderator's privileges extend only to two sections: Bitcoin discussions and Economics.

[CIYAM]

Had hoped to hear more about the Trezor - but *sigh* about time to *unwatch* this derailed thread.

[Mike Hearn]

Yes, guys, please feel free to start a new thread for that discussion in the Politics & Society forum. Otherwise we'll need to get a moderator to move the posts out of this thread manually which would be a shame.

[stick]

Yep, please, stick to the topic and don't hijack the discussion.

[cbeast]

You don't have to do this now. As the software is open source, somebody could contribute such a protocol and then you incorporate it into your signed builds.

Yes, this is very nice concept but currently out of our project's scope. As you write, once the Trezor is released, it is a matter of rather small firmware modification ...

Is this going to have a re-writeable firmware? I thought it was going to be fused to allow only a single access to the memory for backup purposes. Maybe I'm thinking of another project.

[caveden]

Otherwise we'll need to get a moderator to move the posts out of this thread manually

I was actually expecting this to happen.
As it didn't I've done it myself by quoting posts here: https://bitcointalk.org/index.php?topic=138169
I'll delete my off-topic posts of this thread. It'd be nice if other posters quoted on that thread would do the same with theirs.

[2112]

No new information, I'm just quoting the offending posts against further deletions by authors, who may try to follow caveden's & jim618's actions.

Have you considered adding support for a generic trusted path protocol so any server can send a message and get back an A/B answer in a secure manner?

Do you have any proposal how this should work? Actually device is able to sign bitcoin transactions (pay to address, pay to script hash) and sign bitcoin messages.

Yes. Consider the case of online voting. The voting server wants to ensure the following

a) The vote is cast by a human, not a virus.
b) The vote is cast by each human only once.

We can satisfy this use case by using Trezor like this:

• Each Trezor has a unique public key (can be RSA) and a certificate signed by the manufacturers, ie, you and stick.
• When you start the voting process, the Trezor sends its certificate to the computer, which then sends it on to the voting server.
• The server checks that this certificate has not been seen before, and then encrypts/signs a protocol buffer that includes a message to display onscreen (any arbitrary text) and the two options available. That signed message gets sent to the Trezor, which shows it on screen.
• The users answer is then signed with the devices private key and uploaded to the voting server which checks it against the public key in the certificate.

In this way, the server can know that no virus interfered with the communication and there is no fake device at work. Of course, it assumes that the devices are somewhat secure against physical tampering.

If you can already sign/verify text messages in the standard way, then I guess you could add it on top of that, though the need to communicate certificates and things makes it not really necessary to re-use something designed only for text.

You don't have to do this now. As the software is open source, somebody could contribute such a protocol and then you incorporate it into your signed builds.

. . . Mike now goes on to reinvent representative democracy.

Haha, thanks Actually that's an old interest. Here's a paper I wrote about 6 years ago:

https://docs.google.com/document/d/1jidmNJHWAtsPLCUD7EPPm8jOEV93kSXbZOMycqCWOyA/edit?authkey=CN7BnLUG&authkey=CN7BnLUG

It covers how to use secure hardware to build a new kind of democracy where votes can be delegated by topic up a tree of representatives. I think these days people call it "liquid democracy". I never did anything with the idea. The paper talks about smart cards but Trezor style devices are a better fit.

Not content with:
+ redefining the nature of money, see bitcoinj
+ reimagining the meaning of property rights, see video presentation of Bitcoin London 2012

. . . Mike now goes on to reinvent representative democracy.

Please, for the love of God, can someone stop this guy before he redefines the very essence of reality and/ or space-time and we all become - well - the mind boggles.

:-)

yo jim, don't give him ideas!!