Couple of things I just thought about:
1- Presume server should only sweep with two (or more, see below) revocation certificates being present
2- Need to insert something in the flow so that Alice can verify that the uploaded key is actually Bob's (and perhaps vise-versa, given an extremely dedicated attacker with a fast connection?).
Is there a way to do #2 without creating yet another transaction? Admittedly I am still really puzzled about the accessibility of public keys in Bitcoin!
Please remember that the idea is to have two wallets securely exchange a packet of metadata about a transaction beyond the scope of Bitcoin itself (a name, perhaps a small photo, etc) in order to increase usability. This will be my last post here on the topic except to reply in case anyone else contributes.
On Sep 16, 2013, at 4:05 PM, Wendell wrote:
> Luke pointed out that we should not be inserting extraneous data into the blockchain, so this sounds like the best option, Eric.
>
> I'm under the impression that a Bitcoin user Alice cannot see the actual public key of Bitcoin user Bob, so if we had Hive store metadata on a server relating to a given transaction ID, I would not be able to use those public keys key to encrypt. Is this a misunderstanding or is it correct?
>
> Assuming it is correct, the best that I could come up with was storing the transaction ID with a _fresh_ public key on a server, each time a transfer is made. Altogether it looks like this:
>
> - Alice generates a new keypair & revocation certificate for the transaction
> - Alice makes a Bitcoin transaction to Bob
> - Alice sends the transaction ID plus the new public key to server
> - Bob receives the Bitcoin transaction
> - Bob generates a new keypair & revocation certificate
> - Bob does a transaction ID lookup on the server, receives Alice's public key, sends his own new one
> - Bob encrypts his user metadata against Alice's new key
> - Alice downloads and decrypts Bob's metadata
> - Alice uploads her revocation certificate
> - Alice uploads her own metadata
> - Bob downloads Alice's metadata
> - Bob uploads his revocation certificate
> - (Server removes all keys with revocation certificates)
>
> I presume going the extra mile to generate new keys for each transaction is helpful for privacy?
>
> The above seems rather inelegant to me. I really don't like that clients (wallets) are going to be beating down the server all the time checking for keys, or that there is a possibility of a desynchronization so severe that the user receives the data much too late for it to be useful. But, I suppose it can work.
>
> Another thing I'm considering is Alice/Bob validating each other. I suppose we should include some kind of code that we encourage people to read to each other over the phone or some other medium, to ensure that "it really is Alice", before (for example) returning money to a very legit-looking personage.
>
> Any other thoughts? I would love to do this without using any servers at all ("serverless keyserver", anyone?), but I am not quite sure how.
>
> -wendell
>
> grabhive.com | twitter.com/grabhive | gpg: 6C0C9411
>
> On Sep 7, 2013, at 12:47 AM, Eric Lombrozo wrote:
>
>> Why not just use the transaction hash itself for the lookup? Also, presumably you'd want to encrypt the data so that only the recipient of the transaction can do this lookup.
>>
>> -Eric
>>
>> On Sep 6, 2013, at 8:07 AM, Wendell <w@grabhive.com> wrote:
>>
>>> Hi all,
>>>
>>> We're thinking about ways of automatically exchanging contact details between wallets, in order to encourage the proliferation of identifiable names and photos rather than long and hard-to-verify addresses.
>>>
>>> The simplest version goes like this:
>>>
>>> 2 BTC Bitcoin is sent to someone, and a data lookup hash is inserted into the transaction. When it arrives on the other end, it is indeed looked up, and instead of being presented with a dialogue that says "you received 2 BTC from 13Y94z43Nbbb6wevRyk82CeDoYQ5S28zmA", it's "You received 2 BTC from Frank Jones" including a nice photo.
>>>
>>> Now. We can simply delete this data in reference to the transaction ID after it happens (or delete it after a time), but is there any more decentralized way to do it? I would prefer us to run no dedicated servers that would ever put us in a position of being coerced into giving data, or otherwise altering our system to store it.
>>>
>>> Any thoughts about this?
>>>
>>> -wendell
>>>
>>> grabhive.com | twitter.com/grabhive | gpg: 6C0C9411
>>>
>>> ------------------------------------------------------------------------------
>>> Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
>>> Discover the easy way to master current and previous Microsoft technologies
>>> and advance your career. Get an incredible 1,500+ hours of step-by-step
>>> tutorial videos with LearnDevNow. Subscribe today and save!
>>> http://pubads.g.doubleclick.net/gampad/clk?id=58041391&iu=/4140/ostg.clktrk_______________________________________________
>>> Bitcoin-development mailing list
>>> Bitcoin-development@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/bitcoin-development
------------------------------------------------------------------------------
LIMITED TIME SALE - Full Year of Microsoft Training For Just $49.99!
1,500+ hours of tutorials including VisualStudio 2012, Windows 8, SharePoint
2013, SQL 2012, MVC 4, more. BEST VALUE: New Multi-Library Power Pack includes
Mobile, Cloud, Java, and UX Design. Lowest price ever! Ends 9/20/13.
http://pubads.g.doubleclick.net/gampad/clk?id=58041151&iu=/4140/ostg.clktrk
_______________________________________________
Bitcoin-development mailing list
Bitcoin-development@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bitcoin-development