Re: [bitcoin-dev] Named Bitcoin Addresses

From: Kristov Atlas <kristovatlas.lists@gmail•com>
To: Mark Friedenbach <mark@friedenbach•org>
Cc: bitcoin-dev@lists•linuxfoundation.org
Subject: Re: [bitcoin-dev] Named Bitcoin Addresses
Date: Fri, 11 Sep 2015 11:13:18 -0400	[thread overview]
Message-ID: <CAGH37SKyw6hoJaL2bOHChFikW6R9T49rXxL=hCgjndjyuVAwLA@mail.gmail.com> (raw)
In-Reply-To: <CAOG=w-vsp6Oxx3WsjVoQ9xO41SqgUMw97h0Ba1jSd9s=KZL6wQ@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 5653 bytes --]

I applaud your brainstorming efforts! :) But I dislike just about
everything about this proposal.

Bitcoin addresses should never be reused, for privacy and security reasons
documented extensively online.

I definitely do not want my computer to maintain a registry of every
address in the world any more than I want to maintain a list of all email
addresses, web addresses, etc. This cannot be scaled and is wasteful since
most participants in the network don't need to transact.

If we need naming conventions, I'd rather see a name registry system
off-chain attached to bip47 reusable payment codes.

-Kr
On Sep 10, 2015 5:32 PM, "Mark Friedenbach via bitcoin-dev" <
bitcoin-dev@lists•linuxfoundation.org> wrote:

> Are you aware of the payment protocol?
> On Sep 10, 2015 2:12 PM, "essofluffy . via bitcoin-dev" <
> bitcoin-dev@lists•linuxfoundation.org> wrote:
>
>> Hi Everyone,
>>
>> An issue I'm sure everyone here is familiar with is the problem
>> concerning the fact that Bitcoin addresses are too complex to memorize and
>> share. Current Bitcoin addresses can be very intimidating to new users. As
>> Bitcoin grows it's necessary to provide a much more user friendly
>> experience to the end user. I think that having the capability to assign a
>> unique name to a Bitcoin address is in the best interest of Bitcoin and
>> it's users.
>> I've recently come up with a method for assigning a unique name to a
>> specific Bitcoin address. I'm looking to get some feedback/criticism on
>> this method that I have detailed below.
>>
>> Let’s run through Bob and Alice transacting with a Named Bitcoin Address.
>> Bob wants to collect a payment from Alice for a service/good he is
>> selling, but Alice wants to pay from her home computer where she securely
>> keeps all her Bitcoin. So now Bob needs to give Alice his Bitcoin address
>> and because Bob is using a Named Bitcoin Address and a supported wallet he
>> can give her an easy to memorize and hard to mess up address. Bob’s address
>> is simply ‘SendBitcoinsToBob’ which can easily be written down or
>> memorized. Now Alice can go home send the Bitcoin from her own supported
>> wallet and be positive that she sent it to Bob.
>>
>> Let’s look at how Bob’s supported wallet made that address.
>>
>> First Bob let’s his wallet know that he wants to create a new address. In
>> response, his wallet simply asks him what he wants that address to be
>> named. Bob then enters ‘SendBitcoinsToBob’ as his preferred address name.
>> The wallet then let’s Bob know if his preferred address name is available.
>> If it’s available the name is broadcasted to the network and ready to use.
>>
>> Now let’s get a little more technical.
>>
>> When Bob inputs his preferred address name the client has to make sure
>> this name hasn’t been taken or else who knows where Alice will be sending
>> her Bitcoins. The client does this by referencing a downloaded “directory”
>> of names chosen by people using this system. This directory of names are
>> transactions sent to an address without a private key (but still viewable
>> on the blockchain) with the name appended to the transactions as an
>> OP_RETURN output. These transactions are downloaded or indexed, depending
>> on whether or not the wallet contains the full Blockchain or is an SPV
>> wallet. Because of such a large amount of possible address names a binary
>> search method is used to search through all this data efficiently. The
>> names could be sorted in two ways, the first being the first character and
>> the second being the total length of the name (I will being exploring
>> additional methods to make this process more efficient). So now that Bob’s
>> client has verified that the name has not been taken and is valid (valid
>> meaning it's under 35 bytes long and only using chars 0-9 and a-z) it sends
>> a transaction of 1 satoshi and a small fee to the address without a private
>> key as talked about earlier. The transaction's OP_RETURN output consists of
>> two parts. The implementation version of this method (up to 8 characters)
>> and the name itself (up to 32 characters). Once the transaction is
>> broadcasted to the network and confirmed the name is ready to be used.
>>
>> Let’s look at how Alice’s supported wallet sends her Bitcoin to Bob’s
>> Named Bitcoin Address.
>>
>> When Alice enters in Bob’s address, ‘SendBitcoinsToBob’ Alice’s client
>> references the same “directory” as Bob only on her device and searches for
>> the OP_RETURN output of ‘SendBitcoinsToBob’ using a very similar binary
>> search method as used for the verification of the availability of an
>> address name. If a name isn’t found the client would simply return an
>> error. If the name is found then the client will pull the information of
>> that transaction and use the address it was sent from as the address to
>> send the Bitcoin to.
>>
>> Essentially what this idea describes is a method to assign a name to a
>> Bitcoin address in a way that is completely verifiable and independent of a
>> third party.
>>
>> Please ask your questions and provide feedback.
>>
>> - Devin
>>
>>
>> _______________________________________________
>> bitcoin-dev mailing list
>> bitcoin-dev@lists•linuxfoundation.org
>> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
>>
>>
> _______________________________________________
> bitcoin-dev mailing list
> bitcoin-dev@lists•linuxfoundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
>
>

[-- Attachment #2: Type: text/html, Size: 8806 bytes --]