Re: Bitcoin URI

On 09/23/2011 10:57 AM, Amir Taaki wrote:
> I've been discussing with Melvin Carvalho (melvster) about
> the possibility to get a bitcoin URI accepted. Much thought has gone
> into drafting a standard which can be viewed here:
>
>    https://en.bitcoin.it/wiki/URI_Scheme
>
> There's a competing URI scheme:
>
>    https://en.bitcoin.it/wiki/X-btc

Amir and I had a quick chat on Friday about standardizing a Bitcoin IRI, 
the audio and text log of the conversation is available here:

http://payswarm.com/minutes/2011-09-23-bitcoin/

Full text log follows:

Present:
    Manu Sporny, Amir Taaki

Manu Sporny is scribing.

Topic: PaySwarm and Flattr.com

Amir Taaki:  I was checking out PaySwarm in the past, when I
    started getting involved in Bitcoin. Seems like you've been
    working on this standard for a while?
Manu Sporny:  Yes, takes quite a bit of effort to get a standard
    off the ground. Likewise, I've been following Bitcoin for quite a
    while.
Amir Taaki:  What do you think of Flattr?
Manu Sporny:  Flattr is interesting - the main thing we're
    interested in is creating an open and patent and royalty free
    standard. We don't think anyone corporation should own how we
    reward each other on the Web. Flattr is great and is one business
    model that we support with the PaySwarm work, but our focus is
    more on enabling companies like Flattr to not have to worry about
    how the funds are moved around. We want the system to be
    decentralized, we want anyone that can implement the spec to
    setup a PaySwarm Authority. We also want to support multiple
    different types of currencies. That's why we're interested in
    Bitcoin right now - on people's minds. Unfortunately, don't know
    enough about tech to make any solid statements on support for
    Bitcoin.

Topic: Bitcoin Criticisms

Amir Taaki:  Feel free to ask me anything about Bitcoin. There
    are people that say "Bitcoin is a Ponzi scheme" - but banks are
    looking at Bitcoin in a serious way. UK Exchange deals with the
    FSA - so far, no official statement on bitcoin. There will be a
    directive to classify bitcoin as a currency overlay. Banks are
    used as a messaging network - SWIFT - one of our guys is at the
    conference now talking about Bitcoin.
Manu Sporny:  That's the SIBOS / Innotribe stuff at SWIFT, right?
Amir Taaki:  Yeah.
Manu Sporny:  I saw some of the stuff that Donald Norman had
    said.
Amir Taaki:  Many people talk about Bitcoin as a bank killer, a
    subversive thing - but many of these higher-ups are wondering how
    they can use it - they don't see it like that.
Manu Sporny:  That's certainly not our concern - we're more
    interested in the technical aspects - if there are any flaws. Any
    piece of technology can be used for evil.
Amir Taaki:  Right - find a technology that isn't neutral, an
    atom bomb - it can be used for good, it can be used to destroy
    entire countries. A hammer can be used to build beautiful
    carpentry, or it can be used to kill somebody. Blaming the
    technology only shifts the blame away from the perpetrators -
    it's an unhealthy attitude.
Manu Sporny:  That's certainly not the position we take on it. My
    concern is technical - have you spoken with Ben Laurie? He has
    some criticisms of Bitcoin - there are bits of the bitcoin
    algorithm that you can't switch it out. Like switching out
    AES-128-CBC for AES-256-CBC. So - technical concerns.
Amir Taaki:  Is he a skeptic?
Manu Sporny:  Not as far as crypto currencies are concerned - he
    has been heavily involved.
Manu Sporny:  He published a number of papers critisiziing
    bitcoin
http://www.links.org/?p=1164
http://www.links.org/?p=1171
http://www.links.org/?p=1175
http://www.links.org/?p=1179
http://www.links.org/?p=1183
Amir Taaki:  We have a conference in Prague, looking for people
    skeptical about Bitcoin - it's unhealthy to have too many people
    agreeing in the same room. If he is someone that's anti-bitcoin -
    we'd like to have him there.
Manu Sporny:  I'll ping Ben and see if he can join. He's w/
    Google right now - founder of Apache Software Foundation - good
    guy, should have a chat with him.

Topic: The Rise of Bitcoin, MtGox scandal

Manu Sporny:  Getting back to PaySwarm interest in Bitcoin stuff
    - interest in Bitcoin URI scheme?
Amir Taaki:  There is a sudden movement toward standardization.
    If you go to bittorrent.org - it's a spec. We want the same for
    bitcoin.org. That's the eventual direction - standardization. We
    need an RFC for bitcoin IRI RFC. We modeled it after Python PEP
    documents - We are having Bitcoin BEPs. Not sure about the names
    yet.
Manu Sporny:  Are you ready to go for full standardization?
Amir Taaki:  Bitcoin protocol still early days - we don't want
    full-blown standardization. We should start thinking about this
    stuff now though - multiple clients, light nodes - I'm the only
    person that's working on an alternative full client
    implementation. Once it happens, it can't be one piece of
    software deciding the protocol. Lots of discussion around that.
Manu Sporny:  I'm fairly involved in standards work - would be
    happy to help.
Amir Taaki:  Yes, Melvin Carvalho told me about you.
Manu Sporny:  Melvin is a great guy - he pings us about Bitcoin
    from time to time.
Amir Taaki:  He introduced me to Bitcoin some time ago - I was
    complaining about money and he pointed me toward it. It was shaky
    back then, but then looking at it closer - it was incredible.
Manu Sporny:  We had the same sort of opinion in the beginning.
Amir Taaki:  Yeah, I didn't think it was going anywhere in the
    beginning.
Manu Sporny:  I saw a great amount of promise for it - but had no
    idea that it would build into what it is today so quickly. The
    fact that it is trading against the USD so strongly - didn't see
    it happening so quickly. We want to help Bitcoin in whatever way
    we can. Our focus is on helping people buy and sell digital
    content via the Web. PaySwarm is currency agnostic - but we do
    try to find out how Bitcoin could be integrated. How do
    transactions look in pure Bitcoin?
Amir Taaki:  When I first started Bitcoin Consultancy - we had a
    10 year plan - it had multiple stages. 8 years of that plan got
    done within 8 months - so, that was surprising - so much interest
    so quickly. Then this MtGox scandal happened - they got hacked.
    So people haven't realized that MtGox is just one private company
    - but people confuse it w/ Bitcoin. Interest in bitcoin has waned
    - gartner graph shows Bitcoin suffering in all ways - a crash.
    Looking at indicators on network - trading volume, price, nodes,
    etc. - activity in bitcoin is dropping. We are at a low point
    right now before we start growing slowly.
Manu Sporny:  Yes, a very typical trough of disillusionment in
    the standard technology cycle:
    http://en.wikipedia.org/wiki/Hype_cycle
Amir Taaki:  Look at Linux or Wikipedia or Internet - typical
    lifecycle is 10 years - that's how long it's going to take
    Bitcoin to get popular - 5% of people using it... not 100%.

Topic: Standardization and the Bitcoin IRI scheme

Manu Sporny:  Is standardization a part of this growth - or do
    you think you can reach 5% w/o standardization?
Amir Taaki:  Standardization helps development - all these groups
    are doing different things - we need to cooperate. The bitcoin
    URI is something people have been talking about. There are
    competing proposals - so we should standardize early.
Manu Sporny:  You do need implementation experience. You could
    incubate in W3C Community Groups?
Amir Taaki:  Yes, could do that. We could try to get into HTML5
    standard - a property for bitcoin. That's when I raised Bitcoin
    URIs.
Manu Sporny:  Currently chair of the RDFa / RDF Web Apps Working
    Group - markup of metadata in HTML - so we're working on this.
    Browser could go to a page and know that there is a payment URL
    in there... bitcoin URI or some other form of payment mechanism.
    The account name is given using a URI. Everything on the page is
    machine readable - lower-level protocol like Bitcoin could take
    over and process payment at that point. Same direction that
    PaySwarm is going in - ability to express payment endpoints and
    make it machine readable. We would be very happy to work with you
    on that. The Bitcoin URI seems pretty simple.
Amir Taaki:  Yes, it's pretty simple. Bitcoin destination is easy
    - always starts w/ 1, then account, then SHA256 checksum are the
    last 4 bytes - fairly easy to tell it's a Bitcoin address.
Amir Taaki:  about competing standards - there is one simple one
    - it can only send bitcoins and a label - the other one has
    options for everything. Don't know about more complex URI scheme
    - bitcoin is flexible - very difficult, perhaps impossible to
    cover everything. May want to start with something simple - need
    to talk with the person that created it first.
Manu Sporny:  This URI is a pure P2P basis - right? There is no
    HTTP endpoint, there is no server sitting out there that you can
    depend on?
Amir Taaki:  Correct.
Manu Sporny:  The way we approached the PaySwarm stuff is that it
    is decentralized, but you always have an agent out there on the
    Web. Those endpoints can take arguments. So, supporting a bunch
    of different options - do it over HTTP and POST something ot the
    endpoint - you could extend it that way.
Amir Taaki:  That's a bit out of scope right now - bit-pay.com
    could do something like that. Bitcoin payments are not instant -
    if you want more advanced options - you go over HTTP to dedicated
    services. 3rd party service - bitcoin URI is something that
    interfaces directly with Bitcoin.
Manu Sporny:  I saw an e-mail-like scheme for Bitcoin. I know
    current Bitcoin address is a hash on their public key - one
    mechanism is more like an e-mail like scheme. Could be supported
    via DNS - but that's centralized. You have to be able to write
    records into DNS.
Amir Taaki:  Maybe the domain is the extension, but the
    front-part is looked up on a service?
Manu Sporny:  This is the whole problem with decentralized
    identity. DOn't know if you've seen anything about BrowserID or
    WebID - they're trying to solve "identity on the Web". You can
    have pseudonyms, but having payment accounts associated with you
    would be helpful.
Amir Taaki:  Ah, I thought they were only dealing with real
    identities. Ok, I get it now.
Manu Sporny:  The whole PaySwarm spec right now uses
    public/private key for messaging - but you still have to specify
    who will manage your account and identity. We want it to be
    completely decentralized in the future - but for now, because of
    regulatory issues - they're not banks, money processors - but
    PaySwarm Authorities handle identity and accounts for people. So,
    paying someone using their e-mail address is doable because we
    have these PaySwarm Authorities that could look something up. You
    look up user preferences by using these HTTP URLs. Is the ability
    to type it in easily the main concern?
Amir Taaki:  There is an artist that I like, I want to donate to
    him - he accepted bitcoin donations, but it's just text on his
    website. You have to copy-paste. I want them to be able to click
    the link and it opens a link in a bitcoin client. E-mail is
    copy-paste - it's easy. Type an amount, put a label, contact
    info, etc.
Manu Sporny:  The assumption that you're making is that there is
    a bitcoin client that is external.
Amir Taaki:  Same as mailto: - yes.
Manu Sporny:  have you thought about Bitcoin in the browser -
    pure browser-based bitcoin? You wouldn't need a URI scheme if you
    had that.
Amir Taaki:  Yes, but Bitcoin would need to be running already in
    the browser.
Manu Sporny:  What about a 3rd party site that they trust?
Amir Taaki:  People don't like that. There are some people that
    are doing this to offer fast micropayments. The core concept of
    Bitcoin is not to be locked into one vendor. You could run it in
    the browser - the way the protocol works - you can do a lot of
    stuff on initallization - load lots of blocks into memory. Or you
    need to have a database backend and have lots of connections -
    more suited for a Web server. Maybe a possibility in the future.
Manu Sporny:  So the browsers have a database mechanism built
    into them - our CTO has done some pretty amazing crypto stuff in
    the browser. Don't know details about Bitcoin - guess it's very
    processor intensive.
Amir Taaki:  You have to download blocks and process each block.
    There is a school of thought that says that a client doesn't
    validate the blocks - but you have to trust who you're getting
    the blocks from.
Manu Sporny:  You could always keep a Bitcoin tab open in your
    browser. I know that's what I do.
Amir Taaki:  Not me.
Manu Sporny:  It depends on who it is - but having a bitcoin
    client running in a tab - you would be able to validate blocks
    in-browser. I don't see a technical limitation to running a
    Bitcoin client in the browser today.
Amir Taaki:  if it were me, I only keep 2-10 tabs open - don't
    keep  browser open all the time. I'd rather have Bitcoin in the
    system tray.
Manu Sporny:  When you go to standardize - people are going to
    ask questions like this.
Amir Taaki:  Someone has already done Bitcoin in the browser -
    but it's a light node, written in JavaScript.

Topic: Standardizing at IETF

Manu Sporny:  So, we're getting off-topic a bit. What support are
    you looking for regarding the Bitcoin IRI scheme?
Amir Taaki:  I don't know the standardization process - was going
    to read up on standardization process. I will ask people for
    input. Write the initial doc - consult with Melvin - maybe talk
    to IANA then?
Manu Sporny:  Well, it depends on what you're registering.
Amir Taaki:  A simple URI scheme - leave it open-ended - maybe
    tack more things on later on. Don't want to make it too complex -
    don't want to increase the burden on people. We want a simple
    spec. We can expand afterward.
Manu Sporny:  You probably want to go through the IETF - that's
    where people create new URI schemes. Usually go through IETF.
    It's a protocol-level thing. It's not a Web thing, so no W3C.
Amir Taaki:  Really? It's for the browsers, right?
Manu Sporny:  Yes, but the way that you specify the scheme is
    through IETF. It's a fairly straight-forward process. Here is the
    RFC for the mailto URI scheme:
    http://www.ietf.org/rfc/rfc2368.txt You could probably copy that
    almost directly.
Manu Sporny:  I could also take a quick hack at it if you'd like.
Amir Taaki:  I want to ask both of the people working on the
    schemes to get their blessing.
Manu Sporny:  Ok, should be simple to spec - only 4-5 pages.
    There are editorial concerns, but I can edit that for you if
    you'd like. I know some folks at IETF that can help us - Julian
    Reschke - editor of HTTP spec.
Amir Taaki:  Can you e-mail - put us in touch? I'll sort this out
    over a week. There is a conference tomorrow - but maybe next week
    I can start on it? Maybe something done by next week?
Manu Sporny:  Okay, I'll do that. I promised to get you in
    contact with someone else?
Amir Taaki:  The Bitcoin skeptic?
Manu Sporny:  Yes, Ben Laurie - I'll do a quick e-mail intro
    between you two.
Manu Sporny:  Okay - feel free to e-mail/ping if you have any
    questions - we want to help.
Amir Taaki:  Thanks much
Manu Sporny:  Good luck at the conference - great finally
    chatting with you. We'll talk soon, maybe next week or week
    after. Take care.
Amir Taaki:  Ciao.

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Building a Better World with Web Payments
http://manu.sporny.org/2011/better-world/

Received on Sunday, 25 September 2011 19:59:55 UTC