The LTB Network Forum


Reorg: A blockchain based organization protocol

Back to Project Development

Jump to bottom

Comments

Eylrid

The problems with the Bitcoin Foundation being a centralized organization trying to manage a decentralized technology have got me thinking about how to do blockchain based organizations.

The Bitcoin blockchain is an unprecedented tool for decentralized and transparent consensus, record keeping, rule enforcement and accountability. By applying these tools to organizations we can create organizations that are more transparent and democratic and less susceptible to corruption.

I came up with an idea for a protocol to do this, which I call Reorg. My ultimate goal is to eventually use it to create a decentralized alternative to the foundation, rooted in the blockchain and using liquid democracy; but that is a project for another time.

I want to make Reorg open ended enough to support all types of organizational structures, from nebulous decentralized groups to hierarchical organizations with a Supreme Dictator at the top. Even completely hierarchical organizations can benefit from being organized on the blockchain. It provides transparency and can create limits on what the Supreme Dictator and other power players in an organization can and can't do.

Here is the initial draft of my whitepaper. The protocol specification will come soon.

I welcome any feedback.

3 likes
  • monbux

    Following! :)

    Permalink
    0 likes
  • decentral

    Sounds very interesting. Liquid democracy has a hughe future. Not sure why the bitcoin world has not brought more in that area. Will need think more at t before being able to give some feedback.

    Any code yet? Can you mention what you think to solve better than similar concepts/solutions and what similar concepts exist at all? What concepts/other whitepaper/solutions inspired you? What do you do better and what are possible drawbacks?

    Some time ago I read an old (from before 2008/bitcoin) white paper from Mike Hearn on that subject. Can not find it right now, will be link it later maybe

    Permalink
    1 like
  • RockHoward

    An excellent starting point. Here are a few random thoughts:

    Restrict the voting systems to those that make sense for the given type of election. A 'one voting system per group' cannot work for a legitimate organization unless the organization restricts themselves to one type of election.

    Avoid inferior voting systems to the degree possible. (I know that all voting systems are flawed, but some are seriously broken. It pains me to see anyone suggest that instant runoff or plurality voting makes sense for any election much less for the all the types of elections that might arise.)

    I think you have to design the system to support liquid democracy from the start if that is the ultimate objective. If you don't, I fear that you will never get there. Furthermore you need to provide a compelling reason for people to use this protocol over other similar protocols that are under development. You might as well bake in your very best aspirations.

    A templating system should be built in that can capture common organization structures. Think in terms of 'make simple things simple and complicated things possible'.

    The issuance model (like every issuance model I have seen in the bitcoin 2.0 world), is naive. There needs to be support for tiered ownership (shares vs. preferred shares and such) as well as for shelf registration which is the issuance of shares that are not to be actualized and distributed until they are unlocked by certain conditions (typically a vote of the Board of Directors.)

    Speaking of which, most organizations will need a Board of Directors baked in. If you can get that group pinned down in gory detail, you should have a group definition that can work for most other situations where a group might make sense.

    Finally there is the question of 'who issues the tokens'? If it is an individual, then the organization cannot survive the demise of that individual. That may be fine for some situations, but not many. This is a fundamental question that begats others like 'can an organization restructure itself and, if so, how?'

    I am following you on github. If you like I can recast some of the above as issues and/or send you pull requests.

    Permalink
    1 like
  • cryptozio

    This seems a smart project with a promising future, I'm following too, waiting for protocol specs.

    Permalink
    0 likes
  • Eylrid

    @decentral Any code yet?

    I'm working on a protocol specification. When that is fleshed out I will start working on code.

    Can you mention what you think to solve better than similar concepts/solutions and what similar concepts exist at all? What concepts/other whitepaper/solutions inspired you? What do you do better and what are possible drawbacks?

    General purpose over single purpose. The distributed autonomous organizations I have seen so far have been single purpose. A group of developers is needed to bang out and maintain each one. With Reorg I want to create a platform where anyone can easily create new organizations.

    Reorg will be able to support more complex organizational structures than any DAO I've seen so far, for those that need that. Most of the blockchain protocols I've seen focus on ownership, whether that be of a coin, token, share or something else, and what happens to that ownership, e.g. dividends, contracts for difference, etc.

    Reorg, while it still has an ownership element, focuses more on the relationship between people and groups, and how those people and groups make decisions.

    Permalink
    0 likes
  • RockHoward

    More thoughts on the proposal:

    While there is a lot of obvious support for managing ownership and for governance, there is less obvious support for the equally important process of funds dispersal by the org. Now obviously when it gets down to issuing paychecks, there will be third party solutions available, but there are many other aspects involved in managing company funds. One such aspect is that budgets are fluid and subject to revision based on market feedback. The Reorg structure should aid transparency and accountability without being overly restrictive in the face of changing budgets.

    Also there is the potential balance between transparency and 'need to know' matters such as actual salaries, product production costs and other things that are best not revealed in a competitive market environment. Then again, for some orgs full transparancy of assets spent by or on behalf of the org makes sense.

    Of course once the org is making money you get to exercise the fun part of the process: issuing dividends and/or conducting token buybacks. However to get to that point takes a lot of work and generally more assets than you originally plan for. Having some high level framework to handle the corporate assets is worth considering for Reorg.

    Permalink
    1 like
  • Eylrid

    @rockhoward An excellent starting point. Here are a few random thoughts:

    Restrict the voting systems to those that make sense for the given type of election. A 'one voting system per group' cannot work for a legitimate organization unless the organization restricts themselves to one type of election.

    An excellent point. That could be added to the voting rules in group specifications: which voting system is used under which circumstances.

    Avoid inferior voting systems to the degree possible. (I know that all voting systems are flawed, but some are seriously broken. It pains me to see anyone suggest that instant runoff or plurality voting makes sense for any election much less for the all the types of elections that might arise.)

    I agree that some voting systems cause major problems and should be avoided. That being said, I want the protocol itself to be as open ended as possible.

    I think you have to design the system to support liquid democracy from the start if that is the ultimate objective. If you don't, I fear that you will never get there. Furthermore you need to provide a compelling reason for people to use this protocol over other similar protocols that are under development. You might as well bake in your very best aspirations.

    Absolutely.

    Furthermore you need to provide a compelling reason for people to use this protocol over other similar protocols that are under development.

    My highest priority is to have a protocol out there that allows people to easily create and run blockchain based organizations. If there are other protocols already out there that do that, then I would love to hear about them.

    The thing that most compels me to work on this is that I personally want to be able to easily start or join and participate in organizations that offer more transparency and accountability and more of a voice to their members. If other people want the same and find this a useful tool to do that, so much the better.

    A templating system should be built in that can capture common organization structures. Think in terms of 'make simple things simple and complicated things possible'.

    That would be a good thing to include in a client.

    The issuance model (like every issuance model I have seen in the bitcoin 2.0 world), is naive. There needs to be support for tiered ownership (shares vs. preferred shares and such) as well as for shelf registration which is the issuance of shares that are not to be actualized and distributed until they are unlocked by certain conditions (typically a vote of the Board of Directors.)

    Shares vs. preferred shares could be done with multiple tokens.

    Shelf registration could be done with a shelf share token that can be burned to create active shares, when the conditions are met. A provision for that would have to be included in issuance rules for shares.

    Speaking of which, most organizations will need a Board of Directors baked in. If you can get that group pinned down in gory detail, you should have a group definition that can work for most other situations where a group might make sense.

    This comes back to the templates.

    Finally there is the question of 'who issues the tokens'? If it is an individual, then the organization cannot survive the demise of that individual. That may be fine for some situations, but not many.

    That would differ on a token by token basis. When a token is created the creator includes issuance rules. They can also specify initial balances that certain individuals and/or groups get.

    Here are some example rules:

    • No new token units are issued
    • Anyone who sends 1 BTC to a certain address gets 1000 tokens
    • Holders of the token get a 1% dividend once every 30 days
    • Any member of GroupX gets 50 tokens/per week

    This is a fundamental question that begats others like 'can an organization restructure itself and, if so, how?'

    Some organizations need the flexibility to restructure themselves. Others may need to be unchangeable to be trusted, much like Bitcoin gains trust from the fact that the issuance schedule is unchangeable.

    I think the way to go is that when a group is created a set of rules can be included specifying the conditions under which the group can be changed.

    For example BobCorp_Board should not be able to restructure itself, but BobCorp, i.e. the shareholders, may be able to restructure both BobCorp and BobCorp_Board.

    Note that an organization can consist of many groups, but the protocol itself doesn't have any concept of an organization, just groups and how they interact.

    I am following you on github. If you like I can recast some of the above as issues and/or send you pull requests.

    I would happily welcome pull requests.

    Permalink
    1 like
  • Eylrid

    Here is an initial draft of a protocol specification: https://github.com/Eylrid/reorg/blob/master/protocol%20specification

    Permalink
    0 likes
Jump to top

Post Reply

Please Login to post a reply to this thread.

© Copyright 2013–2016 The LTB Network. All rights reserved .