March 1, 2013

The politics of Groups

Filed under: work — Tags: , , , , , , , — admin @ 2:50 pm

One of the great reasons why I started really looking forward to being with Groups and the Q framework for the next 5-10 years is the opportunity it will give me to design solutions to challenges in sociological and political and economic arenas. I believe in the power of tools to shape the way people form organizations, and affect their lives, and I can really make a difference in the world by putting the right tools in the hands of millions of people. More than a politician can, who has to work within an existing system and often is unable to change it. We are building the system, step by step, so we will have a lot of responsibility to do it right.

What do I mean by right? Well, people should be able to accomplish what they need, and we should minimize unfortunate situations where they can’t. We are lucky in that our tools are in the virtual world, so there are greater possibilities and smaller consequences in the real world if something goes wrong. Any real-world enforcement will still be done by actual governments / police / whatever people have. 🙂

Anyway, that is the grand 5-10 year vision for what Groups and Q platform can be. I want to outline the guidelines for the system as I see them so far:
  • Individuals should be able to publish whatever they want, at their own cost — both monetary and in terms of reputation. If you can see a stream, you should know who is publishing it, whether it is a real person’s account, a pseudonym, or a group. That is why streams have a publisher_id, and why the publisher of a stream has complete rights to do anything with that stream. Publishing represents a greater degree of ownership than even admin_level=”own”. The publisher of a stream can never change, but a new “owner” can be assigned.
  • The tools that the individual used to create the stream are the various apps and plugins that they used. Those apps and plugins may have been published by others and simply help create / edit streams that the user publishes, but they may store some of the data for the stream, creating a need for the publisher of the stream to pay them in order to keep serving the data. Alternatively, the publisher of the stream can pay the developer of the app or plugin to get and install it locally, and create unlimited streams themselves. They would sign an agreement not to re-sell the plugin or app, or to contribute any revenue / forks / improvements back to the app owner. That agreement is up to them.
  • Groups should be able to form without “permission” from the government of some bigger group. This eliminates top-down bureaucracy, which may lose touch with the people “on the ground” in various areas of the community they manage. However, bigger communities may set up governments where they have “relations” with individual publishers and groups and other communities. In Q this accomplished via stream relations (which enable searchable indexes), and subscriptions. They may agree to automatically publish anything that any of the groups tell them about. They may subscribe to updates on every stream that they are listing in their index. This creates an aggregator/search engine for example. Certainly each search engine may blacklist any group/individual, but there will be no monopoly on this.
  • Although individuals are free to form any groups they want (e.g. on their own computer cluster), and publish anything they want (at their own cost), each user should have a unique ID. I have seen enough problems with anonymous commenting and other anonymous activities on the internet to know that in the absence of reputation, people start behaving irresponsibly. And when that happens, the entire community suffers the cost eventually, and becomes undesirable for others to join. If we are to create communities people like, one of the biggest signs is that others want to join them, and their own members don’t want to leave. This can be a major measure of how useful a community is.

So where does this leave us? Well, that is what the Q system already does. But it also gives us some information for how we should structure our applications in the coming years, taking advantage of certain Q features more than others. What do I mean?

Well, Q allows you to choose between “each individual publishes their own stream” and some degree of “centralized publishing” by management teams of groups. So who should publish a stream, the individual or the group?

If the individual – the risk is that the individual may have too much power over others who come to rely on the stream. They may suddenly stop publishing it, or cut off access to everyone, which would hurt many people. (I define hurt in terms of needs or strong expectations of people that form over time.)

If the group – then managers may come and go, but the risk is that if the group is too big, it may be out of touch with the individuals. The bigger risk is that the individuals are forced to go along with the group, which may also create a lot of frustration. For instance, the group may give rise to into three sub-groups. They are deciding where to go, but some people want to go bowling, others want to go to the movies, others want to volunteer in a soup kitchen. Even though everyone belongs to the group. Who should publish these activities?

So I think when it comes to publishing streams that others can join, there should be some combination of groups and individuals. And it should reflect the best practices of what happens in the real world: one person starts a group that may later become bigger than him. Then this group grows, gets managers etc. After a while this person may leave. In the future, other individuals may want to start their own groups and invite some members of the old group to join. They may establish relationships between each other, subscribe to each other’s streams, pay each other money, etc.

I think that would be a really good structure for a tool. Q is going in the direction to support all this, except for one thing: if an individual publishes a stream, there is no way to hand it off to a group later. The individual publisher is the complete owner of that group and can dissolve it at any time. That is why I think groups should not be published by individuals. Each group should have its  own publisher. The “contacts” table of this publisher would contain the management team for that group, whether it’s the federal government or a small firefighters union. Neither one should have to care about the entire list of managers of the other. They just need to have relations between each other and get stuff done based on their own, internal policies.

Starting a group is like starting a company. We probably need to provision another user_id for the group, for the central authority, so its identity and reputation is consistent throughout the world, preventing “anonymous abuse” once again. But its internal activities may be hidden and it has full control over what it publishes. Members may come and go, etc. And the group may keep its entire history … it may have started as a chatroom about chess, and became a big society of chess enthusiasts, creating tournaments etc. And we can see by the messages posted in the group’s streams what it history was, if it chooses to preserve it 🙂

So at the end of the day, Q already is already on its way to supporting everything I spoke about above, except we don’t have any mechanisms for individual publishers to join into a group and copy their streams to the group. If a new group forms, then we need to think what its members can “legally” and “easily” contribute so it can get off to a running start. They shouldn’t be able to just take anything from anywhere, but it would be nice if they could contribute something to the group that they join, set up a management structure, and invite/accept new members.

PS: As I’m learning more about economics and the role of money, I’m starting to form a picture of a new system of currency, related to our ideas about reputation and credit. I will write more about this system in an upcoming post. While having a medium of exchange is very useful, and has seen innovations such as BitCoin that decentralize the system, I think it would be interesting for everyone to issue their own currency on their own credit — both groups and individuals. It might actually better reflect what actually happens when “unofficial money” is used, and may be more humane as communities with very little money but compassionate hardworking people will be able to get things done. As it is now, people may hold money drawn on the credit of some larger institution, which may have elements of cronyism or nepotism which this new system might mitigate.

I think this is interesting stuff to think about, and eventually it can help me to fulfill the idea of “tikkun olam”, originated by Jewish rabbis.