Tuesday, August 22, 2006

Informed Consensus - what is it ? part 2

In part 1, I mentioned open-source software development as an example of Informed Consensus (IC). This development model works through systems like "cvs". This cvs is a code repository which can be accessed by any given developer. The developer removes a piece of code that interests him, works on it, and returns it to the repository. There are version numbers assigned to it, and other developers can choose to develop it further or return it to its previous state (this can also be arbitrated by a chosen manager if that's the route decided on).

We see here the use of the concepts of central documents and scalability. Democracy is eminently scalable, because it's just a bunch of votes. You just need to compile as many votes as you need. It may be less obvious to deduce how IC can be scaled to, say, hundreds of thousands of users, or even millions of users. Actually, we already have such systems - Wikipedia is a great example. What Wikipedia and software design have in common is the use of central documents.

IC relies on dialogue, 1-to-1 communication. But you can't have 1-to-1-to-1-to-1... communication between thirty people, let alone millions. You need a way to effect dialogue (to stir the pot of ideas, so to speak) that does not rely on direct communication. So after a certain scale, central documents are the key to IC scalability.

Wikipedia is a great example of this because it is a sucessful example of a gigantic IC system which mostly relies on people's values. People cherish a specific area of knowledge and so desire to share their expertise. Other people come and try to improve on each little area. Of course, disputes arise and sometimes people do come to disrupt, and like any other successful IC system, Wikipedia has an extensive set of policies and rules to resolve disputes. Most of the time, discussion takes place within these parameters, instead of de-evolving into matters of belief or preference. This is the mark of a healthy dialogue.

The power of wikis and other dynamic database driven websites and centralized database driven documentation have seen a rise in business use and will continue to grow. Architects work digitally on centrally stored documents, and many construction contracts are now being reviewed and edited digitially with easy tracking of the changes to the original. The long standing IC based business activities are only becoming easier to manage and highly scalable. The growth of these practices and of the development of networks and software to facilitate them is evidence of their power.

Science obviously works in a similar way, with repeatability of experiments and peer review of papers. It is focused on the facts, rather than a system based on coercion. When scientific papers are published, a group of peers read and agree on the plausibility of the findings. After wider distribution, scientists researching the same or similar topics will attempt to replicate the experiments in their labs in order to prove the legitamacy of the findings. Once there is sufficient evidence that the findings are legitimate, they will begin to be developed further or used to advance related research. It is the informed portion of the scientific community reaching a consensus on research that in turn help advance the goals of that community. There are centuries of evidence to back up the power and effectiveness of this technique, and it also includes the notion of central documents that are edited or expanded upon.

No comments: