Ten potentially useful things to do when you're solving a problem are described by Artificial Intelligence pioneer Marvin Minsky in a series of memos for the One Laptop Per Child project. We can sum them up visually with the following diagram:
We can also see some interesting relationships to the peeragogy patterns we identified earlier. The connections are first described with a picture here, and then in more detail with text below. Some of the nodes in this diagram are clickable, and clicking will take you to the page describing the relevant pattern:
To elaborate in words:
- Simplify things for Newcomers. In practice, this means that we don't expect a newcomer to enter at full speed.
- Use a Roadmapto guide us from one phase to another, while the project's central Heartbeat helps us attend to the central focus.
- Announce changes through a Wrapper who describes the new status or direction of the project. For the Peeragogy project, that often meant summing up the high points that we saw over a given period of time.
- Assemble a Pattern Languagefor the project by first Discerning Patterns, and when we've seen how the patterns relate to each other (as in the diagram above), we can start to build a map like the one above, for subsequent use in problem-solving and design.
- We divide work up not only horizontally among different Roles, but also temporally by using the Roadmap. Someone who is moving ahead with the Roadmap is likely to be working at the leading edge.
- When we find an analogy, we are basically Creating a Guide of some sort. This can be used as a form of "exploration," as we look at how one form of engagement may or may not map onto other forms of engagement.
- When we ask for help, we may avail ourselves of some Moderation service that will decide how to deal with our request. One simple way to ask for help is Polling for Ideas. Obviously once we start to get help, we're working in a regime of "collaborative effort".
- If you know the answer, then you may be able to reuse it (which is the basic idea described in Praxis vs Poesis, though the title is a little bit obscure). Someone who knows the answer and who is good at self-explanation may also have a good idea about how to get from the current state to the goal state; alternatively, this may be broken down into steps in some sub-Roadmap, and moving from step to step would then illustrate "progressive problem solving".
- It is important to give it a rest so as not to over-exhaust oneself, busting one's own Carrying Capacity, or, alternatively, overwhelming the group.
- It seems that one of the things that experts do is Discerning a Pattern. This allows them to simplify their processing.
- Finally, again, if we know why it is hard, then we may be able to Create a Guide that will help get around, or at least better cope with, the difficulty.
- The Society of Mind is a book by Marvin Minsky that talks about how a mind can be made up of many different "agencies" that work together. A relatively recent review by Push Singh is available online, and contains a succinct overview of the key components of the more complex problem solving or "thinking" architecture described in Minsky's book.
- Peer Supported Problem Solving and Mathematical Knowledge is a forthcoming Ph. D. thesis by peeragogy co-author Joseph Corneli.
- Our Style Guide contains some guidelines to use when working on "problems of exposition"