Hand Drawn Graphs

Here we examine the utility of hand drawn graphs in concept and wiki navigation and comprehension.

The first thing we need to examine is the close relationship between comprehension and navigation. A Map is understanding the terrain. And for a wiki, having a good map is a significant part of comprehending the subject or subjects described in the content of the individual pages.

So why "hand drawn" - what is the importance of using the human eye and had rather than an algorithm? The answer to this question may be in the unique social architecture of wiki - this is not a fixed landscape, objectively independent of the writer of wiki, but rather a set of diverse overlapping personal perspectives.

If your interest is in the objective fact of the physical terrain - then satellite imagery, and machine generated representations of the transport system are what you need.

If your interest is to objectively cover or review the mass of information, and find the "best", the "most popular" or some other global metric, then an algorithm will suit your purpose well.

But if I want to find out what you consider most important in the mass of wiki pages you have created, then it may not be the page you visited most, or linked to most. It may not even be something you have written yet, or it may be (and often is) represented by a cluster of pages which lead up to a page that you "want to write" but are currently simply researching, or otherwise have not formulated yet.

Einstein famously developed his core insights and "algorithms" through a creative process that involved imagining or picturing narrative scenarios. Could he have been assisted by search? And if so how would you search for a Thought Experiment?

The expressive power of a sketch is undeniable. From architects, to business plans, to mathematicians - the simple act of sketching out an idea is one of the most effective ways of both capturing and iterating over a Barely Formed Idea.

In wiki our sketch is a graph. It is where we join fragments of ideas together, and give a new idea Form. While this is not the tool of choice for everyone it is the tool of choice of many Synthesizers, and it is one of the most familiar ways to navigate and literally map a domain of thought.

Mind Maps are mostly a simple visual index of ideas. RSA Animate is perhaps a closer representation of how Hand Drawn Maps can express Thought Experiments, but it is not enough.

To leverage Collective Thought, without resorting to the Lowest Common Denominator (no bad thing in itself), we need to be able to combine maps, and to do this flexibly at scale.

Graphs can be a representation of logic. They can be combined, filtered, searched, and recreated using hard logical operators.

While classical ontologies, may not be best suited to sketch, or to the Personalisation of Knowledge, there is more than hope. They are close. Emergent ontologies, and the simple social fact that to annotate a personal Knowledge Map is both Fun and personally useful - we have a realistic possibility of an effective social transformation - albeit with fuzzy edges.

It is however, precisely this fuzziness, not in the statistical sense, but the narrative and social sense, that promises a way forwards. People with similar thinking will have emergent ontologies close to ours. Librarians, and other Systematizers can later organise and classify our thoughts. We can find people and ideas to steal from.

Here we can clearly begin to see a methodology that we can implement. We can put into practice the bootstrapping of a social process, and the development of tools that support this process.

Starting with simple search, and our ability to crawl hypertext, fork, and reference links we can quickly form useful graphs based on a query. These graphs can leverage the power of a number of well researched algorithms, and the resulting image can be both expressive, visually appealing, and useful for navigation.

These simple graphs can then be brought into a drawing programme, where a great deal of joy can be derived from annotating them with logical connections, link types, and multimedia annotations.

Following this stage, and in part through the social connections made by connecting with others through the act of creating the graph, we can bring in well formed ontologies.

These micro-graphs, stored or otherwise linked permanently in the page (see IPFS), can then be crawled and combined to form ever more complex graphs, and powerful search and filtering capabilities. All without losing the contextual and personalised nature of the knowledge representation.

We have a plan.