incorporate some (social) network analysis features to create deeper Brain insights
There are many graph-based network analysis software tools available but, as far as I know, they all require structured electronic data such as email logs or data from participation in online discussions. TheBrain is already excellent for developing and managing unstructured data, for example, business relationships between companies observed over time by monitoring trade journals or Google Finance etc. It can also manage "multi-nodal" networks by using thought types, link types and tags. However, it does not allow anything more than a casual analysis of the resulting networks.
However, it comes close. By using custom reports and then sorting by Activity Level, we can get an impression of what the most important nodes in any network might be. However, their rank is based on how many times the user clicks on them whereas a better approach would be to rank them according to "link popularity" i.e. how many other thoughts link to them.
The Analyze Main Thoughts tool seems to do something like this but does not allow filtering by thought / link types and tags.
The incorporation of standard network analysis measures such as centrality, connectedness and clusters could radically improve the power of TheBrain as an analytical tool. Version 7 seems to have made big improvements with links data but it would be great if there was a way to attach a mathematical weight for each link which would affect the statistical results of these kind of measures. Therefore, as well as performing a visual analysis of my thought network, I could run formal network analyses as well which would offer many new insights.
For example, if I was mapping relationships between people in my company I could capture informal links (e.g. friends with, mentor to, often helps) which would help identify key people in a different and perhaps more useful way than merely showing their status in the formal hierarchy.
Finally, such capability would open a new market for TheBrain as it could also work for structured network data in .xml formats, for example.