Let’s get practical — Part 2

Continuing a series of posts on practical suggestions for converting information into meaning.

Initial capture of ideas and information

I follow the practice of capturing information from sources (mostly online) and saving it in discrete, labeled chunks. I use a similar process for Insights: Write an idea down as a statement and associate it with explanatory text. This is also the initial step of Incremental Formalization recommended by Shipman and McCall many years ago.

There are literally hundreds of tools for this purpose, but I’ve never found a truly satisfactory tool for this initial stage. So sometimes I live with the limitations of FreeMind. At other times, I capture information directly in TheBrain Pro — also a commercial product, but one that is available for most common computer platforms. TheBrain is my “master” knowledgebase tool. It’s far from perfect, but its storage format is XML — which means I can extract information about objects and typed relationships among those objects with XSLT if needed.

FreeMind is free and is a de facto standard for interchange among many tools that handle outline-like structures. It’s very good for capturing information and associated URLs and it supports HTML-formatted notes. But I find it clumsy as an outliner and the folks at TheBrain haven’t fixed their filter for importing FreeMind files. And (like most outliners) FreeMind does not allow you to specify any labels for the parent-child relationships supported in the outline.

But outliners are a good place to start deconstructing information into meaning

While outliners have significant limitations as information-capture and -organization tools, some outliners are useful for supporting the process of deconstructing text into objects in a simple hierarchy. (I use TreePad, an inexpensive commercial product for Windows. I find it very fluid at its primary functionality — moving levels of an outline around. It has simple text-file/CSV import and export features, but they do work.)

If you find an outliner that you feel comfortable with, try grabbing a useful paragraph or two from a well-written resource and breaking down that text into discrete objects — one line per item in a text file, for example — and importing that informaiton into the outliner. When I do this, I find myself explicitly interpreting the linear text, re-wording it for simplicity and clarity, eliminating unimportant details, and deconstructing it into small hierarchies that reflect the logic.

[I even use the “note” (or “article”) text associated with the headings as a way to capture the zones of text associated with an Insight expressed in the heading of an outline element. That’s part of my model and practical implementation. You may find different methods.]

So you have taken the first step: You have converted strings of words into discrete objects that you can point at unambiguously. You can say, “I’m talking about that statement.” If you have imposed a hierarchy, you can say, “This statement subsumes other statements.”

That’s a huge step toward representation of meaning. It’s easy technically, but it requires some serious thinking, because you are, in effect, reverse-engineering the meaning of the author’s words from linear text.

Of course, it’s your interpretation of that meaning. But that’s a different issue.

© Copyright 2017 Philip C. Murray


This entry was posted in aspects of practical knowledge representation, strategies and tactics. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *