There’s one at every meeting, the know-it-all. He or she knows the name of every attendee, the agenda backwards and forwards, the subject matter to a tee, and finishes your sentences better than you could. If that leaves you annoyed (i.e. jealous), then it is time to fight back… electronically! After reading so many books and barely remembering any of them, I knew I had to approach learning differently.

Data Normalization for the Human Database

There are numerous methods for packaging and accessing knowledge, from databases to the simple act of highlighting text. My approach is to mercilessly summarize the content, classify it using tags, and build a table of content that links to those tags for quick navigation and knowledge retrieval.

The process is like reverse engineering a book to get back to the author’s original ideas. There is undoubtedly a cost to such minimization in the form of information loss, but in the overall scheme of things, isn't it better to remember the key ideas than forgetting them all?

The above picture is a snapshot of one of my knowledge documents. The right pane holds the summary and labels while the left pane contains a linkable outline to navigate and bring into focus those labels and corresponding summaries.


This process came together during a technical phone interview for a potential job. After creating many of these documents for work, I decided to apply the same principles in organizing and summarizing personal knowledge. During the interview, I was surprised how quickly and easily I was able to compliment my answers with precise technical terms and rich descriptions with the help of such documents. I did get the job.


The digital revolution has overwhelmed us with so many organizational tools; there is something for every taste. You probably have heard of Evernote, Livescribe, Mindview, OneNote, etc. These are the big ones and they essentially serve the same goal of organizing information but each use differing approaches. I err towards simplicity and rely on Microsoft Word. This old standard may be low-tech compared to what’s out there today, but it gets the job done.


Part of its attractiveness is the ability to do the same thing many different ways. I like to type my notes and quickly categorize them with the help of labels. The 'Styles' tool coupled with the 'Navigation’ pane are my big guns here. In a nutshell, I summarize the main ideas and classify each with a title. I then transform each title into a linkable label by highlighting it and clicking the ‘Styles’ toolbar at the appropriate level (check your help file for specifics about ‘Styles’ and ‘Navigation’ as they vary from version to version). Main ideas are usually categorized as ‘Heading 1’ and are the second top-most level behind the ‘Title’ style. Sub ideas are categorized as ‘Heading 2’, ‘Heading 3’, etc. Styling something not only applies a graphical style to the text, but more importantly, adds it to the ‘Navigation’ pane.

The ‘Styles’ toolbar

The ‘Navigation’ pane is the document map that displays every label in a hierarchical, tree-based list. It offers expandable/collapsible toggles. For the consumer of these documents, this is the control center to navigate through the document and gain rapid access to the subject matter by simply clicking on the label.


The ‘Navigation’ pane


Beyond reading the material, it has to be understood and understood well enough to differentiate key points from fluff. Wordy passages, duplications, examples, anything not applicable to ones needs are all easy targets for the garbage heap. How intuitive you design your outline will determine how useful a tool this will become. It is important to end up with enough summary material to cover the subject but light enough that it works as a memory inducer via minimal clicking and minimal reading. Remember, you may not access this information for a while and need to assume you will not remember any of this when you come back to it. If the top element doesn’t logically guide you down to the next element, your efforts will be wasted. For example, taking a 500 page technical document and reducing it to 20 is ideal but only if the labels flow into each other logically.


Not all material is suited for this type of transformation. Obviously, there has to be a need to revisit this knowledge at least once more in the future, preferably more. Also the material needs to be reducible, if it is highly dense then it may not be worth the effort. You do not what to end up with a one-to-one copy of the original, at that point save yourself the time and buy the eBook version.

A book’s table of content is a good starting point. You can trust the author did his or her due-diligence to break down the subject logically and chronologically. Mark all chapter headings as ‘Heading 1’ with the ‘Style’ tool, and all sub-chapters as ‘Heading x’ (depending on the depth offered by the book’s table of contents). Each label or series of labels should lead to a summary.

Once that is done, customizing the tree structure is very easy. Simply dragging a label up or down will re-order the whole child hierarchy and summary accordingly. A label can easily be upgraded to a higher level or downgraded to a lower one by right-clicking it and selecting ‘Promote’ or ‘Demote’. Entire sections can be deleted directly in the ‘Navigation’ pane by right-clicking on the label and selecting ‘Delete’. Adding additional headings is as easy as right-clicking on the ‘Navigation’ pane and selecting ‘New Heading Before/After’, or typing your label directly in the document and labeling it using the ‘Style’ tool.


If you’re stuck during the summarizing effort, here are some mechanical rules that have helped me: only trim a paragraph after it is read and understood, only allow one resulting sentence per original paragraph, and never use the ‘Cut and Paste’ feature as it leads to bloated summaries and cursory reading of the material. Don’t sweat the small stuff; this should be customized for your intellectual needs, if you aren’t sure, leave it out. Also, if a certain area doesn’t flow, move on and revisit the issue from a ‘Navigation’ pane perspective. Once things are all labeled, it is much easier to judge if something is out of place.

Because everything is collapsible, this method can handle large amounts of information.



This is what it's all about. If the summarizing is well done, finding the right label in the ‘Navigation’ pane and drilling down to its summary will be a cinch. If it is done really, really well, it can be used in a real-time setting (such as at meetings or over the phone). In theory, you should be able to collapse all levels to the top most headers (Heading 1) and effortlessly drill down, in a fluid and logical manner towards your summary.

This doesn’t need to end at the document level; related documents can all be saved in the same folder. Also, tools such as Microsoft OneNote allow you to import multiple documents under a common tab making their collective content searchable as a group from a single location.



Microsoft Word isn’t the focus here as most applications offer similar organizational features. Also, the methodology is just one of the many ways to store and catalog information, the important idea here is to get in the habit of summarizing personal knowledge when encountered so it isn’t lost. I just wish I had started doing this much, much sooner.