Lit Review: User Interfaces for Personal Knowledge Management with Semantic Technologies

Dipl.-Psych. Heiko Haller. 2011


This thesis describes iMapping and QuiKey, two novel user interface concepts for dealing with structured information.

iMapping is a visual knowledge mapping technique that combines the advantages of several existing approaches and scales up to very large maps. 

QuiKey is a text-based tool to author, browse and query graph-structured knowledge bases in a step-by-step manner. It can be seen as an interactive semantic command- line that offers an alternative way to access the same structured information with very high interaction efficiency. 

Both tools are primarily intended for the domain of personal knowledge management, although they could also be applied more generally.

iMapping combines the core advantages of the mind-mapping, concept mapping and spatial hypertext techniques, which are incompatible in their original form.

By taking a zooming and nesting approach, iMapping allows for deep hierarchical structures, which are crucial for dealing with large amounts of information items. Linking these items in various ways – both formal and informal – allows users to build knowledge models at just the level of formalization that is beneficial for their specific needs.

* Interesting: We cannot manage knowledge itself, but we can manage knowledge cues. Anything can act as a knowledge cue: a knot in a handkerchief, a symbol, a keyword, a scribbled note, a checklist or a mind-map. Anything that reminds a user of what it signifies. *

Short-term memory is very limited. In fact, a human mind cannot have more than around four to seven items consciously present at the same time (Miller, 1956; Cowan, 2001). When dealing with complex subjects, this is a problem.
To be able to process higher amounts of items and grasp complex topics, the mind uses techniques of chunking and abstraction (Anderson, 2005). Also, literature on complex problem solving (Dörner, 2003; Vester, 2002) identifies as a core difficulty to understand the interrelations and interactions between things.
These cognitive shortcomings can be partly relieved by the use of external knowledge media that have been given many names like “memex” (Allegedly either from memory extension or memory index, Bush 1945), “augmentation to human intellect” (Engelbart etc

In fact, recent studies suggest that reducing such cognitive overhead (Conklin, 1987) must be of central concern to the PKM researcher.

In knowledge-intensive activities, it is even more crucial than otherwise, to unburden the user of all cognitive overhead in order to leave as much of the user’s limited working memory to the actual task at hand. Cognitive overhead is that part of a user’s cognitive load that is not directly related to the intended action, but rather to dealing with side- issues or the software as such.

Mind-mapping is an established technique for brainstorming, outlining, note taking and clustering. Mind-maps are however not suitable for relational structures between items because they are constrained to the hierarchical model.

The Seven Tasks for Visual Information Environments

The Seven Tasks for Visual Information Environments By Ben Shneiderman (1996):

  • Overview Gain an overview of the entire collection
  • Zoom Zoom in on items of interest
  • Filter Filter out uninteresting items
  • Details-On-Demand Select an item or group and get details when needed
  • Relate View relationships among items
  • History Keep a history of actions to support undo, replay, and progressive refinement
  • Extract Allow extraction of sub-collections and of the query parameters

“Overview first, zoom and filter, then details-on-demand”

Spatial Hypertext

Visualisation of text elements where meaning is shown through the layout. No connecting lines in the pure form but they can be used. An analog equivalent could be a pin-board or lots of paper on a table.

“To fuzzily relate two items, they are simply placed near to each other, but maybe not quite as near as to a third object. This allows for so-called “constructive ambiguity” (Shipman and Marshall, 1999b) and is an intuitive way to deal with vague relations and orders.”

Heiko Haller. 2011