También disponible en Español


The digital magazine of InfoVis.net

Diagrams for Visualisation
by Juan C. Dürsteler [message nº 186]

Should any discipline deserve a diagram summarising it, this is clearly Information Visualisation. In this issue we review three of the most appealing ones as a previous step to the proposal of a synthetic diagram that encompasses elements of the three together with other contributions.
DiagShedroff_en.gif (189993 bytes)
Elementary scheme of the conversion of data into wisdom.
Source: Graphic by the author adapted from the one present in the article "An Overview of Understanding" by N. Shedroff in  the book Information Anxiety 2 by R.S. Wurman.
Click on the image to enlarge it.

I've been wondering for a little while about one of the key points of Information Visualisation: the basic scheme by which data become information and this is transferred to our brain by stimulating our sensory perception building a cognitive experience mediated by previous experience, culture and context.

In this issue and the following I intend to review some of the approaches, proposing an extension to the diagrams currently in use.

From data to wisdom

In order to see step by step the elements of said scheme it's necessary to remember our definition of Information Visualisation as the process of knowledge internalization by the perception of information. (See more info at the glossary and one discussion about this topic at number 100

Nathan Shedroff  considers the process that leads to understanding as a continuum that begins in the data and ends in wisdom, passing through information and knowledge. We reproduce here the scheme that Shedroff proposes , adapted from his articles "An Overview of Understanding" included in the book Information Anxiety 2, by  Richard S. Wurman

In Shedroff's diagram there are four main conceptual elements, that we already mentioned in number 94, and that we have represented as an evolution in time, along 2 axes, the increase of understanding on one side and the increase in the importance that context, understood as culture, experiences and the set of acquired patterns has. 

These are the following ones (quoted from number 94): 

  • Data is simple facts, lacking any context. If data doesn'tinform us, it's not information. Or equally valid, without context data it is simply the raw material from which we depart to understanding. 07012007 is data, that can hold many meanings, a date, a batch number, an anniversary...

  • Information. Information is data put within context. It's a concept bound to that of metadata, data that refers to the meaning of other data. For example if in a table of data one of the columns is labeled as "batch number" 07012007, one string in that same column, gets a particular meaning. Information is the distillation of data or data with a meaning, but this still is not knowledge

  • Knowledge. What differentiates knowledge from Information is the complexity of the experiences that you need to reach it. In order for a set of information to become knowledge one has to be exposed to it in different ways and one has to elaborate a personal experience about it. We also saw in number 105  that knowledge can be expressed as a pattern whose measure of interest for the user is above a certain threshold.  That is, if some information is not interesting for us it's very difficult for it to become knowledge. Knowledge is not transferable, you have to build it yourself by experiencing the information.

    In this sense, Shedroff promotes “experience design” as the way to create the experiences that build knowledge in the most efficient way.

  • Wisdom. The ultimate level of understanding. With it we understand a broad enough set of patterns and meta-patterns in such a way that we can use and combine them in new ways and situations that are completely different to those that served us to learn. Wisdom is, like knowledge, something personal that has to be elaborated intimately, that unlike data and information, is bound to certain people and is lost when they disappear. For this reason it’s almost impossible to transmit it directly. 

In the diagram there are two circles that indicate the scope of the people who build information from data (producers) and the scope of those that consume information and process it into knowledge (consumers). A wider circle of Context encompasses the step from information into Knowledge and from this into wisdom. Knowledge is surrounded by a wider circle of Experience.

Shedroff's diagram is eminently conceptual and it makes neither reference to graphics or other artifacts nor to the way the transformations that provide the conversion of one entities into the others are produced.

The process of visualisation

On his side Colin Ware, in his book "Information Visualization: perception for design" (pages 4 and 5), states that there are four basic visualisation stages combined with a certain number of feedback loops, that can be summarised in the following diagram:

DiagWare_en.jpg (66176 bytes)
The diagram of the process of visualisation  according to Colin Ware.
Source: Graphic by the author adapted from the one present in Information Visualization: perception for design.
Click on the image to enlarge it.
  • The collection and storage of data itself”.

  • The pre-processing designed to transform the data into something we can understand”.

  • The display hardware and the graphics algorithms that produce an image on the screen”.

  • The human perceptual and cognitive system”.

There are basically three feedback loops: data gathering, data manipulation and data exploration. The information analyst can gather a series of data, inject it into the process and after a preprocessing and transformation stage obtain a graphical representation of it that eventually activates her or his visual and cognitive system. He/she can also manipulate the way in which the graphics engine shows the data once preprocessed and transformed, maybe changing the colours or the geometric transformations that display it.

The person analysing the result can also explore the data, selecting between different pre-processes. For example selecting different subsets of data or transforming the data to produce derived magnitudes like differences between data or statistical magnitudes computed on it, showing it finally on a new graphic representation.

The  physical and social environment play, according to Ware, a role in the data gathering loop. As he states "The physical environment is a source of data while the social environment determines in subtle and complex ways what is collected and how it is interpreted"

The construction of visualisations 

On the other hand Card et al in the book "Readings in Information Visualization: Using vision to think” propose the following diagram for visualisations, understood as “adjustable mappings from data to visual form to the human perceiver”.

DiagCard_en.gif (41094 bytes)
The diagram for the construction of visualisations by Card et al. 
: Graphic by the author adapted from the one present in the book Readings in Information Visualization  
Click on the image to enlarge it.

This diagram covers the transformation from raw data to graphics. Raw data in whatever format is transformed into structured data tables that have some meaning represented through metadata.

This is accomplished by using “data transformations”. These data tables are in turn mapped into some visual structure that produces a graphic representation. The transformations that make this possible are called visual mappings

For example a 3D data table can be transformed in a 3D graphic using each one of the columns associated to a particular variable (we can imagine that their metadata is petrol consumption, range and speed of a car). The same table can be used to produce a 2D representation with the 3rd variable represented by the size or colour of the points placed in the chart according to the other two variables.

Finally you can see the representation from different standpoints. This is done by view transformations that scale, translate, zoom and clip the graphic representation. Interaction allows the user to bring feedback to the system by changing the parameters that control the three types of transformations already mentioned.

As we see this diagram is closer to the technical process of transforming raw data into graphic presentations.

Up until here we have seen three approaches to the process of our interest converting data into understanding. Each one places the emphasis on different aspects of the process, conceptual in one case, more perception bound in the other and closer to computer graphics in the third  one.

In the next issue we'll see another diagram that synthesizes traits of all these three along some more contributions that we propose in order to represent Information Visualisation understood as a process of interiorisation of information.

Links of this issue:

http://www.infovis.net/printRec.php?rec=glosario&lang=2#VisualizacionInformacion   Glossary entry Information Visualization
http://www.infovis.net/printFicha.php?rec=revista&num=100&lang=2   Num 100 Visual Metaphors
http://www.nathan.com/   Nathan Shedroff's website
http://www.infovis.net/printFicha.php?rec=revista&num=94&lang=2   Num 94 Knowledge and Information Architecture
http://www.infovis.net/printFicha.php?rec=revista&num=105&lang=2   Num 105 about Discovering the Knowledge
http://www.infovis.net/printRec.php?rec=llibre&lang=2#InfoVisWare   The book Information Visualization: perception for design by Colin Ware
http://www.infovis.net/printRec.php?rec=llibre&lang=2#Readings   The book Readings in Information Visualization... by Card et al.
© Copyright InfoVis.net 2000-2014