esono.com | boris müller: Life Blog Projects Teaching & Research

Cultural Image-Making

17. September 2014

In the wonderful book Design for Information1 by Isabel Meirelles I came across a quote from Ben Shneiderman:

»Like Galileo’s telescope (1564–1642), Hooke’s microscope (1635–1703), or Roentgen’s x-rays (1845–1923), new information analysis tools are creating visualizations of never before seen structures. Jupiter’s moon, plant cells, and skeletons of living creatures were all revealed by previous technologies. Today, new network science concepts and analysis tools are making isolated groups, influential participants, and community structures visible in ways never before possible.«

I very much agree with this vision. As more and more data is collected and made available, the creation and the interpretation of  data visualisations is about to become an essential skill that will help to understand the complexities of the modern world. As Shneiderman says – these visualisations will allow us to see structures that would otherwise remain invisible.

While I agree with the vision, I am slightly uncomfortable with an underlying notion of the above quote. The comparison with telescopes, x-ray and microscopes implies that data visualisation is mainly a technical problem – a problem of resolution. Optical devices magnify small or distant objects. Lenses and mirrors direct and focus light waves and are thus revealing existing physical structures like plant cells or galaxies2. The process of creating optical magnifications is essentially deterministic3. The process of creating data visualisations is not.

This view is also expressed very well in the excellent posting »Worlds, not stories« by Moritz Stefaner. Moritz is referring to the similarities between data journalism and photo journalism:

»This is why I see data visualization as sort of a new photojournalism — a highly editorial activity using a deceivingly objective-looking apparatus.«

I would like to take this idea even further. Data visualisations are essentially cultural images. There is no definite way how to transform abstract data into an image. Even fairly obvious visual solutions like a bar chart are just one option how to represent data. The transformation of abstract data into a concrete form is a creative act. It’s design.

Take personal data – your bank balance, your emails, your travel itineraries, your school grades, your Instagram stream, your health data. If you were trying to create a visualisation of your digital self – how would it look? How would your public data image be different from your private one? We don’t know. We don’t know because we have not decided yet which questions we would like to ask the data. We don’t know because this visualisation does not yet exist. We have the data and the algorithms – but so far we don´t have a meaningful and adequate design.

I like the comparison of data visualisation with photography in terms of the role of the designer / visualizer. As Moritz puts it: »the most important issues today are not photographable«. So we need new image-makers that help scientists and journalists to convey complex information. However, I find the comparison problematic as it suggests that there is some kind of apparatus that allows us to take a snapshot of »reality«. As if we just have to find the right angle and the right spot in order to create an image.4

Data visualisation is about finding appropriate visual encodings that enable viewers to decode and interpret the generated image. The appropriateness of the visual encoding depends very strongly on the data that is depicted. Comparing this process to image-making with optical devices like microscopes, telescopes or photo cameras suggests that we are mainly dealing with a technical problem. It suggests that we just need better virtual lenses in order to see data structures more clearly.

Don’t get me wrong – it is true that the ability to record, store and process large amounts of data is only possible due to the great progress in computing that we are witnessing right now. Furthermore, the development of new visualisation algorithms provides us with new ways to manage the shape and relationship of visual entities. All these elements are great achievements in themselves. But you cannot derive meaning from technology.

In a way, data is like the script of a play. The words are all written down – but it is up to the director to decide how the words are spoken, who speaks them and how the environment of the play is set.

The transformation of abstract data into an image requires creativity. It is essentially a design problem. And I believe it is one of the most interesting and most relevant design problems in the years to come.


  1. Meirelles, 2013. Design for Information. Rockport Publishers, p. 63 

  2. It’s interesting that the further we look out or look in – the less the notion of amplification works. The borders of our scientific knowledge defy the experiences of our everyday life. 

  3. That is the magnification of form – not of colour. It is quite surprising that in many scientific images the colours do not represent the real appearance. 

  4. I have a lot of friends who are professional photographers and I will probably get a thrashing for this paragraph. But I believe there is a widespread naive understanding of photography. Here, I am referring to this understanding. The naive view of photography might be the subject for another posting. 



Continue reading:





RSS Feed | Twitter @borism | Impressum