Name: Mike Grice Title: Data Visualisation Observations.

Course: BA iMedia Module: Reflective Practices Module Code: IDCEM 3004 Date of submission: 12 March 2010

“We thrive in information-think worlds because of our marvellous and everyday capacities to select, edit, single out, structure, highlight, group, pair, merge, harmonize, synthesize, focus, organise, condense, reduce, boil down, choose, categorize, catalogue, classify, list, abstract, scan, look into, idealize, isolate, discriminate, distinguish, screen, pigeonhole, pick over, sort, integrate, blend, inspect, filter, lump, skip, smooth, chunk, average, approximate, cluster, aggregate, outline, summarise, itemize, review, dip into, flip through, browse, glance into, leaf through, skin, refine, enumerate, glean, synopsize, winnow the wheat from the chaff, and separate the sheep from the goats.”
Edward Tufte. Envisioning Information. p.50

CONTENTS
Page 3. List of Illustrations Page 4. Introduction to Data Visualisation Page 5. Brief History of Data Visualisation Page 11. Patterns of Intent Page 14. Visual Metaphors and Visual Literacy, Posterisation and Chartjunk. Page 18. Bibliography and References

2

LIST OF ILLUSTRATIONS
Page 5. Figure. 1.1 – ‘Map of Konya, Turkey’ Page 6. Figure. 1.2 – ‘Planetary movements’ Page 7. Figure. 1.3 – ‘Bill of Mortality’ Page 8. Figure. 1.4 – ‘Nightingale Rose Diagram’ Page 9. Figure. 1.5 – ‘Dublin Passenger Flow Map’ Page 10. Figure. 1.6 – ‘Pioneer Plaque’ Page 11. Figure. 1.7– ‘Stockport Emotion Map’ Page 12. Figure. 1.8 – “Close-up of Stockport Emotion Map” Page 13Figure. 1.9 - “Flight Patterns by Aaron Koblin”

3

1. INTRODUCTION TO DATA VISUALISATION

This paper is about Data visualisation. Data visualisation is the visual representation of data, in order to communicate information clearly and effectively. Data Visualisations or Information Graphics aim to visually/graphically represent information and data in a clear, easily accessible way. They are all around us - everywhere we go, data visualisation follows us. They are used wherever complex information is present; in magazines, business reports, television featuring heavily weather and finance channels, other examples include road signs, maps, education, textbooks and technical manuals. Commonly occurring ‘Info Graphics’ are things like bar, pie charts and line diagrams, these being typically used when data needs to be produced and displayed quickly and effectively. Ferdi van Heerden explaining Data visualisation. “To convey ideas effectively, both aesthetic form and functionality need to go hand in hand, providing insights into a rather sparse and complex data set by communicating its key-aspects in a more intuitive way.”
Ferdi van Heerden. Data Flow

4

2. BRIEF HISTORY OF DATA VISUALISATION

“ …information visualisation takes us back to the earliest scratches of forms on rocks…to the earliest use of diagrams in the history of science and mathematics.”[3]
Michael Friendly. "Milestones in the history of thematic cartography,
statistical graphics, and data visualization". p 2

“The principles of information design are universal – like mathematics – and are not tied to unique features of a particular language or culture.” [4]
Edward Tufte. “Envisioning Information” p.10

The history of Data visualisation encompasses thousands of years of human history and development. With no developed form of written communication, early humans relied on glyphs and symbols to communicate ideas and information. The ancient iconography on rock carvings is an example of data visualisation, it uses primitive graphics to illustrate important information. Cartography is considered as the one of the earliest forms of data visualisation, it requires the clear representation of information such as locations, landmasses, and rivers. This is one of the earliest known examples of cartography, a map of the Turkish town of Konya.

5

Figure. 1.1 – ‘Map of Konya, Turkey’ http://www.math.yorku.ca/SCS/Gallery/images/oldest-map.jpg

Maps like this one help us understand how graphical representation of information has evolved; it tells us something about the early humans who collated geographical information, and how they used signs and symbols to represent their findings. If we take Data Flows’ concept - that there are six distinctive styles (‘Patterns of Intent’) within the field of data visualisation design, it is interesting to note that the ‘Patterns’ have been present throughout the history of data visualisation. This map, produced around 6000 years ago, is a good example of a ‘Datascape’. Using symbols and colour to represent the different elements to help the viewer derive meaning. The height of the ground is graded using two different distinct shades, the black and white shapes are thought to represent man-made structures like buildings, while the dark brown cone shape above is an erupting volcano, complete with ash cloud. As human cognitive thinking progressed, increasingly more complex information was beginning to be visualised. Astrological data, such as planetary movements, was being measured using increasingly sophisticated instruments. This graph comes from Cicero's ‘In Somnium Scipionis’. [5] This is one of the earliest known attempts (circa 950) to graphically depict the changing positions of celestial bodies in the sky. Positions of objects like the Sun and the Moon were recorded through graphs like this.

6

Figure. 1.2 – “Planetary movements” http://www.fi.uu.nl/wiskrant/artikelen/hist_grafieken/begin/images/planeten.gif

X-Axis/Horizontal plane = time of day / horizontal position in sky Y-Axis/Vertical plane = object / vertical position in sky. The symbols marked on the Y-axis represent the objects in the sky. The symbol on the bottom is the Moon. If we follow the Moons’ line, as the time of day increases so does the Moons vertical and horizontal position. It is at its highest point when it reaches the centre of the graph, logically we can conclude that this is Midnight. Applying the same theory we can assume that the symbol at the top of the graph depicts the Sun. As the day goes on the Sun’s horizontal and vertical position decreases, coinciding with the rise of the Moon. This graph could be considered to be an example of ‘Datalogy’. It uses the very literally translation of plotting the objects position in the night sky on the graph, therefore the reader can very easily visualise the X-axis as the horizon, with the objects rising and setting in the distance. As the world was entering the 1600’s, enlightened thinkers began to wonder about physics and measurement, speed, time and distance. Demographical statistics were now beginning to be recorded, leading the way for data visualisations using large quantities of information; population numbers, rates of immigration and all manner of social and economical data was being to be recorded. This table comes from John Graunts’ “Natural and Political Observations Mentioned in a Following Index and Made Upon the Bills of Mortality”[6], showing mortality rates in London through various years – 1592 to 1665.

7

Figure. 1.3 – ‘Bill of Mortality’ http://www.math.yorku.ca/SCS/Gallery/images/dan/mortality_table.jpg

It is the first example to use text, thus I believe that the designer has taken much of the deciphering of the graph away from the reader. This table is a good early example of ‘Datablocks’. It uses the clean grid of 6 primary columns to separate and display the information. All of the columns are headed with the same title: ‘Buried of all Diseases in the Year 1592’ for example, then the columns subheading are listed in this order - dates of burials, ‘Total’ – being the number of total deaths, and ‘Pla’ being those deaths caused by Bubonic Plague. Each column is closed with the statement similar to this one: ‘The Total of all that have been buried is 25886 whereof of the Plague 11503’. Tables like this help historians determine the severity of the Plague at various times of the year, throughout the years of infection. Other interesting facts can be gathered by studying this table, like the number of burials from deaths caused by Plague in the 97 Parishes without walls - 2696. Florence Nightingale is a thought of as a pioneer of information visualisation. Her efforts to improve sanitary conditions for the troops fighting in the Crimean War led to her to creating her own diagrams, referred to as ‘Nightingale Rose Diagrams’. [7]

8

Figure. 1.4 – ‘Nightingale Rose Diagram’
http://www.egge.net/~savory/stus_blog_pix/florence_nightingale2.jpg

This is an excellent example of good design in data visualisation, it is also an example of a ‘Pattern of Intent’’ – the Datasphere’. Nightingale used diagrams to highlight the problems the Armed Forces faced to the British Government. The months of the year each have a corresponding wedge, and are named appropriately. All of the wedges are measured from the centre, the blue wedges represent deaths from preventable or mitigable diseases, the red represent deaths from wounds and the black wedges represent deaths from all other causes. The diagram does not contain any numerical information, thus the reader is left to wonder at the numbers involved, but the contrast that Nightingale draws between the large blue wedges and the red and black wedges is substantial. Numerical data would only clutter and cloud the real issues put across by the diagram, leading the reader away from the important point Nightingale was trying to make. The success of graphs like these were so significant that it led to revolutionary upgrades to nursing practices and medicinal care out in the battlefield.

9

The 1800’s brought about greater means of public transport, graphs like this one, created by Henry Drury Harness [8], depict the train routes and passenger numbers in and out of Dublin. This graph uses lines to show transportation routes with the thickness of the relative to the number of passengers traveling on that route. This map would be considered as an example of the ‘Pattern of Intent’ called a ‘Datanet’. The underlying network can be clearly seen by examining this map, although the towns and cities are marked, the major transportation routes effectively give away their position. The strong thick line connection shown between Dublin and Dundalk tells of the two cities close ties, Dundalk being
Figure. 1.5 – ‘Dublin Passenger Flow Map’
http://www.math.yorku.ca/SCS/Gallery/images/harness-flow.gif

equidistant from Dublin and Belfast, this would be an important, regularly used route.

In the 1970’s Nasa launched the Pioneer 10 and 11 spacecrafts featuring gold plaques with pictorial messages engraved on them. The goal of these plaques was to carry the first message from humankind to outer space, and, if ever intercepted by other life forms, inform them about life on Earth. The ‘Pioneer Plaque’ features a schematic representation of a hyperfine transition of neutral hydrogen, the Earths position relative to the position of the Sun, along with the trajectory the Pioneer Spacecrafts took from Earth.
Figure. 1.6 – ‘Pioneer Plaque’
http://vintageprintable.com/wordpress/wp-content/gallery/rare-bookphysics-mathematics/pioneer10-plaque.jpg

10

Also depicted are two figures, male and female. A simple line drawing of the spacecraft is shown behind the figures; by placing the figures beside the spacecraft the designer gives the (possible) extraterrestrials an idea of a human beings size, relative to the size of the spacecraft. This would be an example of a ‘Datanoid’ as these info-graphics place humans in the scale, to give relevance and impact. Although this is not a graph in the traditional sense, it is an example of how far humans have come. From simple scratches on rocks to interstellar messages, data visualisation is an important, underappreciated art form that has been a driving force of human development for thousands of years, and will continue to push the fields of science, mathematics, architecture and design.

3. PATTERNS OF INTENT
The ‘Patterns of Intent’ are the six styles of Data visualisation that are classified by ‘Data Flow’ [9]. Under the ‘Patterns of Intent’ a pie chart becomes a ‘Datasphere’ and the bar chart becomes a ‘Datablock’, they are new terms coined by the Data Flow team to classify the different styles of data graphics. In this chapter the two most interesting ‘patterns’ are discussed, with examples being evaluated.

3.1 Datascapes The origins of the word ‘landscape’ are ambiguous. Its roots derive either from a combination of

11

‘land’ and the Dutch word for ‘ship’, or the German verb ‘schaffen’ – to create. In datascapes, both meanings suggest the potency and responsibility of the designer in guiding the viewer through a complex sea of meaning. Elevating the reader from ‘Flatland’- the reduced, lessened experience or reality that results from subjecting real experience two-dimensional expressions – they create a journey of context and interaction. Perspective is blended with graphic frameworks to bring depth and meaning to the expression of data.
Data Flow, p.97

‘Datascapes’ are predominately used when data relating to space, distance and environments need to be visualised. Topographical data appears to lend itself well to ‘Datascape’ visualisation, the evidence being that the most common example of a ‘Datascape’ could be considered to be the simple map. The map, detailing the location, its roads, buildings and natural landmarks like trees, rivers and mountains is an excellent example of how complex data can be successfully compressed and easily digested by the reader. This map was created by Christian Nold Biomapping, it is a cartographic depiction of Stockport. This map takes the physical features of the town, streets and building and overlays them with visual explanations of over 200 participants emotions. There are two sets of data depicted on this map, the first being the participants responses to question.
Figure. 1.7– ‘Stockport Emotion Map’ http://www.softhook.com/stock1.jpg

The participants were asked to sketch their responses to a variety of serious and humorous questions and their daily lives, questions like “What really annoys you about Stockport?” These drawings were then placed as close as possible in their geographically correct position. Data like this was used in a number of interesting ways, based on responses to the “Where do you usually meet your friends?” it was determined that the youth of Stockport are particularly marginalised and seem to congregate in the skate-park.

12

The second set of data shown on this ‘Datascape’ is the most interesting, not only because of what it is being visualised, but how the data was collected. The data concerned is emotional arousal based on geographical location; the artists developed a special device that was worn by the participants and the data was gathered as the participant walked freely around the town.

Figure. 1.8 – “Close-up of Stockport Emotion Map” http://infomagination.typepad.com/.a/6a00e554eecbdf8833011570596141970b-500pi

The device measured emotional arousal and traced their location, the red lines, while the emotional arousal is represented as red bars, corresponding to the intensity of the arousal. Interesting patterns appear upon inspection, large clusters of bars appear more frequently in the market area and high street, while the suburban areas are relativity ‘quiet’. The heightened emotions in these areas could be triggered by a number of things; the increased number of people – thus the chances of social interaction and responses to annoyances like litter and graffiti. ‘Datascapes’ are not just in the employ of cartographers. Visualisations like these offer a wide range of opportunities to designers to use spatial, topographical statistics and combine them with deeper meanings. They allow the reader to navigate through the data, catering for a deeper understanding of the subject matter, the readers personal experiences our imposed upon on the landscape.

13

3.2 Datanets When individual data points develop tension and connection with each other, the resulting structure becomes and entity in its own right – the network. It draws life essentially from connection and connectedness, and it is these qualities that a directed explicitly by the designer to show cause, context, or collaboration.
Data Flow, p.55

‘Datanets’ are interesting, they suggest the links between topics and allow the reader to navigate through the information as if it were a narrative. ‘Datanets’ as the name implies, are about networks, they are useful when trying to represent links. The links themselves often provide more information than the nodes; the structure of the network will sometimes allow the reader to visualise underlying relationships often left unspoken by the information graphic itself.

Figure. 1.9 - “Flight Patterns by Aaron Koblin” http://deeperwants.com/ratboys_anvil_2/flightpaths.jpg

This is an examples of a ‘Datanet’ produced by designer Aaron Koblin. ‘Flight Patterns’ is a motion graphic piece of information design that depicts flight patterns in and out of the United States of America. This graphic contains no text, yet the information coming through is so strong. Without any outlines or borders to indicate the landmass the network provides the reader with enough information to build up use their own knowledge and recognise the shape of the country. The underlying network and information is subtly veiled by the design, but becomes apparent on closer inspection, 3.3 Conclusion on ‘Patterns of Intent’
14

These two examples have been chose because they represent the most interest to my independent project. For my final piece of work in response to the ‘Independent Digital Project’ brief, I aim to produce an interactive piece that incorporates my research into data visualisation and the ‘Patterns of Intent’. As the piece will be an interactive map of Merseyside, understanding ‘Datascapes’ and how they can be used effectively has helped greatly, along with the possibilities of showing underlying connections and networking stories that comes with ‘Datanets’, this research and reflection will provide the cornerstone to my independent project.

4. VISUAL METAPHORS AND VISUAL LITERACY, POSTERISATION AND CHARTJUNK.
Data visualisation could be seen as Visual Metaphors, a visual metaphor being defined as the presentation of a person, place, thing or idea by way of visual image, suggesting a particular association or point of similarity. Ferdi van Heerden - “Visual metaphors are a powerful aid to human thinking.” What is a metaphor? It is the direct comparison between two or more seemingly unrelated subjects describing one subject as being alike to another subject in some way. They are useful to illustrate ideas, simplify complex subjects and making people think. They are extensively used in poetry, music, advertising and art. These are two of the most horrendously clichéd metaphors, “You are my sunshine” and “Strong as an Ox”. They are used to describe people; “You are my sunshine” does not mean that you are literally that individual’s source of sunlight, but that you are important to them, like the Sun is to all life on Earth, whereas “Strong as an Ox” alludes to that person being very strong, almost Ox-like, not literally having the strength of an Ox. Like visual metaphors, data and information graphics use imagery to present information, make or suggest association, simplify complex subjects, and allow viewers to extract meaning. Visual metaphors = Data Visualisations = A Picture is Worth 1000 Words When discussing visual metaphors and data visualisations the old proverb “A picture is worth 1000 words” is an interesting concept to look at. “A picture is worth 1000 words” is the idea that a single image can tell the viewer more about the subject matter than 1000 words written on said subject ever

15

could. Data visualisations are fundamentally the manifestations of this proverb. Data visualisations can store vast quantities of information and makes that information readily available to people who would otherwise not have the time to absorb numerous pieces of text. A single highdensity data graphic can be more useful to someone than 10 leaflets on the subject. Given the choice of keeping hold of 10 scattered pieces of text or a highly informative data graphic they would more than likely choose the later. Visual literacy is an important aspect of data visualisation, it is an individual’s ability to interpret or make meaning of information presented to them in an image and is based on the theory that an image can be ‘read’. This is an important aspect to consider when working with data visualisations, how a person interprets or ‘reads’ your graph or diagram can be the difference between a successful visualisation and a failure. Designers must always be aware of visual literacy, but not to the point of oversimplification of the visualisation or assuming that the data is too complex for the audience. Oversimplification of data visualisation leads to data thin designs, the ‘Posterisation’ of an information graphic. Posterisation is a concept developed by renowned statistician Edward Tufte. It is the act of applying graphic design ideologies regarding things like typography, object representation, layout colour, production techniques and visual principles to the visualisation of data and information, and allowing it to dictate the presenting of information. Josef Albers wrote about typography is true for data visualisation and information design: “The concept that “the simpler the form of a letter the simpler it’s reading” was an obsession of beginning of constructivism. It became something like a dogma, and is still followed by “modernistic” typographers. The fashionable preference for sans-serif in text shows neither historical nor practical competence.”[2]
Albers, J. Interaction of Color (New Haven, 1963; revised edition, 1975, p, 4.

Albers makes the interesting point that designers often go against clutter and complexity, and tend to favour the clean/simpler approach of ‘modernistic’ sleeker typefaces. The same can be said about some designers approach to data visualisation and information design. What E.B. White said of writing is equally true for data visualisation and information design: “No one can write decently who is distrustful of the reader’s intelligence, or whose attitude is patronising”
[1] Strunk, Jr. William. White, E.B.The Elements of Style (New York, 1959), p.70

16

High-density designs allow the reader the depth of information needed to interpret meaning, to navigate through the information, personalising the data for their own use - the interpretation of the information has been given over to the reader, the designer is no longer dictating the delivery of the message. “High-information graphics, convey a spirit of quantitative depth and a sense of statistical integrity. Emaciated data-thin designs, in contrast, provoke suspicions- and rightfully so – about the quality of measurements and analysis”
Edward Tufte. Envisioning Information. p.32

A lack of data leads to misleading data. Data-thin designs are misleading and show contempt for the audience. Misleading data can be damaging, they rightly prompt suspicion from the audience, giving rise to questions like “What are they hiding?” and “Is that all they know?” “Clutter and confusion are failures of design, not attributes of information. Often the less complex and less subtle the line, the more ambiguous and less interesting is the reading. Stripping the detail out of data is a style based on personal preference and fashion, considerations utterly indifferent to substantive content”
Edward Tufte. Envisioning Information. p.50

Data posterisations may be aesthetically pleasing but will ultimately lack the density of data required to convey important information. Tufte uses Maya Lin’s Vietnam War Memorial as an example of how information takes precedent over aesthetics without being compromised; the importance of the names of the fallen soldiers outweighs the choices she has made as the designer. Serious information requires the credibility only found within data rich visual displays. “…who would trust a chart that looks like a video game?”
Edward Tufte. Envisioning Information. p.34 .

‘Chartjunk’ is a term coined by Tufte referring to useless, excessive elements in information design pieces that often distract the reader from the information. Examples of ‘chartjunk’ are unnecessary text, complex typefaces, visually ‘noisy’ and cluttered backgrounds. Tufte on ‘chartjunk’: “The interior decoration of graphics generates a lot of ink that does not tell the viewer anything new. The purpose of decoration varies — to make the graphic appear more scientific and precise, to enliven the display, to give the designer an opportunity to exercise artistic skills. Regardless of its cause, it is all non-data-ink or redundant data-ink, and it is often chartjunk.”
Edward Tufte. The Visual Display of Quantative Information. p.21

17

Chartjunk, like posterisation, tends to lead to confusion. The information may be present within the information graphic but is clouded by the aesthetic choices made by the designer. Both show the designers contempt for both the data/information and the audience.

1.1 Conclusion, of sorts There aren’t many arguments in favour of ‘Posterisation’ or ‘Chartjunk’ but as a designer, the words of Ludwig Mies van der Rohe ring true ‘Less is more’. Although Tufte feels this phrase does not apply to data visualisation, there is a subtle difference between a posterisation being effective and still visually stunning information graphic and being a complete failure. Although, as Mies van der Rohe also said “God is in the details”. A successful data visualisation/information graphic is dependent on the quality of information, good design will not hide the fact the data is poor. Although ‘posterisation’ is deemed by Tufte to be degenerative to information design, the subject matter can be the deciding factor in how a designer approaches the data. A designer should respect the data and information and graphically represent it in the most effective way, retaining the integrity of the data. If the information allows the designer to have freedom over the way it is presented then the designer should by all means create the most aesthetically pleasing graphic possible. “The deepest reason for displays that portray complexity and intricacy is that the worlds we seek to understand are complex and intricate” Edward Tufte. Envisioning Information. pp. 33 + 51

18

References [1]Strunk, Jr. William. White, E.B.(1959) The Elements of Style (New York, 1959) [2]Albers, J. Interaction of Color (New Haven, 1963; revised edition, 1975) p, 4. [3] Friendly, M (2008). "Milestones in the history of thematic cartography, statistical graphics, and data visualization" August 24, 2009 [4]Tufte, Edward R.(1990). Envisioning Information. Cheshire, CT: Graphics Press [5] Funkhouser, H. Gray (1936). A note on a tenth century graph. Osiris, 1:260–262. [6] Graunt, John (1662). Natural and Political Observations Mentioned in a Following Index and Made Upon the Bills of Mortality. London: Martin, Allestry, and Dicas. [7] Nightingale, Florence (1857). Mortality of the British Army. London: Harrison and Sons. [8] Harness, Henry D. (1838). Atlas to Accompany the Second Report of the Railway Commissioners, Ireland. Dublin: H.M.S.O. [9]Heerden, F van. (2009). Data Flow: Visualising Information in Graphic Design. Die Gestalten Verlag.

Bibliography Tufte, Edward R.(1990). Envisioning Information. Cheshire, CT: Graphics Press Tufte, Edward R.(1983). The Visual Display of Quantative Information, Cheshire, CT: Graphics Press Heerden, F van. (2009). Data Flow: Visualising Information in Graphic Design. Die Gestalten Verlag.

Strunk, Jr. William. White, E.B.(1959) The Elements of Style (New York, 1959) Albers, J. (1975) Interaction of Colour. New Haven(1963; revised edition 1975) Loos, A. (1908). Ornament and Crime. Innsbruck (reprint Vienna, 1930)

19

Sign up to vote on this title
UsefulNot useful