Sunday, 19 December 2010

Words over time

Google Labs have a tool that exploits the large corpus of words they have built up through digitising all those Google Books.

A chart of the percentage occurrence of the words "art", "design" and "science" from 1810 to 2000. See it live here.
It is worth bearing in mind that none of the example words ("art", "design" and "science") continued to mean the same thing throughout that period.

Something fifhy going on 
All visualisations depend on the quality of the underlying data and this is where this tool falls down.

I thought I would check for occurrences of a word which is unlikely to be much affected by fashion, from 1700 to 2000: I tried "fish". This produced some highly suspect results with very low occurrences before 1800 and a dramatic rise at that time:

A chart of the percentage occurrence of the word "fish" from 1700 to 2000. See it live here.

And then I realised what is going on.

Google have not corrected the long-tailed s's which the scanning software thinks are f's. If you chart the nonexistent word "fifh" you find a steady climb which dramatically drops during 1780-1800 when the long s was replaced by the one we use now.  Charting "fish,fifh" shows both. Together they make a more sensible picture:
A chart of the percentage occurrence of the word "fish" and "fifh" from 1700 to 2000. See it live here.
Charting "defign,design" shows a similar pattern, as does "fcience,science".

This is idleness on Google's part and undermines the usefulness of the tool. I am sure it would be perfectly possible to make their Optical Character Recognition software tell the difference between a long-tailed s and an f, since they are not the same glyph:
Different glyphs for f and long s. From Joseph Priestley, 1764, A Description of a Chart of Biography on Google Books.
The crossbar of a long tailed s extends only to the left, as in "science" in the first line here, whereas the crossbar of an f extends both sides, as in "therefore" in line one/two.

A tool to use with care, indeed suspicion.

Friday, 17 December 2010

Ben Fry's watching the evolution of the Origin of Species

Ben Fry, co-inventor of Processing, has created an intriguing visualisation of the chapters of Darwin’s Origin of Species, showing how they alter between each of the six editions that Darwin produced between 1859 and 1876.
Ben Fry. 2010. On the Origin of Species: The Preservation of Favoured Traces.
Used with permission.
These are not representations of time as such, but representations of change. Ben writes:
The idea that we can actually see change over time in a person’s thinking is fascinating. Darwin scholars are of course familiar with this story, but here we can view it directly, both on a macro-level as it animates, or word-by-word as we examine pieces of the text more closely.
This is where the hidden depths of the project lie. At first sight ‘just’ a visualisation, this is actually an interface to the full text of all the editions, based on van Wyhe et al.’s Complete Work of Charles Darwin Online.

Ben has written a book for O'Reilly Visualizing Data: Exploring and Explaining Data with the Processing Environment based on his 2004 PhD dissertation at the MIT Media Lab Computational Information Design [PDF].

Friday, 19 November 2010

Tracking 18th-century "social network" through letters

Researchers at Stanford have mapped thousands of letters exchanged in the 18th century's "Republic of Letters" and produced a visualisation (partly time-based) of the flow of knowledge.

Sunday, 26 September 2010

Paper for Computers and the History of Art 2010

The Computers and the History of Art (CHArt) conference this year is concerned with Technology and ‘the death of Art History’.

My paper is again based principally on the work of Barbeu-Dubourg and Priestley, but looking at new aspects. Among these I discuss for the first time the spatialisation of knowledge in Priestley’s Harmony of the Evangelists. The most striking feature is the visual gaps, the empty spaces, at times resembling the famous empty page in Tristram Shandy by Priestley’s older contemporary Sterne (1713-1768).  He describes his method: ‘If I should be thought to have succeeded in this work better than the generality of my predecessors, I shall attribute it chiefly to the mechanical methods I made use of’ (Harmony pxvi original emphasis). He goes on to explain how he cut up two copies of the gospels and rearranged them. The physical, mechanical nature of the process was of help to him as well as to his readers: he was able to move the elements about as his ideas changed, before fixing them just prior to going to print (pxvii).

Pages 19 and 207 of Priestley’s Harmony of the Evangelists of 1780. Aligning the four Gospel accounts according to time, using between one and four columns per page. Chetham’s Library, Manchester. Used with permission.
The abstracts for my paper says:
The paper is concerned with the use of computers to represent historical time visually, typically as ‘timelines’. Research into the sophisticated practice and theory of early modern paper timelines in the eighteenth century reveals the weakness of current practice, especially on the Web. Behind the work of the early pioneers lay a vision of mechanising knowledge. At that time, this proved a productive metaphor, but in our own time the mechanistic properties of computers have tended to encourage an approach to visualising history that excludes all but the crudest aspects. Solutions are needed which use computing in ways that do justice to the demands of historiography. 
CHArt 2010 takes place Wednesday 10 - Thursday 11 November 2010 at The British Computer Society, First Floor, Davidson Building, 5 Southampton Street, London WC2E 7HA. In the draft programme my paper is on the Thursday afternoon.

View The British Computer Society, Southampton Street, London in a larger map

Wednesday, 28 July 2010

3D timelines

I discussed in March the use of the depthwise axis to represent the time dimension, highlighting a BBC project that unusually placed most recent time furthest away into the depths of the screen. As you travelled forward into the space you were travelling forward in time. I shall come back on a future occasion to these spatial metaphors of time, epitomised in English by phrases like ‘I look forward to seeing you.’

Another BBC timeline is embedded in a data visualisation created in conjunction with the University of Westminster, 3D Documentary Explorer [external link].
3D Documentary Explorer by BBC and University of Westminster in an AHRC-funded collaboration.

In 1996, Robin Kullberg, a postgraduate student at the MIT Media Laboratory, created a ‘Dynamic Timeline’ using full 3D. She wrote a short paper about it, accessible as a web page [external link], and her video of the system can be seen (and even downloaded) at the University of Maryland Open Video site [external link]. There are several weaknesses in the thinking behind the project – not least that the assumed advantages of 3D over 2D are not really discussed – but it is a remarkable piece of work.
Robin Kullberg: Dynamic Timelines: visualising the history of photography. MIT Media Lab 1996.

Tuesday, 6 July 2010

Paper at Electronic Visualisation and the Arts 2010 now online

The paper I wrote for EVA2010 [external link] with Emma Bevan and Aleksei Kudikov is now online on the BCS website [external link].

The presentation (to be delivered tomorrow, Wednesday 7 July) illustrates the range of ingenious graphical solutions, such as this...
Part of Stream of Time - Chart of Universal History from the German of Strass published by C Smith 1855 
... that have been used to represent time.

We focus on the issue of the Shape of Time, looking at the historic roots of the idea that time is a uniform container for events – Descartes, Newton etc, leading via Helvicus into the pioneering visualisations by Barbeu-Dubourg and Priestley discussed in other parts of this blog.

This leads into a discussion of the rationale for our own Historical Interactive Timeline.

Sunday, 30 May 2010

Change over time: Antarctic

In my work on the visualisation of time, I am above all interested in explicitly representing time graphically and spatially – as calendars, clocks and timelines do. But sometimes it is hard to resist appealing examples that implicitly represent time by showing change.

Nicholas Hutcheson has put online a series of short experiments using time-lapse drawings of the Antarctic Landscape. They can be viewed here in Vimeo.

Nicholas explains:
In 2008 I spent 8 weeks as an Artist in Residence for the Australian Antarctic Division drawing frantically as I journeyed in and around the continent. On my return, the challenge has been to try and capture some of the Antarctica I experienced.  Out there, you have a constant awareness of movement and time. Some of it is so slow - gigantic icesheets flowing towards the sea at seemingly imperceptible rates –  but then, you can also watch the sea water become ice, and weather fronts moving across the horizon.  And the majority of what makes up the landscape is frozen water. It’s defined by this ever-creeping whiteness – in compositional terms, a mass of negative space. How to deal with this in the drawings I was making?
At the start of the year, in response to this dilemma, I began to play with making very short animations, sort of time-lapse drawings of the landscape. 
A page about his drawings is here

Wednesday, 19 May 2010

Jacques Bertin (1918-2010)

I learned today of the death, in Paris on 6th May, of Jacques Bertin. He pioneered an analytic approach to the use of graphic elements to convey meaning. Some of his pronouncements were distinctly strange: for example in his monumental publication Semiologie Graphique: diagrammes, réseaux, cartographie Bertin insists on separating the retinal from the spatial. This becomes very odd when he discusses the difference between the use of lengths and areas to represent quantities, since it involves declaring length as spatial but area as not.

Bertin is ironically at his best – not in his attempt to systematise through textual rules which becomes intimidatingly prescriptive – but in his opposite, graphical, tendency to offer numerous solutions to a single data visualisation task. Even within the narrow domain of a chart of four quantities, Bertin is able to show twenty different representations – an object lesson in not just plumping for the first idea that comes along.

His Semiologie Graphique: diagrammes, réseaux, cartographie is being republished in English (as Semiology of Graphics) this autumn. I cannot easily find out the publisher, but it appears on Amazon.

Thursday, 6 May 2010

An indexical chronographic

Justin Quinnell makes very long exposure photographs using a pinhole camera.

One of Justin Quinnell’s long-exposure photographs, from his pinhole photography site.

This one shows the tracks of the sun over the Clifton Suspension Bridge, Bristol, UK during a six-month exposure from 19 December 2007 to 21 June 2008.

Apparently the pinhole camera was made from an empty drinks can with a 0.25mm aperture, strapped to a telephone pole overlooking the Gorge.

Dotted lines of light are the result of overcast days when the sun was less consistently visible.

Sunday, 25 April 2010

HiT: Historical Interactive Timeline MA project at Lansdown Centre

Emma Bevan and Aleksei Kudikov who studied on the MA Design for Interactive Media at the Lansdown Centre for Electronic Arts created an interactive timeline which will feature in a presentation at EVA 2010, the conference on Electronic Visualisation and the Arts, in London on 5-7 July 2010.

Just in Time: defining historical chronographics by Stephen Boyd Davis, Emma Bevan and Aleksei Kudikov is historical in two respects, both concerned with visual representations of past time.
The paper’s first purpose is to enquire how visual representations of historical time can be used to bring out patterns in a museum collection. A case study is presented of the visualisation of data with sufficient subtlety to be useful to historians and curators. Such a visual analytics approach raises questions about the proper representation of time and of objects and events within it. It is argued that such chronographics can support both an externalised, objectivising point of view from ‘outside’ time and one which is immersive and gives a sense of the historic moment. These modes are set in their own historical context through original historical research, highlighting the shift to an Enlightenment view of time as a uniform container for events. This in turn prompts new ways of thinking about chronological visualisation, in particular the separation of the ‘ideal’ image of time from contingent, temporary rendered views.

A prototype of HiT can be seen here. Special features of the timeline include that it represents uncertainty – no date is marked as a point, as though it were of infinite precision, but as a line – and that multiple searches can be combined. The prototype uses a subset of the collection housed at MoDA, the Museum of Domestic Design and Architecture. We are very grateful to Zoe Hendon, Senior Curator of the museum, and to System Simulation Limited for their help and support.
To see thumbnails of the items in the collection, choose Images On/Off at the top left of the display.
Use the left and right arrow keys on your keyboard to scroll through the timeline.

Sunday, 18 April 2010

Cartographies of Time reviewed

Cartographies of Time 
Daniel Rosenberg and Anthony Grafton

ISBN 9781568987637
8.5 x 10.5 inches (21.6 x 26.7 cm), Hardcover, 272 pages 
268 color illustrations ; 40 b/w illustrations 

[see the book on the publisher’s site]

Cartographies of Time is a milestone book. It makes a cogent claim for the importance of chronography as a neglected aspect of past historiography and as an important cultural form for the present. The most significant recent work in this area is that of the authors themselves, such as Rosenberg’s 2007 article on Joseph Priestley’s ‘graphic invention of modern time’, and Grafton’s two-volume study of Joseph Scaliger of 1983 and 1993, so it is ideal that these two should have come together to create this new book.

As the authors point out, there is a tendency to regard chronology as primitive or incomplete history writing, yet it is the scaffolding on which history as we now understand it relies. As late as the 18th century, the word ‘history’ had connotations of narrative and story to which chronology was seen to give rigour: it brought (various authors argued) meaning, vividness, memorability, an evidential basis, and a unifying framework. Locke argued that chronology’s ability to give history form made it both more memorable and more productive of moral lessons.

Feeney notes in his Caesar’s Calendar of 2007 the near-impossibility of recovering in imagination the character of events before we had a standardised numerical grid for history, and emphasises the recency of its invention. Rosenberg and Grafton’s main theme is similarly the way in which the line, visible or implied as a metaphor for time, is a product of only the last 250 years. They document a series of key influences on chronology and chronography: the explosion of conflicting sources faced by Renaissance scholars, the realisation that astronomical records might be used as a historical clock to correct dates corrupted or lost, the eschatological motive (many chronologies and chronographics mapped the past in order to predict the future), and the distressing difficulties experienced by Christian chronologers as evidence accumulated – in the form of reliable records from other cultures and the growth of deep time through geological investigation – that the Bible could no longer be treated as history.

Profoundly knowledgeable in historiography, the authors write sensitively about the visual, and effortlessly connect the two. For them the graphical is not a childish substitute for the sophistication of words but an essential counterpart. An admirable feature of the book is the synergy of the text and illustrations. The figures are printed close to the relevant text, the captions are informative, and in the text the authors draw out with verve the features they want the reader to notice.

The work of Joseph Priestley, preacher, scientist and radical, is pivotal. The authors give a good account of the ‘crucial transition in the history of chronographic representation’ offered by his Chart of Biography 1765 and New Chart of History 1769. While many of the charts in the book are rich in graphical conceits such as trees, rivers, streams, chains and wheels of time, Priestley’s are wonderfully minimal and egalitarian with their spread of two thousand undifferentiated named lifelines, an arithmetical presentation appropriate to Priestley’s scientific and Dissenting philosophy. Jacques Barbeu-Dubourg gets rather less generous treatment than Priestley, despite his 16.5 metre chart of history having preceded Priestley in using for the first time, as the authors acknowledge, a truly arithmetic scale for time. It is true that Barbeu-Dubourg is a rather simple soul compared with Priestley: his argument for his own pioneering timeline is full of the sort of claims about effortless learning that would be made again about multimedia in the 1980s.

Perhaps the account of Priestley’s achievement could have emphasised another of his vital innovations: the visual representation of uncertainty [see Representing Uncertainty].

The nineteenth century brings chronographical games, mostly intended to be educational. Though occasionally ingenious, these represent in some ways a nadir of chronography, essentially childish and not a tool for investigation or serious thought. Nevertheless, it is nice to learn that a chronographic game was explicitly designed by Mark Twain to be winnable by any player knowing a lot of ‘minor events’ as well as the player who knows the dates of monarchs and battles.

The closing chapters of the book are somewhat cursory, touching on public monumental timelines and timelines as art, including Maciunas’ chart of time and space-based art for Fluxus, John Cage’s graphical scores, and Shapolsky et al’s Manhattan Real Estate Holdings cancelled from the Guggenheim’s schedule for being too political. Some exclusions from the book are a little surprising: Marey’s chronophotography is in, but the comic strip and the film storyboard are out; the exhibition timelines of Charles and Ray Eames, the richest of which mounted physical historic objects onto extended graphical representations of time, do not appear. It is a shame that the work of Michael Twyman in the history of chronographics is not acknowledged, nor Stephen Jay Gould’s Time’s Arrow, Time’s Cycle with its important discussion of the conflicting metaphors of the line and the circle as models of time; and one might have expected more on the groundwork for visualising number as line laid down by Descartes and the crucial proposal by Newton that time is a measure which can be treated as analogous to space. But these are quibbles about a book whose historical depth is its finest feature. Even in the middle of discussing modern grand timelines in museums and public spaces, the authors remind us of Augustus’s carved fasti consulares of the first century BCE; a silly timeline on a folding ruler is juxtaposed with Dürer, Sarah Fanelli’s Tate Artist Timeline with Piranesi.

Credit is also due to Jan Haux, the designer. Though some of the illustrations are smaller than they need to be, this is a beautiful book perhaps inspired by the classic publications of Edward Tufte. As a record of achievement during centuries of chronographic invention it should make most modern designers ashamed, and lays the foundation for the future study of chronographics.

Thursday, 15 April 2010

Cartographies of Time: the book is out now

Daniel Rosenberg and Anthony Grafton’s book Cartographies of Time is out now: my pre-ordered copy from Amazon is in front of me as I write. It looks beautiful and is lavishly illustrated. I’ll be reviewing it here soon.

See the book on the publisher’s site.

Friday, 26 March 2010

That other axis, in a BBC timeline

BBC History of the World. A timeline where time is in the depth (z) axis. From BBC History of the World website.
Which way should time go?
The screen, like a piece of paper, has only two dimensions. Illusions of depth, along a dimension orthogonal to the two real dimensions, can be created using a variety of visual cues – J.J. Gibson counted thirteen cues that our perceptual/cognitive system seems to use to infer depth, many of which can be used pictorially to fake depth on a two-dimensional surface (discussed in Chapter 3 of my 2002 PhD [download 6.5MB PDF file]).

This BBC example is a timeline based on things: it is a history of the world told through objects. Time is represented by the illusory third dimension with, rather unusually, the most distant time being shown as nearest to the viewer. At any given date, contemporaneous objects arrange themselves in a circle around the centre-point.

A nice feature is that information on users’ objects can be uploaded and is fully integrated with the original set of objects contributed by museums and other organisations. I have uploaded a picture of my copy of Priestley’s Description of a Chart of Biography. It seemed right to put a pioneering historical timeline in the timeline of historic objects.

Sunday, 7 February 2010

A small technology timeline

This timeline developed by Mark Light (a graduate of the Lansdown Centre for Electronic Arts at Middlesex University) has a couple of interesting features.
  1. The timescale with slider at the bottom is non-linear, using different horizontal intervals of space for a given interval of time. This is common and generally necessary when dealing with a long historical aera since there is far more data for recent times than for the more distant past. But here this is flagged using colour. The user is alerted to the three different scales - a horizontal unit may represent 1000, 500 or 100 years - using shades of blue. Often designers in the interests of tidy uniformity obscure the different scales they use.
  2. Items in time have a relation to others. These relationships are made explicit using serpentine arcs inscribed between the nodes. There is a potential problem, visible in the illustration here, that where one arc touches or crosses another, it is not possible to discern which line belongs to which node. However, as used here, the arcs all convey roughly the same kind of relationship, so there is little need to see which line is which. 
Like many modern timelines and other interactive diagrams this one makes transitions between different states of the diagram as fluid as possible. In part this is just a current aesthetic, but it also seems to assist in maintaining the frame of reference so that the user sees newly displayed information in the context of what was on screen before.
The timeline is not a dead end. It is an interface to further information which appears below the timeline display and which the user can edit.

The timeline itself

Saturday, 9 January 2010

Continuum: an interesting and complex timeline tool

I’ve just come across an unusually subtle timeline developed at Southampton University, UK, by Paul André, Max Wilson, Alistair Russell, Daniel Smith, Alisdair Owens and m schraefel.

Continuum, a Web2.0 application for visualising faceted temporal data, by Paul André, Max Wilson, Alistair Russell, Daniel Smith, Alisdair Owens and m schraefel of Southampton University. 

Among its many points of interest, sections of a timeline can be brought into close proximity so that relationships can be conveniently mapped, omitting the intervening time. In the illustration above, works by Bach are connected to performances by Glenn Gould.

On the right of the illustration can be seen sliders which control the level of detail of different facets of the data independently of one another. 

A video of the work can be seen here:

A paper about it is here:
Continuum: designing timelines for hierarchies, relationships and scale (2007)

André, P., Wilson, M. L., Russell, A., Smith, D. A., Owens, A., and schraefel, m. 2007. Continuum: designing timelines for hierarchies, relationships and scale. In Proceedings of the 20th Annual ACM Symposium on User interface Software and Technology (Newport, Rhode Island, USA, October 07 - 10, 2007). UIST '07. ACM, New York, NY, 101-110.

Sunday, 3 January 2010


Barbeu-Dubourg in 1753 and Joseph Priestley in 1765 both extolled the merits of uniform timescales for chronographics (which they effectively invented). In fact they presented one-sided arguments for this solution.

A number of aspects of eighteenth century culture reflect this commitment to uniformity.

Jethro Tull and the seed drill

A modern field sown with a seed drill. Uniform parallel rows stretch to the horizon. Copyright Tony Atkin and licensed for reuse under Creative Commons Licence
By this means also the Rows of the whole Field may be kept equidistant, and parallel to one another.   Tull 1762: 374
Jethro Tull was a pioneer of agricultural improvement. In 1701, he developed an improved design of horse-powered seed drill that planted seeds in parallel rows, and in 1714 a horse-drawn hoe for tending such crops. In 1731, he published Horse Hoeing Husbandry which promoted his new farming ideas. During the following century and a half, the seed drill was gradually adopted so that broadcast seed, thrown to the ground by hand, was replaced by regular lines of plants. It must have been extraordinary to see the natural random patterns of seedlings succeeded by uniform parallel rows.

Buildings and pavements

The Royal Crescent in Bath, UK, July 2006. Photo by David Iliff, from Wikipedia.

The eighteenth century saw a flowering of the urban terrace in cities such as Bath and Edinburgh. Though sometimes embellished with central porticoes and other architectural features, they were distinctive for their use of simple repetition and the equality of every unit in the façade. They emphasised the neo-classical ideals of ‘order, regularity, restraint, proportion and reason’ (Lee and Kelley, 1996: 48), though it has been suggested that ‘simplicity, austerity and regularity were as much solutions to economic problems as a selfconscious search for aesthetic effect’ (Ashworth 2005: 43).

Stone paving, perhaps laid in the eighteenth century, in Pearse Street, Dublin. Uniform paving was an eighteenth century innovation. From

Until the mid eighteenth-century in Britain it was the responsibility of individual householders to maintain the pavements in front of their property. Pavements were therefore of different heights, materials and quality. At the same period that Barbeu-Dubourg and Priestley were advocating a uniform scale to represent time, there was a shift to uniform paving funded through local taxation. For the first time, the pavement would have been the same along a single street (Cockayne 2007: 202-204).

The uniformity of time
The very idea of time that we now take for granted – a uniform empty structure which contains events – is of course a cultural construct. Newton, whom Priestley and many of his contemporaries revered, was an early advocate of such a model: 
Absolute, True, and Mathematical Time, of itself, and from its own nature flows equably without regard to anything external...     Newton 1687
Poole describes the shifting conception in this way:
The term ‘calendar’ became less likely to signify the whole cultural edifice and more likely to refer simply to the chronometrical framework.     Poole 1995: 100
This contrasts with the earlier ‘lumpish quality of time’ (Poole 1998: 23) in which ‘time was an uneven succession of periods of different qualities, a cluster of high points and low, rather than a steady stream of being’ (op cit: 21).

Gell (1992: 23 passim) warns that to even refer to a culture’s ‘model of time’ is to beg the question. A society may have a model of the relationship between events, between now and the past, the past and the future, etc., but not necessarily have a model of time per se.

Ashworth, Gregory J. 2005. The Georgian City: the Compact City as Idealised Past or Future Ideal. Global Built Environment Review 4(3). 40-53. 
Cockayne, Emily. 2007. Hubbub: filth, noise & stench in England 1600-1770. Yale University Press.
Gell, Alfred. 1992. The Anthropology of Time: cultural constructions of temporal maps and images.  Berg, Oxford
Newton, Isaac. 1687. Mathematical Principles of Natural Philosophy. Book 1. Scholium.
Lee, D. and Kelly, R. 1996. Georgian Limerick. FAS/ Limerick, Civic Trust.
Poole, Robert. 1995. ‘Give us our eleven days!’: calendar reform in eighteenth-century England. Past & Present 149(1). 95-139
Poole, Robert. 1998. Time’s Alteration: calendar reform in early modern England. London: UCL Press/Taylor & Francis. p23.
Tull, Jethro. 1762. Horse-Hoeing Husbandry or, an Essay on the Principles of Vegetation and Tillage. 4th edn. Millar, London. Available as PDF at