tirau persistence
Persistence and change for the Tirau Wikipedia article.

Over the last few days I’ve been exploring Wikipedia edit histories. I intend to write a more detailed post later this week, but I want to share some initial results. These visualisations are inspired by Fernanda Viégas and Martin Wattenberg’s History Flow work from 2004.

I wrote a Python script to collect Wikipedia page histories (e.g. all edits to the Tirau article as XML), inspect each revision & visualise the results. The code is an embarrassing stream-of-consciousness mess but I’ll get it on Github before the end of the week—one way or another.

Each graphic represents the history of a single article. Time moves from left to right. The varying heights of the coloured section of represent how many lines an article had at each point in time. Articles typically start short and become longer over the years.

This chart represents the history of the New Zealand photographer Fiona Pardington’s Wikipedia article. The oldest unchanged lines from the article are shaded deep blue; newer sections are shaded pale yellow.

fiona pardington time
Persistence and change for Fiona Pardington’s Wikipedia article.

Reading from left to right, we can see that the article was created in late 2013 and remained relatively unchanged until about March of 2014. Around this point it became briefly longer (note the stalagmite-like structure) before shrinking and then undergoing a rapid change where almost every part of it was changed. The article stabilised around May of 2014, undergoing only very minor changes until mid-2015 when much of the article’s second half of was altered.

The following chart represents the same article, but shades anonymous edits red. In this article’s history, one anonymous user made two small edits on the same date. Both changes persisted for many months. One of these edits remains part of the article.

fiona pardington anon
Anonymous edits in Fiona Pardington’s Wikipedia article.

I consider the final style of chart the least successful experiment. The visualisation below allocates a random but persistent colour to each Wikipedia contributor’s edits. In this case there are three primary contributors (shaded dark sea green, light forest green and yellow) and a set of smaller edits from other authors.

fiona pardington people
Edit authorship in Fiona Pardington’s Wikipedia article.

I will write a follow-up to this later in the week when I share the code, but for now here are two more examples: the Waikato town of Tokoroa and New Zealand’s Prime Minister, John Key. I find it interesting how varied the pattern of edit histories are.

tokoroa persistence
Persistence and change for Tokoroa Wikipedia article.

tokoroa anon
Anonymous edits for Tokoroa Wikipedia article.

tokoroa contributors
Edit authorship for Tokoroa Wikipedia article.

john key persistence
Persistence and change for John Key Wikipedia article.

john key anon
Anonymous edits for John Key Wikipedia article.

john key person
Edit authorship for John Key Wikipedia article.

I overslept this morning and missed the first half of the Rugby World Cup Final between New Zealand and Australia. I grabbed my phone and opened the New Zealand Herald’s Live Scoring page to catch up with the score. It greeted me with this baffling timeline.

NZ Herald Rugby Timeline

The graphic shows the score clearly, but otherwise I am uncertain how to interpret it. The reader must work hard to figure out what the icons represent. (A try has the same symbol as a drop goal?!?). Yet even when you work out what those little pictures represent, it is difficult to get a sense of the rhythm of the game.

This evening I started mucking about with an alternative way to represent how a game plays out in a compact space using good old fashioned area charts. The test graphic below needs considerable work, particularly in terms of annotation and colour, but it already provides a better sense of how the game played out. [Source code.]

Alternative Rugby Timeline
An alternative timeline chart for the 2015 Rugby World Cup Final.

The All Blacks scored an early penalty; the Wallabies equalled a few minutes later. The All Blacks pulled ahead in the 26th minute and built a formidable halftime lead. The Wallabies mounted a comeback in the last third of the match, coming to within four points with 16 minutes remaining on the clock. After some tense moments, the All Blacks pulled away again with two drop goals and a converted try.

One reason why I think this chart is successful is that I could write that last paragraph without referring to a match report. The only other place I have seen sports results represented this way is in this excellent Guardian the history of the Ashes Tournament. I don’t read a lot sports journalism, so this may be a very common way of representing results and I am just blissfully unaware.

Tour of England in AustraliaAustralia 2013 - 14
2013-14 Ashes Tournament.

As a bonus, here is the score timeline for the extraordinary Pool B Japan versus South Africa game. Another incredible match.

Japan versus South Africa
Pool B Japan vs. South Africa, Rugby World Cup 2015.

Things go wrong when I make maps or graphics. Usually I fix the code and seconds later the memory fades. But sometimes I feel compelled to take a screenshot and shuffle it into some obscure folder or other.

Here is a collection of mistakes and screw-ups that I found either aesthetically pleasing or interesting in some way. With one exception, the glitch code is long lost. This seems appropriate.

Enjoy. Or not.

Wet Hair Octopus
Read the rest of this entry »

On October 12 I spoke at the National Digital Forum in Wellington. These are my speaker notes.

Unidentified man and woman, surveying a map of stock control and infected areas, conference on diseases in the Wairarapa district (1959).
Unidentified man and woman, surveying a map of stock control and infected areas, conference on diseases in the Wairarapa district (1959).
Alexander Turnbull Library. Ref: EP/1959/1566-F.

There are three pure science Crown Research Institutes in New Zealand. GNS Science studies geological phenomena; NIWA researches water all its various forms; and Landcare Research, an institution I worked in for four years, is concerned with soils, landscapes and living things.

Maps & charts play a very special role in that workplace. I would often find some combination of Garth, Pierre, Carolyn, Evelyn or Marino leaning over a roughly printed topographic map or hand-drawn chart. We would gesture, point, nod and disagree, tracing lines with fingers, pointing out inconsistencies with ballpoint pens. And by “talking through the map”, we developed a shared understanding of our respective problem domains.

These graphical props rarely lived longer than a few hours, although on occasion a landscape profile sketched on the back of a pie packet captured an idea so perfectly that it got filed away in a scientist’s records. The power of these artefacts was largely in their transience. Their ephemeral nature encouraged us to scribble, cross-out and annotate, all the while narrating our thought processes to our colleagues. Read the rest of this entry »

Possible New Zealand Flags

Late this afternoon, the New Zealand Flag Consideration Project Panel announced the official Long List of 40 Alternative Flag Designs narrowed down from 10,292 public submissions. While examining the various flags I became curious about a couple of patterns I noticed. Specifically, it seemed that most of the designs represented either a koru, silver fern or the Southern Cross constellation and that there were a handful of designers who had multiple successful entries.

Curious, I took a couple of hours to turn the webpage into a dataset and visualise the results. I extracted the data using Kimono—a handy tool that turns websites into structured APIs—and imported it into Google Spreadsheets. This quick post shares a few of the rough charts I pulled together in that time.
Read the rest of this entry »

distribution of letters within words
Distribution of letters within words across Māori and English Te Ara articles.

The Māori and English languages have fundamentally different structures. One way they differ is the distribution of vowel and consonant sounds within words. I was curious to understand some of these contrasts through data visualisation. The charts at the top of this post summarise an analysis of letter distributions within words in Te Ara—The Encyclopedia of New Zealand, one of the largest Māori-English bi-lingual corpora in existence (and available under a Creative Commons Attribution-NonCommercial 3.0 New Zealand Licence).

Read the rest of this entry »

A common critique of discourse that occurs on social media is that we tend to interact with people who already hold similar opinions to ourselves, reinforcing one another’s opinions and biases. A few months ago I started wondering about the social networks of New Zealand members of Parliament and whether this holds true for our politicians. To find out I collected some Twitter relationship data and experimented with a small visualisation project. Read the rest of this entry »

Yesterday New Zealand’s Minister of Finance tabled the 2015 Budget. The Budget lays out how the Government will spend $NZ88 billion dollars over the coming year.

Harkanwal Singh prepared an informative interactive visualisation in the New Zealand Herald, Keith Ng produced a treemap breakdown for the NBR and I expect there are more that I have missed. I enjoyed exploring the data through Harkanwal and Keith’s visualisations, but I wanted to see it from a slightly different perspective. After a restless night, I had some 4am inspiration and hacked together a D3 sketch.

Everything matters to someone. I wanted to understand the scale of spending across different areas of concern and government departments in a single glance. Accordingly, I allocated each spending area the same amount of space and then sprinkled dots across them to represent how much money they receive. I find it amazing to look at these figures and realise that each dot represents one million dollars.

There is a lot of work left to do on these, particularly in relation to label legibility and showing finer-grained data. However, given the timeliness of the budget I thought it worth sharing in this early state.

New Zealand 2016 Estimated Budget Spend by Functional Classification

One dot = $1 Million.

Budget spend by vote

New Zealand 2016 Estimated Budget Spend by Department

One dot = $1 Million.

Spend by department

Central Otago river networks

Stream order diagrams are simultaneously beautiful and illuminating. Subtle line width variations form pleasing dendritic patterns that suggest how water moves across a diverse landscape. The technique represents river networks as a hierarchy of segments, weighting downstream stretches of water more heavily than upstream tributaries. Depicting river networks this way is useful in environmental modelling because the technique offers a mathematical approximation of river strength. For cartographers, assigning order values to a network enables rivers to be mapped to line width, providing a visual indication of tributary size.

A visual overview of the effects of different stream order algorithms (source: osgeo)

Stream orders

Read the rest of this entry »


You can play with a full-screen interactive storm map in the blog post here. Read on for details.
As I write this on a Saturday morning in mid-March, a major storm is bearing on New Zealand. The intense depression that was formerly Tropical Cyclone Lusi made landfall late last night and will spend the day passing south across the country. For the last 24 hours folk have been sharing EarthWindMap screenshots through social media. EarthWindMap transforms US National Weather Service’s Global Forecast System data into dancing DayGlo particle streams. The results are extraordinary, enlightening and beautiful.

EarthWindMap March 15 2014

I love how these maps convey the nature of complex atmospheric systems through stillness and motion. Conventional weather maps are difficult to interpret until you learn the visual conventions of meteorological cartography. In thinking through how these maps so successfully communicate complexity, I am reminded of something Paul King wrote recently in response to a Quora question on what UI/UX designers can learn from neuroscientists.

“While we appear to take in the whole visual field as a single complete experience, neuroscience and cognitive science shows that it doesn’t work that way. What we actually process is a narrow region that moves around, and these regions are processed in sequence through visual exploration. So thinking about this narrow window of attention, how it moves, and how the user will interpret and organize information sequentially can help to design more intuitive and effective UIs.”

As EarthWindMap’s swirling currents guide my eyes over familiar landmasses, my mind starts weaving stories about what my eyes see. A  narrative about the state of the world’s weather develops as my eyes sweep through different parts of the map. Country outlines are known reference points. Streams of activity are narrative arcs. Areas of stillness offer stable environments where I can catch my breath. The cartographic design promotes visual exploration gives me a sense of the whole even though I am only ever looking at one small part of the map. Read the rest of this entry »

More Posts

Field pop
January 6, 2014

Field pop

Argumentum Ornithologicum and uncertain cartographies
December 9, 2013

Argumentum Ornithologicum and uncertain cartographies

Markov chains and WonderTao
September 1, 2013

Markov chains and WonderTao

On mapping the constellations
August 24, 2013

On mapping the constellations