An index to quantify an individual’s scientific research output – by Misha Donohoe

The field of bibliometrics employs statistical analyses of written publications in order to quantify the impact of a field, research institution or individual researcher.  In 2005, J.E. Hirsch of U of California proposed the h-index* (or Hirsch index) to quantify the cumulative impact of an individual’s scientific research output. It was developed to provide a focussed snapshot of an individual’s research performance.

In 2011 Google introduced an automatically-calculated h-index for researchers to include on their Google Scholar profile. Google also expanded on the h-index with the i10-index to calculate the number of publications by a specific author with at least ten citations from other researchers.

This poem is a cautionary tale about the loss of contextual information and identity in the age of metadata. I constructed the poem to mimic the automated tables on Google Scholar,  which include an h-index that varies depending on the search engine or database calculating it. The poem shows citations to “my articles” but upon inspection, the reader discovers that all reference to the individual has been removed. Where information about the title and author should be, lines of the poem describe the internal repercussions of the modern research imperative “publish or perish”.

Post script:

In 1934 mathematician Simon Kuznets delivered a report to US Congress on how to respond to the Great Depression. In this report Kuznets proposed the modern concept of GDP, but also warned against using GDP as a proxy for societal welfare or happiness.

*J.E. Hirsch, An index to quantify an individual’s scientific research output, PNAS, Nov 15, 2005. 102(46) 16569–16572

Misha Donohoe bio:
Born 1983, Sydney, Australia. Lives and works Whitehorse, Yukon.

Misha Donohoe creates beautiful, intricate and scientifically observed works utilising a range of media including graphite, watercolour, gouache, prints, calligraphic texts, film, sound, performance and installation.

Donohoe’s work manifests a lifelong curiosity with natural systems and is unique in its embodiment of elastic perception. From the preoccupations of the smallest bug, to those of the scientific mind, and to the expansive experience of the largest land formation, her works invite contemplation of the other in nature.

“Intellect. Reason. Sensation. Intuition. All are paths to truth, and these truths combine to form an intricate web of awareness. I aim to tell part of this story through the work I create.” – Misha Donohoe

web.  mishfish.com

social.  @mishaleena [twitter | tumblr | insta]

Mish Fish [facebook | linked-in]

gallery.  Yukon Artists @ Work

Mercury Rising – a datapoem

 

The difficulty in tackling climate change can be seen not as a failure of science, but of communication. The science is robust, but only speaks to the converted, or those inclined to listen.

I wanted to create a datapoem using a graph of rising sea-levels, because we are all familiar with that theme and have probably seen similar graphs, to the point of saturation. It ceases to shock us. I hoped that by overlaying it with a haiku, it might provide an emotional jolt, to make us take notice and consider the future.

The title of the datapoem ‘Mercury Rising’ refers to both increasing temperatures and the planet Mercury, which is inhospitably hot. Mercury is also the messenger of the Gods, and the poem is intended to communicate a warning.

The 3 lines of the haiku cover record-breaking temperatures becoming the norm, drought and floods wiping out humanity, and our future regret.

Visually, I wanted the poem to feel like it had run out of time and space by the end, reflecting our increasingly cramped world and the possibility that it may already be too late to act.

The data comes from the US Environmental Protection Agency, incorporating historical tide gauge measurements and more recent satellite observations.

Click here to see all datapoems

Laniakea – a datapoem

Laniakea is our galactic supercluster, home of the Milky Way. Its name means ‘immeasurable heaven’ in Hawaiian and honours the Polynesian navigators who memorised the stars to voyage extraordinary distances without instruments.

The poem considers the thread of exploration that runs through ancient seafaring and modern astronomy, always looking to the stars. The use of alternating 5 and 4 beat lines is intended to evoke the lurching rise and fall of a small craft on big waves.

The picture in the background is my attempt at drawing Laniakea based on a Nature article that describes its discovery based on observations of matter and their velocities. Note the scale – it really is immense at 520 million light years across! Our Milky Way galaxy is 180,000 light-years at its widest point. To put that in perspective, it would take around 4.5 years travelling near the speed of light to reach our nearest star, Alpha Centauri, and in practice it would take far longer as we can’t get anywhere near that speed with current technology.

Laniakea is due to be ripped apart by dark energy in the distant future, which was surprisingly sad to learn after writing the poem!

Click here to see all datapoems

Persistence – a datapoem

I wanted to make a poem that explores the web of technologies we inhabit in our modern lives, and the traces we leave behind, even after we’re gone. The ‘data’ in this datapoem is the collection of objects and the links between them – their relationships. This is called a ‘graph‘ – not the kind we are more familiar with like bar charts, but a network diagram. Graphs are an increasingly common way to represent data in the networked world, for example social media, and forgo the use of tables, rows and columns found in a traditional relational database. We are all nodes in a graph, in a constant state of flux.

Click here to see all datapoems

A Green and Pleasant Land – a datapoem

This datapoem looks at the massive deforestation of England since records began and its consequent loss of biodiversity. This manifests not only in the visual landscape, but the auditory one as well, as birdsong is often absent in intensively-farmed areas.

Since the 1920s there has been a concerted effort to reforest parts of England, which is now around 10% wooded. A better understanding of ecology has highlighted the importance of varied habitats for supporting wildlife. This in turn benefits agriculture through more resilient and varied pollinators, and the preservation of species that might have useful traits in future.

The poem is a haiku. Traditional elements of a haiku include a seasonal reference and an unusual juxtaposition of imagery. In this case the endless Autumn refers to the loss of leaves over the centuries, and the green clouds are the canopies of new trees, as seen from the forest  floor. The visual grain is intended to evoke a slight nostalgia, whilst adding the texture of distant leaves to the green graph.

Click here to see all datapoems

Slime Mould – a datapoem

The inspiration for this poem wasn’t slime mould, but graphs of the human population increasing on Earth, combined with decreasing fertility and increased connectivity via internet usage. This suggested a species rapidly reaching the limits of its environment and forming some kind of escape plan. At least, for some lucky individuals.

Comparing humans to slime mould is completely subjective. It’s impossible to prove or disprove such an idea and so it is not ‘scientific’. Its appeal is rooted in pattern matching – certain shared traits between the species – and it’s hard to ignore the pattern once you see it. There’s a good chance the next time you hear about the increasing population, or lack of resources, you’ll think “slime mould!”

I decided to embed the poem inside the mould’s life-cycle, with the circular area suggesting looking down into a petri dish. This resulted in the disembodied narrative voice of a scientist or a lecturer. This was intended to raise the question (I’m not sure how successfully) of whether anything may be looking down on us in our ‘dish’, and where our ‘spores’ might travel to.

More about slime mould

Cellular slime mould are fascinating organisms that are neither plants, funghi nor animals. They live as single cells hunting bacteria, until the food runs out, whereupon they emit a signal that causes thousands of them to band together into a giant slug, which crawls off in search of food.

If the slug can’t find a better location and is going to starve, it develops tall towers in which spores are formed, to drift away on the wind. This is a group endeavour but only the spores escape.

Another weird feature of cellular slime mould, is that the individual cells are a like predatory sperm or eggs. They are haploid, meaning they only have one copy of their genetic code, like sperm or eggs cells. They only get another copy (becoming diploid) when they fuse together with another slime mould cell.

Click here to see all datapoems

Six months in the Med – a datapoem

I was shocked by the continuing scale of the migrant deaths in the Mediterranean (Missing Migrants Project 2017), as it hasn’t really featured in the news recently. It has become a dry statistic, which is part of the issue I’d like to address with this project. To that end, I thought a short, punchy haiku atop the climbing death toll would have more impact than a longer piece.

This format is an experiment but I think it helps to tell the story in the data in a memorable way, perhaps more-so than an accompanying panel of prose. The ‘orange petals’ refer to an image that has stayed with me, of empty life-jackets washed up on the shore.

Click here to see all datapoems

Incantation- a datapoem

I was researching extinct species on Wikipedia, and it struck me how poetic and arcane the Latin names sound. You can imagine Harry Potter reciting them as a spell. I thought that a roll-call of extinct species might form a poem in itself. I wasn’t sure whether to add the common names and dates of extinction, in case it detracted from the flow of the names, but on balance I think it helps to appreciate what has been lost and adds the ‘data’ to the poem. Try saying it out loud three times in a mirror.

Click here to see all datapoems

Family – a datapoem

I wanted to make a piece exploring how closely related we are to different species and to eachother, and had the idea of a bar-chart, containing the lines of the poem. The data was found in this article in National Geographic by Carl Zimmer about the percentage of genes we share with different species. The data originally came from gene sequencing at the European Bioinformatics Institute.

Whilst we are no longer surprised that we share 90% of our genes with chimpanzees (although even a few decades ago that would have seemed incredible to some), it is thought-provoking that we share nearly half our genes with fruit flies, and around a quarter with a grapevine! Given how complex these genes are, it is good evidence that we are all distantly related, with the degree of similarity correlating to our proximity on the family tree.

The percentages in the article’s infographic gave me the idea of using a bar chart, with the lengths of the bars dictating the size of the lines. I worked through several drafts omitting some species until I found a combination that used most of them (sorry, yeast!). I decided to pick out the letters of the nucleotide bases A, T, G and C in a different colour, just for extra geekiness.

The idea of a writing metaphor emerged from consideration of the letters, and the ‘words’ (genes) they made up, with the constant revisions and deletions over time. Poetry too is about the arrangement of a limited number of letters and words, sometimes through happy accident.

Click here to see all datapoems

Anthropocene – a datapoem

A major piece of inspiration was this article in Scientific American by David Biello, about how the human era (the Anthropocene) might be remembered in the geological record.

This set me thinking about the evolution and extinction of species on a geological timescale, and how humans might be a flash-in-the-pan. I drafted the poem without knowing how to integrate it with data, but when I saw that the block-like stanzas looked like geological strata, I decided to bury it in the fossil record.

There are four stanzas, equating to four eras, going 1 billion years back in time. The problem was that the eras are different sizes, with the older ones being much longer. I tried laying out the poem on a vertical timeline and it was ridiculously long; there was a good chance people would just stop scrolling before they reached the end!

To compress the timescale more at the end I tried using logarithmic scales. I experimented with different bases (having the opposite problem with base 10, that made the end segment wafer thin), and it was base-2 that gave the most evenly sized layers. Base-2 is used extensively in genomics and computer science, so using it to format a digital poem on natural history pleases me greatly!

Click here to see all datapoems