“So Yahoo hired a professional ontologist…”

This article by Clay Shirky is a good encapsulation of the computer/GIScientist understanding of ontology that is rapidly entering geography and especially mapping and GIS:

It is a rich irony that the word “ontology”, which has to do with making clear and explicit statements about entities in a particular domain, has so many conflicting definitions. I’ll offer two general ones.

The main thread of ontology in the philosophical sense is the study of entities and their relations. The question ontology asks is: What kinds of things exist or can exist in the world, and what manner of relations can those things have to each other? Ontology is less concerned with what is than with what is possible.

Shirky is right that ontology has many definitions, though he then confuses this statement by offering only one as the definition: in his words ontology is “essence, “Is-ness.”” Of course, philosophers will recognise this definition, and a classic one at that: objects with properties ie., Arisotelian substance or predicate ontology. This tradition had already been challenged in the 1920s by Martin Heidegger who replaced ontology as “is” with “being” and essence with existence. Today we would see Shirky’s definition of ontology as rather covering the “ontic” rather than the ontological.

Nevertheless, it’s important to get to grips with this kind of ontology-speak, because as the rest of Shirky’s article discusses it is being applied to some really interesting problems, namely how we sort information, and by extension how we access and use information.

One fascinating aspect of this problem has emerged over the course of the development of the more interactive web, or web 2.0, that changes the game from just downloading information (web 1.0) to contributing to it. In geography this is playing out as various forms of participatory geography: volunteered geographic information (VGI), geolocated pictures, tracks and locations, geotagging, and what is sometimes called the semantic web.

The journal I edit, Cartographica, has a special section in its latest issue of how ontologists are approaching the problem of redoing the US National Atlas (or any other large scale mapping project), many of whom draw on this tradition of ontology for their schemas. To me, the most interesting aspect of this are bottom-up categories, which need not be commensurable or static, are deeply meaningful to individuals and yet sharable.

As this fascinating article points out though, the problem of the “universal book” and data organization, access, retrieval and interrelationships long predates the internet. There was for example Paul Otlet:

In 1934, years before Vannevar Bush dreamed of the memex, decades before Ted Nelson coined the term “hypertext,” Paul Otlet envisioned a new kind of scholar’s workstation: a moving desk shaped like a wheel, powered by a network of hinged spokes beneath a series of moving surfaces. The machine would let users search, read and write their way through a vast mechanical database stored on millions of 3×5 index cards.

Having just come back from the National Archives, where information discovery and retrieval are often dodgy at best (print-only finding aids, stapled “cheatsheets” for use by the archivists with different but similar dates lying around on battered ring binders, frank admissions by the staff that the online catalog covers only as small fraction of even the top-level categories–and this for a reasonably popular “record group” the OSS), well, I can certainly relate.

The quote in the post title comes from Shirky’s description of Yahoo’s attempt at classifying the web back in the “earlies.” He goes on to pull this system, which the “professional ontologist” had devised, apart. With a powerful search engine like Google, you don’t need categories just good tagging (metadata).

“Ontology” may (or may not) be good for small technical domains (and even then bottom-up looks more useful) but for getting at lived human existence, it seems hopeless. Here in contrast is Jeremy Wood’s lived map of Warwick campus, which he made by walking it with a GPS and then making a map form the traces:

The Warwick campus, produced through GPS traces by Jeremy Wood. A map traversed at 1:1 scale, the scale of lived human experience.

Philosophers have a lot to contribute to these issues, not just formal philosophers such as Heidegger but (obFoucault) people who talk about archives, genealogy and epistemology, and the production of knowledge.

(Thanks to John Krygier for the Paul Otlet link.)

Disclosed location blogging


Pork knuckle special, Köln

Unlike a certain other blogger, I can disclose my location, especially given that I was attending a conference on locative media.

Anyway, turns out I was interviewed by German radio (pdf, audio here, about 10-15 minutes in, though I am given a voice-over in German!).

Three new ground-breaking books on spatial reason

Three new books on spatial reason and politics have recently appeared. Remarkably, they all come from a single publisher, the University of Chicago Press in the last year or so.

The History of Cartography Vol III, Cartography in the European Renaissance (edited by David Woodward).

Abysmal, by Gunnar Olsson.

The Sovereign Map by Christian Jacob.

Continue reading

Politics of maps

I’m thinking of some sessions for next year’s AAG conference in Boston, perhaps on the politics of maps. There are so may ways to go on this topic however: historical, regional or some kind of cross-cutting theme.

A review in the New Yorker highlights this issue. Sixty years ago, India was partitioned:

Cyril Radcliffe, a London barrister, was flown to Delhi and given forty days to define precisely the strange political geography of an India flanked by an eastern and a western wing called Pakistan. He did not visit the villages, communities, rivers, or forests divided by the lines he drew on paper. Ill-informed about the relation between agricultural hinterlands and industrial centers, he made a mistake of enormous economic consequence when, dividing Bengal on religious lines, he deprived the Muslim majority in the eastern region of its major city, Calcutta, condemning East Pakistan—and, later, Bangladesh—to decades of rural backwardness.

It was in Punjab that Radcliffe’s mapmaking sparked the biggest conflagration. As Hindus, Muslims, and Sikhs on either side of the new border suddenly found themselves reduced to a religious minority, the tensions of the preceding months exploded into the violence of ethnic cleansing. It seems extraordinary today that so few among the cabal of Indian leaders whom Mountbatten consulted anticipated that the drawing of borders and the crystallizing of national identities along religious lines would plunge millions into bewilderment, panic, and murderous rage. If the British were eager to divide and quit, their successors wanted to savor power. No one had prepared for a massive transfer of population. Even as armed militias roamed the countryside, looking for people to kidnap, rape, and kill, houses to loot, and trains to derail and burn, the only force capable of restoring order, the British Indian Army, was itself being divided along religious lines—Muslim soldiers to Pakistan, Hindus to India. Soon, many of the communalized soldiers would join their co-religionists in killing sprees, giving the violence of partition its genocidal cast. Radcliffe never returned to India. Just before his death, in 1977, he told a journalist, “I suspect they’d shoot me out of hand—both sides.”

This is what I’m talking about. To some extent this cartographic partitioning is both unusual and extremely consequential (similar events occurred after the two World Wars, in Yugoslavia, Palestine and now possibly Iraq). At such, this might be the “height” of the politics of maps. From being a reasonably varied country, people began to partition themselves:

The British policy of defining communities based on religious identity radically altered Indian self-perceptions, as von Tunzelmann points out: “Many Indians stopped accepting the diversity of their own thoughts and began to ask themselves in which of the boxes they belonged.”

Maps are a way of operationalizing these boxes or categories of thought. Putting into play the knowledges: this is what you might call the politics of maps.

Expressed on a continuum we will find a range of effects from local town planning to the geographical way we tell stories about ourselves. I don’t know if there’s interest in this from the larger community, I guess I’ll just have to see!

The work of Michael Friendly on early statistical mapping

Michael Friendly is a professor at York University in Toronto and has done some absolutely amazing archival work on early examples of thematic mapping. You may recognize some of the names:

Continue reading

Ghost Map author speaks about his new book

Steven Johnson is the author of Ghost Map, a book about the cholera epidemics of 19th century London and the famous work of John Snow. Snow is sometimes described as the father of epidemiology and his maps have been very influential in geography and cartography.

Continue reading

Foucault, statistics and cartography

In the forthcoming English translation of Security, Territory, and Population, Foucault discusses the notion of the “specific qualities” of a territory (as Elden points out, in a phrase omitted from the original 1991 publication, “within its frontiers.”

Foucault observes in STP that this makes the modern era the era of statistics:

Because statistics is etymologically knowledge (connaissance) of the state,* the knowledge of forces and resources which characterise a state at a given moment. For example, knowledge of the population, measured in its quantity, measured in its mortality, its birth rate, estimation of the different categories of individuals in the state and their wealth, estimates of the potential (? virtuelles) disposable wealth in the state: mines, forests, etc., estimates of the value of products, estimates of circulating capital, estimates of the balance of payments, measures of the effects of taxes and imports…all essential for the knowledge of the sovereign (trans from French edn. p. 280)

*Here Foucault footnotes the origin of the term in German as Statistik to Achenwall in 1749. Writing 2 years later in The Taming of Chance, Ian Hacking gives the same etymology.

The things listed by Foucault are all geographical, and during the nineteenth century many attempts were made to map them, especially by Charles Joseph Minard.

Today Minard is well known for his Napoleon map because it is heavily promoted by Edward Tufte:

But Minard drew many other maps and was primarily concerned with Foucault’s list (wine exports, circulation of people via railroad etc.).

Circulation of goods (1844)

Circulation of passengers (1845, his first map, the width of lines is proportional to number of travelers, a favorite Minard design element).

The first choropleth (shaded area) map was also invented at this time (1826) by another Frenchman, the Baron Charles Dupin.

Both of these men, and others such as D’Angeville, Quetelet, etc. were trying to “think out space” as Foucault says elsewhere.