Drilling down into maps of Swahili Wikipedia

The global scale map of Swahili Wikipedia posted on this blog a few weeks ago (also reproduced below) sparked a lot of discussion about what exactly was going on in Turkey. There aren't many articles in the Swahili version of Wikipedia. But, of the articles that have been written, it seems that people really have a lot to say about Turkey.

However, that map only tells part of the story. All articles are treated equally and represented by bright yellow dots. What if we instead shade each article by the number of words that it contains?

The map above tells quite a different story. We see that the great mass of Swahili-language articles about Turkey are almost all stubs. In fact it is likely that most of these entries were created by one very committed Tanzanian editor.

Interestingly, the map at the top didn't spark much debate about why so many Swahili articles have been written about Western Europe. Likely because most of us are simply used to these sorts of North-South informational inequalities.

Another thing that we notice is the cluster of articles in continental Europe (in particular Belgium and the Netherlands) that each contain hundreds of words (indicating that few of them are stubs).

Also worth pointing out is the fact that once you filter out all of the relatively short articles (as in the map above), you see that the Swahili Wikipedia has a core focus on East Africa (as well as Africa more broadly) and England (and to some degree, Western Europe more broadly).

Article length is obviously only one measure of quality, and we've pulled out a range of other metrics such as number of references, images, hyperlinks, contributors and many many other things that we'll be sharing over the next few months.

Ultimately, all of these metrics allow us to get beyond the (very important) question of which parts of our planet are being annotated, and move towards asking how they are being represented.

AAG session: Information Geographies: Online Power, Representation and Voice - schedule announced

The Association of American Geographers has just released the preliminary program for the 2012 Annual Meeting in New York.

Below are the details of the session that I am co-organising with Matt Zook. The good news is that it is on a Saturday (Feb 25, 2012) and we have some great talks lined up for the session. The bad news is that it is at 8am. We hope that the interesting abstracts below are still appealing enough for you to brave the early morning start. 

Session: Information Geographies: Online Power, Representation and Voice

Saturday, 2/25/2012, from 8:00 AM - 9:40 AM in Regent Parlor, Second Floor, Hilton New York City

8:00am Creating an image: Cape Town's tour operators' self-representation on the Web
Bjorn Surborg (Trinity College Dublin)

This paper investigates which electronic media are utilised by tour operators and other actors in the South African tourism sector to gain access to potential clients and how the content is chosen, generated and placed. The web sites of tour operators, hotels or official tourism promotion agencies are often a visitor's first experience with a new a place, but there are increasingly many diverse choices for gaining access to place based information through social media (e.g. Facebook), map sites (e.g. Google maps, including street view), travellers' blogs and others. These are relatively new ways of communicating an image of a place, but in which ways have these new information and communication technologies (ICTs) changed the perception of a place and the way in which it is reproduced? Based on a survey amongst tour operators in Cape Town, South Africa, the paper will explore, if those providing content cater towards the stereotypes and pre-conceived images of Africa or if there is a conscious attempt to provide a more nuanced picture, especially given the diverse independent sources of information that potential tourists can access parallel to a tour operator's web-site. While the paper will focus on content providers, it will also touch upon the question of what on-line tools are being used by tourists to access information about a place before, during and after a visit.

8:20am Augmented realities and uneven geographies: exploring the geolinguistic contours of the Web

Mark Graham (Oxford) and Matt Zook (Kentucky)

Digital geospatial information is layered throughout our urban landscapes; it is invisible to the naked eye, but is a central component of the augmentations and mediations of place enabled by hundreds of millions of mobile devices, computers, and other digital technologies. We not just produce, access, and use all of this geospatial information about place, but also access it whilst we are in those very places. Moreover, due to advances in mobile technology, many people now quite literally have access to this information in the palms of our hands.  But far from uniform and ubiquitous, these digital dimensions of places are fractured along a number of axes such as location, language and social networks.

This paper analyzes how these fractures differ across space and language to both highlight the differences and begin the process of explaining the factors behind them.  While some of the disparities conform to longstanding offline patterns, others highlight the changing fortunes and positions of places in a globalizing economy and highlight the increasingly finer scale of differentiation in which understandings of places are constructed.

8:40am Citizen Cartographers: Tools, Challenges, Opportunities

Jens Riegelsberger (Google), Brent Hecht (Northwestern), Matt Simpson (Google) and Michelle Lee (Google)

Current developments in digital cartography closely mirror the evolution of 'citizen journalism'. The rise of social networking, micro-blogging, and mobile phones that double as video cameras enabled everyone to act as a journalist - either accidentally by being at the right place at the right moment - or by building up an audience and bypassing traditional media organisations.

In cartography, lay people are now actively participating in the creation of maps - a domain that has a long history of being tightly regulated and controlled. Today 'citizen cartographers' add points of interest to public maps using the contribution features available on many online mapping sites.

There are numerous motivations for these contributions: some may want to show their neighbourhood in the best possible light; others may realize that their private annotations can be of use to a wider audience; and others yet may want to give visibility to areas they feel are not sufficiently represented.

These developments raise questions similar to those that were brought up with the rise of citizen journalism. Where does this leave the trained professionals, the cartographers? How can, in this new world, quality be assured - and more fundamentally how can it be defined?

A second question is how, as creators of systems that enable 'citizen cartography', we communicate the unique qualities of this data, e.g. uncertainty or potential bias. Are there ways to help contributors avoid bad cartographic choices or automatically choose good ones for them?

9:00am 'Nobody wants to do council estates' - digital divide, spatial justice and outliers

Muki Haklay (University College London)

The understanding of the world through digital representation (digiplace) and VGI is frequently carried out with the assumption that these are valid, comprehensive and useful representations of the world. A common practice throughout the literature on these issues is to notice the digital divide, and while accepting it as a social and not natural phenomenon, either ignoring it for the rest of the analysis or expecting that it will solve itself over time through technological diffusion. The almost deterministic belief in technological diffusion absolves the analyst from fully confronting the political implication of the divide.

However, what VGI and social media analysis is revealing is that the digital divide is part of deep and growing social inequalities in Western societies. Worse still, digiplace amplifies and strengthen them.

In digiplace the wealthy, powerful, educated and mostly male elite is amplified through multiple digital representations. Moreover, the frequent decision of algorithm designers to privilege those who submit more media, and the level of 'digital cacophony' that more active contributors are creating mean that a very small minority - arguably outliers in every analysis of normal distribution of human activities – are super empowered. Therefore, digiplace power relationships are arguably more polarised than outside cyberspace due to the lack of social check and balances. This makes the acceptance of the disproportional amount of information that these outliers produce as reality highly questionable.

The paper highlights the mass silencing and call for a more critical engagement with digiplace and VGI.

9:20am Spooks, Scholars and Secrets: Geographies of"Volunteered" and Open Source Intelligence (OSINT)

Jeremy Crampton (Kentucky)

In 2010 for the first time ever the USA disclosed its total intelligence budget: $80.1 billion. By contrast the Department of Homeland Security budget is $42.6 billion and the State Department $48.9 billion. Intelligence expenditures have more than doubled since 2001, with $3.5 billion being spent on Iraq intelligence alone. In response the intelligence community (IC) has increasingly exploited open source or unclassified intelligence (OSINT). It has done this in two ways. First, by extending its tradition of using scholarly scientific sources and experts, and second, by exploiting the internet and social media.

This paper examines these developments. On the one hand, we need to improve our understanding of the relationship between intelligence and science. Can scholarly work, traditionally open, co-exist with intelligence, traditionally closed? Will the IC become more transparent or science less so? Can scholars exploit "counter-intelligence" such as WikiLeaks?

On the other hand, what are the geographies of the intelligence landscape--the "alternative geography of the United States" (Priest 2010)? How is the IC exploiting social media and especially the geoweb for intelligence? Does this constitute an extension of surveillance into the everyday, an "infra-power" (Foucault 1977), and if so, to what extent is (geographic) information truly "volunteered"? To ask these questions is to recognize the extent of the information asymmetries of the modern security state: we still know very little about it even as it collects ever more information about us.

Mapping Wikipedia at the global-scale in Arabic, English, French, Hebrew and Persian

Building on the maps posted a few days ago which mapped Wikipedia in Arabic, English, French, Hebrew, Persian and Swahili, I wanted to present a few alternate visualisations of the same data.

In the maps below, the number of articles was grouped by country in order to better understand national-level inequalities in Wikipedia's augmentations of our world. The data were all taken from November 2011 Wikipedia data dumps. Our project team wrote a script to search for coordinate representations in every article (taking into account the language variations between the dumps and the varying ways in which geo-coordinates are expressed). We improved the quality of our coordinates by doing things like eliminating or fixing erroneous coordinates, grabbing coordinates (where sensible) from not just structured infoboxes, and making sure to remove irrelevant coordinates (Wikipedia actually contains a lot of coordinates for extra-terrestrial entities like lunar craters!). We then did some post-processing to make sure each country also contained articles for entities like lighthouses, coastal shipwrecks and piers that sometimes fell just outside of the coastal boundaries of the country-boundary file that we were using. 

And after all of that, we're able to tell you how many Wikipedia articles are in each country!

The map above (click the image to enlarge it) displays the number of articles in English. There are a staggering number of articles in the United States (over 180,000 of them) and tens of thousands in many European countries, Japan, Australia and India. As we saw in our last post, there are also far fewer in much of the rest of the world. In fact, there are only a few countries in Africa that contain more than 1000 articles. 

Below are the Arabic, French, Hebrew, and Persian Wikipedias. Rather than discussing each individually, I will discuss some key themes at the end of this post. 

The most striking pattern that you probably notice in these maps is the significant amount of self- or inward- focus of some languages (including English) (an observation that is supported by the work of Brent Hecht and Darren Gergle). There are more Hebrew articles about Israel, French articles about France, and Persian articles about Iran than anywhere else. 

The same pattern, however, doesn't hold true for the Arabic Wikipedia. In fact, in the top-10 list of total number of articles by country in the Arabic version of Wikipedia, there are only two countries (Algeria being 8th on the list and Syria bring 9th) that can be considered to be predominantly Arabic-speaking (the rest of the list is: #1 USA, #2 Spain, #3 Russia, #4 UK, #5 France, #6 Italy, #7 Greece, and #10 Iran).

The lack of focus on, and contributions by, the Arab world is particularly striking in all of these maps. Arabic is the world's fifth most spoken language and yet only has the 25th largest Wikipedia. There are just over 24,000 geotagged Arabic Wikipedia articles whilst there are over 691,000 geotagged articles in English. 

The scale of these difference ultimately results in some almost implausible comparisons. For instance, there are more articles in English about North Korea than articles in Arabic about Saudi Arabia, Libya, the UAE and many other countries in the region!

But, perhaps most interesting is the question alluded to above. Why is there such a (relatively) large number of Persian articles about Iran, but so few Arabic articles about places like Saudi Arabia (the country in which more than a quarter of all Arabic edits originate) and the UAE? A key goal of our research project is to answer this very question and better understand the barriers to participation. 

In future posts, we'll begin to move beyond these raw data counts and explore patterns of participation and representation in the region. In the meantime, any observations and questions are welcome.

Feodor Vassilyev and Wikipedia's Gender Imbalances

I recently had the opportunity to visit the Wikisym 2011 conference and am in the process of writing a conference report with Han-Teng Liao. Doing so made me want to flag up an interesting debate that happened during the meeting.

There has recently been a lot of talk in the media about gender imbalances in Wikipedia (for instance, the often cited statistic that less than 15% of Wikipedia editors are female), and luckily this was also a major theme at the conference. Two papers in particular demonstrated the gender imbalances not only exist, but also significantly influence the types of information that exist in Wikipedia (the papers were titled ‘An Exploration of Wikipedia’s Gender Imbalance’ and ‘Gender Differences in Wikipedia Editing

Perhaps the most interesting discussion of these imbalances came during a talk by Jen Lowe when she brought up the Wikipedia article on Feodor Vassilyev.

Feodor is apparently notable enough for a Wikipedia article because his wife sets the record for the most children birthed by a single woman. Just to reiterate, it is Mr. Vassilyev and not Mrs. Vassilyev who is deemed notable enough to have a Wikipedia article here!
Given the fact that the Vassilyevs were alive in the eighteenth century, the masculinist biases that shaped how this story was recorded are perhaps not surprising. However, what is more important is for contemporary information creators on Wikipedia to become aware of such biases and actively work to not reproduce them. In other words, while we of course need to make efforts to make Wikipedia editors more representative of the general population, we need to recognise that addressing imbalance is only the first step. The issue is not just a lack of female editors, but also gender biases embedded into the ways in which we discuss and represent subjects in Wikipiedia.

Mapping Wikipedia's augmentations of our planet

We all know that Wikipedia is an immense project. It is an incredibly impressive coming-together of human labour on a scale that the world rarely sees. Over the last few years, we've also seen a few maps of the encyclopedia (including my own) which have shown that the project is far from complete (whatever that might mean). 

That doesn't mean we should stop mapping the project though, and as part of our efforts to answer the first of our research questions looking at Wikipedia in the Middle East, North Africa, and East Africa, I'll present these global-scale maps of every article in the November 2011 versions of the Arabic, Egyptian Arabic, English, French, Hebrew, Persian, and Swahili Wikipedias. 

First, the English project. This encyclopedia is by far the largest, and currently hosts almost 700,000 geotagged articles (click on the image for a larger and more detailed version):

Each one of these yellow dots represents human effort that has gone into describing some aspect of a place. The density of this layer of information over some parts of the world is astounding. Some of our future posts will look more closely at measures of inequality in Wikipedia, but it is still hard not to be awed by this cloud of information about hundreds of thousands of events and places around the globe. 

The French Wikipedia, with almost a quarter of a million geotagged articles, is much smaller than the English version, but nonetheless still another impressive collection of human labour. There are much denser augmentations of information are much denser over some parts of the planet than others, but it remains that there is simply a lot of content about a lot of the world.

However, when looking at some of the smaller Wikipedias like Arabic (and Egyptian Arabic), Hebrew, and Persian, we don't see that same glowing cloud of information over much of the world. 

In the maps above we instead saw limited global focus. This is perhaps not that unexpected given the relatively small size of these encyclopedias (in terms of total numbers of geotagged articles, Arabic has 24,000, Hebrew has 15,000, Persian has 21,000, and Egyptian Arabic has only slightly more than 1000 [these are all approximate figures]). 

But it remains that if your primary free source of information about the world is the Persian or Arabic or Hebrew Wikipedia, then the world inevitably looks very different to you than if you were accessing knowledge through the English Wikipedia. There are far more absences and many parts of the world simply don't exist in the representations that are available to you.

However, one thing that should be pointed out are some of the strange patterns on parts of these maps. If you look closely at the Arabic or Persian maps you might see some interesting patterns (for instance look closely at the patterns in the US). You see a similar sort of unexpected spatial distribution of articles in the map of Swahili Wikipedia below (i.e. why are there so many articles in Turkey?). The answer is simply a few dedicated editors creating stub articles about relatively structured topics such as cities in Turkey (in the Swahili Wikipedia) or every county in the US state of Georgia (in the Arabic Wikipedia). 

What is perhaps most interesting about the Arabic, Hebrew, Persian, and Swahili Wikipedias is that it isn't the Global North that vanishes from the map. It is rather other parts of the South that become absent: an observation that seem to simply imply an entrenchment and a reproduction of the visibility of the already highly visible.

Cyberspace

I've just returned from a conference about Internet unlike any other I've been to before. The London Conference on Cyberspace, organised by William Hague and the Foreign Office, was a meeting that attracted huge names (David Cameron, Joe Biden, Helen Clark, Carl Bildt, Jimmy Wales and many others). The ambitious goal of the meeting was "to develop a better collective understanding of how to protect and preserve the tremendous opportunities that the development of cyberspace offers us all."


The meeting was quite stimulating, and it was interesting to hear people like Cameron and Biden outline their visions for the future of the Internet (even if those visions contradicted some moves by both UK and US governments). But the thing that struck me the most was the constant use of the word 'cyberspace.' People just wouldn't stop saying it.


David Cameron told us that "we can't leave cyberspace wide open to criminals." Joe Biden called it "a new realm." Russia's Minister for Communications was worried enough that he asked that the Internet be made to respect borders and state sovereignty. Continuing the use of the spatial metaphor, Carl Bildt, the former Prime Minister of Sweden, speculated that light would be brought to even the most hidden corners of the Internet by asserting that "there will be no dark spaces for dark acts any more." I could go on with examples, but you probably get the idea.


As I've argued before (in my article on "The Spatialities of the Digital Divide"), the 'cyberspace' metaphor is an inherently geographical concept. It allows the virtual to take on an ontic role. 'Cyberspace,' in this sense, is conceived of as both an ethereal alternate dimension which is simultaneously infinite and everywhere (because everyone with an Internet connection can enter), and as fixed in a distinct location, albeit a non-physical one (because despite being infinitely accessible all willing participants are thought to arrive into the same marketspace, civic forum, and social space). 'Cyberspace,' in this sense truly becomes a global village.

The ontic role assigned to cyberspace is likely also reinforced by the grammatical rules associated with the Internet in the English language. Common prepositions associated with Internet use (e.g. to go to a website, or to get on the Internet) imply a certain spatiality associated with the Internet. In other words, the need to move to a cyberspace that is not spatially proximate to the Internet user. Similarly, it is common practice to treat the word “Internet” as proper noun (hence the capitalization of the word). In doing so, the notion of a singular virtual entity or place is reinforced.

Even before the coining of the term, commentators were speculating that synchronous communication technologies like the telegraph would bring humanity together in some sort of shared space. For instance, in 1846, in a proposal to connect European and American cities via an Atlantic telegraph, it was stated that one of the benefits would be the fact that “all of the inhabitants of the earth would be brought into one intellectual neighbourhood and be at the same time perfectly freed from those contaminations which might under other circumstances be received” (Marvin, 1988: 201). Twelve years later after the completion of the Atlantic telegraph, The Times proclaimed that “the Atlantic is dried up, and we become in reality as well as in wish one country” (quoted in Standage, 1998: 80). In the 1960s, Marshall McLuhan’s philosophy of media posited a future not too different from proclamations about the power of communication technologies a century earlier. He noted that “electric circuitry has overthrown the regime of “time” and “space” and pours upon us instantly and continuously concerns of all other men. It has reconstituted dialogue on a global scale...“Time” has ceased, “space” has vanished. We now live in a global village” (McLuhan and Fiore, 1967: 63).


Such ideas were prevalent in the early days of the Internet. John Barlow, for example, in his Declaration of the Independence of Cyberspace, boldly asserts that “cyberspace does not lie within your borders” and “ours is a world that is both everywhere and nowhere, but it is not where bodies live.” 


Most of us have moved beyond such ideas and recognise the hybrid and augmented ways in which the internet is embedded into our daily lives. We recognise that there is no singular ontic entity of 'cyberspace' that we can enter into to transcend our material presences. And that is probably why few of us actually imagine a movement into 'cyberspace' when we access Wikipedia, log into Facebook, or watch a video on YouTube. So why do the global leaders present at the London conference insist on using this term that is so rarely used by everyone else? Or more broadly, why do they continue to employ a spatial metaphor to imagine a network?


I suspect that part of the reason lies in the fact that states, and their representatives and leaders, are naturally concerned with unregulated activity that is hard to geographically place. When thinking about warfare, hackers, pornography, fraud, and other threats to the rule of law, it is challenging to fully understand the complex geographies of these processes and practices. It is much easier to imagine that they simply happen 'out there' in Carl Bildt's dark spaces of the Internet. 


Another reason is likely the extensive literature on the ‘information revolution’ and the 'networked society.' Most national governments have departments, task forces, plan and policies set up to address issues of digital exclusion. Because of the existence of the ‘global village’ ontology of cyberspace, there is often a pollyannish assumption that once the material ‘digital divide’ is bridged, the many problems attributed to ‘digital divides’ will also vanish. Or, in other words, once people are placed in front of connected terminals, the ‘digital divide’ becomes bridged and the previously disconnected are consequently able to enter 'cyberspace.'


As such, those without access to 'cyberspace' and the ‘global village’ are therefore seen to be segregated from the contemporary socio-economic revolution taking place. This idea of exclusion is powerful, and some, such as former US secretary of State Colin Powell, and the chief executive of 3Com, have on separate occasions gone so far as to term this exclusion “digital apartheid.”


In contrast to these imaginations of a digital global village and an ontic entity of 'cyberspace,' my talk at the London conference argued that there isn’t some sort of universally accessible cyberspace that we are all brought into once we log onto the Internet. The Internet is not a space, but rather a network that enables selective connections between people and information. It is a network that is characterized by highly uneven geographies and in many ways has simply reinforced global patterns of visibility, representation and voice that we’re used to in the offline world.


Imagining the Internet as a distinct, immaterial, ethereal alternate dimension ultimately makes it more challenging to think through the contingent and grounded ways in which we consume, enact, communicate and create through the Internet. The Internet is characterised by complex spatialities which are challenging to understand and study, but that doesn't give us an excuse to fall back on unhelpful metaphors which ignore the Internet's very real, very material, and very grounded geographies.

London Conference on Cyberspace Presentation

"The future is already here, it is just unevenly distributed" - William Gibson.

Slides from my talk at the London Conference on Cyberspace are now available below. The session, on "cyberspace and international development," also included talks from Helen Clark (UNDP), Barbara Stocking (Oxfam), Christele Delbe (Vodafone), and Sarah Jordan (Oxfam) and was moderated by Michael Anderson (DFID).

The talk began by arguing that it is important to not forget the revolutionary and empowering promises and potentials of the Internet for the developing world. People use the Internet to not just non-proximately connect with friends and family, but learn, share information, check market prices, trade, and bypass exploitative economic relationships.

But it is equally important to remember – first - that despite a rapid growth in internet access for much of the world, most people on our planet are still entirely disconnected. And – second - even amongst those two billion that are now online, a significant number are still left out of global networks, debates and conversations.

This may seem like an odd point to make at a conference with the word "cyberspace" in its title, but there just isn't any sort of universally accessible cyberspace that Internet users are transported into. The Internet is not a space, but rather a network that enables selective connections between people and information. It is a network that is characterized by highly uneven geographies and in many ways has simply reinforced global patterns of visibility, representation and voice that we’re used to in the offline world.

The issue isn’t just that some people in the developing world are disconnected, but also that many of the benefits of the Internet don’t automatically arrive into the developing world once Internet connections do. In other words, while the internet is clearly a pre-requisite for a lot of economic development and participation in the 21st century knowledge economy, it is by no means a determinant. 

As always, I'm happy to hear any questions and thoughts on the presentation.