Mashups, crowdsourcing and their impact on the mapping industry

Crawford [1] wrote that computers are being used differently with Web 2.0. This illustrates a shift from the conventional publishing model. She writes: “But what has changed over the past fifteen years is that they (the users) no longer represent the only way to produce and distribute creative work. As computing power has become more affordable and software has become more powerful, the creative potential of what can be done at home – and at relatively low cost – has soared. And there has been an explosion of creative production as a response. It has been described as “mass amateurisation”: the masses now have greater access to the means of cultural production. We are witnessing a crucial shift as the gap narrows between what can be done at home and what is professionally produced; amateur productions take on professional approaches and professional productions make use of the amateur aesthetic” [1 p. 23]. She goes on to say: "Everyone is making something, collaborating on something or distributing something. It doesn’t matter if it’s a zine, a new media installation, a piece of software, an album, a short film or a photo blog” [1 p. 23].

Maps produced through the process of mash-ups include the amateur map producer. This map producer has access to powerful Web 2.0 delivered software and resources, empowering them with the ability to produce and deliver maps that are both professional and current. Geographical information and base maps can be sourced from conventional providers – for example the Ordnance Survey (OS) of the United Kingdom has developed an API called Openspace which provides free data for non-commercial experimentation (http://openspace.ordnancesurvey.co.uk/openspace) – and from non-conventional sources – for example Nokia Map (http://europe.nokia.com/maps) or from OpenStreetmap (www.openstreetmap.org), the organisation providing free data and maps that are produced by individuals who collaborate to provide a free geospatial resource.

However, with Web 2.0 for the provision of maps and geographical information is not without a number of issues. The following section addresses some of these.

The use of Web 2.0 as a means for providing geographical information presents different problems for cartography. There are issues to be dealt with such as:

There has been a movement of data repositories from just government resources to private industry with companies like Google and MicroSoft purchasing massive amounts of geospatial information. The model of data collection, storage and distribution has changed. Non-public organisations now control massive amounts of data and provide it, in many instances for free. But, will this continue to be the case? And, are users being provided with data for now, with future access perhaps attracting a fee? Some uncertainty does exist with this private sector data model.

When accessing geospatial information from public sector repositories users were assured that the data had been properly collected, maintained and updated by responsible authorities. For example the Ordnance Survey states “We make an average of 5000 changes every working day to our large-scale map data of Great Britain” [2]. These government or quasi-government sources of information have been trusted as custodians of this service which is funded by the public purse and by user payments and royalties. But how is the quality/integrity of data assured by commercial or collaborative data provision resources.

The OpenStreetMap organisation incorporates a number of quality checks in its data collection to delivery system. It actively encourages data collectors/mark-up collaborators and users to check the quality of their data and to make changes if necessary [3]. (See the OpenStreetMap Wiki for further information at: http://wiki.openstreetmap.org/wiki/Quality_Assurance.) A quality statement from Google related to their maps could not be found at the time of writing, However, the title of a presentation by Google’s Ed Parsons at the sixth International Symposium on Spatial Data Quality in July 2009, "When Good Enough, is Good Enough: Data quality requirements of the geoweb", indicates that Google are conscious about data quality issues and its maps.

One blog site that looks at mapping provision via the Web and its application to local search had this to say about map accuracy and Web-delivered products: “Geocoding errors and data quality seem to plague most mapping applications, but are especially pernicious in Local Search, store finders, brand finders and other systems that pretend to deliver consumers to buying opportunities. Open almost any internet mapping application and look at some areas you know, and you will find map errors, geocoding errors and other data quality errors. How can it be so bad?” [4]

And, a Google user support forum contributor (mtiffany) had this to say about the quality of Google maps:

“I've been using Google Maps for years and I've noticed a significant decrease in quality in the past several months from both the Web and the Google Maps for Windows Mobile interfaces. I live in New York, NY, not exactly an obscure place, and the road I live on is not even on the map anymore! It certainly was last year.” …“Obviously nothing is perfect, and I assume mapping is hard, but why is Google Maps, overall, worse than it was last year? Before this decrease in quality, I guess I naturally assumed that Google products all get better over time. I think the usability of the interface has gone up, and there are some nice new features, but what's the point, if the actual maps and routes aren't as good?” [5].

This is an area of major concern, and much debate. For example, Mary Spence was interviewed on the BBC in 2008 about Internet maps [6] and she criticised the lack of content in some internet-delivered maps. This caused much talk on blog sites that focus on Web mapping, like Google Earth Design (http://googleearthdesign.blogspot.com/2008/09/gis.html) [7], which conducted a post BBC News telecast email interview with Spence. If users do not trust maps and geographical data delivered via the Web per se, then they may not trust data delivered by any source, including reputable traditional information providers.

Linked closely to the previous topic is the issue of data maintenance. We assume that traditional custodians of geographical data maintain their data to acceptable standards. But, do the "new players" in geospatial information provision also maintain their data to the same standards that the users of ‘traditional’ data repositories expect? This issue also relates to map data user and map user’s confidence in the data supplied or the map generated from non-traditional data repositories.

Data "scraping" is the process whereby a Web page source code is interrogated automatically and data intended to generate screen or printer output is extracted and a new data file created. This bypasses completely the need to access databases directly and altogether payment or authorisation by the data owner. Data is scraped from general websites or from search engines.

This obviously is of great concern to organisations and individuals who generate Web maps and do not wish to have their data copied in this way. For repositories where free access and use of data is possible it is not a problem, but for organisations that sell or licence data this should be of great concern.

Added to this problem is that of attribution. If data can be scraped and another mapping product generated from this data, it would be possible for another product to be generated and published with no reference whatsoever to the original data source or provider.

As covered earlier, there are a number of social, collaborative organisations that make their data and maps available for free. Through a network of Web-connected amateur and professionals, geographical data is collected, scripts marked-up with data attributes, data placed in repositories and maps published. These organisations can be viewed to be either competing or collaborating with traditional suppliers of geographical information.

The Ordnance Survey (OS) has recognised that “…it will face increasing competition from commercial rivals to deliver geographical information services to the public sector and others” [8]. The OS has developed a new business strategy and one of the five key areas that has been addressed is to “Promote innovation for economic benefit and social engagement” (Goal 1). Here the OS will provide additional data and usage rights for “publicly accessible applications” by providing free data at scales from 1:10 000 to 1:1 000 000 as well as official boundaries information [8].

With the massive growth of information that is geotagged and the ability to transmit and receive maps and geographical information via the mobile internet, through the use of mobile telephones and wireless devices, the consumer electronics industry has developed enormously. The general public are now offered a plethora of devices and associated applications that are geo-located or referenced. Location-based services (LBS) and "at location" mapping, where maps are delivered where and when needed using wireless technology, have now become ubiquitous. But, there is a trade-off of information accessibility for privacy [9, 10]. As these devices  are usually always switched on, the user can be continually tracked by a service provider. This issue is perhaps one of the sleeping problems of mobile geographical information services that might cause problems in the future and limit the success of maps delivered via this medium.

Another privacy problem has arisen with data capture for Google StreetView. When collecting imagery not only inanimate objects like buildings are captured, but also people in the street are also photographed. This has led to concerns about privacy and the unauthorised photographing of individuals. Contributors to the blog site Boing Boing were asked whether they would be concerned if the CIA were collecting such information in pubic places, and would the public accept it [11]. In some instances individuals have contacted Google to request that their image or property be removed from the site. In response to the public’s concerns about privacy, Google StreetView has blurred the faces or people captured in its imagery, as well as other identifying items. One interesting sideline is that some street signage, like Kentucky Fried Chicken’s (KFC) advertising which includes an image of Col. Sanders (the founder of the company) was blurred as well. Google Street View said that this was done because he is "a real person" [12].

There are now new players in geospatial information provision whereby consumer electronics companies like Nokia have purchased mapping companies. As well as Google wanting to enhance its advertising potential and Microsoft its profile in computer software through a Web presence, other players are now also part of the geospatial industry. For example, TeleAtlas was purchased by TomTom and Navteq was acquired by Nokia. The potential of enhancing consumer electronics with geographical information is now evident in the advertising about what these devices can now do. As these companies do not have cartography as their main focus there exists the need for the cartographic community to seek ways of collaborating.

When using the Web for map publishing, quality was adjudged by the speed of delivery, circulation figures and screen resolution. Quality was gauged by how the "rules" of computers and communications systems were applied. Users were still seen as consumers, and not collaborators in geographical knowledge acquisition. The use of Web 2.0 as a means for providing geographical information presents different problems for assuring quality. Problems might arise with a conglomerate product related to "self-constructed" Web 2.0 products. For conventional cartographic products these assurances are provided by cartography. A major issue if self-composed products are to be used with confidence might well be quality assurance. Therefore, methods would need to be developed for assuring quality with conglomerate products, assuring quality with user-produced products and the means for informing users about the source of conglomerate information resources.

By using data and information from the general public and by making digital information freely accessible via the Web, outcomes that would otherwise not happen can result. This concept of making data freely available for problem-solving or by “harvesting” information from Web users has been termed "crowdsourcing".

In an article in Wired magazine Howe [13] outlined the phenomenon of “crowdsourcing”. He commented about how the Web had changed where companies now outsource their contract work and how contract workers for certain work can be physically located anywhere, as long as they are connected to the internet. He also noted how even outsourcing via the internet had also changed – from outsourcing to “crowdsourcing”:

“Technological advances in everything from product design software to digital video cameras are breaking down the cost barriers that once separated amateurs from professionals. Hobbyists, part-timers, and dabblers suddenly have a market for their efforts, as smart companies in industries as disparate as pharmaceuticals and television discover ways to tap the latent talent of the crowd. The labour isn’t always free, but it costs a lot less than paying traditional employees. It’s not outsourcing; it’s crowdsourcing” [13].

This type of methodology for problem-solving has been called “The wisdom of the crowds” by James Surowiecki [14]. He says: “Ask a crowd, rather than a pair, and the average is quite close to the truth“.

This approach was formalised somewhat in 2001 when the pharmaceutical company Eli Lilly funded InnoCentive (www.innocentive.com), which invites researchers from outside the company to address the company’s research. Later, companies like Boeing, DuPont, and Procter & Gamble also used InnoCentive to put their research questions to outsiders. "Lakhani", from MIT, studied InnoCentive and said: “The strength of a network like InnoCentive’s is exactly the diversity of intellectual background” …“We actually found the odds of a solver’s success increased in fields in which they had no formal expertise.” He links this method of problem solving to sociologist Mark Granovetter’s “strength of weak ties” [15], where “efficient networks are those that link to the broadest range of information, knowledge, and experience”. A similar application is Amazon’s Web-accessed Mechanical Turk (www.mturk.com:443/mturk/welcome) that invites individuals to make interpretations of imagery. Here these “HITs” (human intelligence tasks) require little time and input to complete. The fee for completing these tasks is also modest.

In terms of mapping and the provision of data in a different manner, Google’s Ed Parsons had this to say: “We should not forget that they are simple and cheap approaches to providing greater levels of information to the citizen by allowing the citizen to carry out the analysis themselves."

Another key point I made was that the next generation of citizens, “Generation Y”? if you like, are in many ways more open to sharing data, having grown up defining their personalities on-line via Myspace and Bebo. However this willingness to share data with others, even government, comes from the fact that as authors of  their own data they are free to modify, correct and update it.

“For anyone delivering the citizen services of the future here is an important lesson – it is NOT your data, it is the citizens’ and they must feel true ownership of it.” (Parsons, quoted in The Guardian, [16].

A mapping application developed at University College London’s Centre for Advanced Spatial Analysis (CASA) undertook a project that used crowdsourcing to map anti-social behaviour in East Anglia, UK [17]. They wanted to map things like “peoples perceptions on: fear of household burglary, quality of local schools, who would you vote for?” [18]. They developed an application called MapTube, which combined the idea of YouTube and their software GMap Creator to produce thematic maps. A pilot study was undertaken to generate a “mood map” of the credit crunch for the United Kingdom. This was done with the UK’s BBC Radio 4 iPM show [19].

Web 2.0 offers the potential for providing geographical information in a collaborative, shared manner. Already the impact of maps via Web 2.0 has been felt by the ever-growing number of maps being published as collaborative products via mash-ups.

For the cartogaphic community this provides both opportunities and issues that need to addressed. The opportunities include the ability to include the amateur cartographer in the map production equation, so as to benefit from these members of the cartographic community who can contribute greatly to mapping endeavours. By sharing resources more effective procedures can result and the amount of geographical information available can be increased. But, there are a number of issues that need to be addressed if the potential of Web 2.0 is to be responsibly exploited. Some of these issues have been covered in this paper. It is hoped that they might be further explored.

[1] K Crawford, 2006. Oh grow up!. Edited extract from the book Adult Themes: Rewriting the Rules of Adulthood, Macmillan Australia, in The Age Good Weekend, 23 September 2006, pp 18–23.

[2] The Guardian, 2006, “Free Our Data: Articles: the Ordnance Survey official response” (Blog), http://www.freeourdata.org.uk/ordnancereply.php, Web page accessed 25 May 2009.
[3] CloudMade, 2009, “About OpenStreetMap“, http://cloudmade.com/about/osm, Web page accessed 25 May 2009.
[4] M Dobson, 2009, “Data Quality and Local Search – Quid Est Veritas?“, Exploring Local (Blog), 13 April 2009, http://blog.telemapics.com/?p=101, Web page accessed 25 May 2009.
[5] Google, 2009, “Why is the quality of map data worse than it used to be?”, Google “Maps Help” support forum, http://www.google.com/support/forum/p/maps/thread tid=2d22d2dac07521e2&hl=en, Web page accessed 25 May 2009. 
[6] BBC News, 2008, “Online maps 'wiping out history'“, Friday, 29 August 2008., http://news.bbc.co.uk/1/hi/uk/7586789.stm, Web site accessed 20 April 2009.
[7] Google Earth Design, 2008, “Ed Parson Calls for a ‘New Cartography’”, Blog, Monday, 1 September 2008, http://googleearthdesign.blogspot.com/2008/09/gis.html, Web site accessed 25 May 2009. 
[8] Ordnance Survey, 2009, "Ordnance Businiess Strategy", Blog, http://strategy.ordnancesurvey.co.uk, Web page accessed 25 May 2009.
[9] R Butler, W Cartwright, Black, M. and C Arrowsmith, 2005, “Will Privacy and Security Compromise the Further Development of location-Dependent, Internet Delivered, Geographically-Related Services?”, Spatial Sciences Institute Conference Proceedings 2005, Melbourne: Spatial Sciences Institute, ISBN 0-9581366-2-9.
[10] W E Cartwright, 2007, "From mapping Physical and Human Geographies to Mapping ‘Personal Geographies’: Privacy and Security Issues", Multimedia Cartography Edition 2, W E Cartwright, M P Peterson, and G (eds) Gartner, Heidelberg: Springer-Verlag, pp. 455 - 470.
[11] Boing Boing, 2007, “Google Street View: would it be more/less evil if it were CIA or NSA?“, Blog, 3 June 2007, http://www.boingboing.net/2007/06/03/google-street-view-w.html, Web siteaccessed 25 May 2009.
[12] The Telegraph, 2009, Google Street View blurs face of Colonel Sanders at every KFC“, The Telegraph online, 20 May 2009, http://www.telegraph.co.uk/scienceandtechnology/technology/google/5356031/Google-Street-View-blurs-face-of-Colonel-Sanders-at-every-KFC.html, accessed 24 May 2009.
[13] J Howe, 2006, “The Rise of Crowdsourcing”, Wired , Issue 14.06, June 2006, http://www.wired.com/wired/archive/14.06/crowds.html, Web page accessed 21 August 2008.
[14] J Surowiecki, 2004, The Wisdom of Crowds: Why the Many Are Smarter Than the Few and How Collective Wisdom Shapes Business, Economies, Societies and Nations, Random House.
[15] M S Granovetter, 1973, “The Strength of Weak Ties”, The American Journal of Sociology, Vol. 78, No. 6., May, 1973, pp. 1360-1380, http://www.jstor.org/pss/2776392.
[16] The Guardian, 2007, “Ed Parsons, ex-Ordnance Survey: ‘data belongs to citizens’” (Blog), 8 April 2007, http://www.freeourdata.org.uk/blog/?p=96, Web page accessed 25 May 2009.
[17] A T Crooks, A M Hudson-Smith, R Milton, and M Batty, 2009, “Crowdsourcing Spatial Surveys and Mapping“, in D (ed.)Fairbairn, Proceedings of the 17th Geographical Information Systems Research UK Conference, Durham University, England.
[18] CASA, 2009, “GIS and Agent-Based Modelling_ Crowdsourcing Spatial Surveys and Mapping“, Blog, 27 February 2009, http://gisagents.blogspot.com/2009/02/crowdsourcing-spatial-surveys-and.html, Web page accessed 25 May 2009.
[19] A Hudson-Smith, A T Crooks, M Gibin, R Milton, and M Batty, forthcoming, “Neogeography and Web 2.0: Concepts, Tools and Applications”, Journal of Location Based Services.

Contact Prof. William Cartwright, International Cartographic Association, william.cartwright@rmit.edu.au