Innovative Auto-ID and LBS - Chapter Eleven Geographic Information Systems & Location-Based Services

Chapter XI: Geographic Information Systems & Location-Based Services



This chapter is about geographic information systems (GIS) and its relevance to the location-based services industry. One might initially ask how relevant GIS is to a book that is predominantly about automatic identification and its future trajectory. The answer becomes apparent quickly as the reader is introduced to the importance of geocoding information, i.e., geographically linking data such as personal details using a unique ID number. In the past data matching programs have received a great deal of attention from privacy advocates, especially those used for the administration of government procedures. Till now, automatic identification has facilitated electronic services (e-services), allowing an individual to be matched to a fixed address, usually their place of residence. But it is one thing to tag and another to track. Today, we are moving towards a model of tracking and monitoring people as they go about their daily business, in real time. We are no longer satisfied with knowing where an individual lives but we want to know their every move- so that we can estimate traffic congestion on a busy road, design 3G mobile networks that have enough capacity during busy hours, and to ensure someone’s safety when adequate supervision is not available.



Geographic information systems (GIS) are playing a crucial role in the success of location-based services (LBS). GIS is defined by Burroughs (1986) as a “set of tools for collecting, storing, retrieving at will, transforming, and displaying spatial data from the real world for a particular set of purposes” (Taylor & Blewitt, 2006, p. 9). Dransch (2005, p. 32) classifies LBS as a subset of mobile geoservices. A location-based service is the ability for an information system to denote the position of a user, based on a device they are carrying or their position in a given context (Gartner & Uhlirz, 2005, p. 159). LBSs have the ability to provide specific, relevant information according to a given “spatial location associated with a physical point or region relative to the surface of the earth (Dawson et al., 2006, p. xv). While a great deal is written about the network technologies that allow for the tracking and monitoring of objects and subjects, GIS is usually considered the add-on feature. However, without GIS, most location-based services would not be plausible as commercial offerings. According to Lopez (2004, p. 171) “LBS consist[s] of a broad range of services that incorporate location information with contextual data to provide a value-added experience to users on the Web or wireless device.” It then follows that GIS is integral to the success of LBS (Brimicombe & Li, 2007). Motivations for using GIS in LBS include: cost-effectiveness, service provisioning, system performance, competitive advantage, and database creation, access, and use (Shiode et al., 2004, p. 363).


What is the Difference between GIS and LBS?

Lopez (2004) is extensive in his book chapter on the differences between GIS and LBS. He stipulates that GIS is about mapping, spatial indexing, spatial operators, geocoding and routing technology. LBS on the other hand, is different to GIS because it makes use of information technology and wireless technology. GIS is about maps, people, places, buildings, points of interest, while LBS is about using that basic knowledge to provide some kind of meaningful application such as “find the nearest Automatic Teller Machine” or “help me, I’m lost”. Performance, scalability and interoperability are three other differentiators of LBS and GIS. LBS requires numerous components to work together to provide an end-to-end solution while GIS is quite localized in that it manages and stores information using proprietary data structures and models. Consider the location service for WAP phone users in shopping centers, such as the Colombo Centre located in the Iberian Peninsula (Câmara & Dia, 2004). GIS provides the spatial background usually in the form of a map form with georeferenced points of interest while the LBS provides the awareness, the reality of the user’s position in a given context.


The Importance of Geocontent to LBS

In his classic text, Location-Based Services: fundamentals and operation, Axel Küpper (2005, p. 35) stipulates that while spatial databases and geographic information systems cover a broad range of applications such as surveying and transportation, it is in the context of LBS that “they are important for indicating the positions of one or several targets with respect to geographical content like borders of cities and countries, road networks, or buildings. They are used for mapping spatial location onto meaningful descriptive location information and vice versa, which is referred to as geocoding or reverse geocoding respectively, as well as for creating digital maps and routing information, or for finding nearby points of interest.” Here Küpper is referring to geocontent. Jensen (2004, p. 117) has written that geocontent is essential to LBS, citing the metaphor that “[u]sers think of the real content as being located in a transportation infrastructure and access the content via the infrastructure.” Jensen uses the example of a point of interest (POI) location which is typically given in terms of a civic address (i.e. containing a street name), and the routing directions for how to reach that destination are given in terms of the transportation infrastructure. However, it is important to note that while LBS uses location as a “context parameter to which the presented information is referred”, it is about “the complete situation in which a user acts and requires spatial information” that is becoming increasingly important (Dransch, 2005, p. 32). It is not only about where they are or where it is, but in the context of the time, the temperature, the lighting, the humidity, the precipitation level, the elevation, the sugar level in the bloodstream, the proximity of the user to a shopping mall, to a friend, to family, to normal patterns of behavior including speed and spending patterns. Shiode (2004, p. 352) predicts that this kind of capability- referring specifically to the ability to analyze objects with respect to a particular location in terms of relationships to each other- will have a transforming effect on “average” LBS solutions in the future.



In location-based services, GIS is used to help end-users visually determine where an object or subject has been (i.e. provide a track or breadcrumb of a subject) or to do an instantaneous locate of that subject (i.e. to pinpoint their location). At the same time, the location headquarters requires a visualization of an object or subject every time a device is triggered putting through a request (e.g. an emergency request for urgent assistance). In this instance, using the GIS to see the “locate” information visually is more beneficial than just receiving a longitude and latitude coordinate (figure 1). Emergency services prefer a civic address or “nearest to” this address than coordinates. The GIS also grants the end-user the ability to check on multiple mobile devices at once with the ability to analyze their spatial relationship. Many industry analysts see GIS as forming the foundation for many LBS developments as a core area in managing, processing and delivering spatial information (Shiode et al., 2004, p. 351). Commercial location-based services that use GIS software include fleet management companies that specialize in all forms of transit, personal locate devices for humans for care, convenience and control purposes, and asset tracking companies which focus on supply chain management and other e-business applications (Michael, 2004). Elliot and Phillips (2004, pp. 14-15) categorize location based services into five groups: product retailing, information, maps, purchasing, and access.


The LBS Value Network

The best way to view GIS with respect to LBS is in the context of a well-functioning value chain or value network. Jensen (2004, p. 116) believes that at a high level of abstraction, the LBS value network “begins with content providers that supply various types of content that can be georeferenced to a content integrator”. According to Paavilainen (2001) however, the LBS value chain includes mobile networks, software development, application development, content and online service providers, branded portals, and end-user devices. Here again, Lopez (2004, p. 172f) makes the distinction between the value chain of GIS which is generally limited to the providers of a desktop or client-server solution versus the complex value network of LBS. “Another major difference is that LBSs impose significant technology and service capabilities that exceed the general requirements of static GIS uses.” There is now a burgeoning list of stakeholders that are involved in the delivery of a sophisticated LBS application using technologies and techniques that are converging such as telecommunications infrastructure (mobile network), positioning methods, mobile in- and output devices and multimedia cartographic information systems (Gartner & Uhlirz, 2005, p. 159). These are the main prerequisites for the development of applications which use position information as a variable in a system (figure 2).


The Rise of IP Location-Based Services

Currently, end-users are able to view the results of their requests in an offline mode or an online mode. Offline modes are not very popular but are useful for non-critical applications or where post-mortem analysis is conducted. On-line modes of response are popular because they can reach end users in a relatively short period of time (from 2-3 minutes to a couple of seconds), depending on the network access type (figure 3). The most popular vehicle for the response to be fulfilled continues to be the Internet, even though the telephone is heavily relied upon in emergency situations. The current state of development is focused on delivering LBS solutions to any access method independent of device type, such as Personal Digital Assistants (PDAs) and cellular phones. The current limitations include device screen sizes and bandwidth availability, although the latter is changing with broadband techniques and third generation networks.

Karimi (2004, p. 7f) emphasizes the impact that wireless networks and Internet-based GISs have had in meeting the requirements of telegeoinformatics, in particular the rising use of mobile location-aware devices. “Wireless networks are the fundamental enabling technology for emerging Internet-based GISs providing access and solutions to both desktop and mobile users.” Meng and Reichenbacher (2005, p. 6) identify map-based mobile services as a special type of value-added LBS that “…afford[s] both descriptive information and procedural knowledge.” Edwardes et al. (2005, p. 11) discuss the extreme demands that mobile mapping has had on GIS. There is an obvious trade-off between the highly flexible all you can have and eat view of GIS (i.e. the endless possibilities one might have through the use of detailed satellite imagery and other rich GIS media) versus the crippling constraints in end-user device screen resolution, limited processing power and irregular patterns of network connectivity. Clearly there is a chasm between the what is possible in GIS and the realities in the current state of development in mobile. Gartner and Uhlirz (2005, p. 160) agree that “[t]he possibilities of transmitting spatial information in the context of a determined position by various presentation forms is primarily restricted by the limitations of the used mobile device”. This is not to say however, that fixed Internet GISs are not hindered by any limitations with respect to mapping or location-based services but the problems in that environment are different and are mainly preoccupied with licensing issues, royalties and suitable subscriber billing models.


Precision versus Proximity Location Technologies

The visual precision of a map displaying someone’s location is intrinsically linked to the level of detail and preciseness that can be extrapolated from the location determination technology being used in the network. For example, indoor and outdoor applications have differing levels of capability. In some instances, precise location information can be obtained, in others proximity information can be obtained down to a couple of meters, and still in others showing that a subject is within a post code may be the level of refined searching that can be done and subsequently displayed. While precise and proximity polling is available today, it comes at a price. Beyond the cost factor, which will inevitably drop over time, there is the obvious limitation that some technologies work well outdoors (like GPS), while other technologies like wireless fidelity (wi-fi) can be used within a building or campus.

Precision versus proximity can also be determined by how a locate of a handset takes place, i.e., is it via a network-based solution by a mobile operator or is it via a handset-based solution (Paavilainen, 2001, p. 169). Handset-based solutions which allow for highly accurate locates, are becoming increasingly popular with mobile handset manufacturers, as prices for GPS receivers fall and innovative software solutions can be built into software on a SIM card. Increasingly humancentric LBS applications are relying on integrated network infrastructures to be able to poll a device out in the field. For example, the Global Positioning System (GPS) is now being used with cellular networks to offer a precise locate using Assisted-GPS (A-GPS) techniques (Taylor & Blewitt, 2006, ch. 4-5). When the device carrying the receiver is out of a cellular footprint range, the service only then reverts to GPS. In this instance, the location headquarters needs GIS beyond that of the purpose of network management.

According to leading researchers in the field, the main research questions in the context of cartographic LBS include: integrative positioning, route information systems, information presentation and visualization (e.g. e-maps and virtual reality). The complementarity between outdoor and indoor positioning will offer powerful surveillance possibilities to governments, employers, and citizens never before seen. Pervasive computing is set to revolutionize the way we live and work and is closely allied to LBS. Jensen (2004, p. 118) also believes that the dynamic union between what he calls geocontent and real content will offer giant leaps in LBS VAS. In actual fact, he is referring to live geocoding practices which will likely be at the centre of true fourth generation intelligent mobile applications (Timpf, 2006). He writes “[i]n LBS, the special-purpose GIS world and the general-purpose relational data management world need to be integrated.”


Case 1: Mobile Location tracking Applications

The data for the following cases was collected between 2002 and 2004 from three company web sites. The cases stand as a historical snapshot of three early deployments of LBS. Some of the more representative LBS applications on the market in 2004 included: iMode by NTT DoComo, mMode by AT&T Wireless, the Personal Locator by WherifyWireless (figure 4), and the VeriChip by Applied Digital Solutions. Five years later, in 2009, the capabilities of LBS throughout the world have grown in sophistication and functionality. In 2004, iMode and mMode offered consumers and business users a diverse range of mobile commerce applications, including LBS functions to find people nearby, find facilities nearby, and get directions, weather and traffic reports. The Personal Locator used a GPS wristwatch and additionally took advantage of the wireless operator’s footprint within the coverage area to identify an individual’s latitude and longitude coordinates (figure 5). The VeriChip device on the other hand allowed for identification of a user in a building and could be used for offender monitoring and patient-supplied healthcare-related information. VeriChip’s VeriTrack application offering was marketed as the “who, what and where of your company… VeriTrack is designed to track, monitor and protect all assets within an organization or company, including people.” Other niche LBS are those such as the DestronFearing Corporation offering for animal ID, Skye-Eye for asset tracking, SnapTrack for fleet tracking, Starmax’s Startrax monitoring system, and CarCom as a locator for cars.



AT&T Wireless was the first mobile carrier to launch m-Commerce applications in the US in July 2001. Following the success of NTT Docomo’s i-mode and c-mode in Japan, mMode provided a value-added data-centric package to AT&T’s voice and SMS basic plans. Subscribers to mMode could use numerous devices to communicate including IP-enabled phones, PDAs, handhelds and even vertical devices such as the Panasonic Toughbook and Microslate Sidearm. The service was carrier-grade and based on a GSM network architecture that used new network elements, namely the Gateway Mobile Location Centre (GMLC), Serving Mobile Location Centre (SMLC), and the Location Measurement Unit (LMU). The accuracy of the specific location-based applications was dependent upon the general location of the mobile transmission tower most recently contacted by the customer’s device. For example, the IP device could be right next to a tower or some fifteen kilometers away. In metropolitan areas the accuracy is greater given the number of base transceiver stations is higher than in less urbanized areas.


“My mMode: this time it's personal”

mMode is heavily oriented towards the consumer market, although AT&T Wireless also offered package deals to business users specifically for the purposes of email (plus attachments), web access, and remote access. mMode was marketed as the beginning of mLife, next generation services that ‘one could not live without’ (McDonough, 2002). Among its mCommerce suite that includes news, music and finance services are a number of LBS solutions. mMode’s LBS applications were diverse- everything from a mobile traffic report to directions ‘to the nearest’ and find people nearby (AT&T, 2003). Some of the more creative LBS applications were featured in chat and date, and travel and dining. There were four plans subscribers could choose from including: mini, mega, max and ultra. The plans were charged monthly ranging from $2.99 to $19.99 USD and included a limited megabytes (MB) download. Additional usage fees were charged at between 2c and 0.6c per extra kilobyte (KB) received or sent, dependent on the plan. These fees did not include voice calls and SMS. The mMode service was bundled allowing the subscriber maximum personalization to choose from any application they required. The myMode web site allowed the subscriber to customize their preferences and settings.



mMode’s location identification was not pin-point such as in the Wherify Personal Locator solution that was based on a combination of GPS satellites and code division multiple access (CDMA) PCS network triangulation methods (figure 6). The Personal Locator wrist-worn device was accurate within 30 meters of the wearer, possibly even as close as a meter. The GPS device could be controlled by both the subscriber and individual wearer, allowing the parent subscriber to track the wearer, and for the wearer to alert the parent subscriber and/or location centre headquarters in case of an emergency. Coverage was available throughout the US given the GPS capability but was dependent on the PCS network coverage footprint. The Wherify (2004) frequently-asked-questions (FAQs) page stated: “[i]f a GPS signal is received, but the Locator is outside the digital wireless coverage area or does not receive a digital wireless signal, no location report will be provided. If the Locator receives a digital wireless signal, but no GPS signal is available, a CDMA tower-based location report will be available for emergencies.” On December 30th 2003, Wherify unveiled its new GPS Universal Locator Phone which was targeted at all age groups of both the consumer and business market. After its initial market testing phase, Wherify is now marketing the WheriPhone which has the same functionality, and is small and compact.


Personal Locator “Just For Kids”

In contrast to AT&T Wireless, Wherify strategically chose to enter the market with a niche LBS application for a Personal Locator Just For Kids, specifically targeted at parents of children between the age of four and twelve (figure 7). The device previously cost $399 USD but was slashed in 2004 for a “back to school special” of $199. Monthly plans for the LBS application ranged from an average of $19.95 to $44.95 dependent on the plan chosen (liberty, independence or freedom). There was a one-time activation fee of $35 USD plus usage fees related to additional page requests above the included locates, additional operator assistance calls and subsequent emergency calls. Wherify (2003) made it clear that it was looking to diversify to other niche applications including Alzheimer’s and law enforcement, even though the Locator for Kids was the only marketable application demoed on the web site at the time. In 2009, Wherify was marketing the following location-based applications: family finder, social networking, fleet management, mobile search and advertising and asset tracking, among others.

Wherify’s location service centre (LSC) was at the heart of its product innovations. A carrier-class server and software hub, the LSC manages and presents location-based information. Unlike mMode, Wherify utilizes wireless data and aGPS. Consider the following scenario where a parent wants to be reassured that their child made it to school alright after missing the bus. The parent requests a location report via the Internet using a Microsoft IE browser (or ringing the toll-free telephone number). The LSC contacts the child’s Personal Locator via the PCS network (if within the footprint), and then downloads the current GPS data and requests a location. Using the data from the LSC, the device that is identified by an electronic serial number (ESN), finds the closest satellite and then computes the longitude and latitude coordinates of the child’s location. The Personal Locator then communicates location information to the LSC and the LSC generates a location report for the parent via the Internet. The whole process from request to report takes about sixty seconds. The parent is able to look at the report visually on a scalable map which shows streets and other feature points in a vector or aerial view, using geographic information systems (GIS) capabilities (figure 8). Each report requested by the parent is logged in the customer’s event file database for billing and subscriber profiling. The location database included a time stamp along with the longitude/ latitude coordinates (figure 9). The wearer’s profile was also stored including: age, gender, height, weight and features.

In 2004, Wherify made no secret of their technology partners. They included an impressive list of companies: SiRF who provide the GPS chipset that is integrated into the Personal Locator based on a-GPS; Qualcomm for the CDMA chipset; Baldwin Hackett & Meeks who are applications developers, Conexant who provide the RF board; Advanced Micro Systems who specialize in flash memory; Compaq for the server technology; Intrado for emergency communications; and GlobeXplorer Online for the component of aerial photography. Security firewalls are paramount in the Personal Locator system as is redundancy and fault tolerance. During an emergency situation for instance, the LSC is even able to interact with public safety answering points (PSAP) through Wherify’s emergency operation service. There are customer care representatives available 24x7x365.



While mMode required the subscriber to carry a device, and the Personal Locator required an individual to wear a device, VeriChip was radical in that it required the subscriber to be implanted with a microchip. The campaign to “Get Chipped” was launched in early 2003, and the first person to do so formally was implanted in September of that year. The chipping procedure only lasts a few minutes. In 2003, there were a number of Veri centers where the procedure could take place in the US. There was even a high-tech ChipMobile bus fully equipped to perform the implant procedure, ‘on the road’.

Applied Digital Solutions (ADSX) initially invested heavily in another product they called the Digital Angel in 2002, which resembled the Personal Locator solution but aimed at a broader market base than just children. The Digital Angel wristwatch was more slim-line but required the user to carry an additional wallet with battery power. While remnants of the Digital Angel web site were still operational in 2004, it was VeriChip which had become the flagship product of the VeriChip Corporation (a subsidiary of ADSX). About the size of a grain of rice, the VeriChip was the world’s first subdermal radio-frequency identification (RFID) microchip. According to an ADSX (2003) press release: “[t]he standard location of the microchip is in the triceps area between the elbow and the shoulder of the right arm.”

In theory an implantee could be identified in a wi-fi network, such as in a workplace or university campus. Whereas GPS has limitations in-building locations due to construction materials used, RFID thrives in a local area network (LAN) setting, allowing walkways and door entries to act as scanners. RF energy from the scanner triggers the dormant VeriChip and in turn sends out a signal containing the unique verification number. The exchange of data is transparent and seamless in the case of RFID, there is no need to physically stop to verify a biometric feature- the network is ubiquitous. In another scenario, an individual could be identified by the RFID implant, giving emergency services access to the implantee’s medical data and history that could be potentially life-saving. Unlike other fixed services, m-Commerce applications grant the subscriber access to services twenty-four hours a day, seven days a week. In the case of the VeriChip it is not only “always on” but “ever-present” inside the body of the subscriber. Unlike physical biometric attributes, the VeriChip is inconspicuous to the naked eye.

In a recent comment provided by a VeriChip spokesperson, Ms Allison Tomek said to Channel 9 News in New South Wales, Australia, that VeriChip was not pursuing “an implantable microchip with GPS technology” (Asher, 2009). While this statement is in fact probably true, one need only look at the service mix of today’s Digital Angel (2009) Corporation which has suddenly made a resurgence back into the global market, after selling its stakes in the VeriChip Corporation late last year (RFID News, 2008). It is hard to see what went wrong in just a little over a year since the two organizations agreed to develop an “implantable glucose-sensing microchip, negating the need for diabetics to draw blood to monitor their blood glucose levels” (Sensors, 2008). According to the homepage of the Digital Angel Corporation (2009), the company currently specializes in “GPS and RFID products… utilized around the world to save lives, ensure the safety of food supply, reunite loved ones and improve the quality of life.” For Digital Angel it is clear that accurate identification, location tracking and condition monitoring are important humancentric applications that will converge.


“Get Chipped” with VeriChip: “technology that cares”

There was little information on the ADSX web site about the pricing of the VeriChip, however it was stated that the global VeriChip subscriber (GVS) registry subscription fee is $9.95 USD monthly. There was a cost for the implant medical procedure as well, although this was not provided. In 2002 the first one hundred pre-registered persons were granted a $50 USD discount on the chipping procedure (ASDX, 2003). The pricing for the new VeriPay and VeriGuard services had yet to be published on the WWW. The “Trusted Traveler” and residential security programs (i.e., prisoners serving their sentence from home) are two examples of VeriGuard LBS applications. One desirable feature of VeriGuard was that it could operate in conjunction with other auto-ID technologies like smart cards and biometrics, rendering customer legacy systems reusable.

VeriChip was based on RFID. RFID networks are usually small in scale when compared to nation-wide or global networks. They include the following components: the RFID transponder, a reader that captures information, an antenna that transmits information, and a computer which interprets or manipulates the information gathered. In the case of VeriChip, there was a requirement that each subscriber registers their personal details (and other relevant information they desire) on the GVS database. In 2004, all the transponders issued by VeriChip were passive but it is likely that active transponders will be issued in the future, despite the fact that they require on-board battery power to operate internal electronics. When an individual passes an associated scanner, information is read and sent to the computer via an antenna. Dependent on the application, a log may be retained or the implantee’s location updated a predefined number of times in a set period.



Geocoding with the intent of conducting geodemographic analysis for business applications is the process of linking intelligence data (e.g. customer counts) to longitude and latitude (x and y) co-ordinates based on a spatial reference (Harris et al., 2005). This reference may be a street address, a post code or state. Geographic coding is like placing a pin on a map for each record in a database thus linking each pin to the corresponding data (Kennedy, 2006; MapInfo, 1998). Geocoding has become synonymous with converting street addresses to longitude and latitude coordinate positions (Schiller & Voisard, 2004, p. 32). Geocoding methods include: the direct survey and property of boundaries, simple database queries, and other specialized geocoding options (Cowen, 1997). Motivations for geocoding for business applications may include for strategy, logistics support, operational support, marketing, and service (Keenan, 2005, ch. 1; Pick, 2005).

One can geocode against almost any level of detail whether it be linked to region, line or point objects. Almost all data has a geographic unit it belongs to, even data that seemingly looks unstructured at first glance. Commonly defined boundaries include: administrative, political, telecommunications, social and land-use. Data linked to line objects include: roads, rivers, contours, flight paths, teletraffic and transport routes, fiber links. Data linked to points include: landmarks, buildings/ dwellings, points of interest, civil services, and utility points. In human geography, analysts often use geocoding as a way to link spatial information with a new set of data they have generated or manipulated. This can then be used in the process of thematic mapping or districting.


The Process of Geocoding

On the process of geocoding, Cowen (1997) writes that a feature “must have a field which can be linked to a geographic base file with known geographic coordinates”. For instance, a database of information can be used to join the geographic coordinates of the base map with address details. The geocode table is that foreign table the end user wants to import into the GIS. It can take the form of a text file (comma separated value .csv, or .txt), spreadsheet or database. The user must specify which column is being used to geocode (if there are several keys defined or various location data). The boundary column is a column that allows the user to run a geocode on a partial set of data, for example, statistical local areas (SLA) within a given statistical division (SD). The search table is that table which is already native to the GIS. It consists of a base map that has associated data stored in a table. When geocoding the user also has the option to signify which regions, lines or objects have been successfully geocoded by displaying a designated symbol on the centroid of the object after the parse has been completed.

Geocoding can be done in a manual, interactive or automatic fashion. When a user is not confident about the quality of the data they are trying to attach to a spatial reference they will probably select the manual geocoding process. This requires the user to hand select an item from the search table to match the base map. This is a detailed process and while accurate is unfeasible for large batches of geocoding. Automatic geocoding occurs when the unique keys of each table are compared and only those records with an exact match are processed. Those records that remain ungeocoded remain unlinked. Interactive geocoding works to bring to the attention of the user, only those records that the software has not been able to automatically link. The user is then required to manually intervene to link each record in an external dataset and match it correctly to a record on the base map. Geocoding requires that at least one field in each table is defined as the unique key. In the base map, it then follows that the unique key must have a geographic context. Simple geocode results will often stipulate how many records were geocoded, how many remain ungeocoded and how many have been previously geocoded.


Geocoding Software

Geocoding is a fundamental feature of all geospatial applications. Most GIS software has built-in geocoding functionality and in addition some offer add-on specialized geocoding engines built for larger tasks. The add-on geocode modules allow for the conversion of the geocode files into a database format, then software records are parsed and geocoded. An output file is also generated indicating the geocode status of all records, and subsequently map points are created (MapInfo, 1998, p. 2). More advanced geocoding software can help users match individual credit information down to an individual in a dwelling. By locating individuals to their home, a bank can use a Marking Customer Information File (MCIF) to determine for instance, which branches are running efficiently and which may be closed (Reider, 2003, p. 44). Retailers or election campaigns can also target market direct mail based on the results of geocoding (Harris et al., 2005, pp. 63f).

Schiller and Voisard (2004, p. 164) believe that a geocoder service should support the following features: (a) given an address, the address-matching geocoding algorithm should be able to determine the x and y position, (b) handle more than one address in a single geocode request, (c) fuzzy address matching where an incomplete address is given and data returned after normalization, (d) count the number of exact matches as a subset of the total number of records parsed, (e) provide information on the quality of the result using a special pre-defined match code (e.g. where an exact match was not found, indicate whether the record has been locked to the centroid of a street or a post code or other). Table 1 shows the table structure of a file that is about to be geocoded and the field names of where the results post geocoding will be stored. Specialized geocode software like MI’s MapMarker Pro allow the user to save settings within configuration files, so as to ensure options have remain unaltered when future geocodes are performed.


Limitations of Geocoding

The process of geocoding is dependent on the number of records in each table. It takes longer to geocode databases that contain millions of records than smaller databases with only a couple of hundred records. Central processing speed, the amount of random access memory (RAM), and storage space are all important factors in geocoding large amounts of data. Processing can occur in batch mode or single file mode, and different batches compared based on a pre-defined configuration. Different geocoders have different thresholds for processing based on the way they work but a rule of thumb is not to geocode in excess of 1,000,000 records at a time. Commercial off-the-shelf data is also limited in how much assistance it can grant the user who tries to reconcile the “ungeocoded” records with actual physical locations. Newly erected dwellings and new estates are just some of the issues users encounter when dealing with movements of people and business locations.


Old, Dirty, Incomplete, Incorrect, Unstructured or Foreign Data

The success of geocoding all depends on how clean the data sets are. “It is not uncommon to have a geocoding hit rate of under 50% on your first attempt” MapInfo, 2002, p. 140). Consider variations in the way names, companies, dates, phone numbers are recorded and the resultant issues that would arise in a custom GIS (SearchSoftware 1997, pp. 7-8). The old adage, Garbage In/ Garbage Out (GIGO) could not be more appropriate than in the context of geocoding. Dirty data can have a large impact on the number of geocoded records and on the length of time the user spends debugging. This is true in particular of street-level data (see Table 2). Are records incomplete or incorrect? Have certain keywords been omitted or abbreviated in data entry? Is data free of spelling errors? What happens in the case of street corners, or post office boxes, or floors and suites, shopping malls, or roads that are so long they are differentiated by the words East or West, North or South? What happens if the same postcode has multiple streets or roads with the same name? Some conventions and standards do exist but they are hardly followed. The problem with geocoding is that the user is seldom in charge of the data they are given to geocode. The data has either come from a third party organization, a department who has kept records on an ad-hoc basis, a piece of telecommunications equipment whose database is almost impossible to decode, or a customer who is willing to divulge only so much of their intellectual property.

Even when the data is clean, how to join it to spatial references that were created before or after the data set is a complex problem on its own. Trying to match records from different vintages almost always affects the overall number of successful hits obtained (Sears, 2004). For example, census-defined boundaries like statistical local areas (SLAs) may change names from one census to the next or grow to encompass more collection districts (CDs) within their bounds. Any change in numerical values or alphabetical content in the unique key also means that a match does not occur. Consider post codes in the United Kingdom. Harris et al. (2005, p. 114) make the point that even post codes do not stay constant. And with that inconstancy, almost always comes changes to postal geography. Consider another scenario also, where the creation of industry-specific boundaries have been created for historical reasons, such as telecommunication exchanges. One might be able to geocode the number of telephone lines for each exchange, but as soon as one attempts to geocode post code-level information to exchanges a problem arises in an irregular spatial fit.

Many companies are now trying to tap into unstructured data sets as a form of rich intelligence. Mapping this data into a physical space is an admirable aim of any organization but is a long process. One does not always know what they are looking for when they start out, so one way to narrow the possibilities down is to consider the question of one’s digital map granularity. Data in a language other than English (particularly non-character based languages) may also present a stumbling block for users, despite the fact that the process of geocoding remains unchanged in this context. One can always revert to using ASCII or Unicode during the pre-geocode stage of data trimming, cleaning, concatenation and searching, and then reapply the appropriate font style post-geocode. A precautionary measure in this situation would be to hire a local planner who has the language skills to check the results of the analysis in the event that errors have crept through without notice. Anthes (2005, p. 61) goes one step further and calls the process of geocoding from external and internal sources “tricky” and emphasizes that “[t]here is a limit to what you can do with GIS technology and how much you want to trust maps” (Anthes, 2005, p. 61).

The success of geocoding accuracy is also about the type of application requirement one has- if it is a critical application it must be precise- but if it is not, then interpolating based on dwelling frontage or floor space, is an equally valid approach. For example, a user could begin by gathering the start and end coordinates of a street segment, identify the address ranges for that street, and then approximate a “fix” for a given address based on the range divided by the number of households (Schiller & Voisard, 2004, p. 60). Interpolation is a good approach for estimating demand in a given area, but it would not be useful for precision planning such as in the case of local councils who need to be able to overlay geocode map points with cadastre plots and perhaps other raster imagery.



Geocoding becomes extremely powerful when multiple sources of intelligence are brought together telling the story of a bigger picture (figure 4). In private enterprise internal intelligence is used alongside external intelligence (Michael, 2003, p. 496). In this instance, customer address detail records may be geocoded and used in tandem with government statistical information. By using this approach, one can make certain aggregate assumptions on household income, household size, and telecommunications needs.


From Postal Codes to Pinpointing Households

McCurley (2001, p. 223) writes that although the standardization of mail addresses is a fairly well studied problem, it is complicated by the fact that each country has its own set of rules. Each postal address is typically made up of several fields- the street number and street name, the post code ID, and the post code name with any number of variations on format and content. Compare for instance post codes in the United Kingdom, with post codes in the United States, China, Australia, and Greece. Even when a successful geocode has occurred it does not imply accuracy. University campuses in Australia are renowned for being spread over many kilometers, each with their own local mail facilities. Mapping data to a centroid of the campus cadastre is not good enough for most applications.

Enter the global street databases which are usually maintained through government bodies bestowed with the authority. In Australia, G-NAF is the Geocoded National Address File which brings together the resources of thirteen government agencies. G-NAF is considered the ultimate solution to a problem that has been plaguing large organizations like telecommunications operators who have spent in some cases hundreds of thousands of dollars on this problem. In the past, wrongly embedded data in the base search map subsequently meant that wrong data was being geocoded. The Postal Service in the United States distributes a similar product to G-NAF called Tiger/Zip+4 containing 35 million records, and the United Kingdom has the Ordinance Survey that has 25 million address points. In addition, the U.K. have a product called Code-Point that “gives coordinates for the 1.6 million different postal codes within England, Scotland, and Wales. Each postal code contains an average of fifteen adjoining addresses. The resolution of this data is 1 meter” (McCurley, 2001, p. 224). Much of the trend toward geocoding at the address level was borne out of the federal census (Brassel et al., 1977, p. 79). Data is also being resold in packages like BusinessMap Pro that show customer lead information by size and type of business based on the Standard Industry Classification (SIC) code. Data also typically found in Dun & Bradstreet MarketFind sources exists within a GIS. For example, BusinessTracker provides a list of more than 10 million businesses that you can search and display by company name, location, SIC code, size by sales volume or employees, or classification” (Hollander, 1998, p. 78). TeleAtlas and Navtek are two more companies that have a global focus on the collection of location information.


The Risks of Geocoding in Mission Critical Systems

In his book Geographic Information Science: mastering the legal issues, Cho (2005, pp. 356-358) masterfully discusses the importance of quality and liability. Schiller and Voisard (2004, p. 32) echo these sentiments- accuracy is critical to the quality of the results. Unfortunately the first indication that there are problems with geocoded information is when something goes wrong- either mail is sent to the wrong address, or even worse, a major accident takes place (Cho, 2005, pp. 357f). Embedded data is becoming more and more invisible to the user and error detection harder and harder. Take for example the 1999 bombing of the Chinese Embassy in Belgrade by United States military forces. The target had previously, and wrongly, been identified as the head office for the Yugoslav Federal Directorate for Supply and Procurement (FDSP). Instead, what actually stood in the target location was the Chinese Embassy. The bombs, all GPS-guided missiles reached their geographic target successfully, killing three Chinese journalists and injuring twenty embassy staff. In later press releases and formal statements of apology, the U.S. government admitted that the bombing was both an error and an accident. The positioning technology had functioned correctly but the knowledge systems supporting it had failed (Michael & Masters, 2006, p. 211).



Thematic mapping is that part of cartography that deals with mapping data on a particular theme (Carter & Icove, 1976, p. 164). The word thematic refers to the theme of the map, that value which is chosen for a given location, illustrated by a selected color or pattern (Reider 2003, p. 43). Thematic maps link a data element to an underlying geographic unit, such as a building location, major road, a cadastre parcel, census district, post code, region or province (see Table 3 and 4). Items that may be mapped in telecommunications range from an operator’s market share for data services, the number of potential customers by market segment by census district, and the types of assets owned by a bandwidth provider across regions. A selling point of most geographic information systems (GIS) software today is that they allow for fast manipulation of tables and maps, granting even novice users the ability to analyze and visualize their data using thematic mapping. The technique does more than just graphically show you your data however, it lets you see it mapped to a real world context, revealing patterns and trends that are almost impossible to detect in lists (MapInfo 2002, p. 209).

In 1977, Brassel et al. (pp. 84f) defined twelve types of thematic maps including: choropleth mapping, graduated circle maps, ring diagrams, graduated rectangle maps, frame diagrams, graduated segmented circle maps (pie charts), point symbol mapping, contour mapping, 3-D mapping, fence maps, dot maps, and cartograms. It is important to note, that just one year prior to that, Carter and Icove indicated there were only three main types: dot (e.g. density in terms of dots per area), choropleth (e.g. based on a fixed area unit of data collection such as a census tract) and isarithmic maps (e.g. smooth continuous surface maps such as elevation or population density). Advancements in the cathode rode tube (CRT), terminals, printers, and plotters have revolutionized what can be achieved with thematic maps. Users today have the capability to plot detailed ISO A0 formatted paper maps in a matter of minutes. Yet in the late 1970s, papers were being published about resolving screens into a “matrix of 25 rows by 80 columns with colored characters on colored backgrounds” and describing the limitations of terminals “in terms of the selection and brightness of the eight colors, the dimensions of the cells and the ability to mix colors in small cells” (Carter and Icove, 1976, p. 163).


Different Types of Thematic Maps

Malerba et al. (2001, p. 291) distinguish between thematic maps and composite maps, the former being concerned with one geographic feature and the latter with several separate thematic maps (one for each layer). Today’s GIS software allows for thematic maps to take the following form- district maps, dot density maps, graduated maps, grid maps and graphic maps such as pie charts and bar charts (figure 5). The MapInfo GIS, for example, allows for a variety of ranged value thematic maps including: equal count, equal ranges, natural break, quantiling, and standard deviation. Depending on the type of data being analyzed, or the desired resultant effect, the user will choose accordingly. It should be stated however, that choosing the wrong ranged value technique will result in a misleading visual result, even if a cartographic legend accompanies the map. Fotheringham et al. (2000, pp. 72-92) describe some of the varying techniques used in exploratory spatial data analysis and emphasize the need to produce maps that do not misrepresent reality. For the greater part it is about understanding your dataset well enough to know how it should be displayed as well. MapInfo (2002, p. 250) also describe the use of bivariate thematic mapping which “uses point or line objects to represent two thematic variables.” For example, a star can represent one variable, such as the number of mobile phone users, while a gold fill for the star represents their annual number of SMS messages.


Data Sources for Thematic Maps

Data for thematic maps can come from a) the same table, b) different tables such as in the case of bivariate maps, c) by joining different tables together based on a unique key, or d) by creating expressions which change the values in your table for a meaningful way. Sources of data for thematic maps can come from in-house business intelligence systems such as customer datawarehouses, or external intelligence such as free public data, or even commercially available data. The better the source of information in terms of vintage, data detail at the smallest object or region level, the better the value of the thematic maps. Network planners for instance find it useful to create thematic maps at varying layers of detail- by region, census tract, administrative boundary, clutter type, road, cadastre, building, longitude/latitude coordinate. Brassel et al. (1977, p. 83) make the point that users are restricted in creating particular types of thematic maps based on the data they have in their dataset. “For instance, specific thematic data (aggregated-disaggregated) require particular spatial-referencing information, whereas geometric data (centroids-segments-area outlines) govern the kinds of thematic display possible.”


When to Use Thematic Maps

Thematic maps are ideal for executive strategic level presentations when the fine details are not a requirement, but an overall big picture view needs to be gained. Many planners and designers rely on the results of thematic mapping for timely decision making (table 5). For example, thematic mapping can assist with the decision of where to place a base station tower- in terms of its proximity to human (e.g. schools) and natural features (e.g. crown land). Sales people also require thematic maps for identifying new customer leads and for market segmentation purposes. Telecommunications companies also have a large base of assets in terms of equipment and other pieces of infrastructure that age over time. Asset managers are beginning to rely more and more on thematic mapping, especially for managing customer installations for contractual or compliance purposes. In an informal survey of a global telecommunications switching vendor in 2001, it was identified that out of the 80,000 employees hired by the company, about 2,000 employees were “developer” users of geographic information systems software, and even more were recipients of the information produced by the GIS.


Steps in Producing a Thematic Map

Some software programs offer automated features that allow users to create thematic maps using wizard-like procedures. Initial steps require the user to choose from a template of options, including size, color, type, and effect. The next step requires the user to indicate which table and attribute is to be used for the thematic mapping. At this stage the user has the option of creating an “on the fly” join between a base map and an imported table or using the expression function to do some limited additional calculations on specific data. The user may also choose whether to ignore null values or blank fields from the overall dataset, especially if what they are trying to limit options to a few choices and there are too many zero values. The final step requires the user to check whether the settings they chose in step one are still suitable for the display of their map. For example, if the symbol size chosen in the template is too big or tool small, this can be changed at this final stage. The legend also allows for titles, subtitles and custom labels to be recorded and to determine the label ordering in ascending or descending order. Some GIS software even grants the user the option to save templates that they have custom created. For an overview of how to map geodemographic information with GIS with commercial examples, see Harris et al. (2005, pp. 88-102).



Consider the possibility of a Google Street View ‘in motion’. Imagine not only being able to view your house or that of your neighbor’s but with the potential of subscribing to a service that let’s you ‘watch’ breadcrumbs of people’s behavior. While the ability to geocode large numbers of records has been around a long time, the telecommunications infrastructure to support real-time location based services is only just beginning to sprout. The problem with too much data, e.g. location information being captured every three seconds, is that the margin for error is great dependent on the measuring instrument in question. Not only is there the possibility of errors creeping into geocoded data due to the data matching principles employed but also we cannot be certain that because an individual is located somewhere during the day, that they are/are not engaged in a particular act. For example, my location profile may show that I am at the university library but in no way does that mean that I am actually studying or researching.

The problem with constant location tracking is that commercial entities and the government (e.g. law enforcement agencies) will inevitably use the advanced capabilities to generate minute by minute thematic reports either for the conduct of social sorting (Lyon, 2004) or for evidence in a court of law. The danger with the creation of thematic maps based on pre-defined categories is that assumptions about a given individual will be based on data pertaining to speed, distance, time and altitude. Even worse still will be the co-location of persons in a questionable zone of activity that may have absolutely nothing in common and never met previously, adding new meaning to words like “suspect” or “alleged” or “possibility”. The scenarios in this context are endless. Today the mobile phone has become an extension of the person, just like one’s DNA or ID number. Increasingly, people are taking notice of the potential for surveillance and some are even opting out altogether, preferring to have their mobile phones permanently switched off, save for emergencies (Renegar & Michael, 2009).



ADSX. (2003). Implantable Personal Verification Systems. Applied Digital Solutions   Retrieved 15 April 2004, from

Anthes, G. H. (2005). Beyond ZIP codes. Computerworld. 39(38), 56, 58, 61.

Asher, J. (30 January 2009). Humans 'will be implanted with microchips'.   Retrieved 1 February 2009, from

AT&T. (2003). Feature and Services User Guide. AT&T Wireless   Retrieved 15 April 2004, from

Berkin, M., Clarke, G., Clarke, M. & Wilson, A. (1996). Intelligent GIS: location decisions and strategic planning. Cambridge: GeoInformation International.

Brassel, K. E., Utano, J. J. & Hanson, P. O. (1977). The Buffalo crime mapping system: a design strategy for the display and analysis of spatially referenced crime data. ACM Siggraph, Proceedings of the Fourth Annual Conference on Computer Graphics and Interactive Techniques, 78-85.

Brimicombe A.J. & Li. C. (2007). Location-Based Services and Geo-Information Engineering. Chichester: Wiley.

Bureau of the Census. (1970). Census use study. The Dime Geocoding System, 4, 1-38.

Câmara, A.S. & Dia, A.E. (2004). Location-based services for WAP phone users in a shopping centre in J. Stillwell & G. Clarke (eds) Applied GIS and Spatial Analysis, West Sussex: John Wiley & Sons, 55-70.

Carter, J. R. & Icove, D. J. (1976). The application of the Intercolor 8000 terminal to thematic cartography. ACM Siggraph Computer Graphics, Proceedings of the Third Annual Conference on Computer Graphics and Interactive Techniques SIGGRAPH ’76, 10(2) 163-166.

Cho, G. (2005). Geographic Information Science: mastering the legal issues. Australia: John Wiley & Sons.

Cowen, D.J. (1997). Unit 016 - Discrete Georeferencing. URL:, Australia.

Dawson, M., Winterbottom, J. & Thomson, M. (2006). IP Location: Geographic Location Measurement, Delivery and Conveyance. New York: McGraw-Hill Communications.

Digital Angel. (2009). Digital Angel.   Retrieved 1 February 2009, from

Dransch, D. (2005). Activity and context- a conceptual framework for mobile geoservices in L. Meng, A. Zipf & T. Reichenbacher (eds) Map-based Mobile Services: Theories, Methods and Implementations, Munich: Springer, 31-42.

Edwardes, A., Burghardt, D. & Weibel, R. (2005). Portrayal and generalisation of point maps for mobile information services in L. Meng, A. Zipf & T. Reichenbacher (eds) Map-based Mobile Services: Theories, Methods and Implementations, Munich: Springer, 11-30.

Elliot, G. & Philips, N. (2004). Mobile Commerce and Wireless Computing Systems. England: Addison Wesley.

Foody, G. M. (2000). Image classification with a neural networks in P.M. Atkinson & N.J. Tate Advances in Remote Sensing and GIS Analysis, New York: John Wiley & Sons, 17-38.

Fotheringham, A.S., Brunsdon, C. & Charlton, M. (2000). Quantitative Geography: Perspectives on Spatial Data Analysis. London: Sage Publications.

Gartner, G. & Uhlirz, S. (2005). Cartographic location-based services in L. Meng, A. Zipf & T. Reichenbacher (eds) Map-based Mobile Services: Theories, Methods and Implementations, Munich: Springer, 159-171.

Harris, R., Sleight, P. & Webber, R. (2005). Geodemographics, GIS and neighbourhood targeting. West Sussex: John Wiley and Sons.

Hollander, G. (1998). BusinessMap Pro turns information into clear images. InfoWorld. 20 52(1), 78.

Jensen, C. S. (2004). Database aspects of location-based services in J. Schiller & A. Voisard Location-Based Services. Amsterdam: Elsevier, 115-147.

Jones, K. & Hernandez, T. (2004). Retail application of spatial modelling in J. Stillwell & G. Clarke (eds) Applied GIS and Spatial Analysis. West Sussex: John Wiley and Sons, pp. 11-53.

Karimi, H.A. (2004). Telegeoinformatics: current trends and future directions in H.A. Karimi & A. Hammad (eds) Telegeoinformatics: Location-Based Computing and Services, London: CRC Press, 5-25.

Keenan, P. (2005). Concepts and theories of GIS in business in J.B. Pick (ed) Geographic Information Systems in Business. Hershey: Idea Group Publishing, pp. 1-19.

Kennedy, M. (2006). Introducing Geographic Information Systems with ArcGIS: featuring GIS software from Environmental Systems Research Institute. New Jersey: John Wiley.

Küpper, A. (2005). Location-Based Services: fundamentals and operation. England: John Wiley & Sons.

Laserna, R., Landis, J. & Strategic Mapping. (1989). Desktop Mapping for Planning and Strategic Decision-Making. New York: Strategic Mapping.

Lopez, X. R. (2004). Location-based services in H.A. Karimi & A. Hammad (eds) Telegeoinformatics: Location-Based Computing and Services, London: CRC Press, 171-188.

Lyon, D. (2004). Surveillance Technologies: Trends and Social Implications. In OECD (Ed.), The Security Economy (pp. 127-148): OECD.

Malerba, D., Esposito, F., Lanza, A. & Lisi, F.A. (2001). Machine learning for information extraction from topographic maps in H. Miller & J. Han (eds), Geographic Data Mining and Knowledge Discovery, London: Taylor and Frances, 291-314.

MapInfo. (1998). GeoLoc. Australia: MapInfo Australia.

MapInfo. (2002). MapInfo Professional: User’s Guide v7.0. New York: MapInfo Corporation.

MapInfo. (2004). MapInfo MapMarker UK 2.0 takes address cleaning & geocoding to new levels of accuracy and consistency. M2 Presswire. 16 November 2004, 1.

McCurley, K. S. (2001). Geospatial mapping and navigation on the web. ACM, WWW10, 221-229.

McDonough, B. (17 April 2002). AT&T Wireless Pushes mLife with mMode. CIO Today   Retrieved 6 April 2004, from

Mendez-Wilson, D. (2001). Plotting the location points. Wireless Week. 7(7), 28.

Meng, L. & Reichenbacher, T. (2005). Map-based mobile services in L. Meng, A. Zipf & T. Reichenbacher (eds) Map-based Mobile Services: Theories, Methods and Implementations, Munich: Springer, 1-10.

Michael, K. & Masters, A. (2006). The advancement of positioning technologies in defense intelligence in H. Abbass & D. Essam (eds), Applications of Information Systems to Homeland Security and Defense, IDG Press, 211.

Michael, K. (2003). The importance of conducting geodemographic market analysis on coastal areas: a pilot study using Kiama Council, in C. D. Woodroffe & R. A. Furness (eds.), Coastal GIS 2003: an integrated approach to Australian coastal issues, Wollongong: Centre for Maritime Policy, 481-496.

Michael, K. (2004). Location-based services: a vehicle for IT&T convergence in K. Cheng et al. (eds), Advances in e-Engineering and Digital Enterprise Technology, London: Professional Engineering Publishing, 467-477.

Paavilainen, J. (2001). Mobile Business Strategies: Understanding the Technologies and Opportunities. London: Wireless Press.

Pick, J.B. (2005). Geographic Information Systems in Business. Hershey: Idea Group Publishing.

PSMA (2006). G-NAF. URL:, Australia.

Reider, S. (2003). Map your market with GIS. ABA Bank Marketing. (35)7, 42-46.

Renegar, B.D. & Michael, K. (2009), Privacy-value-control harmonization for RFID adoption in retail, IBM Systems Journal, 48(1), in press.

Robinson, A.H. (1982). Early Thematic Mapping in the History of Cartography. Chicago: University Of Chicago Press.

RFID News. (13 November 2008). Digital Angel sells stake in VeriChip. RFID News   Retrieved 1 February 2009, from

Schiller, J. & Voisard, A. (2004). Location-Based Services. Amsterdam: Elsevier.

SearchSoftware (1997). The Math, Myth and Magic of Name Search and Matching: see how to improve your business applications. Connecticut: SearchSoftware America.

Sears, B. (2004). Geocoding challenges: why accuracy matters. Directions Magazine. URL:, 20 April.

Sensors. (24 May 2007). Digital Angel, Verichip to Design Implantable Microchip. Sensors Magazine   Retrieved 1 February 2009, from

Shiode, N., Li, C., Batty, M. et al. (2004). The impact and penetration of location-based services in H.A. Karimi & A. Hammad (eds) Telegeoinformatics: Location-Based Computing and Services, London: CRC Press, 349-366.

Stolz, P. (2000). Voice a ‘killer app’ for E911. Wireless Week. 6(50), 73.

Taylor, G. & Blewitt, G. (2006). Intelligent Positioning: GIS-GPS Unification. West Sussex: John Wiley & Sons.

Timpf, S. (2006). Wayfinding with mobile devices: decision support for the mobile citizen in S. Rana, & J. Sharma (eds) Frontiers of Geographic Information Technology. Heidelberg: Springer, 209-228.

Wherify. (2003). Wherify Wireless GPS Locator For Kids. Wherify Wireless   Retrieved 15 April 2004, from

Wherify Wireless. (2004). Frequently Asked Questions.   Retrieved 15 April 2004, from


Key Terms & Definitions

3D Maps: Using a process of rendering, three dimensional shapes are projected in two dimensions in computer graphics. 3D thematic maps are becoming increasingly popular with large companies who are even utilising “fly-bys” over terrain.

Automatic geocoding: Automatic geocoding occurs when the unique keys of each table are compared and only those records with an exact match are processed.

Bivariate thematic maps: Allows comparisons to be made between records and to draw conclusions about variables in the entire dataset.

Cadastral Map: Maps defined by local councils indicating land ownership of a given area.

Context-aware systems: These are systems that support location-based services that rely on the dissemination of contextual information such as temperature.

Data Cleaning: Data sets to be used for geocoding may contain embedded characters that affect the geocoding process, even if they cannot be seen by the user. Cleaning prepares data for geocoding by applying standards, trimming, concatenation or separation of some content in the table.

District maps: Displays common data elements in the same symbol, line or region colour. For example, shading for regions that are managed by the same salesperson is color-coded identically.

Dot density maps: Displays the data in your table as dots on a map. Each dot has a corresponding value. When the number of dots is multiplied with that value, the total value for that region is obtained. This is usually used to show the number of consumers or employees in a given area.

Exact Match: Occurs in geocoding when the source and target information are exactly the same- word for word, letter to letter.

Geocode: The process of assigning X and Y coordinates to records in a table so that the records can be shown on a map.

Geocontent Provider: A geocontent provider is a type of content provider who specialises in the management, storage and distribution of spatial content at multiple levels of detail- including topographic information, points of interest, and postal information. Content providers need to ensure their databases are up-to-date especially when distributing information to mission-critical systems in the government sector.

Geodemographics: Geodemographics is the analysis of people by where they live and work.

Geoparsing: Geoparsing is the recognition of geographic context, whereas geocoding is the process of assigning geographic coordinates (McCurley, 2001, p. 222).

Geosorting: Once data has been geocoded then the GIS can perform a number of spatial queries allowing users to conduct geosorting (Birkin et al., 1996, p. 31).

Graduated symbol maps: Displays a symbol (such as a person, or dollar sign) for each record in the table. The size of each symbol is directly proportional to the data value.

Interactive geocoding: When you geocode interactively you are not changing the data record, you are redirecting the software to look for different information.

Interpolation: A simple method of geocoding which constructs a set of new data points from a discrete set of known data points.

Joining Tables: When data is stored in two separate tables and required for the one thematic map, a join must occur to bring the data together. The join feature adds a temporary column of information to the data set.

LBS value-chain/network: Various stakeholders (including geographic content providers, handset manufacturers and network access providers) work together to offer an end-to-end LBS solution. The value chain of LBS is considered complex and meshed.

Localisation: Localization (noun), locate (verb), is the determination of the locality (position) of an object or subject.

Location-based services: Typical LBS consumer applications include roadside assistance, who is nearest, where is, and personal navigation. LBS business applications differ in their focus and many are linked to core business challenges such as optimising supply chain management (SCM) and enhancing customer relationship management (CRM). Some of the more prominent LBS business applications include: fleet management (incorporating vehicle navigation), property asset tracking (via air, ship and road) and field service personnel management (i.e. people monitoring).

Longitude and Latitude Coordinates. A coordinate system for representing geographic objects on a map.

Parsing: Parsing is the recognition of geographic context.

Pervasive Computing: Pervasive computing is a term that is being used to describe future LBS applications in 4G mobile networks. These services will be rich in location-based information, in context awareness, and will depend on intelligent agents and intelligent devices with unobtrusive access, anywhere, anytime.

Pull services: A type of on-the-spot location-based service. These are generally considered easier to implement than LBS push applications because a polling request only occurs when the services are needed, i.e. the network does not have to track the user continuously. For example, locate your friends or family, emergency services, yellow pages inquiry services, Go2Systems.

Push services: A type of location-based service based on real-time monitoring. For example, LBS applications such as retail alerts based on the position of the subscriber, cell advertising, traffic reporting.

Ranged maps: Display data values across different types of objects. For example, the annual telecommunications household expenditure by collection district.

Telematics: Global Positioning System (GPS) chipsets are being increasingly installed in computers and mobile communication technology to allow them to be tracked and monitored. The word telematics is synonymous with vehicles and intelligent road systems, but is beginning to be used in the context of people tracking.

Thematic templates: Thematic templates allow users to reuse the same values and settings (e.g. colors) when they create new thematic maps. They can be considered thematic defaults which users can alter at any time.

Thematic values: Any value (string, number or other) that is used in a thematic map to represent something meaningful.

Theme legends: Legends are integral to thematic maps. They contain textual descriptions of what is being represented in the map including headings, names of attribute columns, symbols used, and other optional information such as counts and the number of objects. The metadata aspect of legends is paramount.

Ungeocoding: Ungeocoding is the process of removing objects that have been attached to data records either because the relationship is no longer valid, or an error in attachment has occurred. Users have the option of ungeocoding the whole table or selected records (MapInfo 2002, p. 162).

Workspace: A workspace saves an instance of a user’s “work” at a given point in time. It means that any previously opened tables and windows can be automatically reopened at the same location, with one menu selection only.