ABSTRACT
		Only 1.5 billion of the estimated 6 billion land parcels world-wide 
		have land rights formally registered in land administration systems. 
		Many of the 1.1 billion slum dwellers and further billions living under 
		social tenure systems wake up every morning to the threat of eviction. 
		These people are the poor and most vulnerable and have reduced forms of 
		security of tenure; they are trapped in poverty. Increasing global 
		population and the rush to urbanisation is only going to turn this gap 
		into a chasm.
		
		This paper explores one potential solution to the security of tenure gap 
		through establishing a partnership between land professionals and 
		citizens that would encourage and support citizens to directly capture 
		and maintain information about their land rights. The paper presents a 
		vision of how this might be implemented and investigates how the risks 
		associated with this collaborative approach could be managed.
		1. INTRODUCTION 
		Land Administration Systems (LAS) provide the formal governance 
		structures within a nation that define and protect rights in land, 
		including non-formal or customary institutions. Their benefits range 
		from guarantee of ownership and security of tenure through support for 
		environmental monitoring to improved urban planning, infrastructure 
		development and property tax collection. Successful land markets depend 
		on them.
		
		Despite this pivotal support of economic development, effective and 
		comprehensive LAS exist in only 50 mostly western countries and only 25 
		percent of the world’s estimated 6 billion land parcels are formally 
		registered in LAS. This leaves a large section of the world’s population 
		with reduced levels of security of tenure, trapping many in poverty. 
		Missing and dysfunctional LAS can precipitate problems such as conflicts 
		over ownership, land grabs, environmental degradation, reduced food 
		security and social unrest. Rapid global urbanisation is exacerbating 
		these discrepancies.
		
		This security of tenure gap cannot be quickly filled using the current 
		model for registering properties that is dominated by land 
		professionals. There are simply not enough land professionals 
		world-wide, even with access to new technologies. To quickly reduce this 
		inequality we need new, innovative and scalable approaches to solve this 
		fundamental problem. This is one of our fundamental global challenges.
		
		This paper explores one potential solution to the security of tenure 
		gap: ‘crowdsourcing’. Crowdsourcing uses the Internet and on-line tools 
		to get work done by obtaining input and stimulating action from citizen 
		volunteers (www.crowdsourcing.org). It is currently used to support 
		scientific evidence gathering and record events in disaster management, 
		as witnessed in the recent Haiti and Libya crises, for example. These 
		applications are emerging because society is increasingly spatially 
		enabled. Establishing such a partnership between land professionals and 
		citizens would encourage and support citizens to involve themselves in 
		directly capturing and maintaining information about their land rights.
		
		Although citizens could use many devices to capture their land rights 
		information, this paper advocates the use of mobile phone technology. 
		Due to high ownership levels (5 billion licenses world-wide) and 
		widespread geographic coverage (90 percent of the world’s population can 
		obtain a signal), especially in developing countries, mobile phones are 
		an excellent channel for obtaining crowdsourced land administration 
		information. Frugal innovation is making them affordable for all, 
		especially in developing countries where a new generation of information 
		services in health and agriculture, for example, is turning the mobile 
		phone into a global development tool. 
		
		Mobile phones are progressively integrating satellite positioning, 
		digital cameras and video capabilities. They provide citizens with the 
		opportunity to directly participate in the full range of land 
		administration processes from videoing property boundaries to secure 
		payment of land administration fees using ‘mobile’ banking. But even 
		today’s simpler phones offer opportunities to participate in 
		crowdsourcing. 
		
		A key challenge in this innovative approach is how to ensure 
		authenticity of the crowdsourced land rights information. The paper 
		explores applicability of the approaches adopted by wikis (a piece of 
		server software that allows users to freely create and edit Web page 
		content using any Web browser), e-commerce and other mobile information 
		services and recommends the initial use of trusted intermediaries within 
		communities, who have been trained and have worked with local land 
		professionals. This approach has the potential to provide a good level 
		of authenticity and trust in the crowdsourced information and would 
		allow a significant network of ‘experts’ to be built across communities. 
		To optimise the scarce resources, these intermediaries could be involved 
		in a range of other information services, such as health, water 
		management and agriculture. 
		2. ARE CURRENT LAND ADMINISTRATION SYSTEMS DELIVERING THE EXPECTED 
		BENEFITS? 
		Despite the clear link between effective LAS and efficient land 
		markets (Al- Omari, 2011), sustainable development and the other 
		benefits, their current adoption and effective implementation are 
		limited to about 50 and found mainly in western countries and in 
		countries in transition in central Asia (Enemark et al, 2010). A number 
		of factors limit their scope of implementation:
		
			- Costs are significant and national solutions can take from five 
			to over 20 years to implement.
- Overly complex procedures lead to high service delivery costs 
			and end user charges, excluding the poor and the vulnerable.
- Lack of a supporting land policy framework ensures that the LAS 
			do not deliver against the main drivers of land tenure, land markets 
			and socially desirable land use.
- Insufficient support for social and customary tenure systems 
			excludes large proportions of the  population.
- Lack of transparency encourages corruption in the land sector, 
			lowering participation through lack of trust.
- Communication channels to customers are either office or 
			Internet based and lead to geographic  discrimination or exclusion 
			through the ‘digital divide’.
- A mortgage requires a bank account and credit rating, which is 
			difficult for the poor and those remote from financial services to 
			obtain.
- Cadastral surveys using professional surveyors are normally 
			mandatory and generate higher fee rates, e.g. in the USA a typical 
			residential land parcel costs $300 -$1,000  
			(http://www.costhelper.com/cost/home-garden/land-surveyor.html) to 
			survey depending on local rates and the size and type of parcel.
It is estimated that there are around 6 billion land parcels or 
		ownership units world-wide. 4.5 billion parcels are not formally 
		registered and of these 1.1 billion people live in the squalor of slums. 
		With urbanisation predicted to increase from the current 50% to 60% in 
		2030 and a further 1 billion being added to the world’s population in 
		this timeframe, the security of tenure gap will become a chasm. This 
		will be impossible to fill in the foreseeable future using the currently 
		available land administration capacity. The International Federation of 
		Surveyors (FIG) currently represents 350,000 land professionals 
		world-wide. The current LAS paradigm cannot be scaled up quickly enough 
		to meet the demand.
		
		The lack of effective, affordable and scalable LAS solutions conspires 
		to limit access to land administration services by large sections of 
		society, especially the most vulnerable, leaving them trapped in 
		poverty. There is a pressing need to radically rethink LAS: simplify 
		procedures, reduce the cost of transactions, and open new channels for 
		participation. Crowdsourcing through ubiquitous mobile phones, for 
		example, offers the opportunity for land professionals to form a 
		partnership with citizens to create a far-reaching new collaborative 
		model and generate a set of LAS services that will reach the world’s 
		poor. The rest of this paper explores how citizens can be empowered to 
		support the delivery of LAS services through crowdsourcing.
		3. A NEW CITIZEN COLLABORATION MODEL FOR LAND ADMINISTRATION
		This section provided a vision of how citizens armed with mobile 
		phones, with the help of land professionals, could effectively capture 
		and manage their land rights.
		3.1 The Increasingly Pervasive Mobile Phone  
		Although citizens can provide their crowdsourced data through a 
		number of traditional channels, including paper, mobile phones are 
		progressively proving to be the device of choice. Mobile phones have 
		made a bigger difference to the lives of more people, more quickly, than 
		any previous communications technology. They have spread the fastest and 
		proved the easiest and cheapest to adopt. In the 10 years before 2009, 
		mobile phone penetration rose from 12 percent of the global population 
		to nearly 76 percent. It is estimated that around 5 billion people 
		currently have mobile phones and 6 billion will have them in 2013 
		(http://www.itu.int/ITU-D/ict/statistics/) 
		
		Recently the fastest growth has been in developing countries, which had 
		73 percent of the world’s mobile phones in 2010, according to estimates 
		from the International Telecommunications Union 
		(http://www.itu.int/ITU-D/ict/statistics/). In 1998, there were fewer 
		than four million mobiles on the African continent. Today, there are 
		more than 500 million. In Uganda alone, 10 million people, or about 30 
		percent of the population, own a mobile phone, and that number is 
		growing rapidly every year. For Ugandans, these ubiquitous devices are 
		more than just a handy way of communicating: they are a way of life 
		(Fox, 2011). Not all phones in the developing world are in individual 
		use, but are actually used as a communal asset of the household or 
		village.
		
		Due to their high ownership levels and widespread geographic coverage, 
		especially in developing countries, mobile phones are therefore an 
		excellent channel for obtaining crowdsourced land administration 
		information. But are they affordable and do they have the necessary 
		functionality?
		3.2 The rise of smart phones and tablets
		Telecommunications has developed exponentially. Phones have changed: 
		there is a big shift from holding a phone to your ear to holding it in 
		your hand. Smart phones have emerged that are able to browse the web, 
		send and receive email, and run applications - as well as storing 
		contacts and calendars, sending text messages and (occasionally) making 
		phone calls. See figure 1 for the range of Cyborg (an organism that has 
		enhanced capabilities due to technology) functionality provided by smart 
		phones. Smart phones represented 24 percent of all mobiles sold 
		worldwide in the first quarter 2011 – up from 15 percent a year before. 
		The tipping point when they make up 50% may only be a year or so away. 
		Although smart phones may cost around US$600 today, volume of sales and 
		frugal innovation will drive the cost down to an estimated US$75 in 
		2015. A US$100 smartphone has already arrived on the streets of Nairobi. 
		Before the end of the decade, every phone sold will be what we'd now 
		call a smartphone and cost US$25 (Arthur, 2011). 
		
		Although smart phones have combined an array of technologies onto the 
		mobile phone platform to significantly increase its functionality and 
		its applicability in a wide range of new applications, regular mobile 
		phones can still be used to support information services and gather 
		crowdsourced information, through text messaging services (SMS) for 
		example.
		
		The emergence of tablets is also providing an opportunity for 
		effectively supporting crowdsourced information, especially graphical 
		information. This technology will play a significant role in the future 
		of crowdsourcing.
		
		
		Figure 1: Smart Mobile Phone 
		Cyborg Functionality
		3.3 Vision of an effective crowdsourced Land Administration 
		solution
		This increase in functionality of the mobile phone, its migration to 
		lower cost devices through frugal innovation, its increasing 
		pervasiveness across developing countries and its connection to Internet 
		and information services is opening up significant opportunities for its 
		use in delivering more effective and accessible land administration 
		services. The possibilities are explored below:
		
		Accessing Customer Information Services - A whole new generation of 
		innovative information services, such as agricultural and health, are 
		being provided to users of mobile phones in developing countries. A good 
		example is the use of mobile phones to record and transfer water quality 
		or water source inspection data from the field to a central database 
		where water sector professionals can then view the data collected and 
		identify hazardous water sources
		(www.bristol.ac.uk/aquatest/about-project/workplan/ma6/). A similar set 
		of land administration services for users could provide explanations of 
		procedures, electronic forms for completion, standard applications and 
		best practice for land registration and cadastre, for example. This 
		remote guidance and support will be essential when there is more 
		significant citizen participation in land administration services and 
		could be provided by tiers of citizen intermediaries with guidance by 
		Land Professionals.
		
		Recording Land Rights - The mobile phone will allow citizens to directly 
		record the boundaries of their land rights. This can be achieved in 
		several ways:
		
			- Marked up paper maps digitally photographed with the phone.
- A textual description of the boundaries recorded on the phone.
- A verbal description recorded on the phone.
- Geotagged digital photographs of the land parcel recorded on the 
			phone.
- A video and commentary recorded on the phone – this could 
			include contributions from neighbours as a form of verification 
			(mobile phone numbers of neighbours could be provided).
- The positions of the boundary points identified and recorded on 
			imagery using products such as Google Maps and Bing, for example.
- The co-ordinates of the boundary points recorded directly using 
			the GNSS capability of the phone.
 
In all cases the authenticity of the captured information would be 
		enhance by passively recording the network timestamp at time of capture. 
		This information is not something that most (99.999%) of users can 
		tamper with.
		
		The results of this crowdsourced or self-service information could then 
		be submitted electronically to either the land registration and 
		cadastral authority or open data initiative for registration. Although 
		there are limitations in the quality and authenticity of the ownership 
		rights information provided, it could form the starting point in the 
		continuum of rights (UN-HABITAT, 2008) being proposed by UN-HABITAT. 
		This recognises that rights to land and resources can have many 
		different forms and levels. 
		
		To increase the authenticity and quality of the registration 
		application, the concept of the ‘Community Knowledge Worker’ created by 
		the Grameen Foundation (Donovan, 2011) could be adopted. The ‘Community 
		Knowledge Workers’ are trained members of communities supporting 
		agricultural and health information services who act as trusted 
		information intermediaries to those who have limited skills and access 
		to information. A similar model could be used for crowdsourced land 
		administration services to record or check ownership rights prior to 
		their submission. In fact, the ‘Community Knowledge Workers’ model could 
		be extended to also support land administration information services. 
		This model is similar to the administrative roles of the Patwari in 
		India and the Lurah in Indonesia.
		
		This engagement of local communities is also being highlighted as a key 
		success factor by crisis mapping projects. They realise that without 
		community buy-in, the valuable crisis mapping tools will not be used. 
		Communities must be engaged at all stages of the project and technical 
		design to ensure that crisis mapping efforts are in line with local 
		incentives and capacities. For example, this community led approach 
		brought fourteen organisations into a network in Liberia contributing 
		data to a multi-layered map that served as a central nervous system for 
		early warning signs of conflict in the run up to the national elections 
		in 2011 (Heinzelman et al, 2010).
		
		When the captured land rights are submitted to the property register 
		(see section 4 for a discussion on an alternative shadow property 
		register based on an open data initiative) a variety of quality checks 
		could be applied to the submitted information, including: random checks 
		in the field; comparisons with other applications submitted in the same 
		proximity; checks on ownership of the mobile phone; review evidence for 
		the location of its owner through the log showing that the phone is 
		frequently used within a location; network time stamping of captured 
		information; and contact the client and their neighbours on their mobile 
		phones to ask for clarification. Further details of approaches to 
		managing the authenticity risk are contained in section 5 ‘Managing the 
		Risks’.
		
		Obtaining Title - The submission of an application for registration 
		usually involves the payment of a fee. This is normally paid as cash 
		over the counter or a financial transaction through a bank or post 
		office. However, in the context of mobile phones, the payment could be 
		made by the client through ‘mobile banking’ on the mobile phone.
		
		Mobile phones are currently being used to manage identification 
		information. In Finland chip ID cards for government employees are being 
		adopted throughout Finnish central government. It is therefore feasible 
		that encrypted forms of land title could be incorporated into clients’ 
		mobile phones and used as proof of ownership.
		
		Accessing Land Information - Effective LAS are supported by Land 
		Information Systems. These are initially developed to support the 
		internal operations of the land registration and cadastral authority. 
		However, the next development stage is to make them outward facing and 
		accessible by customers either by Extranet or Internet. However, with 
		mobile phones directly supporting Internet access, these information 
		services can now be accessed by mobile phones. This new channel, which 
		will be the only access to the Internet for many countries, creates much 
		more accessibility for the citizen, bringing land administration 
		services to a wider range of society, many of whom are currently 
		excluded.
		
		Paying Mortgage Instalments - Securing a mortgage normally requires the 
		property owner to have a bank account to support the mortgage payments 
		transactions. However, the mobile phone offers opportunities to provide 
		secure payment of land administration fees with the increasing use of 
		‘Mobile Banking,’ simplifying the procedures and again potentially 
		opening up the means of wider property ownership.
		4. IMPACT OF NEW CITIZEN COLLABORATION MODEL ON THE EXISTING LAND
    ADMINISTRATION SECTOR
		The introduction of this new LAS model will likely be perceived by 
		most land professionals working in the land administration sector as 
		radical and by some as a serious threat. However, the current generation 
		of mobile phones and other devices are increasing the potential range of 
		participants in land administration. We are seeing the rise of the 
		‘proamateur’, somewhere between the professional and the amateur, caused 
		by this easy to use and accessible technology. Disruptive technology has 
		caused professional realignments in the past: total stations allowed 
		surveying technicians to perform more tasks, more accurately than 
		before. Crowdsourcing by ‘proamateurs’ is not a risk to land 
		professionals, but allows a wider range of participants to be involved 
		in land administration and more quickly address and solve our global 
		challenges.
		
		Land professionals’ attitudes towards this new model will determine how 
		land administration is shaped in the future. Here are two scenarios of 
		the potential impact of the new model on the land administration sector.
		
		Rejection by Land Professionals: Shadow Property Register - In countries 
		where there is little citizen trust in poorly performing or corrupt land 
		administration services provided by the government, an alternative 
		property register may be created through crowdsourcing. This ‘shadow’ 
		property register would be similar to the OpenStreetMap crowdsourced 
		model that has successfully provided an alternative source of mapping 
		for many countries. An ‘OpenCadastralMap’ (Laarakkar and de Vries, 2010) 
		or ‘OpenLandOwnership’ open data initiative would emerge. Despite not 
		having the usual endorsement and guarantee from government, its 
		legitimacy may progress over time as quality and trust evolve. It may 
		even be embraced by the informal market as a trusted repository to 
		support transactions more affordably and effectively than the formal 
		property register. The real test will be if financial services use it to 
		judge risk in the mortgage market. Ultimately, it may either replace the 
		government land administration service, reinforcing the informal land 
		market, or be adopted by government once it has reached a critical mass 
		and quality.
		
		Acceptance by Land Professionals: Supplement to the Formal Property 
		Register - Other countries may embrace this new model as an opportunity 
		to accelerate the number of properties being registered across the 
		country and support a much more inclusive solution to land 
		administration. If land professionals work in partnership with citizens 
		and communities and grow a network of trusted citizens to record and 
		register land rights then this source of land information could be 
		managed directly by the formal property registers. Initially these 
		crowdsourced records could have a provisional status that would be 
		formalised following checks on authenticity. This could be performed 
		directly by land administration staff or accepted directly from trusted 
		community experts or quality checks achieved through crowdsourcing. The 
		approach to and judgement of authenticity would evolve and improve over 
		time, just as has happened with the maintenance of all wikis. This would 
		involve a changing role for land professionals, working with citizens 
		rather than for citizens.
		
		In emerging nations where there are insufficient land surveyors or land 
		surveyors do not wish to embrace a crowdsourced approach, the lawyers, 
		assessors or even bankers may eventually try to remove or at least 
		reduce the need for land surveyors in the property transaction by either 
		resorting to direct crowdsourcing or identifying another type of 
		intermediary to facilitate crowdsourcing in different communities in 
		exchange for some cash or in-kind consideration.
		5. MANAGING THE RISKS
		As with all radical changes to long standing approaches, vested 
		interests will be jeopardised and entrenched opposition will inevitably 
		be encountered. Here are some of the risks that will most likely be 
		raised to attempt to keep the status quo.
		5.1 Can crowdsourced land rights information be sufficiently 
		authenticated?
		One of the most contentious issues surrounding crowdsourced 
		information is the authenticity or validity of the information provided. 
		Without the rigors and safeguards associated with formal professional 
		and legal based processes, crowdsourced information is of variable 
		quality and open to potential abuse. Crowdsourced information has 
		provided input to wikis, feedback of quality of services and counting 
		birds, for example, but is not normally used to capture information as 
		critical and legally binding as property rights in an authoritative 
		register. So what techniques could be used to quality assure the 
		authenticity of the information to a level that would be acceptable for 
		inclusion in a property register? Some alternatives, including lessons 
		learned from leading wikis and e-commerce, are discussed below. However, 
		the most appropriate crowdsourcing approaches to authenticity assessment 
		will only be identified through testing in the field.
		
		Grameen Community Knowledge Workers as Intermediaries
		This approach would avoid open, direct crowdsourcing at the outset and 
		only allow information to be provided by trusted intermediaries within 
		communities who have been trained and have worked with local land 
		professionals. Initially, there would be comprehensive quality assurance 
		of the crowdsourced information, but over time as trust is established 
		with the intermediaries the level of quality assurance sampling could 
		significantly decrease. These initial intermediaries could then train 
		further experts to build a significant network of ‘experts’ across 
		communities. Each expert would be continually checked and appraised to 
		determine the level of expertise and trust in the associated 
		crowdsourced information. To optimise the scarce resources, the 
		intermediaries could be shared with a range of information services, 
		such as health and agriculture.
		
		Community based Quality Assurance
		Quality assurance could be directly provided by members of the local 
		communities who take direct responsibility for authenticity. The 
		crowdsourced land right claims could be posted for communities to review 
		and comment on. Some form of local or regional land tribunal could be 
		established to arbitrate on conflicting claims. Once a critical mass of 
		land rights information is obtained it is then easier to identify 
		anomalies and conflicting claims. Levels of trust and accuracy of the 
		land rights would be upgraded over time as more evidence and cross 
		checking validates the claims.
		
		Wiki and e-Commerce Solutions
		Beyond local involvement in quality assurance, a centralised user 
		reputation system based on feedback from crowdsourced registrations, 
		similar to the buyers’ ratings of the sellers used in eBay, could be 
		used to assess the credibility of contributors and the reliability of 
		their contributions (Coleman, 2010). Leading wikis, such as 
		Wikipedia.org, originally relied solely upon the "wisdom of the crowds" 
		to evaluate, assess and, if necessary, improve upon entries from 
		individual contributors, usually with great success. However, recent 
		contributions of deliberate misinformation to specific entries have 
		caused Wikipedia to re-assess its approach. Beginning in December 2009, 
		it has relied on teams of editors to adjudicate certain "flagged 
		entries" before deciding whether or not to incorporate a volunteered 
		revision (Beaumont, 2009).
		
		Although the data that are contributed to VGI projects do not comply 
		with standard spatial data quality assurance procedures and the 
		contributors operate without central co-ordination and strict data 
		collection frameworks, research of VGI is starting to provide methods 
		and techniques to validate quality and also the needed evidence to show 
		that this data can be of high quality. Recent research (Haklay et al, 
		2010) supports the assumption that as the number of contributors 
		increases so does the quality; this is known as ‘Linus’ Law’ within the 
		Open Source community. Studies were carried out using the OpenStreetMap 
		dataset showing that this rule indeed applies in the case of positional 
		accuracy.
		
		Crowdsourcing Quality Assurance
		Some elements of the quality assurance process do not require local 
		knowledge of the land rights claim and could be crowdsourced to a 
		network of informed consumers and world-wide professionals or could even 
		be automated.
		
		Passive Crowdsourcing Quality Assurance
		Mobile phones can also be used passively to collect evidence that 
		supports validation of user entered information. For example, the use of 
		a mobile phone is continually logged and this log can be analysed to 
		show where the phone is frequently used, inferring the location of the 
		owner. The network timestamp is another robust piece of evidence that 
		could be associated with collected land rights data, such as images or 
		videos. This is not something that most (99.999%) of users can tamper 
		with.
		
		The extent to which control is held by the contributor, by the 
		institution, or by "the crowd" of contributors assessing each other's 
		contributions may be different across different implementations of 
		crowdsourcing.
		
		5.2  Will openness lead to more corruption in 
		the land sector?
		Land administration is often perceived as one of the most corrupt 
		sectors in government. Although individual amounts may be small, petty 
		corruption on a wide scale can add up to large sums. In India the total 
		amount of bribes paid annually by users of land administration services 
		is estimated at US$ 700 million (Transparency International India, 
		2005), equivalent to three-quarters of India’s total public spending on 
		science, technology, and the environment. However, one of the best means 
		of reducing corruption within a good governance framework is through 
		transparency of information and the ability to have two-way interaction 
		with clients.
		
		Data collected by the public must be validated in some way, otherwise 
		the crowdsourced information is open to abuse, and in the case of land 
		rights, corruption through false claims. However, transparency, which is 
		at the heart of the crowdsourced philosophy and the increasing use of 
		the mobile phone to check authentication, should support a fight against 
		corruption.
		
		5.3  Will Land Professionals form a new 
		partnership with citizens?
		This new partnership model implies that Land Professionals will have 
		a different relationship with citizens or ‘proamateurs’. The increased 
		collaboration with citizens opens up the opportunity for new services to 
		train citizens and community intermediaries and to quality assure their 
		crowdsourced information. It should therefore not be perceived as a 
		threat to their livelihoods and profession. But will Land Professionals 
		accept this new role and will sufficient citizen entrepreneurs provide 
		land rights capture services and become trusted intermediaries? 
		Disruptive technologies have and will continue to challenge the 
		relationship between ‘proamateurs’ and land professionals, but these 
		drivers of change also present significant opportunities for all 
		stakeholders.
		
		5.4 Will crowdsourcing just reinforce the informal land market?
		There is a danger that the emergence and acceptance of crowdsourced 
		land rights information by citizens will just reinforce the informal 
		land markets in countries where there is ineffective land governance, 
		poorly performing land administration systems and weak formal land 
		markets. Lack of trust in the formal land administration system will 
		persuade citizens to try crowdsourcing alternatives that are attractive 
		due to their transparency and citizen involvement. The final outcome of 
		the informal or formal market will depend on the Land Administration 
		agencies’ reaction to crowdsourcing and whether they reject or embrace 
		it.
		
		5.5 Who will provide the ICT infrastructure to support this 
		initiative?
		The implementation of crowdsourcing in land administration requires 
		technical infrastructure to support the uploading, management and 
		maintenance of the land rights information. The implementation could 
		mirror the voluntary support model of OpenStreetMap. OpenStreetMap's 
		hosting, for example, is supported by University College London’s VR 
		Centre for the Built Environment, Imperial College London and Bytemark 
		Hosting, and a wide range of supporters 
		(http://wiki.openstreetmap.org/wiki/Partners) provide finance, open 
		source tools or time to support the initiative.
		6. CONCLUSIONS
		Crowdsourcing within the emerging spatially enabled society is 
		opening up opportunities to fundamentally rethink how professionals and 
		citizens collaborate to solve today’s global challenges. This paper has 
		identified land administration as an area where this crowdsourced 
		supported partnership could make a significant difference to levels of 
		security of tenure around the world. Mobile phone and personal 
		positioning technologies, satellite imagery, the open data movement, web 
		mapping and wikis are all converging to provide the ‘perfect storm’ of 
		change for land professionals. The challenge for land professionals is 
		not just to replicate elements of their current services using 
		crowdsourcing, but to radically rethink how land administration services 
		are managed and delivered in partnership with citizens. Land 
		administration by the people can become a distinctly 21st century 
		phenomenon.
		REFERENCES 
		
			- Al- Omari, M. 2011. “Land Administration Systems and Land Market 
			Efficiency”, FIG May 2011, Morocco FIG Working Week.
- Arthur, C. 2011. “How the smartphone is killing the PC.” 
			Guardian Newspaper, 5th June 2011. Retrieved from 
			http://www.guardian.co.uk/technology/2011/jun/05/smartphones-killing-pc. 
			(Last accessed 10 August 2011). 
- Beaumont, C. 2009. "Wikipedia ends unrestricted editing of 
			articles". The Telegraph. 26th August 2009. Retrieved from 
			http://www.telegraph.co.uk/technology/wikipedia/6088833/Wikipedia-ends-unrestricted-editing-of-articles.html. 
			(Last accessed 11 August 2011). 
- Donovan, K. 2011. “Module 6: Anytime, Anywhere: Mobile Devices 
			and Services and Their Impact on Agriculture and Rural Development”, 
			“ICT in Agriculture” e-sourcebook. World Bank. To be published 
			September 2011. 
- Enemark, S., van der Molen, P. and McLaren, R. 2010. “Land 
			Governance in Support of the Millennium Development Goals: 
			Responding to New Challenges,” Report on FIG / World Bank Conference 
			Washington DC, USA 9-10 March 2009, FIG Publication.
- Fox, K. 2011. “Africa's mobile economic revolution”, The 
			Observer Newspaper, 24 July 2011. Retrieved from 
			http://www.guardian.co.uk/technology/2011/jul/24/mobile-phones-africa-microfinance-farming. 
			(Last accessed 12 August 2011).
- Haklay, M., Basiouka, S., Antoniou, V. and Ather, A., 2010. “How 
			Many Volunteers Does It Take To Map An Area Well? The Cartographic 
			Journal, 47 (4), pp 315 – 322. 
- Heinzelman, J., Sewell, D.R., Ziemke, J. and Meier, P. 2010. 
			“Lessons from Haiti and Beyond: Report from the 2010 International 
			Conference on Crisis Mapping”. Retrieved from 
			http://www.usip.org/files/resources/PB83.pdf. (Last accessed 17 
			August 2011).
- Laarakkar, P. and de Vries, W.T., 2010. “www.Opencadastre.org - 
			Exploring Potential Avenues and Concerns”. FIG May 2011, Morocco FIG 
			Working Week. 
- UN-HABITAT. 2008. “Secure Rights for 
			All,” United Nations Settlement Programme (UN-HABITAT)2008, ISBN: 
			978-92-1-131961-3.
- Zimmerman, W. 2011. Private correspondence. 
BIOGRAPHICAL NOTES 
		Robin McLaren is director of Know Edge Ltd a UK based, independent 
		management consulting company formed in 1986. The company supports 
		organisations to innovate and generate business benefits from their 
		geospatial information. Robin has supported national governments in 
		formulating National Spatial Data Infrastructure (NSDI) strategies. He 
		led the formulation of the UK Location Strategy and has supported 
		similar initiatives in Kenya, Hungary, Iraq and Western Australia. He 
		has also supported the implementation of the EU INSPIRE Directive in the 
		UK and was recently a member of the UK Location Council. Robin is also 
		recognised as a world expert in Land Information Management and has 
		worked extensively with the United Nations, EU and World Bank on land 
		policy / land reform programmes to strengthen security of tenure and 
		support economic reforms in Eastern and Central Europe, Africa, 
		Middle-East and the Far-East.
		CONTACT
		Robin McLaren
		Director
		Know Edge Ltd
		33 Lockharton Ave
		Edinburgh EH12 1AY
		Scotland, UK
		Tel: +44 (0) 131 443 1872
		E-mail: 
		robin.mclaren@KnowEdge.com 
		Web: www.KnowEdge.com 
		
		© RICS & Know Edge Ltd, 2011
		
		
		