By Mike Powell, Development Information Specialist, facilitator of monitoring and evaluative processes with the Land Portal and volunteer member of its Technical Advisory Group
The Context: Conflicts over land have shaped human history. Policies as to its desired use are shaped by ever-changing political, economic and environmental concerns. Their effectiveness depends on the local realities in which they are applied. One rare point of consensus is that secure access to land is central to achieving a wide range of development goals for a variety of land users. However, defining, maintaining and measuring what secure access to land and land use involves has proved elusive. From efforts to agree comparable indicators with which to track global development goals, to resolving competing demands at local level, to assessing investment risks, finding information which is reliable, complete and broadly accepted by all stakeholders continues to be a major challenge.
Some areas of difficulty relate to the nature and use of land itself:
- Rights relating to land are more complex than for any other asset. It is possible to have rights relating, for example, to minerals, firewood, grazing and free passage to land that is owned by someone else. These rights sometimes relate to reciprocal obligations, such as labour, which may not have been consistently applied and therefore may or may not be legally enforceable.
- Many countries, for historic reasons, have a variety of means of formalising land ownership or other rights to land even within their legal canon. In addition, claims based on custom, common rights and/or on indigenous practice often remain significant in both practical and legal terms.
- Land use and land values are related to a variety of topographical features (soil, rainfall, vegetation etc.) which may or may not be properly recorded or be subject to any legal protections.
Other difficulties emerge from stakeholders’ differing interests and requirements for information about land:
- Governments and, more recently, policy makers wish to analyse and manage land issues at scale. They are looking to aggregate and compare information over large areas. Historically, and even more so with the emerging technologies for linking data, this has heightened their interest in large scale data, collected and maintained to established standards. All governments in the world struggle to maintain complete records of the land in their territory. Incomplete or inaccurate information can fuel conflict and may be of less value than none at all.
- By contrast, land users, at almost any scale, are interested in the detail regarding the land they wish to use. This detail is likely to take many forms, few of which will be standardised or lend themselves to aggregation. It may not be in the public domain or, if it is, may be hard to access and use.
- The challenges extend beyond official data. The largest, most accurate and most up to date data on populations in developing countries now comes from their mobile phone use. This provides big data, not necessarily representative of the population as a whole, on who is making calls/ texts to whom and where. There are also an increasing number of apps which claim to provide users with useful services or information about specific areas of life, such as health care or farm prices. Such apps generate further and more specific data for their providers. The potential for similar apps related to land ownership and use is obvious. However, these possibilities raise as many questions as opportunities. The data is held by the phone companies, which are usually privately owned. Is it potentially another revenue source for its owners or can it be used in the public domain? What issues does its use raise for privacy, civil rights or commercial confidentiality?
Contrasting needs and uses lead to asymmetries in how information about land is collected and resourced, by whom and which technologies are applied to processing it. These asymmetries can exacerbate long standing mistrust between stakeholders rather than build bridges between them. It is easier to navigate the (possibly inaccurate) view from the top than the (invariably incomplete) view from the bottom. Information even about the same land is often highly fragmented and hard to assemble. No-one is fully satisfied with the information available to support their work on land and development, whatever their role.
The Land Portal and its approach
The Land Portal is a not-for-profit foundation, registered in the Netherlands. It aims to be the leading online destination for information, resources, innovations and networking on land issues. Its work is based on the assumption that secure, sustainable and equitable access to land requires good quality information, as complete as possible. Although giving particular attention to the potential of secure access to land to support the rights and wellbeing of the poorest and most marginalised land users, the Portal seeks to fulfil what it perceives to be the necessary role of an impartial provider of all potentially relevant information from all sources to all stakeholders. It does not engage in advocacy about land issues.
Its work is based on three key principles – open and standards based information, local ownership of information processes with information shared more widely through negotiation and consent, a collaborative approach to building and maintaining an information eco-system of value to the land and development community. Its experience to date has highlighted a number of key themes.
- New technology should not repeat historic exclusions. The Land Portal, along with a number of national governments and large multilateral organisations like FAO and the World Bank, champions the use of Linked Open Data (LOD) because of its openness, its foundation of agreed standards and its interoperability. However, within the development LOD community, the Portal is keen to ensure the emerging platforms and their associated vocabularies do not reproduce the kind of asymmetries between different users seen in previous systems.
- Data alone is not enough. Most land information, especially that produced by smaller and more local stakeholders, is produced in document formats, most often pdf. Even if such documents are made available on-line, their dispersed nature and the small volume of documents published by each stakeholder, makes such material almost invisible to mainstream search engines. The land portal has therefore focused on developing its capacity to produce metadata, that is information about these other sources and their location. Linked open metadata may be of more value to the land community than LOD
- If issues of their quality and inclusivity are resolved, LOD technologies present the possibility of highly decentralized and distributed models for both the collection and sharing of information. The Portal is working with a number of projects in developing countries, essentially offering technical capacity building and support in exchange for access to new sources of information. The needs for continuing system maintenance and support does not disappear with this model, but it should offer a more resilient and sustainable information system than one based on a costly centralized apparatus.
- Finally, the Land Portal has come to see the collaborative production, sharing and discussion of information as a vital constituent part of a productive and inclusive discourse on land. To this end it, jointly with partners, hosts on-line debates on key land issues, commissions or republishes relevant blogposts and amplifies the proceedings of land related conferences and workshops by providing social media services. This active approach to information sharing creates opportunities for policy formation and collaborative practice which are simply not offered by approaches which treat information as a commodity to be harvested.
Towards a Sustainable Land Information Eco-System
The Land Portal offers a single site ‘destination’ for all those looking for data and other information about land. However, it does so in a way which re-uses data and material from other sites in a form which facilitates their further re-use elsewhere. Information which used to be have to be found, can now be published in such a way that it will appear automatically, wherever it may be valued, on sites dedicated to providing information on a particular theme or location. This is what the Land Portal means when it talks of a land eco-system. It offers a way of publishing, sharing and maintaining information of value to all parts of a community, such as the land sector, which is characterised by a multiplicity of users, with different practical and scientific interests, located across the globe.
Realising this vision requires progress in three key areas. The core standards of linked open data are well established but they depend on agreed definitions with which to describe the data entered so that it can easily be found and retrieved. Historically, most such systems have been developed by experts for use in a limited disciplinary field. Their use in a multi-user, multi-disciplinary setting is far more challenging. Language is key, not just actual language but issues of two-way translation between colloquial language and official or scientific terminology. Important too is the diversity of understandings which link human beings to the land and its features. Imposing one understanding on another will not contribute to lessening disputes over land. Instead the Land Portal is considering an ‘open definition’ policy where diverse definitions of the same term can be posted on-line whilst their use within the system is overseen by an editorial committee. In seeking to address these challenges in the development of its own service, the Portal has created a specialist Land Vocabulary – and translated it in pilot projects with local NGOs in South East Asia and Latin America. The opportunity now exists to establish a broader base for the governance and development of this LandVoc and encourage its adoption across the sector.
The use of tools like LandVoc offers an opportunity for anyone wishing to share information about land on-line. Properly labelled documents will automatically be discovered and made accessible by specialist sites, such as the Land Portal, or in professional repositories. At one level, this produces an effective route for even the smallest and most remote publisher to make their material visible globally. At another, it offers a technical solution for bringing together information at a local level, enabling local actors to share and discuss information about the land which most concerns them. This, of course, also raises issues of trust and governance which need to be addressed by those directly concerned. The Land Portal, however, has been developing support and training services to enable such initiatives where the conditions exist for their implementation.
Finally, many groups become involved in land issues because of their commitment to development more broadly. This means that information systems focused on land issues need to work with and contribute to the wider development information eco-system. With this in mind, the Land Portal engages wherever possible with others working on the use of linked open data and other semantic web approaches within the development context. It is part of the GODAN partnership, which works on the potential of open data in the field of agricultural development, is contributing land vocabulary to the Agrovoc Thesaurus, maintained by FAO, and has also worked on issues connected with land data and indicators with GLTN and Habitat for Humanity.
All these initiatives offer opportunities for further development and for collaboration with other agencies. To pursue this further the Land Portal aims to schedule the workshop described above late in 2017. To express interest in participating in this process and workshop, please contact Stacey Zammit at Stacey.firstname.lastname@example.org.
 For example, the Land Registry in England and Wales (there are separate registers for other parts of the UK) has registered land transactions since 1862. Its records, which are available for a fee on-line, still only cover 80% of the land area and can only be searched by reference to the location of the property, not by other criteria. They may or may not contain information about any rights of access or use by others, depending on the quality and date of the documentation submitted to the registry. Data on common land or land with common rights is not included, being maintained in a different system by local government.
 This issue was identified and discussed in the context of development in Jessica Seddon Wallack and
Ramesh Srinivasan, 2009, ‘Local-Global: Reconciling Mismatched Ontologies in Development Information Systems’, Proceedings of the 42nd Hawaii International Conference on System Sciences