MICHAEL BRAND ^[1] BA FRICS FInstCES FIS FRGS et al

President EUROGI

Chief Executive & Director, Ordnance Survey of Northern Ireland

Colby House, Stranmillis Court, BELFAST BT9 5BJ

E-mail: mbrand.osni@nics.gov.uk

This is the major "theme" paper prepared in advance of the international seminar on Global Spatial Data Infrastructures (GSDI) which will be held between the 17-19th November in Canberra, Australia. The meeting will be the third in a series (previous ones were held in Bonn, Germany and Chapel Hill, North Carolina, USA) and its objective is to build on the results of the previous meetings concentrating on the theme of "Policy and Organisational Framework for GSDI". It draws heavily on the "theme" papers from the previous meetings and is designed to inform and hopefully provoke. Organisations are not created overnight and those that have taken part in the birth of such organisations understand and know well the torment, anguish, heat, etc, that these events provoke and engender. The period of gestation is necessarily long with the many players differing in their perspectives, requirements and agendas. The UK Association for Geographic Information (the AGI) was at least two years in its formation while EUROGI (the European Umbrella Organisation for Geographic Information) took considerably longer.

Readers may not necessarily agree with all that is in this paper, it may not be radical enough for some, and there will be omissions but delegates to Canberra are asked to rectify these and to come prepared to participate positively in both the formal and informal debate. For those who will not be able to make that journey please let us have your views so that they can be taken into consideration. However this paper reflects the contributions of a number of individuals of the Steering Committee for GSDI3.

It is assumed that readers are familiar with the papers from the previous meetings and these, as essential reading, can be found at http://www.gsdi.org/docs1998/canberra/ .

The first two GSDI meetings concentrated on providing a forum for exposing national and regional initiatives in the formation of Spatial Data Infrastructures (SDI). This has enabled many to share and compare experiences. It has also enabled the definition of the problem to be tackled, to identify the actors involved and to start to draw lessons from national spatial data use initiatives such as NSDI in the US.

Central to the problem is the definition of what is meant by a Global Spatial Data Infrastructure. Although a certain consensus was reached at the Chapel Hill meeting, this was only after much debate and with the acknowledgement that the definition should be revisited from time to time and the concept of GSDI would be periodically redefined.

Thus the following definition was adopted:

Since Chapel Hill, a number of actions have been begun to further inform debate within the Global Community. Primary among these is a baseline survey currently being assembled by Harlan Onsrud ^[2] (Department of Spatial Information Science & Engineering and National Center for Geographic Information and Analysis, Maine, USA) of the various national and regional infrastructures which will complement and expand on the work published by Professor I Masser ( An International Overview of Geospatial Information Infrastructures: Lessons to be Learnt for the NGDF ^[3] ). This will provide a fundamental information resource for future activities and a starting point for some of GSDI future actions.

Professor D Rhind ^[4] last year formulated a number of practical steps that could be taken to facilitate and encourage the efficient linking, combing and more widespread use of spatial data. These would include:

• the facilitation and encouragement of collaboration in the collection, provision and use of spatial data
• the facilitation and encouragement in the use of standards and best practice in the collection, provision and use of spatial data
• the facilitation of access to spatial data

These "practical steps" are actually the three key issues currently being investigated regarding best, most efficient, most beneficial use of geographic information at some national and regional levels epitomised in Europe by the "GI2000" ^[5] process. One task of a GSDI secretariat and any allied working groups would be to define and implement some practical solutions.

This all pre-supposes some organisational structure or structures to move things forward. Indeed, this fact was identified at Chapel Hill where the first resolution recognised the need for an informal organisational nucleus to encourage the creation, development and linkage of local national, regional and global geospatial data infrastructures. This need has been emphasised by a number of complementary, and to a degree, overlapping initiatives such as those of the United Nations Regional Cartographic Conference for Asia and the Pacific and the work of the International Steering Committee for Global Mapping. The goal should be 'co-ordination' and 'harmonisation' not 'centralisation'.

It was recognised, at the March meeting of Delegates and Experts from the United Nations Regional Cartographic Conferences in Aguascalientes, Mexico, that there was a need for effective ways of organising international efforts and co-operation concerning geographic information bearing in mind the process of globalisation in all fields of human endeavour.

Al Gore ^[6] in postulating "The Digital Earth - an understanding of our planet in the 21st Century" clearly identified the importance of Geospatial Information and its associated technologies and challenged experts in government, industry, academia, etc, to help develop a strategy for realising his vision.

The general purpose of a GSDI organisational nucleus is essentially to facilitate the key elements of the GSDI definition which is inescapably linked to data. If we were to judge the success of the initiative we have embarked on with so much enthusiasm, looking back from a vantage point of some 10 years on it would be summed up in the word Data! and more Data! Data! in the hands of users that was cost effective, efficient and fit for purpose together with the means of using it.

In other words, the goal of the GSDI is to facilitate the development of policies and actions that ensure access to information and the delivery of data to users that ensure fitness for purpose. This is a key point. Industry will deliver the tools and governments, industry plus government/industry partnerships will deliver the data - but it is the policies and or lack of them that are holding back development.

These are many and varied and would include:

• Public sector at local, state, federal, national and regional levels (as both users and data collectors/owners)
• Not for profit organisations
• Research community
• Aid/development organisations
• Educational organisations
• International maritime, navigation and meteorology community
• Science and environmental community (including agricultural, forestry and fisheries)
• Private sector information/providers (including the new high resolution remote sensing providers whose technologies and products will offer alternatives to more traditional mapping technologies)
• Private sector hardware and software providers
• Private sector end users
• Private citizens

The public sector includes the interest of the military and the national mapping organisations not to mention national statistical bureaux, census bureaux and other holders of 'spatially related' datasets all of whom have an important role to play.

In trying to put together a structure to support the establishment of a GSDI we must balance the interests of the developed countries with those of the developing ones, or those whose economies are in transition. A potential downside of GSDI, if due cognisance is not made, could be to aggravate the current dependency of the less developed parts of the world on the developed ones. We need to be aware of these problems and the activities and structure of the GSDI should work to avoid or limit the effect of their consequences.

It is important to keep in mind that over 75% of the world's population is excluded from technology and one of the goals of the GSDI must be to reduce this gap and not increase it.

One vision of the future is where there is a very real possibility of a citizen (non-GI expert!) being able to pose a meaningful question (whose answer requires spatial analysis), in near natural language, using an 'Internet' access point that could be either personally owned or a community shared resource (such as the mobile phones now being installed in poor communities across India and Pakistan so that every village at least has a telecom connection point). The question will be posed either to an intermediary (human expert) or to an automated service which will locate and use only the bits of information needed to answer that query, applying the appropriate tools to process the query, then displaying the answer in a meaningful way (which may, in fact, have nothing to do with a 'map' based presentation) via the telecom link (Web, Internet email, voice answer, fax - whatever is most appropriate). The enquirer will only have paid for the pieces of GI actually used and the services of the intermediary, at much lower cost and zero learning curve compared to trying to come to grips with even the simplest GIS and related decision making tools available today. Competition between intermediaries, their combined bulk buying power for data from data owners/holders, ability to process (sell!) thousands of requests with minimal overheads (teleworking could feature large here) would all lead to low "per transaction/per answer" costs.

Representation and involvement must also reconcile the interests of those countries and or regions which have not yet created organisations which are fully representative of the Geographic Information community. This not only reflects the situation in the "developing" world but includes others in more "front line" economies. It is a dynamic situation full of challenge and change. Europe when EUROGI was first envisaged had only eight national associations representing the GI community. Now there are more than 23, no doubt encouraged and facilitated by the formation and existence of the umbrella organisation.

There are a number of policy conflicts between different parts of the globe which must be recognised, addressed and eventually resolved or at least the impact of the differences accepted at the political level. One example of this is the philosophical differences between the US and Europe concerning the pricing for access to public sector data.

Coleman/McLaughlin ^[7] identified in their Chapel Hill theme paper a number of GSDI components and the problems that would arise at the interface between them. In their analysis of the interface between people, policy and agreement standards and technologies, they identified a number of issues and posed a number of questions.

• Is a single basic infrastructure even appropriate? Is the concept too ambitious or even naive at this time?
• What is the real influence of pricing on the nature and direction of GSDI development?
• How much of this infrastructure will grow regardless of plans, policies or actions?
• Can we recommend appropriate combinations of prescriptive and enabling policies to promote effective application of geospatial data infrastructure in support of substantial and sustainable economic development? In particular, a requirement to support the actions envisaged in Agenda 21 of the Rio Summit?

The consensus at Chapel Hill to the first question was that we needed to define a general framework for different types of infrastructure across the many sectors of the so-called "GI Industry". The concept is ambitious but 'naive' only if we think this is going to be easily achieved in the near future.

With regard to whether we need to do anything or not, it is certain that infrastructures will grow regardless of what we say or do. The real point is, that it will grow in a non-harmonised way, increasing the costs to every one and not capitalising on opportunities made possible by data sharing, etc.

Should the solution of these be a result of "top down" or "bottom up" or as is often the case in these matters a combination of both. It will be both by default because solutions are and will continue to be developed and implemented in response to current needs. Whatever happens these matters should be addressed.

We have over the past 2 years discussed the definition of GSDI and have at least some consensus of what it should be. But what is the real goal of GSDI? A difficult question to provide a concise answer to but would we argue with a paraphrase of

If we could guarantee this at the global level many of us would be content!

There are a number of factors, however, which stand in the way of this vision all and would need to be overcome or at least addressed. Many are similar to those experienced at the local, national or regional level. These would include:

• the lack of data
• lack of adequate metadata
• existence of policies that prevent sharing
• lack of access to communication networks
• lack of common standards
• lack of awareness at all levels local, national and regional
• lack of education
• lack of effective search engines
• the impact of security
• lack of freedom of access
• the issues surrounding pricing
• the cost of data acquisition and/or conversion

Central to the discussion on the type of organisation structure most suited to support the development of GSDI are the possible functions it should carry out. The possibilities are potentially endless but all should be linked to the development of a viable system or systems that we can all identify with and share ownership. As a result a non-definitive list could include the following:

• Information sharing: newsletter, web page, publications (doable today)
• Awareness: promotion of GSDI by presentations and outreach (which will require the "conversion" or support of some high level politicians in all countries)
• Education: workshops, training courses and material
• Promotion of model strategies and policies (is possible and doable today - at least the 'model' bit)
• Promotion and monitoring of standards efforts: development of strategies for standards (doable today and already underway?)
• The provision of a forum: for debate, analysis and the identification of issues relevant at the global level

Another way of approaching what GSDI should deliver initially would be to identify some of the desired outcomes of GSDI in (say) three to four year's time. These could include:

• specification of core global spatial datasets (based on benefit analysis)
• specification of technical standards and infrastructures
• specification of global policies for access to the core datasets
• adoption of these specifications and policies by nations
• funded programmes commenced to deliver the core data

Is this vision just wishful thinking? Many would aspire to it and in theory support it. But how many would be in a position to deliver the resources necessary to undertake such a program. In the end it will come down to money and perhaps we should be pragmatic and realistic in our initial activities.

In looking for models and structures which could overcome the obstacles and most effectively deliver the functions envisaged, certain principles can be identified. These, if met, would provide a sensible framework for further development. We may well question if they can all be met in practice but at least the list set out below outlines an optimum situation we should aspire to.

As a result the GSDI structure should aspire to:

• be inclusive of all stakeholders
• add value
• build on, facilitate and support existing initiatives
• command respect and authority
• support sustainable development
• be flexible and adaptable to change
• command support and financial resources
• facilitate new initiatives especially those relating to the use and sharing of data
• be as simple, transparent, open and democratic as possible
• enhance democratic decision making processes
• engender partnerships

What type of organisational model is going to deliver most effectively the results identified, be best for overcoming the obstacles and is compliant to the greatest number of the principles that there is consensus that we should adhere to?

All organisational types for which examples exist around the world have both advantages and limitations and have developed either for historical reasons or to suit local conditions. It may well be that we will have to consider something uniquely constructed to meet the current requirements.

To assist, therefore, the discussion three main organisational types have been identified. These are "Government orientated" epitomised by the UN Regional Cartographic Conferences or at the national level by CNIG (France) or RAVI (The Netherlands), "Business orientated" with OGC (OpenGIS Consortium) as an example and the "Umbrella" type of association for which examples abound like EUROGI (European Umbrella Organisation for Geographic Information) or some of its many members at the national level in Europe, e.g., AGI (UK), PROGIS (Finland), AFIGEO (France), etc.

The latter example is interesting in that the connection between AFIGEO and CNIG in France is extremely close with all the members of CNIG also members of AFIGEO but the latter embracing all parts of the GI community wishing to participate.

Each of these models has features to recommend them but also down sides.

Government organisations have official political recognition, some official sources of revenue or revenue in kind from the supporting agencies. However, they are potentially ponderous and bureaucratic and at the global level membership would not represent all stakeholders. There are strong possibilities of an over involvement of the richer and more developed regions or countries.

Business orientated potentially have more resources, with more commitment from those involved, often more focused and can potentially be more productive. However, they, almost by definition, will not represent the various interests of the GI community.

The Umbrella model on the other hand has the potential to involve all the GI community, at different levels of involvement, speed and commitment. In other words it can be designed to be very flexible and inclusive. Potentially it is less threatening to existing initiatives which in many cases will need facilitation and support especially to ensure implementation. On the other hand they traditionally have been difficult to resource especially if they are not from the outset officially recognised and supported by those with access to resources, i.e., governments, their agencies and the private sector.

Other possible GSDI structure models that could be considered include a regional approach, a country approach or an interest group approach. Included at Annex A are worked examples of some of the main models identified based on existing examples. In identifying the best structure, which may need to embrace a number of the facets of the models above, we must ensure that the issues of representation i.e., Inclusivity, Equality (developed and developing areas of the world), Geographic (regional as well as national and Local), Sectorial (e.g., public, private, not for profit, academia, etc), Organisational (e.g., International Federation of Surveyors (FIG), International Society for Photogrammetry and Remote Sensing (ISPRS), North Atlantic Treaty Organisation (NATO), OpenGIS Consortium (OGC), United Nations (UN) and its Agencies, World Bank, International Hydrographic Organisation (IHO), International Standards Organisation (ISO), etc) are all taken into consideration. It is important that all are comfortable with the vision, structure and goals of the initiative and the means of taking a programme forward.

One very important issue will concern the relationships between the GSDI structure and existing initiatives of which there are many both at the regional and global level. Under the GSDI banner anything that is done must add value, be supportive and facilitate the development of these and new initiatives but it must be done in a way that acknowledges all the sensitivities that exist.

One which in particular has been identified for such treatment is the International Steering Committee for Global Mapping (ISCGM).

Having said that they key activity of GSDI is making data available to users we have to consider this major international project which is:

• already making progress to that goal
• plans to establish a permanent organisation to maintain and build on that first dataset
• already has substantial political and financial support
• includes many of the key players in GSDI

Perhaps the global mapping dataset could become the "version 0" for GSDI and the ISCGM a major committee of the GSDI structure.

Ways must be found to support their work as a vital component of the GSDI concept.

When this paper was proposed it was hoped that it would lead logically to an optimum solution which would appear almost magically as systematic consideration was given to all the issues as they were identified. In practice, as consideration was given to the issues delving deeper into the detail the number of questions multiplied most, which can only be answered by debate, and the consensus that we hope will ensue.

Questions that need to be addressed must include:

• Leadership and a recognised mandate for that leadership which at the global level will come from the inclusion in the formal agenda of an accepted global body
• Secretariat (where situated, staff, role, separate budget)
• Committee structure & membership
• Work groups (policy and technical groups)
• Budget and funding mechanisms (selecting as few national goals as possible to see real progress in the next few years)
• Communication mechanisms (the web providing low cost communication highways)
• Initial workplan
• Priorities
• Implementation milestones and performance criteria for the initiative itself

Many think that GSDI must be more than a talking shop or a forum but even this would be a good start and provide a focus for information exchange, the identification of issues and the start of co-operative action. The more GSDI is required to deliver, the more complex the structure will need to be and the more dedicated resource that will be required.

Already suggestions for GSDI working groups are surfacing. One concerning technical development potentially involving all nations involved with hands-on SDI implementation programmes has been mooted. The formation of a GSDI Technical Group would allow for the implementation of many of the policy directions that emanate from the main GSDI Policy Group. If there is sufficient support for this it is intended giving official GSDI "sanction" to arrange a first meeting of relevant interested parties immediately following the GSDI meeting itself. The initial aims of this GSDI Technical Working Group would be to build on the commonalities of existing national SDI implementation approaches, develop a set of GSDI technical business principles and vitally, develop a common GSDI Glossary of Terms in order that we can understand one another. A brief outline of this working group is set out in Annex B.

Al Gore, Vice President of the United States of America, in his vision of the "Digital Earth" challenged governments, industry, academia and the GI community at large to work together to develop a strategy to realise this vision. A vision which in one of its many possible manifestations we all share. Although his challenge was largely made to the US nation its applicability globally is unquestioned. A successful creation of a GSDI and its implementation will go a long way in answering that challenge and in turn help to solve many of the pressing problems facing society in the world, inspire its citizenry and underpin sustainable development.

In reviewing this presentation I would like to return to my central theme, that of data. As global communication becomes ever more efficient and accessible to more citizens the need for GSDI will become, in my opinion, an ever more pressing tool to complement the decision making process. The establishment of GSDI is a significant but commendable challenge to all of us. It will be a world first, in my experience and to the best of my knowledge I cannot find an equivalent initiative or organisation which is so "multi-disciplinary". I would, therefore, conclude that it will be unique both in its formation and its undertakings. The greatest challenge to GSDI is therefore, to deliver results that affect people's lives in a positive and constructive manner.

[1] The views expressed in this paper are the personal views of the contributors . They do not necessarily reflect the views of their employers.

[2] Sample questionnaire responses as completed by a respondent for the United States URL http://www.spatial.maine.edu/questionnaire/USNSDI.html

[3 ] An International Overview of Geospatial Information Infrastructures: Lessons to be Learnt for the NGDF, Version 1, Ian Masser, 1 July 1998 URL http://www.ngdf.org.uk/whitepapers/mass7.98.htm

[4] Theme Paper 2 : Implementing a Global Geospatial Data Infrastructure, David Rhind, October 1997 URL http://www.gsdi.org/docs1998/canberra/ggdiwp2b.html

[5] GI2000 Geographic Information for Europe home page URL http://www2.echo.lu/gi/en/intro/gihome.html

[6] Al Gore, The Digital Earth: Understanding Our Planet in the 21st Century, February 1998 URL http://www2.nas.edu/besr/238a.html

[7] Theme Paper 1: Defining Global Geospatial Data Infrastructure (GGDI) Components, Stakeholders and Interfaces, David Coleman and John McLaughlin, October 1997 URL http://www.gsdi.org/docs1998/canberra/ggdiwp1.html