A Summary of the Seminar hosted by the Association for Geographic Information at the RICS in London on 14 May 1997 with further recommendations from the Working Group for research and development actions
| Prof Mike Worboys Department of Computer Science Keele University Staffordshire ST5 5BG |
Dr David Parker Department of Geomatics University of Newcastle Newcastle upon Tyne NE1 7RU |
Initial draft prepared by:
Richard Honey
The Royal Institution of Chartered Surveyors
12 Great George Street
Parliament Square
London SW1P 3AD
CONTENTS
Executive Summary
Section 1 Introduction
Section 2 Setting the NGDF Scene
Section 3 Defining the Research Headings
Section 4 Research Requirements
Section 5 Way Forward
Section 6 Summary
Section 7 Recommendations
Appendix 1 Working Group 3 and Seminar Attendees
Appendix 2 A Non-definitive List of 'Problem' Issues
for NGDF
Appendix 3 What is the NGDF - A 'User' Survey'?
Appendix 4 NGDF-like projects in other countries
This report is a summary of the proceedings and conclusions of the National Geospatial Data Framework (NGDF) Setting the Research Agenda seminar hosted by the Association for Geographic Information on 14 May 1997. The seminar was convened by the NGDF Working Group on Research (NGDF Working Group 3) and chaired by Prof. Mike Worboys of Keele University.
The stated NGDF mission is 'To develop an over-arching framework to facilitate and encourage efficient linking, combining and widespread use of geospatial data which is fit for purpose.' The aim of the seminar was to identify areas where NGDF-enabling research is required and produce a written report of the proceedings, with recommendations for further work. This was achieved by gathering opinion and ideas from a range of experts with an interest in NGDF, deciding on the main headings under which research can be classified, listing relevant research that has been done, is being done, and needs doing, and developing an action plan for taking the work forward.
In the sense in which the NGDF is a data framework, it is a unified collection of digital data held in many databases, and the associated supporting services to facilitate provision and usage. For such a framework to work effectively, the collection should at least be described in detail and ideally be integrated, consistent, contain the 'best' data available (quality), and have institutional and economic frameworks to facilitate interchange. A rough categorization of research areas immediately emerges as institutional, technical, and usage/markets. The seminar considered this categorisation, and some of the research topics under each heading. It was clear that prototypes targeted to particular markets could provide testbeds for research activity in all areas. It was also felt that the technology already existed to get a basic 'level one' NGDF into operation, but that institutional and usage/market issues still needed to be resolved. However, for anything but the most rudimentary framework, much technical research, focusing on integration, interface and temporal issues, needed to be addressed urgently.
In brief, the broad research areas and the core research issues within these areas identified at the seminar are as set out in the table:
Technical
Users/Markets
Appendix 3 on 'What is the NGDF - A "User" Survey?' summarises responses collected at the meeting to the question 'What is the NGDF?'. These well-informed experts showed a wide range of responses, indicating that whole idea of the NGDF is still very fluid and ill-formed.
Following the Research Agenda meeting, the Working Group prepared some recommendations to the NGDF Board for implementation and research. Section 7 of this Report sets out the recommendations in more detail. In outline, the Working Group recommend:
This report is a brief summary of the proceedings and conclusions of the National Geospatial Data Framework (NGDF) Setting the Research Agenda seminar hosted by the Association for Geographic Information at the Royal Institution of Chartered Surveyors in London on 14 May 1997. The seminar was convened by the NGDF Working Group on Research (NGDF Working Group 3) and chaired by Prof. Mike Worboys of Keele University with Dr. David Parker of the University of Newcastle upon Tyne. A list of working group members and seminar attendees is included in Appendix one. For more information on the National Geospatial Data Framework itself visit the website at www.ngdf.org.uk or contact Liz Hobman, NGDF Programme Manager at Ordnance Survey (Tel: 01703 792936, Email: ehobman@ordsvy.gov.uk).
1.1 Aim of the Seminar
The aim of the seminar was: to identify the areas where NGDF-enabling research is required, and produce a written report of the proceedings, with recommendations for further work.
1.2 Objectives
The seminar was arranged to achieve the aim by meeting the following objectives:
The agenda for further work would then be developed by summarizing the work done to date in NGDF-related disciplines, taking note of current research, comparing this to the needs identified by this seminar and developing the areas where work is required.
The stated NGDF mission is 'To develop an over-arching framework to facilitate and encourage efficient linking, combining and widespread use of geospatial data which is fit for purpose.' The purpose of the meeting on which this is a report was to consider the research effort that is required to make this mission a reality. The approach was to attempt to place the research topics into broad categories, and for each category to consider what research had already been done that was germane to the NGDF, and what was still required to do. There was little attempt in this meeting to deal with priorities and time-scales. The NGDF Working Group 3 will take this report as the foundation for later and more detailed work.
In the sense in which the NGDF is a data framework, it is a unified collection of digital data held in many databases, and the associated supporting services. For such a framework to work effectively, the collection should at least be described in detail and ideally be integrated, consistent, contain the 'best' data available (quality), be up-to-date, and have institutional and economic frameworks to facilitate interchange. The supporting services should enable transactional updating by data providers, network access for data users, data integration for data users, multiple quality levels (cost dependent), and metadata support. The business and policy context should ensure that user constraints are explicit, quality descriptors are provided, a pricing structure is in place, as is also a liability structure.
The development of the NGDF will depend on a number of key issues. The NGDF vision includes many data linkage applications, many new services and consequently many more uses for data. Data linkage depends upon integration, and this key principle must act in all the different facets of the NGDF, including technology, institutions and economics. It is not only the data that must be integrated to make NGDF work (and that itself is no mean task), but even more problematic are the services, interfaces, business contexts, institutional factors, and data management.
Examples of the types of issues involved are shown in the table below:
Supporting Methods of updating data
Services Providing access to data
Metadata services for NGDF datasets
Business Context Constraints within the data
Quality of data
Price of data
Liability for the use of data
Facilitation Easy identification of relevant data
Efficient searching through metadata and data mining
Security Confidentiality
Authenticity
Integrity
Intellectual property rights
Technical Barriers Volume and disparity of data
Schema organization
Requirements for and reporting of data quality
Complexity of queries allowed by the data
Ease of use of data services, given a range of differing
users
Security, including authorization for use and
interoperability
Outside of the specific research agenda for NGDF there were many fundamental issues which needed to be better addressed to ensure progress within the initiative. These include:
It was proposed that research for the NGDF could be organized under three headings: Institutional, Technical and Use. The seminar split into four small groups to discuss the proposal and reported back through a plenary session. The conclusions are summarized below.
3.1 Institutional
It was felt that this heading would encompass research into issues such as the legal, political and economic context of geospatial data. These issues would be important in the development of NGDF and would provide the impetus for the facilitation of geospatial data services, enabling greater data availability and promoting the growth of new markets. A significant proportion of the current institutional problems in regard to data availability and use arise within government, both central and local, despite the best efforts of groups such as the Interdepartmental Group on Geographic Information (IGGI).
3.2 Technical
The seminar agreed that of the three areas, technical issues will be the easiest to resolve, at least for setting up a starter framework in the short-term. In terms of software and hardware, developments must follow those in mainstream IT, but with the geospatial data and GIS community having an strong influence on the direction of general IT developments. It was agreed that there is a very real difference between data as a whole and geospatial data in particular - from a lack of user familiarity with geospatial data through to the technical problems of integrating geospatial data.
3.3 Use
It was agreed that use was not the best heading for a section of research, and that users would be better, and indeed this should include markets. It was felt that any research undertaken to facilitate the NGDF must be driven by the need to develop viable and sustainable markets for the benefit of all current and potential users of geospatial data. A sense of perspective would also be required from an expert community undertaking research to support end users, who will have no knowledge of the underlying issues of the NGDF. A wide scoping study would be a valuable exercise to establish the extent of the possible future market, and end user's needs, for NGDF products and services.
It was also noted that while users are paramount, the data producers and potential service providers must also be considered, especially as they may be a major source of resource for research to establish the Framework.
3.4 Commentary
It was felt that whilst these headings were useful for the development of the research agenda, there was considerable overlap and that they should not be viewed as separate programmes. There was some feeling that the headings categorize researchers rather than research, and that a more fluid and evolutionary approach was needed.
It will be vital to ensure communication and feedback between the three different groups. Research into the relationships between the three groups would also be needed, especially into how they inter-relate and influence each other. Research undertaken must be driven by a real need from the NGDF end users, must not attempt to cover areas which will be dealt with by mainstream IT research, and should, as far as practical be 'future-proof'.
There was discussion as to whether the overall approach to research should be top-down (from the 3 headings to specific topics) or bottom-up (from specific topics which may transcend the 3 heading areas). It was felt that a top-down approach would, despite some faults, provide the overview of the situation which was vital to the successful planning of a research programme for NGDF.
Given broad agreement on the three headings under which research could be organised, the seminar re-formed into three groups to consider each heading, and what research would be needed in each area - drawing up a shopping list. The discussion was aided by an impromptu, unstructured list of possible research topics drawn up by Neil Smith (Ordnance Survey) and others (see Appendix 2).
4.1 Institutional
A brain-storming session allowed four issues to be identified as core priorities. These could then facilitate other developments in downstream areas, with a snowballing effect. It was felt that the line between many institutional issues on geospatial information was hard to define as separate from mainstream information issues. The proposed Freedom of Information Act could have a very significant impact on geospatial information, but it would not be in this specific arena that decisions would be made.
The first core priority would be a review of legislation as it relates to the management of geospatial information - used to highlight barriers or gaps, which could then be addressed by further research. A compendium of geospatial information law for the UK, reviewing major acts such as those which govern Agencies (such as the Ordnance Survey), Data Protection and Copyright, and setting an agenda for developments such as a Freedom of Information Act, could be invaluable.
The second core priority would be research into the policy for the development of geospatial information services. This would effectively be a broadly based cost/benefit analysis of the relative advantages and disadvantages of following different models for the development of NGDF services (such as the National Land Information Service or ScotLIS®) - ranging from a government based scenario, through to a fully commercial scenario, driven by the market. A strong business case will be required for any development effort from NGDF partners and stakeholders. This project should also assess the need for, and merits of, a regulator for trade in data and information products, and any associated issues of privacy.
The third core priority was for research into the problems of intellectual property rights for digital geospatial data particularly within large scale datasets, addressing the flaws of the application of conventional law to digital information. This would be a process of evaluating the options available, including the impacts of European initiatives in this area, and mapping the future implications of today's framework for copyright administration and protection of IPR. Liability within a geospatial data framework will need to be considered, especially in relation to onward use of original datasets.
The fourth core priority would be for research into the comparison of national (geo)spatial data infrastructure initiatives, so that NGDF can learn from a worldwide experience. Parallels could also be examined with other infrastructure systems such as highways.
The seminar felt that as a research area relatively little funding would be required, as most work would be evaluation and comparison, rather than research and development of systems etc. However, because the issues to be examined did not easily fall within the traditional roles of researchers working in the fields of geography, surveying or computing, there could be a problem in securing resources, financial and human. It was felt that it was important to include cross-disciplinary researchers, from the business, economics and legal communities, to ensure the work was effective.
4.2 Technical
Based on work undertaken in the first session, the group looked at what was needed technically to get up and running and what was required to permit the NGDF to grow beyond what the current consortium can provide. Nothing was really identified that was essential before a service based on the NGDF could get underway. There is a need to standardize spatial data models, indirect position referencing and metadata requirements before any service is started but really to such a basic level that no new research is required. To take the NGDF significantly beyond what it can achieve with the current consortium then does require significant research in the six areas identified below.
The first core area of required technical research is a collection of
issues clustered around the concept of integration. This has
several components: data model integration, including heterogeneous
data semantics; incompatible spatial units;
data integration, including handling errors and imprecisions; multi-resolution,
and other aspects of data quality.
In geospatial data modelling, what is required is to understand the differences
in current data models as used and proposed, to rationalize the model components
needed to describe the different data models and then to standardize the
methods by which we describe all this.
The multi-resolution data modelling issues are key in the long term. The
NDGF might well start as a series of access routes to largely non-redundant
and homogenous datasets: i.e. complete coverages of information to the same
resolution and the same time frame and without competition. To move on to
a service that accesses live datasets, maybe different ones for different
areas, and overlapping datasets with common data but to different resolutions
we will have to take our abilities in multi-resolution modelling much further.
The second core technical area identified was that of metadata; what is the basic information needed for an effective metadata service with further work needed on the needs of the metadata life-cycle. Metadata is a large area to cover and it was the belief of the meeting that no-one will really know what is required until the first release on metadata standards have been implemented in the first installment of the NGDF
The third core area identified was that of data quality in the sense of 'fitness for purpose'. This is a hard concept to define, but critical for data providers to understand. There was seen to be significant overlap between data quality issues and metadata issues. However particularly notes here was the need to research error propagation whilst making inquiries on multifarious datasets.
Interfaces was the fourth area for consideration. This
included issues such as information retrieval, data mining and flexibility
to account for wide ranges of types of data, markets, users and providers.
For a minimal service to be up and running, the group felt that little further
research was required. However as more and different datasets are offered
for purchase then new approaches to retrieval and data mining will be required.
The fifth core area of technical research was that of temporal data management, including data currency, archiving, maintenance, access to historic data. The group noted that archiving is more than just `long term storage'. Many will be interested historical tracing. Also, there is the problem of handling updates and corrections (distinguishing between them and other kinds of changes).
The last of the six core technical research areas is that of the development of prototype services for NGDF, based on particular market areas. Prototypes should be considered sooner rather than later. There was discussion, however, as to whether prototypes were a research or educational issue.
Security and access mechanisms were also raised, but it was thought that they would be addressed outside of the NGDF community, as they were not pure geospatial data issues. Working from existing technical platforms will allow both the flexibility required of NGDF services and allow access to markets.
It was observed that many of these issue are not purely technical problems. For example, the handling of data as they cease to be current has technical and institutional dimensions.
4.3 Users/Markets
The group decided that research was needed to find out who the potential users of NGDF services would be, how their markets operated and what they would want from NGDF. This discussion was aware that many data consumers could also be data providers, but concentrated only on their role as customer and consumer. The group considered a number of topics and ranked them from a high priority to a low priority, with issues of liability, quality and error propagation at the top of the list. Other significant topics included mechanisms for delivery, generalisation and scale requirements, ownership of integrated data and the incompatibility of spatial units. Standards issues were considered to be too uninteresting to users for direct consideration here, and although security mattered to users, it would not form part of this research agenda.
It was felt that the best way to address these issues was a comprehensive user survey, tailored to the differing market sectors and leading to an authoritative market analysis. A precursor to this would be a range of demonstrators, along the lines of NLIS and ScotLIS®, to fully illustrate the possibilities of NGDF services. The search for users should be as exhaustive as possible, and should try to identify new customers, different from existing customer bases of organisations such as Ordnance Survey and the Land Registry. The group felt that the survey / research process could be more efficient if it focused on identifying communities of users, e.g. from education, the utilities, environmental organisations, property professionals etc.
The allocation of individual organisations into communities will not be without its problems, for example, from the basic definitions of community groups and criteria for membership, through to how they inter-relate and overlap. The identification of datasets core to a majority of user communities would be a valuable step forward in asserting the commercial potential of services developed through work undertaken by NGDF, as well as streamlining much work undertaken today in commercial environments. Whilst the development of metadata systems is a mainly technical problem the link with users is fundamental.
4.4 Commentary
The session was effective in categorizing possible research topics which were agreed as being important to the development of NGDF, and there was general agreement on what was proposed by the three sub-groups. The recurring theme was to be aware of the limits of NGDF, and how these are constrained by developments in mainstream computing and information issues. Close liaison between all research initiatives would be required to avoid any overlap or duplication of effort.
Given broad agreement on the need for research, and what topics should be included, the major hurdle for the development of an NGDF research agenda will be securing support and funding for subsequent projects. The seminar discussed the issues of reporting and funding in a plenary session.
5.1 Reporting
It was felt that a collection of NGDF research reports, although unlikely to be exhaustive of all the issues, would be an important way of recording the issues and raising the profile of research in this area. It was agreed that pilot studies focusing on particular application areas would provide a way of assembling the real requirements from users of the NGDF.
5.2 Funding
Bodies likely to provide funding fell into two broad categories: traditional funding bodies for research, and organisations who would benefit from facilitation of NGDF. The former category included EPSRC, ESRC, NERC and the European Union programmes. The research agenda crosses the range of traditional research disciplines and would therefore encompass a broad spread of sponsor bodies. Individual organisations were seen as unlikely to provide funding alone for an NGDF project - a collective approach to funding would be more successful. The potential was discussed for funding from the European Fifth Framework, which had some money reserved for GI research into policy issues, following the GI2000 initiative. An international dimension to NGDF research was felt to be beneficial.
Some organisations who would benefit from NGDF, and who could therefore be approached for funding, include current data providers, GIS vendors, GI service providers and major customers such as local government and the utilities. Commercial organisations in particular could be approached to sponsor the proposed user survey, based on the salable value of the information collected. The use of time, resources and personnel from government and commercial organisations has been successful in developing NGDF so far but would not be suitable for long-term or sizable projects. Significant funding from umbrella bodies such as the Association for Geographic Information (AGI) could not be realistically expected, but they may have a role in organizing funding, such as forming an NGDF research trust fund. It was suggested that AGI might take a lead in exploring funding possibilities, without committing themselves to any further funding of the project.
5.3 The Next Steps
Among the next steps outlined for the development of research in this arena was the creation of a research directory to act as a source book for all geospatial data stakeholders. The publication should list recent, current and planned research into NGDF agenda related topics, as well as researchers and their fields, and the interests of possible funding agencies. The NGDF Working Group on Research will need to address the development of the three headings for research, possibly by creating informal research groups to bring together the leading experts in each area. The use of similar seminars over time could be an important way of tracking progress, increasing communication and building the cooperative spirit needed for the success of NGDF.
Maybe the key contribution that the Working Group can make is in coordination and facilitation of research for the NGDF. Investigations would need appropriate funding from NGDF, companies who perceive benefits, and research funding councils.
We must take care not to reinvent the wheel. It would be useful as an early step to consider other work in this area, with NSDI being an obvious starting point, as well as similar European ventures (see Appendix 4). It would be useful to fund a fact-finding mission along these lines.
The core research areas and issues as identified by the seminar are as detailed in the table:
Institutional
compendium of geospatial information law
policy for the development of geospatial information
services
intellectual property rights for digital geospatial data
comparison of national (geo)spatial data infrastructure
initiatives
Technical
integration: model, semantics, units, multi-resolution
metadata: basic information and life-cycle
data quality: error propagation
Interfaces: information retrieval, data mining and
flexibility
temporal data management
development of prototype services
Users/Markets
comprehensive user survey and authoritative market
analysis
The overriding message from the seminar, and one which is illustrated by the definitions included in Appendix three, is the need for a clearer conception of NGDF, and one which is understandable throughout all levels of the current geospatial data stakeholder community, including some relatively uninformed end users. The second major message was the need to understand users and their requirements, both now and in the future.
A cautionary statement echoed by many was self-control within NGDF research, to understand what existing research can be applied and where other parties are already leading the way to solving some of the important (particularly technical) problems. The ability of NGDF to overcome the institutional issues, as a mainly public sector project to date, was questioned - government bodies cannot operate without reference to the wider political climate. An unquestionable independence was perceived as being vital to the successful development of NGDF. Organisations such as the Association for Geographic Information and the Royal Institution of Chartered Surveyors could provide the lead on political issues, and undertake the task of lobbying on behalf of NGDF.
The private sector was seen by most as the best group to lead research into the different user communities, where commercial opportunities in the exercise could be exploited to reduce the cost of the project to NGDF. A concerted effort should be made by the current NGDF partners to ensure that vital private sector companies, such as GIS systems vendor and information solution providers, are aware of the value of investment in the framework.
The fact that NGDF would cover the whole of the UK was welcomed, and a necessity to concentrate on developments in the UK accepted, but the need for close cooperation with international initiatives was emphasized. Although the use of three headings for organizing research was supported by the majority, it was acknowledged that the overlaps and inter-relationships would be complex and should always be kept in mind. Setting a detailed research agenda for the next three, five or ten years at this stage is an impossible task, due to the changing needs of users and developments in technology. An open approach, around a dynamic agenda of aims, will be the best approach to the successful management of NGDF-related research. Setting measurable success criteria will be required to audit the progress and cost-effectiveness of the NGDF and any supporting research work.
To summarize, the working group must move forward with achievable aims, providing a clear case of the costs and benefits involved for all research partners, without losing sight of the needs of the users and developing markets for the future success of geospatial data services and products in a national and international context.
This section arises from discussions of the Working Group following the Research Agenda meeting. Recognising that there is little resource within the Working Group itself, these recommendations are only likely to be implemented if external resources can be identified. It was noted emphatically both at the Research Agenda meeting and at subsequent meetings of the Working Group that the NGDF cannot effectively be constructed on a shoe-string budget - this is certainly not the case for data infrastructures being currently set up abroad. However, there are no pots of gold, and the Working Group feels that the best move is to begin a phased implementation of the NGDF (outlined below) that will attract funding as it grows and is seen to be useful to geospatial data providers, brokers, and users.
The Working Group therefore recommends that:
Recommendation 1:
The Board to fund the initial stages of the development and implementation of a phased architecture for NGDF, and to create an office with dedicated personnel who can begin setting up web pages, etc. This will be part of the funding of the existing Web site.
The key principle about the phased approach is that experienced gained with each phase will determine what is required and involved in the next phase. This will make the overall planning coherent. We do what we can at any phases within the requirements, technology and available resources at that phase, while planning for the future.
Phase 0: 'search by name' NGDF
Phase 0 will consist simply of a Web page containing a structured list of sources of spatial information and services based on this. The list would contain names and links to the source web sites. It would be easy to provide access pages structured in several different ways - by supplier, name of source, etc.
We envisage that the service might generate income from its inception, by letting advertising space on the Web site to service providers and others. The idea is that the service is self-funding in all its phases. However, there was a fear expressed by some members that NGDF should not become an advertising agency.
Questions immediately to be answered at Phase 0 will be:
Access to the site will be monitored, and data fed into next phase.
Phase 1:' search by content' NGDF
Phase 1 will evolve from Phase 0, and will probably allow access to data/service providers by specifying a concept (rather than the name of the source). For example, search on 'providers of data about sheep' rather than specifying the 'Society for Sheep and Lambs'. Metadata becomes very important at this phase.
Later phases will allow integration of data from different providers. All sorts of economic/legal issues will emerge at these phases. Further increments will only become clear as the project proceeds. However some ideas are:
Recommendation 2:
The NGDF Board commissions the following research studies.
We also envisage a report associated with each phase evaluating the previous phase, noting incremental requirements, new technology, resource bounds, and setting out the architecture for the next phase.
There would also be a recommendation for a research study on economic aspects of the NGDF, but such a study we believe has already been commissioned by the NGDF Advisory Council.
With written submissions from:
NGDF Working Group Three: Research
(with acknowledgment to Neil Smith, Ordnance Survey)
|
|
This section includes the written responses to the above question given by attendees at the seminar, near the end of the session. It is interesting to note the wide variance in what the perception of the NGDF is, and this from some of the leading authorities in the field. It strongly suggests the need for a clear definition of what the NGDF is, and what it is not. This must be linked to a campaign of promotion and a proper launch of the initiative. An attempt has been made to categorize the responses.
The NGDF is…
... some kind of database
A means for providing geospatial data to user communities in an easy, efficient and intelligent manner. It is the central point of access to both metadata and data itself. The data may be centralized or distributed. The accreditation of datasets, that may be provided by the NGDF, as meeting a set of defined standards is secondary to the service of a centralized access mechanism.
... mainly about standards, protocols and best practice
A set of protocols which make spatial data available to the widest possible community for appropriate uses. Protocols in this context include; transfer standards, advice on use, methods of analysis etc - without being any physical entity.
An underlying framework of standards and best practice for GI, developed through collaboration between commerce and the GI industry, designed to guarantee that all current and potential users of GI are able to gain the maximum benefit from geo-referenced data, with the minimum cost and trouble.
... a data integrator
A vehicle for facilitating the availability of GI from diverse sources and enabling the integration of this GI.
A concept of sharing spatial data - a data mixer. The concept involves a variety of technical, institutional and user issues concerning making spatial data both compatible and accessible. There is a huge gap between a concept and even a prototype however, and moving to this next stage is an admirable and ambitious step.
... multiple datasets
A small number of core datasets which are available at approximately the cost of dissemination - principally entities from which other data can be produced or to which other data can be added. Likely to include ADDRESS-POINT; basic postal, administration and electoral information, road centre lines.
A core set of geographical units/references which is maintained to a known (high) standard and disseminated at minimum cost to encourage its use for the management of geographically-referenced data. My set would be: master address file (BS7666), street centre lines, cadastral boundaries, administrative boundaries, statistical boundaries, building foot prints, Postal Address File, field boundaries, a map of rivers and the coast etc.
The collection of geospatial data that conforms to a series of protocols and standards which enable access, integration and use of the data by a diverse series of user communities.
... distributed computing with spatial data
NGDF should provide a mechanism for stimulating the rapid development of electronic commerce based on geospatial information. It should encompass a significant number of data and service providers - far more than the current main data suppliers. The current preoccupation with access to OS/AA type data if probably misguided. A transformation is needed which is equivalent to moving from stand-alone mainframe batch computing (of the 1970s) to the Internet based computing of the 1990s.
A mechanism for distributing the UK's geospatial data to all interested users, in as simple and straightforward a means as possible. Depending on the evolution of networking systems this may involve selection, payment and retrieval via the Internet. It is hoped that sufficient metadata would also be available to enable users to have full knowledge of the characteristics of the datasets. It is hoped that formats will make the data almost immediately importable into any GIS typically in use. It is expected that security will be adequate to protect suppliers.
NGDF is not a mega-database. It is a cooperative environment of data providers, service providers and users. The environment consists of agreed standards and practices governing technical and institutional aspects. The vision is of a transparent interchange of information in the terms understandable to any individual.
Means of facilitating and promoting the dissemination and use of spatially referenced data within the UK. To do this by creating an institutional and technical framework within which data can flow. To rationalize the collection of data and availability and reduce data redundancy.
Framework to enable access, flow and availability of geographically-related information in a consistent form across an unrestricted range of systems, mediums and organisations.
Data-warehousing-type application layer which provides centralized access to a variety of spatially aware data sets or databases. Coordinating resource providing information about, and links to, available data for applications which have been developed for specific market areas. It is not an application itself in the sense that it has no particular user interface with which a user interacts. Only exception to this may be demo type application to show capabilities of linking disparate datasets.
A framework which enables all sorts of users and data/service providers to communicate and have shared access. Not a system (hardware) but providing common languages/models, cost models, etc so that data can be shared and services can be facilitated.
... a philosophical/political approach to sharing spatial data
The NGDF - its role: (i) to foster/encourage sharing of spatial data, (ii) to provide examples of services that can be used to enable data providers to participate in the framework, and (iii) to make users (current and potential) aware of the possibilities of easy/efficient access to geospatial information.
A one-stop-shop for finding out about data and obtaining it, including update.
A way people can trade geospatial information and services which is:
An enabler; a focus; a mobilizer. To encourage the key players in the geospatial information arena to work together in creating a consensus approach towards meeting the conflicting needs with regard to the adoption of geospatial and other information. The outcome:
A group whose mission is to enable access to spatial information and services to the UK community. To achieve this it must identify the needs of the user base both currently and into the next millennium. It must act as a focal point to influence public policy. It must also provide a forum where data and service providers can agree on the way forward in interoperability.
... anything you want it to be!
A definition - and realization - of the core national interest mapping. A federated structure of discrete data holdings (above and beyond this core). A manifestation of best practice in data collection description, use, re-use etc. A facilitator for re-use of data. A forum in which to agree and implement distributed resource discovery/description models for geospatial data.
NGDF includes:
The problem is to transmit information from A to B and retain content and understanding.
(with acknowledgment to Steve Reeve, Ordnance Survey)
| Kort og Matrikelstyyrelsen (KMS) (Denmark): NGDI Initiatives |
Source Date Details
CERCO 5-95 The Map Series of Denmark is the nationwide CIS (Coordinated
meeting Information System) that the KMS is establishing. The series is a collocation of georelated databases and administrative registers
and rests on four cornerstones that are: reference systems,
cadastral maps, topographic maps and charts. TOP10DK forms the
topographic element to the concept.
| Finland National Land Survey: NGDI Initiatives |
| ADv (Germany): NGDI Initiatives |
Source Date Details
GISRUK97 3-97 The MERKIS programme defines in minute detail the
technical and organisational aspects to be considered
for developing municipal GI frameworks (Junius,
Tabeling, and Wagener, 1996. Germany: a federal approach
to land information management in Masser I. Campbell H
and Craglia M (eds) GIS Diffusion: the adoption and use
of GIS in local government in Europe. London: Taylor and
Francis.
A 'PLAN FIRST, DO LATER' approach.
| Instituto Geográphico Nacional (Spain): NGDI Initiatives |
Source Date Details
Presentation 11-96 They are starting to promote the concept of a National Spatial
Data Infrastructure in Spain.
| The Czech Republic: NGDI Initiatives |
| USGS: NGDI Initiatives (US National Spatial Data Infrastructure) |
| AUSLIG: NGDI Initiatives |
Source Date Details
Visit by 5-96 David Abel, from the Australian Commonwealth Scientific and
David Industrial Research Organisation (CSIRO) Division of
Abel Information Technology, discussed the Australian Experience
in establishing a Geospatial Data Infrastructure (GDI). The
evolution of the Australian GDI has been based on local
political conditions and solutions have focused more on
building organisational partnerships across tiers of
government and between private agencies, than on technical
issues. The report on his visit, together with its
implications for our NGD concept, also includes issues on
standards and metadata.
GISUser 2/3-97 The collaborative roles of ANZLIC (Australia New Zealand
Land Information Council), ICSM (Intergovernmental
Committee on Survey and Mapping), the Australian Surveying
and Land Information Group (AusLIG), and the academic
communities are reported in this article.
On the ANZLIC side, guidelines for metadata have been
published and a directory is being established by the
Commonwealth. The Spatial Data Transfer Standard has been
adopted through the activities of the Australasian Spatial
Data Exchange Centre, and is now a joint Australian and New
Zealand Standard (AS/NZS 4270). They have also published
research into where the spatial data industry thinks R&D is
required, and a training needs analysis. Academic
communities in particular are providing trained educators
and appropriate courses.
The model of the NSDI comprises four interlinked
components. An Institutional Framework, which defines the
policy and administrative arrangements required to
implements and maintain the infrastructure; Technical
Standards that define the technical characteristics of the
fundamental data sets; Fundamental Data sets that are
produced within the institutional framework; and
Distribution Methods by which the fundamental data sets are
made accessible to the community.
| Japanese Geographical Survey Institute: NGDI Initiatives |
Source Date Details
D. 11-96 GSI began in 1995 preparations of a 'Spatial Data Infrastructure'.
Rhind The following information reports from the presentation to OS by
the Chairman of the National Spatial Data Infrastructure
Association (NSDIA) in spring 1996. NSDIA, as a new organisation of
69 members, aims to promote the rapid development and circulation
of national data by supplying framework data for GIS. They are
currently researching other nation's NSDIs. NSDIA have recognised
the OS' information infrastructure strategy and our technical
support in helping them to begin to set up their own
infrastructure. They are looking for a further exchange of
information, particularly regarding their need to accumulate
appropriate technology.
Existing large scale maps prepared by GSI and local governments
(analogue and digital) are collected and digitised into a certain
format by GSI. This complements the GSI target of the 'Geographic
Information Database' project began in 1990 which is to construct
an integrated database consisting of base map data (raster
1:200,000), geographic name data and various social geographic
data.
N. 3-97 The National Geospatial Data Framework (NGDF) is being released in
Smith May 1997 when the first of 30 CD-ROMs is published. The NGDF has a
simple structure consisting of 9 layers:
Administrative boundaries and coast lines
City blocks
Road centre lines
Road edges (roads < 3 metres and highways)
River centre lines and banks (major rivers only)
Railways
Specified areas-inland water, parks, airports, etc.
Buildings (inner urban only)
Geodetic control points
Data are acquired from local government, consolidated by GSI and
distributed on CD-ROM. The format is proprietary to GSI. The data
are being distributed free of charge to local government; other
organisations will have to pay 7500 yen per CD (£37.50).
| Mexican Instituto de Estadistica Geografia e Informatic (INEGI): NGDI |
Source Date Details
Cambridge 7-95 Information Technology was introduced into INEGI in a programme which
Conference began in 1980.The Sistema Nacional de Informacion Geografica, known as the Mexican
National Geographical Information System, was introduced under the
INEGI Modernisation Programme began in 1990. The main features
include conversion of data to digital form into the Geographic
Database (GDB), production, updating, visual display, and spatial
analysis. Within the GDB, satellite, raster, vector, and alphanumeric
data is held.
| OSNI (UK): NGDI Initiatives |
Source Date Details
Brochures 3-96 OSNI, as part of the UK, approaches the NGDI as part of our NGD
1995 concept.
The topographic database underpins the development of the NI GIS
- a distributed but integrated network of databases involving all
the major government departments and public utilities within NI.
This is still in its development phase. Pilot studies with Land
Registry, Roads, Water and Planning Service of the DoE show
favourable cost/benefit analyses, and the DoA has a GIS Research
Programme with a telecommunications link to the OSNI database.