INSPIRE has the potential to achieve its original vision and much more. However, there needs to be a willingness to change the culture to reflect modern and evolving end user needs. There needs to be less focus on compliance and more on exploitation of the existing and new geospatial resources. There needs to be a shift from the complex to the simple. Forget the 100% solution and implement applications that provide near term ROI and end user satisfaction. Show successes! Change the speed of response to changing technical and political realities. These suggested changes can result in a more agile and flexible policy and technology infrastructure suited to meet the needs to the European (and broader) geospatial producer and consumer community now and into the future. I strongly encourage the INSPIRE community to engage in and actively support the “What if” sessions. Collaborate for change to ensure the success of the INSPIRE vision!
While working for the OGC, I often used the INSPIRE vision as an exemplar of interoperability for sharing geospatial data and services across jurisdictional, cultural, and legal boundaries. INSPIRE is based on a great vision, one that is incredibly important but also extremely ambitious and complex to implement. In recent years, the complexities and related barriers of implementing INSPIRE have multiplied with the continued technology evolution with increasing use and focus on cloud computing, sensors, LIDAR (3D), linked data, the semantic web, internet of things, and more. This natural evolution of technology and related user expectations has created a tension – some even say a wall – in the INSPIRE community between the geospatial world and the much broader web world.
Organizations throughout the European Union have invested resources to implement INSPIRE compliant metadata, network services, and portals. According to the INSPIRE Roadmap[1]between now and 2020, organizations need to implement compliance with Annexes I, II, and II (data specifications) of the legislation. Hundreds of individuals have invested expertize, time, and energy in developing technical guidelines, supporting tools, validation and conformance testing capabilities, and more.
However, the community needs to step back and consider the big picture and what issues need to be resolved in order to capitalize on the investments to date and to achieve the INSPIRE vision. This includes future proofing INSPIRE and being able to leverage in a positive way new advances in technology and policy. Here is also an opportunity to better engage end users and talk about end user requirements. The concept of end user needs to be expanded from the traditional SDI and GIS community to the much broader consumer and web communities. Finally, success stories need to be documented and communicated to individuals with budget authority so that financial sustainability is ensured.
Most of the EU members have implemented INSPIRE compliant metadata. Fewer have implemented compliant monitoring and reporting, spatial data services, and network services. Fewer still have implemented compliance with the 34 data specifications. In 2013, the European Commission included the INSPIRE Directive in its Regulatory Fitness and Performance Programme (REFIT) to assess whether the Directive remains fit for purpose at the halfway mark of its implementation[2]. While the REFIT evaluation confirmed that the overall relevance of the INSPIRE Directive to meeting policy needs in an efficient manner remains high, the Commission report does identify a number of issues and provides recommendations. Some examples of the identified issues are:
- Due to the technical complexity of the interoperability implementing rules and guidelines that require the application of IT tools and skills that are often absent, the deadlines which were valid when the Directive was adopted may not be pertinent anymore for all Member States.
- Member States also questioned the effectiveness of future data harmonisation due to the significant effort and cost involved in transforming existing datasets to meet the new requirements (deadlines in 2017 and 2020). They called for a flexible, pragmatic and user-driven application of the existing data specifications.
As an outsider looking in, INSPIRE is at a crossroads. The issues may run deeper than the REFIT report suggests. All the hard work and commitment may be for naught. Unless the INSPIRE implementing rules, architecture and technical guidance can rapidly evolve, the INSPIRE vision may never be fully achieved. This evolution may require the difficult task of modifying the underpinning INSPIRE legislation. There is a tension in the INSPIRE community. Many recognize that INSPIRE needs to evolve to remain relevant and to provide value to governments and citizens but the question is how.
Focus
There are outstanding issues related to open data and licensing. The current focus on compliance and validation appears to override other concerns. The “cloud” did not exist in 2007. The capabilities of the cloud have fostered geospatial services and data that are delivered as software-as-a-service, giving organizations a hosted portal along with services-based content hosting. This new pattern eases the discovery, management, sharing, and use of content anywhere on any device. Back in 2007 many viewed Open Street Map (OSM) as of little use and an academic exercise. Collection of location enabled sensor data by citizen scientists was in its infancy. Today end users increasingly think of their mobile phone as the primary access when dealing with geospatial content and services. What are the implications of these technology and market trends? Can INSPIRE evolve in an agile way or will geospatial and IT worlds make INSPIRE a great idea that never achieved its vision?
There is definitely a recognition of these issues and others within the INSPIRE community and the Commission. For example, there is the 2017 INSPIRE hackathon, the “What If” sessions, Open Data initiatives, and more. These are great initiatives and will provide excellent feedback into the system. The question then becomes whether the system and related bureaucracy can change in an agile and flexible way to evolve and future proof the INSPIRE infrastructure.
Key Issues
The following are what I consider to be some of the key issues that need continued discussion and resolution in the next year or two.
The INSPIRE Data Specification Technical Guidance defines detailed rules and requirements for modelling and encoding 34 themes of geospatial data. The idea is that defining consistent data models enhances the ability to share data, including semantic context. However, these data models also significantly increase the complexity and cost of implementation and maintenance. There is a need for simple profiles of key elements of the data model. For example, the majority of applications requiring buildings may only need the footprint and the height of the building. This is an example where 95% of use cases do not require full implementation of a data specification. These simple profiles could easily be implemented in GeoJSON. Currently, there is too much reliance on GML and XML. Perhaps INSPIRE needs more reliance on open, volunteered geographic data, and citizen science. This relates to the point of how to deal new data sources or methods of collecting and processing spatial data. Obviously, there is the need for authoritative sources for content such as the cadastral theme. For other themes, such as transportation, and for many applications, such as navigation and emergency response, sources such as OSM have been shown to be more than adequate. Perhaps the INSPIRE community should change focus more to what citizens want rather than just government to government interactions.
A major focus is on compliance. Compliance is about whether an implementation of an INSPIRE service or data specification follows the rules (requirements). While compliance – and compliance testing – may be good in theory in practice compliance testing has many issues. There is the 2020 deadline looming where all spatial datasets shall be conformant to IR-ISDSS and available through network services. This includes performance criteria for service provision, which may no longer make sense for cloud based deployments. In support of compliance, an expanding suite of tools and supporting materials are being developed. Development and maintenance of these tools is an expensive, time consuming proposition. Any Member Country that works to implement all 34 data specification models and the transforming software necessary to map from their internal models to the INSPIRE models faces a very expensive development and maintenance process. What happens if the models change? Many of the data themes, such as atmospheric conditions and environmental monitoring, are collected in near real time using sensor networks. End users want these data accessible in near real time in easy to use formats, often fused with other data themes. For these situations, collecting metadata, mapping to some detailed data model, and testing for compliance or validating the data is viewed by many as a waste of time, resources, and money. Instead of forcing compliance testing against some data model, make sure that the sensor systems are properly calibrated and provide access via standard interfaces. Instead of investing in compliance, perhaps redirect energies and resources into evolving the infrastructure, addressing non-GIS use cases, resolving data sharing agreements and licensing for data access, and enhance end user outreach.
Metadata is critical for discovery, determination of fitness for purpose, and more. INSPIRE specifies an extensive list of mandatory elements using an extended version of ISO 19115 and ISO 19119. The Technical Guidance discusses 58 possible metadata elements, many of which are mandatory and many others conditional based on various rules. An additional set of mandatory elements is specified in Annex B. All of the guidance, examples, and validators targeted to this approach and are limited to XML instances. Conversely, for eGovernment portals the EU has standardized on DCAT-AP as their abstract metadata model. DCAT-AP is an application profile of DCAT[3] which is an RDF vocabulary designed to facilitate interoperability between data catalogs published on the Web. DCAT-AP specifies 19 elements for dataset metadata, 2 of which are mandatory and 5 are recommended. This is much more lightweight than required for INSPIRE and as such easier to implement and maintain.
There is a requirement to harmonize metadata across portal applications. A joint JRC and ISA2 work activity as developed a geospatial extension to DCAT-AP known as GeoDCAT-AP[4]. The driving use case is to make spatial datasets, data series, and services easily searchable on general data portals using standard web search engines. Such an approach makes geospatial information more easily discoverable across borders and sectors. The INSPIRE community should vigorously support this activity and work to simplify the metadata implementing rules and guidance to allow a more flexible and cost effective approach. The question then becomes, “How does INSPIRE evolve to a simpler, more flexible metadata approach?” Perhaps fewer mandatory elements should be required. Perhaps other encodings, such as JSON, need to be accommodated.
Persistent Identifiers are necessary for linked data, spatial data on the web, uniquely identifying network services, proper service and dataset metadata and more. While the INSPIRE commission feels that this is a State’s responsibility, I would recommend that JRC and others step up and work with the States to make sure that persistent IDs are generated consistently, have the proper governance, and are duly registered and discoverable. Without persistent identifiers, moving INSPIRE into modern web data technologies will be difficult if not impossible.
Issue of Equality
There is an issue of equality. There are actually two aspects of equality. One has to do with how can all EU members achieve the same level of capability and value from implementing INSPIRE? The other is how can all citizens have equal access to geospatial content and services available via the INSPIRE infrastructure? To truly realize the INSPIRE vision, all Member countries need to be at similar levels in terms of provision of data and services. This vision needs to extend to all citizens in the EU and beyond. To truly leverage the value of location enabled data, any citizen should be able to access data and services that help them in their day to day decision processes (within the constraints of confidential or privacy considerations).
A consistent theme that occurs in each of the issues discussed above is simplicity versus complexity. “Everything should be made as simple as possible, but no simpler.”[5] For as long as I can remember the geospatial industry has struggled with this issue. Unfortunately, in many cases there are not 100% solutions to interoperability requirements! While many would like the 100% solution, quite often the 90 or 95% solution resolves the vast majority of user requirements. For example most GIS implementations assume static vertical and horizontal datums. In recent years, there has been increased standards focus on how to model and encode dynamic datums. Tectonic plates move. Land subsides or rises. Sea levels are changing. A 100% solution for encoding geospatial content requires that these dynamics be considered. At this time, doing so is probably economically impossible. So there is no 100% solution. The INSPIRE community and commission needs to vigorously debate these issues and consider evolving the implementing rules and related guidance and compliance requirements.
A major forcing function impacting INSPIRE is how technology has changed since 2007 and with those changes what consumers expect for the provision and use of geospatial data and services. Consumers like simple! This places the onus for change on the producers. Within this context, the question is how to future proof INSPIRE, accommodate changing technology, policy, and consumer needs while at the same time protecting the current and future investment? Consumers expect their primary access to location enabled apps to be their mobile device. Most (all?) of these consumer applications exist in the cloud. The complexity is completely hidden from the end user. The market focus is on consumer needs including enhanced visualization, content fusion, real time feeds (traffic, weather), and more recently expanded exploitation by providing access to big data analytics. Perhaps INSPIRE needs to look at an advertising model to generate revenue for long term enhancements and maintenance!
In terms of protecting current investments and implementations and remaining as compliant as possible with the INSPIRE Directive and rules, one solution is that the current infrastructure could be “hidden” behind a façade layer. On one side of the layer are the “traditional” OGC and ISO based service instances and data resources including metadata. On the other side of the façade is an indexed web of spatial services and resources based on link data principals, accessible via Web APIs, integrated with “live” sensor feeds, and using simple profiles of data themes. The façade then provides the ability to define an application layer suitable for open data access, a pervasive and consistent end user experience, a development environment that web developers are familiar with, an infrastructure that is accessible and useable by search engine crawlers, and more. Facades can be used to future proof INSPIRE.
The community should glean lessons from the R&D efforts of the vendor community as well as many of the OGC testbed experiments such as cross community interoperability, portrayal, and 3d streaming. INSPIRE needs to be able to quickly review and leverage new standards efforts, such as vector tiles and Geopackages, that would allow the infrastructure to be more responsive to end user needs. These testbeds and based on well-defined end user requirements and showcase many of the new OGC initiatives (that may not yet be fully mature standards), but will nevertheless become important in the coming months and years. Instead of waiting for “full” standardization, INSPIRE needs to provide a mechanism to accommodate much of this work more quickly.
In summary, I believe that everyone involved wants to achieve the INSPIRE vision. To do so requires a shift in emphasis from complexity to simplicity, from compliance to end user requirements, from perceived dogma to agility and flexibility. Work together as collaborative community for change and achieve the vision!
[1] https://inspire.ec.europa.eu/inspire-roadmap/61
[2] https://inspire.ec.europa.eu/tags/refit
[3] https://www.w3.org/TR/vocab-dcat/
[4] https://joinup.ec.europa.eu/asset/dcat_application_profile/asset_release/geodcat-ap-v10
[5] Quote attributed to Einstein, most likely a paraphrase of a longer quote.
