A vision for INSPIRE: from a traditional SDI to a self-sustainable data ecosystem
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 351 | |
| Author | ||
| License | CC Attribution 3.0 Unported: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
| Identifiers | 10.5446/69156 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language | ||
| Production Year | 2022 |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
| |
| Keywords |
00:00
Machine visionJames Waddell Alexander IIEscape characterSoftware frameworkState of matterIntegrated development environmentDigital signalContext awarenessElectric currentMetadataComputer networkService (economics)InformationImplementationWindows RegistrySoftware testingOpen sourceSoftwareComponent-based software engineeringClient (computing)Raw image formatStandard deviationExtension (kinesiology)Time evolutionStrategy gameGroup actionTelecommunicationZugriffskontrolleProcess modelingComputing platformCorrelation and dependenceComputer configurationCore dumpOpen setLarge eddy simulationInternetworkingComputer-generated imageryDigital photographyArchitecturePoint cloudSolid geometryMachine visionSpacetimeGreen's functionRepository (publishing)Service (economics)State of matterMachine visionCentralizer and normalizerMathematicsRepository (publishing)Different (Kate Ryan album)Process modelingIdentifiabilityValidity (statistics)Point (geometry)Basis <Mathematik>Internet service providerFunctional (mathematics)Latent heatRight angleContext awarenessImplementationGroup actionStandard deviationSoftwareRepresentation (politics)Client (computing)MetadataLibrary (computing)Level (video gaming)Self-organizationComponent-based software engineeringSoftware testingSingle-precision floating-point formatSoftware frameworkResultantField extension2 (number)Complex (psychology)SpacetimeIntegrated development environmentKey (cryptography)Arithmetic meanShared memoryWeb 2.0Projective planeNumberGraph (mathematics)Data modelRegulator geneAddress spaceVector potentialOpen setInternet der DingeVapor barrierCartesian coordinate systemCASE <Informatik>Video gameOpen sourceStrategy gameCoordinate systemMoment (mathematics)Traffic reportingSlide ruleRevision controlFeasibility studySoftware developerTheory of relativityExpert systemSpeech synthesisPresentation of a groupGoodness of fitPerspective (visual)Windows RegistryOcean currentProduct (business)Set (mathematics)Link (knot theory)GeometryBitExtension (kinesiology)UsabilityWeb serviceLimit (category theory)Server (computing)Inclusion mapSoftware maintenanceEvoluteSource codeGreen's functionCloud computingView (database)Observational studyStatuteSearch engine (computing)Control flowDiallyl disulfideOperator (mathematics)Computing platformCuboidZoom lensPoint cloudDampingPortable communications deviceRule of inferenceMultiplication signTerm (mathematics)Decision theoryScaling (geometry)Information privacyGraph (mathematics)ArchitectureFeedbackElectric generatorGreatest elementCapability Maturity ModelMeeting/InterviewComputer animation
Transcript: English(auto-generated)
00:00
Thanks a lot, Agnes, good morning again. Yeah, I think there is no Phosphor-G conference without a talk on Inspire, so that's the one for this year. I'm speaking on behalf of the Inspire JRC team. There's Jordi, my colleague here, and Alex. It's a talk on the vision for Inspire. So it's a talk on the future, how Inspire should evolve
00:23
from a standard traditional SDI to a modern data ecosystem. Let me start with a quick introduction on Inspire, if there's someone which is less familiar with that. So Inspire is a directive, first of all.
00:41
It's enforced since 2007. And it defines the legal, technical, and organizational framework to basically create a European spatial data infrastructure in support of environmental policies. Probably Inspire is the largest geospatial data sharing effort that has been ever undertaken.
01:02
In a nutshell, it defines requirements for interoperability on data, on services, and on metadata. And it's not something to be built from scratch, but it is to be created by making interoperable the spatial data infrastructures that already existed at the member state level. More practically speaking, Inspire is complemented
01:22
by other legally binding documents called the implementing rules, which define what the member states must do, again, in relation to the three components of metadata. Data, we have conceptual data models in Inspire, and network services. In Inspire, we have discovery services to discover the data through metadata. We have view services to view, to access the data,
01:42
and download services to download the data. Then we have the technical guidelines, which are not legally binding. But again, they say they specify for the three components from the technical point of view how member states can implement the legal requirements. Now, there are at least three reasons
02:01
why we are here today to speak about the future of Inspire. The first is the end of the Inspire legal roadmap that was initially defined and the need to inform the revision of the directive that will happen in the near future. The fact that there is a new political context in the EU, where priority number one is the Green Deal,
02:21
the environment, and the fact that we have a completely new technical context, technological context, compared to 2007 when Inspire was initially adopted. So we feel this is the right moment for us as the GRC. GRC is the technical coordinator of Inspire
02:40
to first reflect on the state of play and the lessons learned from the Inspire implementation happened so far, and then based on the new context, policy context, technological context, to present our vision on how we can modernize, we can streamline the infrastructure, and also some concrete actions to achieve that. Of course, all of these by also thinking
03:01
to the role that open source, that those geo can play. So let me start from the state of play, the lessons learned that will be, of course, both positive and negative. And the first one is actually a negative one. Inspire implementation is still heterogeneous in Europe. There is no single member state that has reached full implementation.
03:20
But at the same time, we can observe that the number of data sets that are made available by member states, currently around 80,000, 85,000, and also the level of interoperability of the data set is increasing on a daily basis. It's a continuous process. So Inspire should be seen really as more as a process rather than as a final product.
03:42
This is what we have seen. Now, I said before that Inspire should ideally be built on top of the infrastructures from member states. That is what you see on the right. But what has sometimes happened in the past is what you see on the left. That is, member states have built parallel infrastructures.
04:02
Often, they share only a limited amount of data sets that they have at the national level, and that is in scope of the directive. For example, they may not want to create metadata for all the data sets or to harmonize all the data sets or to set up services for all the data sets. The main reason is that the requirements are different from Inspire and at the national level.
04:23
Data models are different. Inspire requires additional, for example, service metadata that are not already provided by the software used at the national level, and so on and so forth. But the result, whatever is the reason, the result is that we have a patchwork of available data. Often, we have data with different scales, granularity,
04:42
spatial coverage, license, and some are not harmonized yet. And this, of course, is an obstacle for not just interoperability, but even pan-European coverage. And we also experience issues with usability. We have observed, for example, cases of services serving
05:00
huge amounts of data sets that are almost impossible to consume for clients or services that are not monitored and that are always or often or sometimes down. Now, next point is governance. Inspire, since the beginning, was conceptualized with a very inclusive and open governance approach in mind,
05:21
involving stakeholders since the very initial draft of the technical guidelines, which, of course, was key to ensure the future success of the implementation. Today, the main governance body is the MIG. We love acronyms, of course. The maintenance and implementation group that is the body where all member states are represented and that deals with all kinds of organizational, legal,
05:42
and technical matters. The issue here is that the MIG is exclusively represented by data providers. And as Ana said, change is hard. It takes time. So it has not always been flexible and fast enough to accommodate changes and novelties in the infrastructure. That is exactly why, a couple of years ago,
06:00
we started to introduce new flexible, agile governance approaches to actually change, to evolve the infrastructure. So the idea was to come up with community-driven approaches and to operationalize them using online platforms, and mainly GitHub, that we are now strongly using in Inspire. Let me mention two examples.
06:21
The first is the good practices. That is the way that we introduce to include new approaches, like new standards, new technologies for data encoding, data sharing in Inspire. On the right, you see what are the currently candidate and endorsed Inspire good practices. You may, for example, see OGC API features
06:40
that is already usable by member states to serve data in Inspire. And there is a governance process that specifies how to propose, how to discuss, how to provide feedback, and finally, how to endorse a good practice. Similarly, we define a governance approach for the artifacts that are the technical guidelines, so the very same technical documents
07:02
on Inspire implementation, the schemas, and the UML models. Again, there are different GitHub repositories that operationalize the whole workflow, from the initial proposal to the feedback, the discussion, the endorsement, and the final implementation of the changes.
07:20
Moving now to software, here you see what we call the Inspire central infrastructure components that are the software tools that the GRC develops and maintains and operates on a daily basis to make Inspire implementation possible. So you see the Inspire geo-portal. There is the point of access to all the data shared by member states under Inspire, the Inspire reference validator
07:42
that offers a test to check the compliance of data, metadata, and services, and the Inspire registry that manages all the identifiers in Inspire. What is important to say is that they are all based on open source components. From the right to the left, the Inspire registry is based on the registry, open source software.
08:02
The reference validator is based on the ETF framework. And the Inspire geo-portal is currently in the process of being migrated to geo-network open source. If you're interested, there will be dedicated talks to each of these components. So the one on the geo-portal is exactly here after the keynote, after the coffee break, so in one hour.
08:21
And the other two, ETF and registry, are tomorrow, just one after the other. So if we look at the software used by member states to implement Inspire, you can see some few proprietary solutions. And you can also especially see open source projects, and a lot of those geo projects.
08:41
We could probably say that almost all of these solutions have implemented at a certain point some specific functionality because of Inspire, so either to provide ETL functionality, for example, to harmonize the data, or to serve the data, or to consume the data.
09:02
So Inspire is agnostic from the technological perspective, but it has always been strongly based on international standards, mainly from the OGC. Probably, Inspire implementation represents the largest uptake of OGC standards ever. And of course, using standards benefits everyone, not just the users and the providers, but also the standardization bodies
09:23
themselves. Because they get feedback from the implementation, they can improve the standards. There is an uptake because of the participation of Inspire representatives in the technical committees of the standardization bodies. Sometimes we have experience problems, for example, due to the mismatch between a standard
09:42
and the existing implementation of that standard. So the lesson learned here is that it is very important to consider only standards that are already mature, already well used by the community for Inspire, of course. So not just standards that satisfy the legal requirements, but that really add value to the technological stack
10:03
of the data providers, and of course, especially that make it easier for users to actually use the data. Now, something else that we learned from the implementation of Inspire is that the technical infrastructure is only as good as the social infrastructure underpinning it. In a similar way as the phosphor gene community,
10:22
we also rely on a very active community in Inspire, mainly composed of representatives of the public sector, but also we have the private sector, we have academia, we have other organizations. The community meets at the annual Inspire conference. You can see on the map where the conference has happened in the past.
10:41
Unfortunately, no conference in the last few years due to COVID. Now, something that didn't really work in Inspire was the assumption done at the very beginning that the requirements that were created at the beginning and that were not yet supported by existing software and libraries would lead to the evolution of those software
11:02
and libraries to support them. But this actually not always happened, and I can mention two examples. The first is the support that is still limited to complex GML datasets in GIS clients, and the second is the support for the extended capabilities of Inspire network services in GIS servers.
11:23
So the lesson learned is always to make sure that the technical requirements, even the possibly future ones, can be already implemented by solutions out of the box, of course, by solutions that already exist. And then we have the licensing issue. There is a high heterogeneity in licensing approaches
11:44
that are adopted by the different member states and sometimes even by different providers within the same member state. Of course, the result is that not just, not all Inspire data is open data, but it's even worse in the sense that a number of different
12:01
and non-standard licenses are used, which of course pose an obstacle to the actual reuse and uptake of the data. So how to solve this? It's not an easy one to solve. We need a common European approach based on a standard licensing framework such as Creative Commons, and this is what the open data directive,
12:23
that is a directive from 2019, is and will try to solve, at least for some of the datasets. Now let me quickly zoom on the context, political context and technological context that were actually the future of Inspire should happen. And speaking about policy context,
12:42
we need to mention the European Strategy for Data. That is a document from 2020 that you see at the top right. It's an important one because it envisions the establishment of a European single market for data to exploit the whole potential of European data provided by different actors
13:01
and it anticipates some legal documents that you see on the left. Just to mention some, these are important for data sharing in general with of course impacts on Inspire. And these address the licensing. As I mentioned before, there's the open data directive and an upcoming implementing regulation on high value datasets
13:22
and also the governance of data sharing in data sharing. I can mention the Data Governance Act and also the Data Act. Finally, the strategy is also envisioning the birth of so-called European data spaces that in all strategic societal sectors in order to ensure and improve data sovereignty in Europe.
13:43
And of course, among these sectors, there's also the environment and the data space is called the European Green Deal Data Space. And this of course is where Inspire will have a role. From the technological point of view, of course, we are speaking about a completely different situation compared to 2007.
14:01
We no longer have only the data from the public sector. We currently have a huge number of new data sources and actors that we should consider. We have private data. We have data from the Internet of Things. We have earth observation data. Of course, Copernicus here plays a key role in Europe. We have citizen generated data,
14:21
think to open street map, which is increasingly considered even by some governments, citizen science projects. We have open research data and research data is extremely important and there's an increasing attention to the failure, the findable, accessible, interoperable and reusable principles. Not just data of course, but technology in general
14:42
that we should consider first and foremost APIs, web based APIs, modern APIs, agile standards that are co-created by different actors. Again, think to OGC APIs, think to stack, for example, for data sharing, for encoding, think to geo-package, think to cloud optimized GOT,
15:00
for example, we have more and more major tools to serve the data, to use the data, to ETL the data and we have novel architectures, federated cloud infrastructures. Here I can mention IEX, that is a promising example in Europe, edge and fog infrastructures. We have specifications like solid to improve the portability of data.
15:23
So with all these ingredients, let's come finally to the vision. There is a vision for the future of Inspire, but for the future of any SDI, so we can extend this, we can generalize this. First point, data sharing is not a goal in itself, it is a means to an end and the end is always to take better decisions,
15:42
to innovate, to improve citizens' life. In order to remain fit for purpose, for the current European policy context and to be sustainable in the long term, we believe that Inspire, again, and SDIs in general should blend in with the broader IT ecosystem
16:01
of spatial, but also non-spatial data, policies, infrastructures, technologies. Only in this way, we believe Inspire can really attract a broader community and can address an increasing number and a different number of use cases and applications. Of course, this means that we should, first of all,
16:21
we should find ways to lower the entry barrier to the infrastructure. We should come up with more open, more agile governance approaches. We should facilitate the technical complexity of the SDI. How to achieve this in practice? We have a number of actions from the legal, the organizational and the technical perspective
16:41
and of course, based on what I said before, we are, it's clear that we are already working on many of them, you will recognize where. From the legal perspective, first of all, is that legislation should avoid over-specification. So technical details, technical complexity should be always left out from legislation
17:00
which didn't really happen in Inspire. So that when technology evolves, legislation can stay the same. And then the use of a simple licensing framework as I mentioned before, based on a common, a standard licensing framework. Organizational actions, here we speak about governance. We don't want a top-down approach.
17:21
We don't want a bottom-up approach. We want a decentralized approach, shared across multiple levels based on co-design, agile, where multiple actors and again, not only the public sector can interact with each other and can generate value. Of course, this is where we see a role for Osgeo. Osgeo is a strategic partner for us.
17:42
We are already working and collaborating on a number of aspects. Again, follow the other presentations if you want to know more. But this is where we actually see an important role in the future, in this ecosystem. Finally, technological actions, of course. We should continue improving discoverability, accessibility, interoperability of data. Ensure technological neutrality.
18:01
We want to avoid vendor lock-in, of course. All the standards and technologies that we include should be well adopted, already mature. Avoid custom extensions of standards so that the requirements can be covered by already existing software implementations. And again, we are pushing, as JISCO is doing,
18:23
on the use of APIs, because really APIs are the key to make Inspiron SDIs accessible and usable, not just by geospatial experts as it was in the past for traditional SDIs, but to everyone. Optimizing data for, or data discoverability for search engines, because again, everyone,
18:42
or even more and more people, non-geospatial people, look for data using search engines. And finally, leverage on developments of federated European cloud infrastructures. I mentioned GAIA-X. We have just concluded the first pilot looking at the feasibility of using GAIA-X, which does not yet really exist in an operational way
19:03
for the purposes of Inspire. Very last slide to mention that we have written a report in the broader context of the European Green Deal with colleagues from the Commission and from GeoNovem, the Netherlands. Again, we're reflecting this report on lessons learned from Inspire, our experience, and our vision
19:22
for the future. So of course, the detail will be a bit higher than the one I gave in the presentation. There's the link here, but if you are interested, there are also some paper printed copies here. So feel free just to pass and keep one at the end if you are interested. And that's it. Thank you.