INSPIRE 2010 Conference Workshops (22 June) Krakow, Poland

INSPIRE logo smaller

The INSPIRE Conference started with workshops on Tuesday (22 June); the conference summarised under a separate blog. In general, these were organised as four presentation for each session with six sessions for the day. Really not workshops, but decided to attend the following:

  • GIS4EU, a Cooperative Project to Apply the INSPIRE Directive
  • European SDI Best Practices in Dialogue
  • Data Quality and Metadata for Evaluation and Use within INSPIRE (two sessions)

The GIS4EU sponsored workshop provided a summary of activities to date with the announcement of the project ending soon.  Multiple partners, with Intergraph playing a key role, the aim has been to develop a common data model which European local, regional and national mapping agencies can use to implement and deliver cross border accessible and interoperable base cartographic datasets for aggregation.  The INSPIRE Directive Annex I data themes (administrative units, hydrography, transportation networks and elevation) were targeted.

The presentations delivered during this workshop session described the methodology and the results of GIS4EU process impact analysis, which provided a review of the operational implementation steps developed for European SDIs to implement the INSPIRE Directive. This analysis used 50 datasets coming from 10 local/regional/national data providers distributed in five different European countries.

COMMENT: Are these issues for spatial data deposited on ShareGeo and displayed on Go-Geo! using WMS? The main theme of many of the papers presented is about the harmonisation of data (eg: scale, format).

The eSDI-NET hosted the  ‘European SDI Best Practices in Dialogue’ workshop. Presentations were from several finalists for the eSDI-NET award.

Alan Moore from Forth Valley GIS presented as one of the nominated organisations. His emphasis was on approaching organisations and institutions first before doing the technical implementation.  Engaging the users was critical and the reason for success of Forth Valley GIS.

The Iderioja Project presentation provided an overview of the Spatial Data Infrastructure successes achieved for the Government of La Rioja (Spain).

X-Border SDI presentation covered collaborative SDI developed between North Rhine-Westphalia, Germany and the Netherlands. The aim of the programme is to develop and implement a structure for the supply of cross-border geographic information.

COMMENT: These presentations demonstrated that there can be collaborations in terms of data management and sharing delivered through portal services, but it is important to approach users first to understand their demands before any implementation takes place. We should never assume that we understand what they want without approaching the users first. The last presentation also showed how there can be cooperation at the organisational level as well, which would be good to achieve with UK academia at the institutional and departmental levels.

The Data Quality and Metadata workshop was intended to be invitation only for those involved with this project to receive feedback from EU countries about the importance of the use of metadata to describe the quality of spatial data for the purpose of fitness. Basically, there was a to be a delegate from each of the 27 EU countries. The task for him/her was to disseminate a position paper/questionnaire  to key people in each country, then report reactions/responses at this workshop; however, of the 27 countries, there were only 15 people at the workshop, some of whom the original contact delegate must have appointed to represent him/her.

I won’t go over the details of each report, but then the following countries were represented, but had not disseminated the position paper yet, or were planning to do so later:  Austria, Denmark, Finland, France, Italy, Norway, The Netherlands and Slovenia.  The UK (Environment Agency), Sweden, Spain, Poland, Belgium and Czech Republic did provide feedback, but then a one or two from a personal perspective rather than feedback. Actually, the UK had the most respondents with 9 reporting back to the Environment Agency.

In summary, without the details of the questions posed, quite a few didn’t understand, and found problematic, the conformity elements. The Belgian (Flemish) delegate suggested the creation of a tool or ‘stamp of approval’ to address conformity; the Czech Republic delegate suggested that INSPIRE provide some sort of  certification to simplify; both seem like excellent ideas considering that almost a majority of those who commented, had issues with understanding conformity, though there seemed to be general agreement of the need for it.

Regarding other DQ elements, most noted the importance or relevance of lineage, many thought it would be useful to have case studies to provide greater clarification of the DQ elements to users. Data transformation was covered in this position paper, but almost no one had any comments to make with regards to this other than keeping it simple- ‘first iteration, description only’ was one comment.

Spatial accuracy wasn’t really noted other than one from Swedish delegate who noted that their data was very accurate, so didn’t need to be addressed.

Another question in this position paper pertained to the support of additional DQ metadata elements, but most delegates did not believe this should be the case; French delegate believed so, and Slovenian, Spanish and Belgian delegates noted that their profiles already had some of the DQ elements.

The discussions that ensued reinforced the notion that ‘more’ isn’t better, and should be left to individual countries. Someone from the OS suggested that only title and contact details necessary for metadata- at least quoting Mike Goodchild, but then I talked to him at GISRUK and he said metadata was important, and though he would suggest that less is better, he said it was important to describe how a spatial dataset was used, along with contact details, title, description and the other core elements.

This is a separate matter, but point is that there seemed to be an aversion to having more metadata, especially the DQ metadata, but would this mean making it mandatory within existing INSPIRE guidelines or optional? It would seem that making these optional would be the best road to take, but then there were the comments of leaving this to each country to decide, but I’m guessing that no one would object if these were included as long as they were not mandatory elements, which is the case for UK AGMAP 2. Granted, the lineage element is already mandatory, so uncertain why this question was included in the position paper?

There was support for INSPIRE setting the framework for DQ and conformance if DQ were to become a reality, but conformance is already an issue with some confusion about it and how to address it?

Antti Jakobsson, from Eurogeographics, and involved with ESDIN Data Quality project, suggested to organisers that the ESDIN work could be reviewed and considered for DQ. He noted that the ESDIN working group were finding some errors in this DQ for INSPIRE metadata project.

Is it necessary to have these two separate DQ projects? Perhaps the ESDIN DQ project’s results will suffice and serve the INSPIRE community, but again, assuming none of the elements is mandatory.

There should also be discussions about creating quality statement support for services? Downtime and service performance are important and critical to know in the event of accessing data during emergency response mapping. I’m not sure if ISO 19119 supports this sort of detail, but it was mentioned time and time again about how many people live so close to borders and how so many are vulnerable to natural disasters such as floods, so it seems that cross-border responses are very dependent on services in adjacent countries and local authorities, so something to explore.

In the meantime, the focus is on providing support for data quality statements.

This was the it for the INSPIRE workshop programme on Tuesday. There were a couple of other interesting sessions in the afternoon, but with this ESDIN DQ project, this last one seemed more relevant to attend? There was one on  ‘Implementing Rules for the INSPIRE spatial data services‘, but it was invitation only as well, but as mentioned, the delegate turnout was so low that the rest of us were allowed to stay and participate after presentations had been completed.

This entry was posted in Conference. Bookmark the permalink.

One Response to INSPIRE 2010 Conference Workshops (22 June) Krakow, Poland

  1. Dan Cornford says:

    Thanks for the very useful write-up, particularly on the DQ side. I really struggle with this issue of fitness for purpose – this requires such a lot of semantic / domain understanding to describe what purpose etc, and the purpose is a property of the user not the data. I see things as being more simple – data quality is about the relation of the data to reality (maybe traditionally called accuracy, but might be called quality or uncertainty). This information is a property of the data (thus is metadata, or sometimes it might actually make more sense to think of the data as being inherently uncertain). It is up to the users to define their own requirements for data quality and decide whether it is fit for purpose. You cannot predict all the uses that a data set will have in advance, and what accuracy requirements users will have. This desire for simplicity is understandable but a false economy. It significantly reduces the value of the quality information.

Leave a Reply

Your email address will not be published. Required fields are marked *