My experience of the presented results of the testbed Geo4Web 17 March
Last Thursday the 17th of March I visited the presentations of the results of the testbed Geo4Web organized by Geonovum in Amersfoort. Prior to the session I did not know much about this so called testbed. I heard about this session of freelancer and former colleague Jelmer Oosthoek of gispla.net. A quick Google search brought me to the website of Geonovum where the descriptions caught my interest. The participants of the testbed work in an open way and have set up a github.
When I arrived at the location in Amersfoort my first impression was not that good. I did not see any familiar faces from my personal GIS network, most people were at least 10 years older than me and most organizations present were governmental. At that moment I imagined a few sale orientated private organizations presenting with no real new content to me. I could not have had a more wrong first impression! I had a great day; it was incredibly interesting and full of content new to me.
The presentations started out with an introduction of Linda van den Brink of Geonovum. This introductory story was recognizable. Web developers do not know how to work with geodata and there is an imaginary wall between the two groups. In addition to this imaginary wall we are expecting the new Environmental Act (Omgevingswet) in the Netherlands. This new Environmental Act aims on the cooperation of multiple disciplines. Multiple (all?) disciplines make use of geodata, which makes it logical (to me) that the data should be standardized and available from a central location. It should be as easy as possible for the user of the data making choices with economical, ecological and societal implications.
This introduction started my enthusiasm and understanding of the testbed. On to the content!
A usable spatial data publication platform (Spotzi)
The first result of the testbed was an interactive web platform where data can be shared and interaction can be set up between multiple parties. This was presented by Teun van Sprundel of Spotzi. Spotzi was present with multiple young ICT/GIS professionals. At the start of the presentation I (again) made a terribly wrong assumption. A lot of organizations make geo-webviewers nowadays and I figured this would be just one out of many. With some nicely presented use cases it became clear to me that this was something different.
Some well thought about features make the webviewer of a high quality. First of all the interaction between a municipality and citizen was easy to use and clear. In the presented use case a citizen could simply and spatially suggest to change the location of a bike shed. Then the municipality automatically receives an email and can accept or refuse the proposition. The municipality employee, who was played by an employee of Spotzi, could then in the webviewer adapt the shape. This without the need of getting the data in a local GIS viewer and re-publish it again to a server.
I also found it a strong aspect that social media was taken into consideration (implemented and suggested) and have the opinion that they handle data in different coordinate systems well.
Crawlable spatial data (Apiwise)
The second researched part was presented bij Joost Farla of Apiwise. This presentation was quite an eye opener for me. I never heard or thought about crawlable spatial data. I am not sure if I fully understand what was meant with it but it sure proved to be important. Geodata is not found with current search engines. Google does not find spatial data sets. Nowadays you need to be familiar with an SDI to be able to find the datasets. Apiwise showed that Google and Bing do not correctly index geodata published on the web. Only a minimal percentage of their testdata was indexed, which apparently is linked to the searchability of the data.
Furthermore I found a remark made in the presentation important to mention. The remark was something like: ‘Web developers are spoiled and will only use data with documentation’. I forgot which dataset was used as an example, but lack of documentation caused Apiwise to not make use of it. This is a recognizable issue with datasets in the Netherlands. There is data available where only a small amount of people are familiar with the meaning of certain columns within a dataset.
Finally I found a planned next step of Apiwise a bright idea. Apiwise will publish a questionnaire for developers to get international feedback. Developers can teach us how they want to work. This was the first time I saw an international component of the testbed. I imagine the challenge of the testbed to be a global one and definitely not a Dutch one.
A question asked after the presentation is in my opinion also worth mentioning. The question asked was something like: ´Who should apply standard formats of the data? Should that be the government? I can’t image a municipality having the knowledge of multiple formats.´ The question was left unanswered, as it will stay in this post.
Spatial data on the web using the current SDI (GeoCat, Interactive Instruments, Linked Data Factory)
The last presentation of the day aimed on spatial data on the web in the current situation and was presented by Paul van Genuchten of GeoCat and Clemens Portele of interactive instruments. It had some overlap with the presentation of Apiwise and was directly connected to their subject. Similarly it was stated that search engines and indexing of search engines is a black box to us, and that there was no documentation for the Dutch AAN (Agrarisch Areaal Nederland) dataset.
Furthermore I wrote down statements presented or talked about during the presentation. All these statements are in my opinion incredibly interesting, and thus worth mentioning here:
- In current SDI’s the data provider decides how data is published
- How to get large datasets indexed in a search engine is a question that remains
- Suggestion: Maybe we can learn what open data works best by looking at available open data now
- The search engine world is hardly compatible with the linked open data world
First of all I want to come back to my initial impression described at the start of this post. The technical level of the presentors and the audience was amazing. It was far above my expectation. I could hardly follow some technical parts of the presentations and asked questions by the audience, while I worked in the GEO-ICT.
Also, I believe that the scope of the testbed is an important one in our changing society. The new Environmental Act will bring and need change of current procedures. Next to that developments in IT constantly change our work and life.
I do have one critical note that popped up in my head multiple times during the day. I missed the involvement of large scaled governmental organizations. There was financial involvement but no content wise involvement. It put the presented results far out of reality to me. I can understand this for the technical part of relating geodata to web developers. I think that this part is correctly placed out of the scope of governmental organizations. But not for the involvement of the Environmental Act!
I think that the limitation, willingness and possibilities of the existing governmental institutes delivering the data should be a big (the biggest?) component of this testbed regarding the Environmental Act. Just to put it in perspective of reality. What is realistic? Should we force governments to do it in a certain way (standardized dataformats, metadata, one platform)? Can we do that? Can they live up to those standards? How far can we go with standards? Who is the we I am talking about now? How do they perceive this inevitable SDI coming in the line of the Environmental Act?
If I personally philosophize I would imagine the best scenario would be to create a new governmental agency. The Dutch data agency. Which takes its form by taking over, centralizing, standardizing and improving Dutch data platforms like PDOK, Statline and Nationaal Georegister.
Furthermore I found it unfortunate that international comparison was left out. I am curious to how other countries approach these technical challenges we face. Although they will probably not have our Environmental Act. They do have national SDI’s, web developers and geodata. Maybe it is the case that the Netherlands is a front runner in most of these topics. I like to believe we are. But maybe, just maybe, we could find another country where a web developer automatically makes use of geodata. Where the imaginary placed wall between GIS and web developers has been broken down.
Otherwise it would be nice to know that we, this small country, act as this front runner that perhaps will improve and educate the world!
Overall I had a great day. I loved to be able to freely join a session like this with all the interesting content, interesting people and just a general good organization. My compliments and thanks go out to the organization of the session.