Posts Tagged ‘Europeana’

ePSI Platform Conference 2012

March 21, 2012 Leave a comment

On Friday the 16th of March, the European Public Service Information (ePSI) Platform conference was held in Rotterdam. More than 300 guests from all over the world gathered for what turned out to be a very busy and interesting day. The big turnout of the conference showed the huge current interest in Open Data.

The ePSI platform is an organisation working to stimulate and promote Public Service Information (PSI) re-use and open data initiatives. They work to achieve the goals of the PSI Directive, which was created in 2003, and encourages EU member states to make as much public sector information available for re-use as possible. Now, almost 10 years later, there is still a lot of work to be done. Instead of embracing the idea of open data, many large public organisations are fighting to maintain the right to charge costs for their information. It is in response to this that the European Commission proposed its ‘Open Data Strategy‘ in December 2011. It includes the following proposed changes to the European PSI Directive:

  • All data made available by government institutions must be able to be generally used for commercial and non-commercial purposes;
  • In principle, the costs charged by government institutions may not exceed the costs involved in the individual request for information (marginal costs – in practice usually free of charge);
  • an obligation for government institutions to provide data in common machine-readable formats to ensure that information can actually be re-used;
  • Member States must introduce regulatory supervision to monitor compliance with the aforementioned principles;
  • information from libraries, museums and archives will also be eligible for re-use.

From the Open GLAM perspective, the last change of the directive is of course very interesting. It would mean that all the European cultural memory institutions have to make their publicly funded work freely and openly available. It is important to notice here that this will only include their metadata, that is the data about the actual cultural objects that they hold. This includes author/year/location etcetera. By making this data freely available for re-use, data from cultural institutions can be linked to other collections and also be reused in new and innovative applications. A lot of traditional institutions are still anxious about this idea since they fear that they will lose control over their data, and this is just one of the concerns. The whitepaper “The problem of the Yellow Milkmaid” shows a more thorough study about the potential benefits and perceived risks of open metadata for cultural institutions.

When cultural heritage institutions are included under the purview of the PSI Directive, this will improve citizens access to our shared knowledge and culture and should increase the amount of digitized cultural heritage that is available online. At the end of 2011 however, the Dutch government expressed some concerns about the idea of including libraries, archives and museums. The main reason for this is that they believe that it will become too much of an administrative burden for the institutions to conform to. The Dutch government suggested instead that institutions should make their data available on a more voluntary base through for example the Europeana project.
During the presentation sessions about cultural heritage, Richard Sweetenham (Head of Unit, Access to Information at the European Commission), gave his response to Dutch government’s line on the matter. He said that he could not think of a reason why cultural institutions should not be included in the directive; the data is already there and the institutions are not only are funded with public money, but also have a public mission. The content of an archive, museum or library only has value when it is found and used. It gets even more value when the data is formatted in such a way that it can be linked with data from other cultural institutions from around Europe and all over the world.

After his talk, Harry Verwayen, business development director at Europeana and David Haskiya, product developer at Europeana, showed the value proposition of open cultural heritage metadata. To make the most out of this data, institutions should not be afraid to publish their metadata under a CC0 license. Waiving away all rights of the data sounds scary, but it actually enables them pursue their public mission more successfully, while still controlling the copyright of the actual digitised object. A more thorough study about the impact of the proposed amendments of the PSI directive has been done by the Communia association and can be found here.

The next couple of months will be crucial for the PSI directive. All updates can be found on the ePSI platform website.


MA Thesis: Europeana Building a European Identity

November 8, 2011 2 comments

Last month I was finally able to receive my Masters diploma from the University of Amsterdam. My research about Europeana and the European identity has been found interesting enough to let me pass. However, there are still a lot of questions remaining after finishing this research. My supervisor Theo Thomassen commented that in order to really build a complete study, one more year of study is probably required.

Hereby I want to thank Theo Thomassen for supervising me, as well as many other people I have met during this year in the MA New Media at the University of Amsterdam. Beforehand , I actually did not expected it be so interesting and fun. The focus on many different skill sets, like blogging and theoretical analysis, but also on more practical research in the Digital Methods class, really gave me a better understanding in so many different fields. The last course about datavisualization was very extensive, especially working together with so many different people from different studies, but very interesting and it really opened my eyes about the possibilities of this kind of analysis when studying huge datasets.

At this point I can put MA in front of my name, which is a good feeling. I still however believe that I am just only starting to delve into the material and hopefully I will be able to continue exploring a field that has so many interesting aspects.

Anybody who is interested in my MA Thesis: It is freely available under a CC-BY licence and can be found here.

Categories: Europeana Tags: , , ,

Europeana Thesis Conclusions

August 3, 2011 Leave a comment

The last few weeks, I have been working non-stop on my final thesis. At some points, it has been pretty hard to come to a good conclusion. A lot of expectations I had in advance appeared not to be correct.
In the end, I believe I have made a good analysis of Europeana in relation with a European identity and the cultural policy of the European Union.

In short: here are a few of my conclusions.

1. The European Union must not try to become a ‘United States of Europe’. The cultural strategies the EU uses to construct a European identity are similar to the strategies that national states use. For example the introduction of a European anthem and a standardized passport. In terms of cultural heritage: it is easy to claim that all of the German and French history is also a part of European history. This assumption however will not lead to an idea of a European idea of a shared history. The European identity consists out of the variety of all the different states in it and is constantly changing. The EU must try to show its ‘unity in diversity’ and the Europeana project is more than suitable for this since it combines all of Europe’s heritage.

2. The current interface of Europeana is not able to show the cultural objects in a European context. A search query leads to thousands of results presented in a 4×3 grid in no particular order. When an object is clicked, it is shown individually on the website of the contributing institution. This way, the European context is not present. In fact, all context has disappeared. Europeana should strive to show this context. This can be done in two ways. The first is to let experts like historians and archivists create a new European story. By combining primary and secondary sources from different countries and institutions, new relations and contexts can be shown. The second option is to create spaces where users can combine, view and discuss their own objects. The recently released Europeana API proves to be an excellent tool for this.

3. Fix the metadata. At the moment, tests with the Europeana API show that a lot of the objects do not have standardized metadata. This makes it hard for programmers and users to use and combine different cultural objects. For example, one institutions uses ‘1867’as a date mark, while an other uses ’19th century’. This makes it impossible to combine these objects in for example a timeline. Luckily Europeana has also noticed this problem and recently released the ‘aggregators handbook’ and a dummy space where aggregators can test if all fields are filled in correctly

4. The current trend is that the idea of feeling European is decreasing. In fact, it is now lower than at the start of the cultural program in 1992. However, it appears that this idea is very much related to economical and political factors. At the moment, during the economic crisis, less people have the idea that their country benefits from the European Union. At the same time, more people tend to feel less European.

5. The Europeana project fits perfectly in the cultural goals of the European Union. Because of its role as an aggregator of aggregators, it can also very easily adopt new institutions, aggregators and even new members countries. This allows Europeana to show the diversity and commonalities of the European culture, as well as new stories and insights in the history of the world. This way Europeana can become the representation of the diversity that unites Europe.

Right now, an English peer reviewer is taking a look at my thesis. After it is enhanced, I will put the entire thing on my blog.

Categories: Europeana Tags: , ,

Europeana and the possibilities of data-visualization

June 2, 2011 2 comments

I just received my grade for the paper I wrote for the datavisualization course.

In this paper, I discuss Europeana and their problems of showing their objects in a European context. By visualizing the data in different ways, the objects can tell a story about the history of Europe. Here, I also discuss our visualization project and how we tried to achieve this goal of showing the European context of the Europeana project. I left the comments of the lecturer, Bernhard Rieder, in there to show some critique points.

Click here to Download

What’s that on the map? Problems with geo-visualization

May 17, 2011 1 comment

The last couple of weeks, my data-visualization team and I, have been working on our Europeana project. Europeana is a big heritage-digitization project funded by the European Union. Their goal is to digitize all of Europe’s heritage objects and to make them available online. There are several reasons why the EU wanted to create such a huge and expensive project. One if them is a political one. The idea of Europeana is to let European citizens discover the history of Europe. By showing the cultural treasures of each country in all its glory, the public should get more admiration for the rich history of Europe. At the same time, a project on an European scale allows both the user and Europeana itself to show links between several European countries. Europeana has the potential to go beyond the national histories of each country and show the relations each country had with each other throughout history. This way it aims to encourage a more European identity, next to the national identity of the user.

One of the problems at the moment of Europeana, is that some countries have contributed far more objects than others. For example, the French have at the moment a total share of 19% of all objects within Europeana. At the same time, a country like Spain has contributed only 9%. This means that at this point, the European identity, as seen through Europeana, is rather French. At this moment, the Europeana home page, which is designed as a portal, does not show the differences in size between countries. It only allows the user to search through their entire database. Our project group decided that by visualizing it, these differences would become much clearer and our project was born: Visualize a huge dataset from different countries in such a way, it shows the differences and links between countries. To solve this problem we decided to visualize the data of Europeana on a geographical map of Europe. By doing this we would emphasize on the fact that this is a European project and at the same time it would become possible to show all the differences between the countries in one screen. Because of this choice of a geographical map, I will now emphasize on some of the benefits and problems of using maps for datavisualization.

When using a geographical map, the creator should think about the implications this has on the information that it wants to show. Placing artificial elements on map can easily be misinterpreted for several reasons. A good example of how this can go wrong is given by Kaiser Fung on his own blog Junkchart. In this visualization, the creator wants to compare different earthquakes by visualizing them.

Here he makes clear how artificial data placed on a geographical map can be both confusing and wrong. First of all, it looks like one of the earth quakes was in the middle of the Pacific Sea. The creator means however, that there were two earthquakes both in Christchurch in New-Zealand. By creating a line from the bubble in the Pacific to Christchurch, this is made clear. The viewer however, sees first of all something he recognizes, a map of a part of the world. In that element recognized by the the user, is an artificial element placed. The creator of this visualization wants to clarify these two earthquakes in Christchurch by drawing a white line to it. This makes it even more confusing as it now appears that there is a weird stick popping out of New-Zealand. An other element in this visualization is the size of the bubbles. It looks like the earthquake in South-America has a damage range that covers almost the entire country. This is however, not what the creator wants to show. The size of the bubble represents the magnitude of the earthquake, something that becomes clear when one of them is clicked. An other disadvantage that geovisualizations have, that becomes clear in this visualization, is the fact that the circles are not next to each other. This makes it even harder than circles already are to compare with each other. It is hard to tell which earthquake was the most powerful when looking at this map. This visualizations shows that adding artificial elements to a geographical map can be confusing and misleading.

The most obvious, but also one of the biggest problems is the fact that a geographical map has areas in different sizes. As for example in this map where trending twitter words are shown on a geographical map.

The map is completely covered with words, showing almost nothing of the countries. It looks like there are far more trending words in bigger countries than in for example, the Netherlands, this while in the Netherlands, far more tweets are send than in Spain, despite its difference in population. In this map, it becomes very unclear where the tweets are coming from. The word ‘omkleden’ can be found in the middle of Poland, this while it is clearly a Dutch word and it is not very likely that the Polish people suddenly adopted a Dutch word in the morning. In the case of showing worldwide trending topics, it is not very useful to use a geographical map without adding any other information to it.

To overcome this problem, Stephen Few argues that it is useful to combine several quantitative displays next to the map in one view. This way it allows the user to look at the data from different perspectives, simultaneously. In the example of the Twitter map, it would be useful to add a screen where the user could select a certain country to show its current trends.

A map has the benefit that it shows a lot of information at the same time, however, when creating a visualization one must always keep in mind what the map is not showing and think of the consequences of this. In this example about the job loss in the United States it appears that a lot more jobs were lost in the area of New York and California. However, this map does not hold in account the population size of each state. At the same time, the amount of jobs lost is counted cumulative. This way the states with more citizens will appear always as the biggest losers/winners.

Using maps for in a visualization is a very appealing option. Because of the knowledge the user already has of a geographical representation of the real world, it is easy to add artificial elements to it in order to create a new story. However, as several examples show and as we found out during our Europeana project, lots of things can be done wrong when using a map. Because of this mixture, it quickly becomes hard to see what this visualization wants to show. When creating a visualization, the creator should carefully select the elements it wants to include in the map and if necessary, combine other visual displays within or next to the map in order to get a better understanding of the data.


Few, Stepehen (2009). “Introduction to Geographical Data Visualization”. Visual Business Intelligence Newsletter March/April 2009.