Images for the Future: Lessons learned from 7 years of digitisation

At the end of 2014, the biggest digitisation project ever seen in the Netherlands came to a close. Images for the Future was a 7 year project to digitise hours of audiovisual material and millions of photographs. In March, the closing event was held at the Netherlands Institute for Sound and Vision. Results and successes were celebrated, but the lessons learned along the way were also discussed. Here we present the main outcomes. 

The idea for Beelden voor de Toekomst, as the project was called in Dutch, started more than 8 years ago. The project proposal was a direct response to a call for digitisation from a number of cultural institutions. It aimed to save audiovisual material that was particularly old, such as early movies and newsreels, and which were being damaged by the ravages of time. The nitrate films that were once used were slowly disintegrating; waiting much longer would mean that these fragments of Dutch cultural history would be lost forever. In 2007 the project started with a total budget of 154 million euros, and a schedule to run for 7 years. 

After the allotted time, the desired results are there for all to see. An amazing amount of material has been digitised and restored, made safe for future generations. The rescue operation saved:

• 90,000 hours of video
• 20,000 hours of film
• 100,000 hours of audio
• 2.5 million photos

Besides the digitisation work and all the related challenges it entails, such as the storage and metadata, a major part of Images for the Future was to generate economic revenue out of the digitised material. In the proposal, it was agreed that the project would generate at least 64 million euros from the exploitation of the audiovisual material, leaving total costs at 90 million. 

Kaalgeschoren moffenmeiden by Nationaal Archief, on Flickr. No known copyright restrictions.

This target, however, turned out to be a lot more difficult than expected. Various efforts were made to sell the digitised material to various user groups. A Netflix-like platform where users could watch Dutch films for a monthly fee was built to give just one example. Numerous discussions were held with educational publishers to see if the material could be used in classrooms, where schools would pay a fee. Though many people researched and worked on these platforms, none turned out to be economically sustainable, and would cost more to maintain than they would generate.

In 2010, therefore, the conclusion was made that it was not realistic to earn back 64 million euros with the digitised material. The public were simply reluctant to pay for online content as most if it is already free. The project was therefore renegotiated and the payback obligation was dropped. This meant that the total budget also reduced to 115 million euros.  

Besides the long term goals, Images for the Future also ran a number of small pilots with open content. These so called ‘Pearl projects’ took a small set of the collections not restricted by copyright, and pushed them out to a wider audience. One example was the upload of black and white news photos from the National Archives to Flickr Commons. This resulted in thousands of views, with many users commenting on the pictures and thereby enriching what the archives knew about the images. Another example was Celluloid Remix where users were invited to remix public domain movie clips into new works of art. You can see an example below, by Victor Ponten.

How the world changed - a Celluloid remix by Victor Ponten

Even though the full report is only available in Dutch, some great lessons can be learned from the Images for Future project. Never before there has been a digitisation project of this scale. It shows the difficulties of making a direct financial profit out of a digital collection. More and more institutions come to realise this and instead use their digital collection for other purposes such as exposure or because they simply believe that giving the public access is part of their mission as a public institution.
 
This is what the Netherlands Institute of Sound and Vision has shown with their Open Images platform where videos can be found under open licenses or without any copyright restrictions. Although the 5000 videos that can be found there are a very small part of their collection (about 0.007%) it has generated millions of pageviews because they can allbe used in other places such as Wikipedia and Europeana. 

In the end, a lot of material has been saved for future generations, but most of it cannot be accessed by this generation. Copyright regulations mean that the majority of the material is not  made available online. The panel therefore concluded that it would be good to look again at copyright laws and make an exception for educational use of the material at the very least. Besides that, it should also be made easier for cultural institutions to publish works where the rights holders are unknown: so called ‘orphan works’. 

In summary, here are the lessons learned from 7 years digitising and making available content:

• It can be difficult to generate a direct income out of digital content. While there is some material which still has a commercial value in every collection, most of it will not be paid for. 
• While there is often no direct economic value, there is a great deal of social and educational value in digital content, but to achieve this, people need to be able to find, access and use it effortlessly. 
• Copyright turned out to be a far bigger obstacle than expected. Audiovisual material is especially problematic as it is not usually as old, and there are often multiple people who can claim copyright.
• Small and quick projects can have great impact. There is not always the need to wait for everything to be absolutely perfect before you can start sharing material.
• Even though large parts of a collection might not be suitable to publish openly, less is sometimes more: great results can be achieved with small, well-curated and open sets of content.