Vous êtes ici

Born to die? Selection policies and digitisation as common concerns

Title (author1): 
First names (author1): 
Surname (author 1): 
National Library of Norway
Presentation type: 
spoken paper
30 Sept Wednesday
Start time: 
Salle 70

Over the last years there has been a rapid increase in the number of digital files produced in several archives around the world. That is no surprise and the growth is naturally related to developments in new technology and the even more widespread use of Internet services. The latter also reflects the number of web documents stored in national archives responsible for harvesting the web. Even though there is a common consensus regarding digitisation as the primary tool for preserving our audiovisual heritage, you may raise the question if all digitisation really is needed – or even wanted.
Selection policies are based on various criteria, but generally it seems to be easier to acquire material for preservation rather than deselect as such. Furthermore, several archives (including the National Library of Norway) have simply stated that all their collections should be digitised. Such programmes will take several years; therefore priorities have to be made right from the beginning. Deteriorated and obsolete carriers are obviously a starting point, but equally important is the content itself. Even if you choose the latter; infrastructure and the importance of workflow may cause a large number of documents to be duplicated (even though the source or digital files may differ slightly). This in turn requires high-level metadata and sophisticated storage systems for retrieval and access. Any lost or hidden data file usually means the non-existence of such! Therefore the need for advanced search engines is as important as the content itself. Generally, however, coping with duplicate files and countless media files is a major challenge irrespective of available tools. Focussing on bibliographic control, cataloguing any analogue or digital document may be more time consuming than the digitising process itself. Unfortunately, I have also experienced that retrieval of digital documents may prove to be very time consuming.
Broadcast archives seem to be in a special position simply due to their focus on reuse and access more than long term preservation. Several broadcasters have already ripped a major part of their CD collections and carry on digitising vinyl records, 78s and other formats of audio and video, including even rebroadcast programmes. Obviously, a large number of duplicates are being produced. Furthermore, imagine how many duplicated audio files that exist of a Beatles or Madonna song in archives around the world (and you may even add the number of various downloads from the Internet).
Deselection is likely more difficult in the digital domain compared to analogue material. But it may be possible to map out duplicates even when comparing metadata and media files from various archives around the world. For instance, this makes sense if you look for the best available shellac disc from a sound perspective. And, what about master tapes or other production material? Should we preserve this kind of material? The paper will also discuss selection policies in the light of recommended audio and video formats. Finally, I will present a new tool for dealing with born digital material, the National Library of Norway’s Media Editor.