The National and University Library of Iceland has systematically collected Icelandic web pages since autumn 2004. Our web archive is far more thorough in Icelandic material than the comparable Internet Archive, which collects web pages from all over the world. The web archive is enormous and now holds almost two billion items (harvested URLs minus duplicates). As well as letting people look at websites as they were in the past at vefsafn.is (for example the websites of political parties before the last election), this large collection of Icelandic websites offers all kinds of possibilities for research on Iceland’s digital contemporary history and for digital archaeology.
For example, it is interesting to look at the share of different file types in the archive. It turns out that a large majority of the digital items are web pages (text, html, css, xml and so on), almost 70% of the harvested URLs. Next come images of various kinds, at 23% of the harvested URLs. PDF files approach 1%, while other file types, such as video and audio files, are far below that.
If you look at the volume of data instead, web pages/text make up almost 60% of the archive, images about 14%, PDF files about 6%, video about 10% and audio files about 4%. For example, the archive holds about 750,000 different audio files, or about 2.5 terabytes of audio, with an average file size of 3.5 megabytes. It is also striking that there are many more video files than audio files in the archive. There are 1.7 million video files, with an average size of 4.2 megabytes. In total there are therefore almost seven terabytes of video in the web archive.
The largest file category apart from text and images is PDF files. The web archive holds about 8.2 million different PDF files, 3.7 terabytes of data in all. The average size of the PDF files in the archive, however, is only 477 kilobytes.