Development of Prototypes
The National Library’s web archive is currently developing four different prototypes aimed at providing researchers with access to materials in the web archive collections. These services offer URL search, replay of archived web pages, full-text search, and map visualization.
URL Search
The URL search allows the user to search for material that we have indexed based on the resourcer’s web address when it was archived. In addition to searching for exact URLs, the service has advanced features for filtering by, among other things, media type.
Playback Service
![Stortingets forside fra 17. november 2006, rekonstruert i prototypen for visningstjeneste.](https://beta.nb.no/content/uploads/2023/02/Skjermbilde-2023-02-17-kl.-16.58.36-1200x842.png)
The front page of Stortinget (the Norwegian parliament) from November 17th, 2006, reconstructed in the prototype for the playback service.
The playback service is a so-called “Wayback Machine”. It reconstructs web pages based on resources we have archived. For researchers, the playback service offers both opportunities and challenges.
The strength of the playback service lies in the ability to emulate web pages, providing a user experience closely resembling how they were once viewed on the internet. However, academic use of the service requires a critical evaluation of the page being reconstructed and its various components. When you retrieve a webpage “online”, your browser collects the different resources within seconds and assembles them. The various resources in your browser are, in this sense, synchronous. When you reconstruct a webpage using archive material, it can contain elements that were collected over a longer period, spanning hours, days, and sometimes even months in terms of harvesting time. Therefore, the relationship between the different elements being reconstructed on the webpage can potentially be quite diachronic. If your primary focus is to analyze the text within an HTML file, this may not be a significant issue. However, for researchers examining the webpage as a whole and the relationship between its various elements, there are exciting opportunities for the development of analytical methodologies.
Not unlike how archaeologists view the reconstruction of objects as an experimental method for generating hypotheses about the past, the playback service is also a form of experimental digital archaeology that requires critical examination of the origins and context of the various fragments.
Full-Text Search
By extracting natural language from HTML files and indexing them in a database, we have developed the capability for full-text search within the web archive collections.
The service locates documents that match the search query you provide and returns information about the harvesting timestamp, original URL, and a brief text excerpt (similar to Google). For relevant research or documentation purposes, it will also be possible to access and inspect the documents in the playback service.
Map Search
![Kartsøk etter norske nettaviser (2005-21)](https://beta.nb.no/content/uploads/2023/02/Skjermbilde-2023-02-17-kl.-17.29.10-1200x815.png)
Map search for Norwegian online newspaper (2005-21)
We are experimenting with maps and geographical presentation as an alternative interface for finding historical websites. In the prototype, Norwegian online newspapers published between 2005-21 are presented on a map, with links to URL search and the playback service.
If the map solution proves to be useful for end-users, we can expand the interface to display publications from entities such as public institutions, journals, and political parties, for example.
Contact
Are you a researcher and would like to learn more about the prototypes? Please feel free to contact us at nettarkivet@nb.no