The API allows anyone to search for and retrieve the metadata that describes records in the archive in XML format. The data can then be used without restriction or charge. But the archive, which is simultaneously an executive agency of the Department of Justice and a government department in its own right, continues to charge £3.50 per document to retrieve actual records online.
Aleksandr Drozdov, enterprise architect for the National Archives, said it had been speaking with academic historians to help them find ways to use the API in their research.
"It's completely open," he said. "It gives access to all meta data we have in the system. I have spoken to some scientists who do historical research and they are interested in getting access to our data in this way."
The API allows search and analysis of meta data such as a physical description of the physical record held at the National Archives' site in Kew, along with a summary description, an administrative history and details of its origin. The National Archives is talking with other public and private bodies about merging their archive catalogues with its own.
Aleksandr Drozdov's team is consolidating numerous existing electronic archives, either by porting them directly or putting them in a service wrapper that can communicate with its unified system. The unified system uses MongoDB, the open source, document-oriented database used by the New York Times, and is built on a Microsoft software stack. It has used a service oriented architecture based on neuronESB.NET, an enterprise service bus for the Microsoft platform.
The National Archives is also working an e-commerce system. The API may further boost the market in genealogical services. Private companies have paid the National Archives £52m since 2002 for the digitisation of paper records they could then resell through websites. Findmypast.co.uk paid for the National Archives to digitise the 1911 census last year and pays a share of its profits on resale of the data back to the National Archives.
A spokesman for the National Archives was unable to say how much money it made from such deals or what it cost to digitise the 80 million pages of the archive it has already put online. He said the 1911 census comprised 16.2 million images. Most of the National Archives' 11 million records are still on paper, each containing pages that can number in the thousands. It contains collections unlikely to be digitised for lack of public interest. Access to paper records is still free of charge at the archive's site in Kew.
The API would theoretically lead to the creation of such services as the geo-location of entries in the Domesday book from a mobile phone. Recent additions to the online archive include files from MI5 and government UFO files, both of which were made available free of charge for one month before the imposition of a £3.50 fee per file.