Metadata for one archived document Which NWA collection is this document stored in. Element should start with country code (no, se, etc). Then an optional ':' plus an archive dependent string. Characterset in the archived document. The URI by which this document can be retrieved from the archive If the document references other documents using frameset each frame src= link should be listed If the document contains inline multimedia references reference should be listed All <a href= links should be listed All <a href= linknames should be listed The text one wish to present in the query result list e.g. a part of the text between and DC metadata in document. If non-existent use content of html title-tag DC metadata in document. DC metadata in document. if non-existent use the content part of : <META name="keywords" content=.... DC metadata in document. If non-existent use content part of : <META name="description" content="...."></META> DC metadata in document. DC metadata in document. DC metadata in document. DC metadata in document. The MIME-type of the archived document e.g. "text/html" DC metadata in document. DC metadata in document. DC metadata in document. ISO.639-1 language code. If non-existent use the language detected. DC metadata in document. DC metadata in document. DC metadata in document. DC metadata in document. If non existent use last-modified from http-header else date of harvest. The textual content of the archived document i.e. the text between and in the html document. All markup should be stripped off Datatype for storing a list of links Root element