The Anker Kirkebye Collection **

Context

The Anker Kirkeby collection is a historical collection comprised of recordings of cultural value, these are from the first half of the 20th century and are digitized from LP records and 45rpm records.

It contains voice recordings of important Danish figures, sound clips from the 1930's, Danish journalism from London during the 1940-1945 occupation, etc.

The collection is already digitized and the data originates from a relational database, thus it heavily influenced by the Old Collections pattern

Description of pattern

The original data is structured around the relational database from which it originates, there is several exported database files, these reconstruct the structure of the database tables, from the following fields can be extracted:

Danish

English

Dublin core

Notes

ID

(identification string)

not reused

varighed

(duration)

sprog

(language)

1..*

udgiver

(publisher)

titel

(title)

ophav

(origin)

rolle

(role)

1..*

beskrivelse

(description)

datooptagelse

(date-of-recording)

datodigitalisering

(date-of-digitalization)

kilde

(source)

resourcetype

(resource-type)

bidragsyder

(contributor)

0..*

emneord

(keywords)

0..*

alttitel

(alternate-title)

0..*

klassifikation

(clasification)

0..*

lydfiler

(sound-files)

1..*

omraade_spatial

(geo-spatial-area)

1..*

område_temporal

(time-range)

1..*

rettigheder

(rights)

bidragsyderrolle

(contributor-role)

The ID will be discarded since it contains no semantic data.

Any given entry may contain multiple sound-files, alternate-titles, contributors,

Problem description

To a great extend the problem consists of being able to migrate the original database to the DOMS, this is a technically trivial task since the ECMs facilitate this quite well. The main difference is that the DOMS solution will combine all the data about a given recording into one document, fields which have multiple entries can be represented in the XML format.

Solution

In bold font

Consider next