Empowering multiple sources of data
Ross Gardler
collated under a single entity which is given a URI that is local to
Simal. This works well, however, it creates a problem when we create
multiple sources of data. Specifically, it becomes difficult to track
the original source of each data item.
This is important as we look at reuse of Simal data in third party
applications. For example, OSS Watch's use of the data is different from
the JISC's use of data, which in turn is different from Cetis's use. It
is feasible that all three of these use cases would want a project
description targetted at their user community whilst all three will use
the same home page information. It is therefore important that we are
able to record the original source of a description (for example).
I propose to change the import of data to allow each tuple to be
attached to its original source document. The Simal entity will then
collect references to all available data sources.
Users will then be able to configure Simal to have a hierarchy of source
preferences. For example, OSS Watch may say that the OSS Watch
descriptions should take priority whilst the JISC contact details are
more important.
I will extend the data access classes to ensure that the hierarchy of
sources will be navigated to ensure the most appropriate available data
is returned. This means that OSS Watch, for example, need only record
data of particular relevance to themselves, allowing Simal to discover
other data elements held by other sources.
If anyone has any thoughts about this approach please step forwards.
Ross