|
|
|
It seems big names are moving (again) toward "federated" or distributed databases workingl with different types of data. This time with a solution based on XML. See an interesting page on c|net and the Xperanto announcement of IBM. This seems a pure XML - based approach, no things such as RDF mentioned there.
Problems were and will always be the same there but, XML can simplify data handling and light protocols can simplify query distribution.
I believe this is still "half" of the story. RDF (or something similar) will be needed to describe with an unique language different (up to extremely different) resources available on the network in a real-world application. This seems powerful than usage of XML only.
At least, we are touching the real problem :-)
Posted by zac at January 23, 2003 02:12 AM | TrackBackThis was actually what I worked on for my masters degree - Federated databases... a while a go but still..
Posted by: David on January 23, 2003 03:42 AMyes federated databases, distributed indexing and search of data is one of the big challenges for RDF. Semi-structured databases or graph databases are also a much more appropriate technical words for it. Big chanllenges are in the indexing, retrieval and query model, which are completely general and different from traditional RDBMS.
I will blog some stuff about semi-strucutred data and about on-going efforts about SQL and RDF to integrate and query different databases
Posted by: Alberto on January 23, 2003 06:58 AM