Reading Notes for 10/04
Database Wikipedia
The whole concept of database infrastructure and management operates to create categorizational partitions to break data down into manageable chunks for search and retrieval, particularly the use of indexing, and the initial structure has to have flexibility built in for growth or the speeds of data maintenance and/or retrieval will decline exponentially. The locks are an easy and ingenious way to make sure you don't have different versions of the same object floating around, creating confusion about what data is the most current and accurate. It seems that the external levels of databases are similar to the GUIs of operating systems.
Intro to Metadata
It would seem advantageous for the different information and preservation communities to put a greater effort into adopting and enhancing a system like the MOAC for greater cross-disciplinary access. The development of different methods of description and searching in hierarchical metadata is a positive development to create a balance between expert and amateur researchers rather than adopting a dumbed-down system to increase accessibility for all to the detriment of the expert user.
While the amount of user-created metadata and folksonomies has become very popular, it seems unlikely that standards can be applied to those descriptions unless they arise organically in a social community. And does a full-text digital surrogate qualify as metadata? I'm not very clear on that.
Overview of the Dublin Core Data Model
The DCMES, built upon the RDF, is creating a semantic description system utilizable across disciplines. Using this schema will hopefully increase the ease of cross-disciplinary research and the interoperability of existing discipline-specific databases. The ability for semantic refinement is a large part of what makes the DCMI relevant, by linking its generic terms in relation to more specific terms used by a discipline's standard metadata system.
No comments:
Post a Comment