- This article provided a lot of information about the history and development of databases. It helped me to think of a database in terms of its information structure, rather than simply as a product (i.e. a Microsoft Access database).
- Most databases today are in the relational model, with rows and columns: "The ordering of columns is immaterial, Identical rows are not allowed in a table, Each row has a single (separate) value for each of its column."
- Users request information from a database using a query, which is usually in the SQL language. The database then provides a result set (list of rows).
- Normal forms?
- Object databases are a result of the object-oriented trend.
Setting the Stage
- Metadata= data about data.
- All information objects have three features that can be expressed through metadata: content, context, and structure.
- "Library metadata includes indexes, abstracts, and catalog records created according to cataloging rules and structural and content standards such as MARC (MAchine-Readable Cataloging format), as well as authority forms such as LCSH (Library of Congress Subject Headings) or the AAT (Art & Architecture Thesaurus)."
- Types of metadata: administrative, descriptive, preservation, technical, use.
- As I understand it, Dublin Core is a metadata model (like MARC records) meant to be used for diverse, often electornic information items, including online resources.
- RDF = Resource Description Framework - metadata architecture for the web.
- "A resource can be anything that can be uniquely identified" - interesting definition. Also, an item does not have to be "retreivable" to be a resource.
- Dublin Core records are written in XML - a list of fields would be helpful.
- This article was a bit difficult to understand - it might be possible to find a better introductory description of Dublin Core.
4 comments:
In regards to the Dublin core model I wonder what the author means exactly by the fact that a resource does not need to be retrievable. Doesn't it need to be retrievable in order to be used?
What did you think of the database and metadata articles? I thought the database article needed a bit more explanation on the improvements to databases. The metadata article was the best of the three.
Elise -
I understood that statement to mean that a resource does not need to be an actual, physical object that can be found on a shelf somewhere. For example, a Dublin Core record could be created for a website, a digital file, even a community resource. I think the Dublin Core model makes it easier to create records for resources like this, when compared to other record models. That's just my understanding of it, though.
I agree with your explanation, Valerie. There are times that one can not actually access the material "in their hot little hands" (due to a number of reasons- including the condition of the original material), but the information contained in the material can be obtained through a copy or scan that has been made of the (in some cases) fragile original.
Post a Comment