Metadata standard that is universal. Can be given a flavour such as for legislative documents.
Most documents can be described by their title, author, who published them, what format, and many other characteristics. Dublin Core is a system for standardizing this information also known as metadata within a document. By including the information in a standard format, the documents can be more easily found and manipulated. However, the standard does not generally specify how to include the information within the Dublin Core fields so it is important to try and develop ones that best considered a standard especially through industry based consensus.
And within the One Big Database the Dublin Core data stored in XHTML, RSS and other XML formatted documents provide a hook that can relate disparate documents. For example, if you have a movie review, you can use Dublin Core within the head of the XHTML document to associate the review in a definitive way with the movie. In the example below, the Dublin Core subject of the document is the URL for the movie that I considered authoritative (and perhaps the industry standard now).
<link rel="DC.subject" href="http://www.imdb.com/title/tt0033467/" />
Then using some search engines or your own XQuery/XPath (see example below), your review can be easily associated definitively with other movie reviews about the same movie as opposed to every document that merely references the movie or coincidently has the same words as the movie title.
declare namespace q = "http://www.w3.org/1999/xhtml";
//q:link[@rel = "DC.subject"][@href = "http://www.imdb.com/title/tt0033467/"]
See RSS 1.0
Alternative: MODS (Library of Congress metadata standard based on MARC standard)