Please keep
voting for us
as a Top 100
Web site!
|
|
|
Above and Beyond: Understanding Metadata
(Part 2)
By Nate Zelnick
Learning to SMIL
SMIL provides two ways to group media objects: parallel and sequential. These are described using the elements <PAR> and <SEQ> which operate as you would expect. For example, to create a presentation that displays an image while an audio file plays, you would use the Sequential element to group these, like so:
<SMIL> <HEAD> </HEAD> <BODY> <SEQ> <PAR> <IMAGE SRC="/old?u=http%3A%2F%2Fwww.webdeveloper.com%2Fxml%2Ffoo.gif&y=1999"> <AUDIO SRC="/old?u=http%3A%2F%2Fwww.webdeveloper.com%2Fxml%2Ffoo.wav&y=1999"> </PAR> <PAR> <IMAGE SRC="/old?u=http%3A%2F%2Fwww.webdeveloper.com%2Fxml%2Fbar.gif&y=1999"> <AUDIO SRC="/old?u=http%3A%2F%2Fwww.webdeveloper.com%2Fxml%2Fbar.wav&y=1999"> </PAR> </SEQ> </BODY> </SMIL>In the example above, the "foo" picture and voiceover are grouped as a parallel element, which means the two objects are shown together and presented until the "foo" audio file finishes playing. Then the "bar" objects are presented until the "bar" audio file is finished. While each of the combined objects play together (in parallel), they are contained in a larger sequential element, which means they play one after the other.
This is a very simple demonstration of SMIL. The language includes very sophisticated timing and switching mechanisms as well, which I'll outline in a future article.
The interesting thing about metadata and SMIL is that it allows binary elements that otherwise could not be easily categorized by their internal content to be described in a straightforward, human-readable fashion. This is the key concept in how XML will help the Web to run more efficiently--particularly by making it easier for search engines to categorize anything--including pages and sites--so that searches are more likely to return the document that users are looking for instead of the all-too-common information overload of today.
Today, HTML offers a way to describe a site using the meta element Keywords, but this is too loosely structured to provide real benefit, and has been horribly abused by companies that include hundreds of irrelevant words to increase the likelihood that they will be found in a search. Search engines today often parse the contents of a page and then weight a document based on the number of times a word appears in it. When a user enters that word in his or her search query, the document with the most appearances of that word appears as the first hit.
A meta content format that defines a fixed set of content descriptors and their relationship to each other would make that process much more efficient. Of course, this will require that Web authors follow conventions and include that description in their documents and sites. There is an XML proposal that will make this possible--and a whole lot more--called the Resource Description Framework that is now working its way through the W3C. We'll take a look at that next time.
Copyright © 1999
internet.com CorporationAll Rights Reserved. Legal Notices.
Contact the WebDeveloper.com staff
Last modified: Friday, 29-Oct-1999 05:18:20 EDT
Refresh Daily
Join Editor-in-Chief David Fiedler
and find truth, justice, and a clue or two.
ActiveX / VBscript Animated GIF Archive Browsers CGI / Perl Database Connectivity E-Commerce HTML-Advanced: DHTML, CSS HTML / Site Authoring Tools Intranet/Groupware Java JavaScript Multimedia: Audio / Video / Streaming Technologies Opinions Refresh Daily: Editorial Column Security Servers & Server Tools Site Design / Graphics Site Management / Marketing / Log File Analysis VRML / 3D XML
