Skip to page content or skip to Accesskey List.
Search evolt.org
evolt.org login: or register

Work

Main Page Content

A brief introduction to metadata and the semantic

Posted on January 25, 2001

in Backend

by peter van dijck (pedrito)

Rated 3.77 (Ratings: 6) (Add your rating)

Log in to add a comment
(4 comments so far)

Want more?

 
Picture of pedrito

peter van dijck

Member info | Full bio

User since: October 22, 1999

Last login: August 30, 2005

Articles written: 23

(terms used, links and metadata about this article at the bottom)

The problem

Web documents lack metadata.

Sure, some sites use some of the metatags, and the title tag gets used a lot, but even this limited amount of metadata is poorly defined (what is supposed to go in that title tag anyway?) and poorly used.

This lack of metadata makes it really hard to search the web, and limits it's use. It's really hard to search for related documents for example.

The solution

The core of the World Wide Web Consortium's (W3C) metadata vision is a concept known as the Semantic Web. The semantic web means that all the stuff out there has metadata with it that describes it.

But is the W3C then going to define thousands of metatags for every possible subject? Don't think so, Jose.

Enter the Resource Description Framework (RDF).

The resource description Framework is an XML-like standard for defining vocabularies that define metadata.

These vocabularies can then be used to describe documents. There already are a bunch of vocabularies out there (mostly for describing books), that can be used, but anyone could invent a new one. Market forces decide which ones get popular.

The future

This is all of little consequence to web developers right now. But anyone involved in content management should watch this space, interesting times are ahead.

Terms used

Metadata: metadata is data about data. I'll illustrate with a few examples: the author of a book is metadata. The category a library classifies a book in is metadata about that book. The title argument of an tag is metadata.

Related links:

The power of metadata

http://openp2p.com/pub/a/p2p/2001/01/18/metadata.html
(a really good introduction at http://p2p.com)

What is RDF?

http://www.xml.com/lpt/a/2001/01/24/rdf.html
(really good as well at http://xml.com)

Some metadata about this article (you may have noticed, more is provided by evolt):

Author: Peter Van Dijck, aka pedrito
(for example, for meta data on the author, click on the author link at the top of the article)
Title: Introduction to metadata and the semantic web.
Inspiration: the above articles
Date of writing: Friday, January 2001
keywords: metadata, semantic web, p2p, Resource Description Framework, rdf, xml, content management.

Peter Van Dijck is an Information Architect with an interest in localization, accessibility, content management systems and metadata.
  • poorbuthappy.com/ease Weblog
  • petervandijck.net Portfolio
  • Easytopicmaps.com
  • liga1.com Accessibility and localization
    • 4 comments on this article. Log in to add your comment | Rate this article: Rate this node 1 star.Rate this node 2 star.Rate this node 3 star.Rate this node 4 star.Rate this node 5 star.

    RDF? Not better yet..

    Submitted by bezalel on January 30, 2001 - 02:03.

    Please correct me if I'm wrong but RDF is just another model for metadata. Why is this better than the common used meta tags? Maybe I'm not so aware of this new XML definitions, but how many annotation systems are there to use RDF databases? A couple? It's so new that many people who would like to call themselves "web designers" don't have a clue about. Anyway, I like Pedrito's opinions and I don't miss a chance to read what he has to say. Thnx for the Info anyway. It was enlighting. Maybe useful.. in the future.

    login or register to post comments

    RDF Metadata

    Submitted by obershaw on January 30, 2001 - 13:40.

    I don't know if I would say RDF is a model for metadata. RDF is an XML based schema for marking up the metadata. RDF uses XML namespace facility to point to the standard for the metadata contained within an RDF record. Therefore you could use any existing metadata schema or make your own as necessry and put the URL for that standard in the namespace. If you look closely at a simple RDF document (http://purl.org/dc/index.htm.rdf) you will see that it is in fact very simple. It just has some XML and RDF declarations at the top and is followed by the actual metadata between typical <> and tags. The document ends by closing the RDF document. It's no harder than the basic HTML shell. There aren't any really good automatic RDF editors/creators available yet, but they are being developed check here for one such project: http://www.ukoln.ac.uk/metadata/dcdot/ How is RDF better than common tags. For one thing, RDF can point to a specific metadata standard much more efficiently than a tag. Through an XML namespace this could also be used to validate the RDF document. Second, RDF is more structured. It is extensible in that you can use more than one schema to mark up the metadata of a given document. This part would take too long to explain so if you're interested check out Eric Miller's great introduction to RDF: http://www.ukoln.ac.uk/metadata/resources/dc/datamodel/WD-dc-rdf/ The Mozilla project is also doing some fascinating things with metadata. http://www.mozilla.org/rdf/doc/ RDF is simply a means to the ends of helping people get at information they need with higher precision in a much more efficient manner. If implemented correctly, searching the internet would be as accurate as searching well structured database such as a library catalog. While it will be a long time, if ever, before metadata is employed on a large enough scale to make searching the internet any easier, it is being used more than people think for niche applications in digital libraries and on corporate intranets (the subject of my own research). One well known example is the Linux Documentation archive at ibiblio.org (formerly Metalab, formerly SunSite). They are using an XML schema they call OMF which is almost exactly the same as RDF and in fact are considering making a switch to RDF. http://www.ibiblio.org/osrt/omf/#search

    login or register to post comments

    The future (agreeing with obershaw)

    Submitted by pedrito on February 1, 2001 - 03:54.

    As far as I can tell right now, the only use for things like RDF is if you're building content / knowledge management systems, and only if you're working with documents that are going to have a long life span. In that case it is definitely worth looking into. For webdesign per se it will be a few years until you need to concern yourself with this I believe.

    login or register to post comments

    Content/Knowledge management is a broad subject

    Submitted by MartinB on March 28, 2001 - 05:20.

    You could view content/knowledge management as a much broader area than you seem to be implying. If you view your content as important (and who doesn't?) then the taxonomy of that content also becomes important to enable other systems to classify information... which implies the entire Internet potentially.

    As far as RDF is concerned, I'd be very interested in it if I was at all keen to have my content syndicated by other services.

    The long-running and broadly accepted standard for metadata is the Dublin Core Metadata Initiative (standard typology of elements and usage guide). Yes, the work I'm currently doing will all be involved in a formal Content Management process, but simply for improved search engine knowledge of your content, this kind of methology is useful.

    login or register to post comments

    The access keys for this page are: ALT (Control on a Mac) plus:

    evolt.orgEvolt.org is an all-volunteer resource for web developers made up of a discussion list, a browser archive, and member-submitted articles. This article is the property of its author, please do not redistribute or use elsewhere without checking with the author.