RE: [xml-dev] Semantic Web permathread, iteration n+1 (was Re: [xml-dev] InfoWorld agrees with Elliote Rusty Harold)

["Hunsberger, Peter" <Peter.Hunsberger@STJUDE.ORG>]

Elliotte Rusty Harold <elharo@metalab.unc.edu> writes:

> Beyond the issue of lying metadata, there's a much more important 
> distinction between metadata based search engines like the semantic 
> web and data based engines like Google. Providing metadata requires 
> extra effort which is roughly proportional to the quality and amount 
> of the metadata to be provided. There is a noticeable cost for a site 
> to add metadata. 

Maybe not.  The system we're doing here is almost entirely metadata
driven in the first place.  You can't add anything to it without first
defining the metadata for it.  We're collecting data, not managing
content, but for those with similar requirements I think this is the way
to go. 

> By contrast, there is negligible cost to provide 
> data for Google because this is the same data you're providing 
> anyway. Metadata isn't free. The semantic web is going to need to 
> provide really significant benefits to content authors to justify the 
> added costs of supplying useful metadata. Even if it does, I wouldn't 
> be surprised if much metadata is created by tools that screenscrape 
> the data and thus guess the appropriate metadata.

Defining the metadata is still work for a system like ours, but better
to define metadata than to write code.  Which of course points you at
your other opportunity for metadata extraction: reverse engineering of
the code....