SAX RFD: ModSAX Predefined Features

Dan Brickley Daniel.Brickley at
Mon Mar 8 13:13:13 GMT 1999

On Mon, 8 Mar 1999, David Megginson wrote:

> MikeDacon at writes:

>  > Since some finite set of SAX features will not approach a global naming
>  > problem, I strongly urge not to use a URI.  
> I disagree here -- if third parties want to be able to define feature
> names, they need a way to avoid collision (i.e. we want to make
> certain that both Oracle and Sun can define properties like
> 'normalize' without blowing up the whole system).
> That said, the Java package naming scheme also provides DNS-based
> uniqueness, as in 'org.xml.sax.features.validation'.  It's simply a
> matter of taste:
> - org.xml.sax.features.validation is more of a Java flavour.

Yep... but might not feel so natural for developers working with
versions of SAX translated for Perl, Python and so on.

> - is more of an XML/Namespaces
>   flavour

...and RDF [1]. Giving interesting entities URIs makes them more
fully a part of the Web, and means we can take advantage of
URI-oriented metadata. Eg. you might search a software database for
resources that were of type 'Perl Module' and that implemented the 
feature  known as ''. (There's
already a Linux Packages Database[2] along similar lines...). I'm not
claiming that this would be impossible using the Java naming scheme,
just that a Web oriented approach might make it easier to do certain



Daniel.Brickley at                
Institute for Learning and Research Technology
University of Bristol,  Bristol BS8 1TN, UK.   phone:+44(0)117-9288478

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at
Archived as: and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at

More information about the Xml-dev mailing list