xml-dev - Re: Question About Namespaces and DTDs

["Simon St.Laurent" <simonstl@simonstl.com>]

At 01:03 PM 7/26/00 +0100, Matt Sergeant wrote:
>Since then many people have gone away and actually produce working
>programs that use namespaces effectively. Witness the applications of RDF
>and XSLT to name just two.

Working applications on that level, though promising in their own right,
are barely a sign of a functioning standard.  

We're still not entirely sure how to handle relative URI references, though
the W3C seems to be moving toward temporarily banning them.  We're
definitely not sure what if the entity body retrievable from the URI
reference should contain or mean anything, and there's not even very much
agreement about which URI schemes are appropriate for namespaces.

Complex compound documents are barely getting started.  I'm looking forward
to seeing them, but at the same time, I have some looming concerns that the
philosophical issues surrounding namespaces, especially the URI issues, are
not likely to become less controversial.  I expect that to have a sizable
impact on interoperability as well, but we'll see what happens.

>Using the same argument almost every concept in computing is broken to
>some degree or another. Java is broken, HTML is broken, HTTP is broken,
>MIME is broken, ASCII is broken, Unix is broken, Windows is broken, etc
>etc. All of those things are true. But some of us just build applications
>working around the defects, rather than striving for perfection.

Perhaps, but it seems reasonably clear that namespaces do much of the
'breaking' in XML.  I'd love to know (and don't have time to figure out)
what share of posts on XML-Dev over the last two years have been
namespace-related.  Seems to be a recurring cycle.



Simon St.Laurent
XML Elements of Style / XML: A Primer, 2nd Ed.
http://www.simonstl.com - XML essays and books