[
Lists Home |
Date Index |
Thread Index
]
Thomas B. Passin said:
> That is a different kind of thing, Eric, that's all. I wasn't trying
> to identify document types.
I must have read this out of context, sorry but I thought that's where we
had ended up...
And I do think that we have a problem for defining document types even if
this not the point of the discussion ;-) ...
> But even in this example, say I wanted to
> be sure what dc:publisher really was supposed to be, so I can ut the
> right data in there. "dc" is pretty familiar - I would probably say
> to myself, "hey, I bet that is Dublin Core, let's see, the URL has
> purl.org/dc, so I am going to pull out my Dublin Core info and check it
> out" (and the first thing I will do is to check the namespace, because
> I don't remember it exactly).
I agree that prefixes are easier to identify and memorized for human
readers, but that's also dangerous: that's what makes it difficult for us
to spot XSLT transformations using the old namespace for instance!
I see namespaces as a hack which is doing what it's been designed for (no
more, no less) and has unexpected side effects (such as the fact that we,
human readers, attach more importance to the prefix than to the URI and
also such as QNames in values which abuse the xmlns declaration).
Eric
--
Freelance consulting and training.
http://dyomedea.com/english/
------------------------------------------------------------------------
Eric van der Vlist http://xmlfr.org http://dyomedea.com
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------
|
