[
Lists Home |
Date Index |
Thread Index
]
> <Quote3>
> where Creo is the authority (identified by 'creo') for the namespace ID
> 'ppsg'.
> </Quote3>
>
> I would *strongly* advise not mixing these - let the registry tell you
> who the authority is. If the authority changed, all namespace prefixes
> would have to change as well (but I realize that you're not recommending
> registering namespace prefixes).
Well, maybe. I see your point, but in fact I occasionally come across
metatags in other types of documents (not XML) which carry the imprint of
their now-defunct originators. Doesn't seem to affect parsing or semantics
any... the markup still performs as required. Just looks a little dated is
all.
> Also, I think you meant to say "for the
> namespace ID that is represented in the XML document by prefix XYZ". It
> appears that there's an unintentional mixing here of namespace prefix/ID
> terminology.
No, it was quite intentional. I've been arguing that the ID *is* the prefix
in the proposed scheme. If, in your registry, prefixes are mapped against an
old-fashioned <g> URL namespace ID, is that mapping one-to-one? Or can
multiple prefixes be assigned to the same ID? Very different bird if it's
the later.
The point of my simultaneously humble and vain proposal was to 1) make
namespace IDs short strings, rather than URIs; and 2) prepend those to the
local names in a document(like namespace prefixes are now), thereby binding
_by proximity_ the local name to its namespace identifier.
Why? Because such a binding is about as portable and trouble-free as a
binding can be. You can't lose a name's namespace without really trying
hard. They're almost (if not quite) atomic. The namespace authority part of
the ID is merely there as syntactic device, void of any real meaning (but
yes, I suppose somebody would do some processing based on it, just like some
people are now treating Namespace URLs strings as web locations; no plan is
perfect or immutable).
Differences aside, a registry along the lines of what you're suggesting
seems worthwhile. It's a lot more (and different) functionality than what I
was suggesting, but at least yours is heading toward version 3.0. Not bad...
I wish you all the success in the world-- at least until you impact mine.
Take care.
|
