[
Lists Home |
Date Index |
Thread Index
]
----- Original Message -----
From: "Elliotte Rusty Harold" <elharo@metalab.unc.edu>
To: <xml-dev@lists.xml.org>
Sent: Saturday, January 19, 2002 10:55 PM
Subject: Re: [xml-dev] The task to be solved by RDDL. Re: [xml-dev] RDDL
(was RE: [xml-dev] Negotiate Out The Noise)
> At 10:58 AM +0100 1/19/02, Nicolas Lehuen wrote:
>
>
> >-- namespace-centricity, limiting RDDL to describing namespace meta-data,
> >whereas what we really need is to describe document meta-data (think of a
> >DOCTYPE declaration but not limited to DTDs).
> >
>
> The you need something other than RDDL. This is not what RDDL was
> meant to do, nor did RDDL ever claim to do this. A RDDL document
> describes a namespace, not a document. What you propose might be
> useful in and of itself, but it would replace RDDL.
Yeah, I do understand that RDDL is no what I need. The problem until now is
that it seemed that other people on this list didn't understood this. So the
whole purpose of this thread was to try to prove that describing a namespace
is not the same thing as describing the documents that make use of this
namespace.
> >While there are many namespaces that, right or wrong, can stand alone in
> >documents, and thus have associated schemas, there are schemas that
define
> >document with a namespace mix. Those schema don't have a particularly
> >associated namespace, so they can't appear in a RDDL document.
> >
>
> That's true in some schema languages, including DTDs. However, it is
> decidedly not true in the W3C XML Schema Language. Regardless,
> there's no reason I can't include a mixed schema as a resource within
> a RDDL document for just one namespace. I can even include a schema
> for a totally different namespace as a related resource within a RDDL
> document. RDDL is very flexible.
Yeah, but it is too flexible. You can list the DTD or RELAX NG schema for
RDDL in the RDDL document found at http://www.rddl.org/, but it will be
usable only by human beings, not computers, because there is no clear
algorithm to deduce the schema to use. If you take a RDDL document, you have
three namespaces, so three RDDL document to load, with possibly more than
three references to schemas (DTD, RELAX NG or other). How do I know which
one is the one I should use to validate the document ?
Anyway, I think that you have understood my point, now, so I won't need to
repeat it over and over in the future. I'm coming back from week-end, so I
won't reply to all mails from this thread to say the same thing over and
over :).
> >- As an illustration of the previous issue, it is rather funny to think
that
> >RDDL itself is out of its own scope. Granted, the RDDL documentation is
an
> >RDDL document. But like I wrote before, you cannot use the RDDL
resolution
> >mechanism to go to http://www.rddl.org/RDDL from a RDDL document.
> >
>
> Not sure why you would want to go there, but you most certainly can
> use it to go to http://www.rddl.org/, which is the RDDL namespace URL
Duh, my mistake. But this does not change my point, see above.
> >Show me the algorithm. I have asked for it twice, but nobody answered.
I'm
> >not asking for heuristics (like "I have a look at the namespace
> >declarations, see those XHTML, RDDL and XLink namespace names, and decide
to
> >go for the RDDL one") or science-fiction ("I fetch all RDDL documents
from
> >each and every namespace name I find, and merge them together"). I want
an
> >algorithm that can be easily implemented today, if the fate of RDDL is to
be
> >used today.
> >
>
> An algorithm to do what exactly? The algorithm to find the RDDL
> document associated with any given namespace URI is trivial, the
> algorithm to find all RDDL documents associated with any namespace
> URIs in the document only slightly less so. Of course, there is no
> algorithm to find the RDDL document associated with a given XML
> instance document because there is no such RDDL document.
That's why I suggest we should go further than RDDL, carefully explain to
the XML community why RDDL is not enough and why another solution is
required, then work all together on the problem.
> >And once you've shown me the algorithm, show me the same algorithm for
> >XHTML+SVG, WAP 2.0, and my proprietary schema that mix namespaces. Hint :
> >the algorithm has to be exactly the same.
> >
>
> Trivial, in each case load all the namespace URLs. (You'll probably
> get some exceptions because not all of these will resolve to RDDL
> documents.)
Once again, this set of RDDL document won't help you to find which schema to
use to validate the document, which is the idea RDDL was conveying in its
examples by listing schemas as resources.
> >-- describing document meta-data may not be the intended purpose of RDDL,
> >its author may purposely have restricted their scope to describing
> >namespaces meta-data. But in that case, this has to be written clearly
> >somewhere in the RDDL documentation, and a lot of use cases for RDDL
should
> >be marked as non implementable. There are people casting silver bullets
from
> >RDDL, which is the main reason why this thread is so big.
> >
>
> The authors never intended RDDL to manage nuclear power plants or
> describe airplane controls, either. Should they explicitly deny
> RDDL's use for those purposes? Do they have to deny everything you
> can't do with RDDL? RDDL's about describing namespace names. I never
> heard anybody except you and Paul T. claim it was for anything else.
As soon as the RDDL examples contain references to schemas, people can start
believing that RDDL is an indirection mechanism to find schemas for
documents. This enforces the apparently popular but definitely false
assumption that a namespace = a document type. The authors of RDDL may not
have thought of that, but the enthusiasm generated by RDDL seems somewhat
linked to the hope that RDDL can solve this indirection mechanism problem.
It cannot. So I thought, like other people on this list, that this should be
stated precisely, first on this list, and then somewhere in the RDDL spec.
Best regards,
Nicolas
>
> +-----------------------+------------------------+-------------------+
> | Elliotte Rusty Harold | elharo@metalab.unc.edu | Writer/Programmer |
> +-----------------------+------------------------+-------------------+
> | The XML Bible, 2nd Edition (Hungry Minds, 2001) |
> | http://www.ibiblio.org/xml/books/bible2/ |
> | http://www.amazon.com/exec/obidos/ISBN=0764547607/cafeaulaitA/ |
> +----------------------------------+---------------------------------+
> | Read Cafe au Lait for Java News: http://www.cafeaulait.org/ |
> | Read Cafe con Leche for XML News: http://www.ibiblio.org/xml/ |
> +----------------------------------+---------------------------------+
>
> -----------------------------------------------------------------
> The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
> initiative of OASIS <http://www.oasis-open.org>
>
> The list archives are at http://lists.xml.org/archives/xml-dev/
>
> To subscribe or unsubscribe from this list use the subscription
> manager: <http://lists.xml.org/ob/adm.pl>
>
>
|
