[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: RDDL (was RE: Negotiate Out The Noise)

• To: 'Jonathan Borden' <jborden@m...>, 'Paul T' <pault12@p...>, "'Bullard, Claude L (Len)'" <clbullar@i...>, 'Leigh Dodds' <ldodds@i...>, "'xml-dev@l...'" <xml-dev@l...>
• Subject: RE: RDDL (was RE: Negotiate Out The Noise)
• From: Nicolas LEHUEN <nicolas.lehuen@u...>
• Date: Fri, 18 Jan 2002 16:44:58 +0100

>-----Message d'origine-----
>De : Jonathan Borden [mailto:jborden@m...]
>Envoyé : vendredi 18 janvier 2002 16:09
>À : Nicolas LEHUEN; 'Paul T'; 'Bullard, Claude L (Len)'; 'Leigh Dodds';
>xml-dev@l...
>Objet : Re:  RDDL (was RE:  Negotiate Out The Noise)
>
>
>Nicolas LEHUEN wrote:
>"...I'd like to remind you that RDDL is not the only way to do this,
>far from it, RDF being another serious candidate. RDF is not 
>perfect either
>for this purpose..."
>
>[jb]
> RDF, TM, XML Schema, RELAXNG, and natural language text. are 
>all better
>than RDDL at describing "Resources". RDDL is _simply_ intended 
>to serve as a
>mechanism that allows all the above to happily _coexist_. I believe the
>'Semantic Web' will not be dicted from a single source and use a single
>language or protocol, rather pieces of it will spring up much as the
>internet itself arose from a patchwork of connected networks. 
>RDDL is simply
>a 'router' that allows an intelligent machine (or human) to 
>pick from any or
>all of the above ways to describe a 'resource'.

I do retain your term 'intelligent machine', because in its current
incarnation RDDL is going to required truly intelligent machine to handle
some documents containing mixed namespaces, beginning with RDDL.

>------
>Human-readable documentation is another important problem, with lots of
>specifics consequences, amongst which the need to deal with
>internationalisation. Come on, you can't be serious about 
>using xml:lang.
>Browsers don't support this, and the whole objective of RDDL is that by
>pointing my browser to a RDDL URL, I can see some 
>documentation appear in my
>browser, right ? So either you tackle this problem with 
>browser friendly
>solutions, or you drop the idea of stuffing ALL resources and 
>documentation
>in the same document.
>
>[jb]
>Please solve this problem and I am more than sure that XML-DEV 
>will be happy
>to modify RDDL to include this solution to the problem of
>internationalisation. We work with what we have. "xml:lang" is 
>what we have.
>Not perfect I agree.

1) Nothing to do with internationalisation, but please drop the 1 document =
1 namespace axiom. To link a document to a RDDL document, don't pollute the
namespace spec and simply use specific attributes just like XML Schema does
with its xsi:schemaLocation (xsi mapping to the proper XMLSchema namespace)
attribute. RDDL can then be seen at the ultimate catalog of meta-data
resources associated to the document, amongst them the DTD, schemas and
human-readable documentation.

2) Have a translation of the human-readable documentation in as much
languages as you need or want, all written in XHTML if you want (but plain
text is really an option). In your "root" RDDL document (the document found
at the URL specified in the special attribute mentionned above), have
arcroles of "human-readable-documentation" kind, with, yes, xml:lang
attributes to qualify them. Since the "root" RDDL document is no longer
readable by a browser (or could be, with a clever CSS or XSLT stylesheet),
there is no problem in using the xml:lang attribute anyway. What is
important is that the human-documentation resources found at the resources
referenced by the proper arcroles are displayable in a browser, not the
"root" document.

I know this doesn't allow the simple pleasure of typing a namespace URI in
your browser and seeing something appear. But first if you really want to
see something appear at the end of a namespace URL, it should be a NDL
document, not RDDL, NDL being much more restricted in scope than RDDL (well,
in fact, restricted to describing meta-data about namespaces, not XML
documents).

Second, we could make the browsers support RDDL (through CSS and/or
stylesheets) much easier than make them support xml:lang, which could
eventually lead to the followig scenario : when pointed to a RDDL document,
the browser either displays it in XML format, or style it with a CSS or XSLT
stylesheet, or it could even find the "human-readable-documentation" arcrole
with the proper xml:lang attribute (corresponding to the language settings
of the browser and/or client OS) and display the human readable
documentation at once.

Regards,
Nicolas

• Follow-Ups:
  • Re: RDDL (was RE: Negotiate Out The Noise)
    • From: "Rick Jelliffe" <ricko@a...>

• Prev by Date: RE: RDDL (was RE: Negotiate Out The Noise)
• Next by Date: Re: RE: Silver Bullets and Partial Solutions
• Previous by thread: Re: RDDL (was RE: Negotiate Out The Noise)
• Next by thread: Re: RDDL (was RE: Negotiate Out The Noise)
• Index(es):
  • Date
  • Thread