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

Re: XSchema question

• From: "Thuy-Linh Nguyen" <tln@i...>
• To: xml-dev@i...
• Date: Fri, 7 Aug 1998 20:50:26 +0000

> >Hey!  Nice idea!  XSchema certainly supports this -- it's just a bunch of
> XML
> 
> 
> Thanks.  Support is, I suppose, unintentional?;-)
> 
Hi !

Could I bring this topic back again ? :-)
I'm a novice in this X-World in that I haven't implemented 
anything, and I have to "run at full speed to just stay still" with
all the new specs :-) (In fact I haven't read about XSchema yet ! :-)
Could somebody give me the reference ?). But I'm thrilled to hear
somebody have a similar idea to mine :-) Don, you are certainly not
alone in this world ! Now I would like to contribute some of my 2c :-)....

>> 1) You've probably already realized this, but a DTD for such a file
>> would be of little or no use.  Because each XSchema section can
>> introduce new elements and redefine old ones, the DTD would probably
>>consist of a bunch of elements with content models of ANY.  This is of
>> no use either for validation or determining storage structures on the
>> fly.

>You are right but it makes perfect sense for transitory documents
>which exists only while it is moving from one place to another. 
>Ability to redefine default attribute values should be enough of a
>benefit I think.

Do you have an example of this Don ?
Also I think it makes a lot of sense with regards to
schema evolution. I think this theory will give a lot of inputs for
the design of an evolvable system. I'm working on this and have
also got some papers in the last WWW8 and SCI'98 conferences. We
do have some believers ! :-) As people are now free to create their 
own DTDs I think we would expect to have many DTDs created even in 
the same domain. The XSA and OSD just posted here is one example. Why 
don't we reused the elements that have been defined in other DTDs ? 
Why don't we have OO in XSchema so we can inherit features from other 
DTDs ?... And the problem of old XML document with new (dynamically 
generated) DTD that Peter mentioned in the log file example can 
probably be resolved using the conformal and polymorphism rules. And 
the dynamic schema will come in to automate the evolution process...

Just a few thoughts...

TL
****************************************************************************
Thuy-Linh Nguyen
CSSE, PO Box 197, Monash Univerisity, Caulfield East, VIC 3145, Australia
Ph: 61-3-9903-2041, Fax: 61-3-9903-1077, http://www.sd.monash.edu.au/~tln/
****************************************************************************

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@i...
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo@i... the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@i... the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@i...)

• Follow-Ups:
  • LISTRIVIA (Personal Prefixes, Namespaces and Unique Personal Identifiers)
    • From: Peter Murray-Rust <peter@u...>

• Prev by Date: XML DMS and indexing ?
• Next by Date: LISTRIVIA (Personal Prefixes, Namespaces and Unique Personal Identifiers)
• Previous by thread: Re: XSchema question
• Next by thread: LISTRIVIA (Personal Prefixes, Namespaces and Unique Personal Identifiers)
• Index(es):
  • Date
  • Thread