RE: [ubl-cmsc] Summary

[Dan Vint <dvint@acord.org>]

At 09:44 AM 4/17/2003 -0400, Grimley Michael J NPRI wrote:
>Greetings,
>
>Aside from the risks identified on the W3C site,
>
>         http://www.w3.org/TR/xmlschema-0/#Redefine
>
> 
>http://www.w3.org/TR/2001/REC-xmlschema-1-20010502/#element-redefine ,

I belive most of the issue from these sources is in regards to if you have 
used all the various ways of modifying content.

Even if it is focused on just multiple derivations of say Address, I think 
this problem exists for any of their methods for changing content - the 
warning is you need to know what you are doing and how your schemas are 
built. Maybe that should be put on the cover - sort of like the 
Hitchhiker's Guide to the Galaxy "Don't Panic".


>my main concern is the fact that, in the XML instance, there is no direct 
>way of telling that someone is using a 'redefined' UBL component rather 
>than the 'original'. I'm not sure this is a showstopper, but things like 
>that bother me...

I am trying to pull the sample together.

This issue is the one that I was sort of rediscovering yesterday on the 
call. I stopped on the sample and did a task I was planning to do for a 
while now which is to try and document in one place all the methods for 
changing content and address it as extension methods. This will be an ACORD 
document and will have stuff from our discussions, but I think it will help 
us have a single place to reference and refer to the methods.

It turns out the adapter schema method (multiple files) might apply as a 
way to manage the extensions for redefine as well as what was described in 
the original documentation.

..dan


>Thank You,
>Michael Grimley
>MJGrimley@acm.org
>
>-----Original Message-----
>From: Monica J. Martin [mailto:monica.martin@sun.com]
>Sent: Wednesday, 16 April 2003 23:31
>To: Lisa Seaburg
>Cc: Eduardo Gutentag; Context Methodology
>Subject: Re: [ubl-cmsc] Summary
>
>
>As an aside, I think that UCC decided in their architecture not to allow
>the use of redefine, but I don't know why (if my memory serves me
>correctly).
>
>Monica
>
>Lisa Seaburg wrote:
>
> >Eduardo,
> >We had a short discussion about redefine in NDR about a month or more ago.
> >We didn't know enough about it to make an informed decision, and so let it
> >go.  I think it is very appropriate that you bring it forward now, with use
> >case and purpose.
> >
> >Lisa
> >
> >----- Original Message -----
> >From: "Eduardo Gutentag" <eduardo.gutentag@sun.com>
> >To: "Context Methodology" <ubl-cmsc@lists.oasis-open.org>
> >Sent: Tuesday, April 15, 2003 12:06 PM
> >Subject: [ubl-cmsc] Summary
> >
> >
> >
> >
> >>I will not dignify what follows with the name "minutes" ;)
> >>
> >>We met this morning at 8am Pacific.
> >>
> >>Present were Eduardo Gutentag, Dan Vint, Michael Grimley, Matthew Gertner.
> >>
> >>We discussed Dan's suggestion of using xsd:redefine as the enveloping
> >>
> >>
> >methodology
> >
> >
> >>for implementing the CM; Dan explained the concept and its use a bit more,
> >>and all present agreed there is a lot of merit in it.
> >>
> >>Eduardo's memory appears to be faulty (like we didn't know that
> >>
> >>
> >before...),
> >
> >
> >>as there seems to be no record of NDR ever agreeing not to use redefine,
> >>
> >>
> >or
> >
> >
> >>even talking about it.
> >>
> >>Dan will send a complete example (hopefully short) of the use of what he
> >>has been calling adapter schemas, following in the steps of the vocabulary
> >>used in xml-dev.
> >>
> >>Eduardo is expanding the Introduction to include things we take for
> >>
> >>
> >granted
> >
> >
> >>but should not be all left to references to other papers. He will send a
> >>new version soon, like today if possible.
> >>
> >>Eduardo is still waiting for Arofan to fulfill his AI from last week,
> >>
> >>
> >namely
> >
> >
> >>send references to other papers or specs.
> >>
> >>Eduardo will present the concept of redefine and adapter schemas to the
> >>
> >>
> >NDR
> >
> >
> >>group during their meeting on Wednesday 23 (but only if Dan is there
> >>
> >>
> >too...)
> >
> >
> >>--
> >>Eduardo Gutentag               |         e-mail: eduardo.gutentag@Sun.COM
> >>Web Technologies and Standards |         Phone:  +1 510 550 4616 x31442
> >>Sun Microsystems Inc.          |         1800 Harrison St. Oakland, CA
> >>
> >>
> >94612
> >
> >
> >>W3C AC Rep / OASIS TAB Chair
> >>
> >>
> >>
> >
> >
> >