Re: Minutes from XML Schema Patterns for Databinding F2F 27-28 Feb 2006

----- Original Message From: <paul.downey@bt.com>

Excellent notes (thanks).  I just had a few comments:

> ISSUE-19: Advice against using the 'all' model group
>
...
> pauld: are we suggesting 'all' doesn't appear in our basic patterns?
>
> yves: well, everything is ordered in memory
>
> jonc: do toolkits generate 'all'?
>
> pauld: not sure

A number of tools do.  Typically an array records the order of elements 
within a document.

> ISSUE-23 - Use of Mixed Content datatype
>
> dezell: what do we do about sending HTML, mixed='true'

FWIW - we would only suggest XHTML in XML and then doing <xs:any 
namespace='???xhtml???'>...

> msmq: proposition - suggest mixed content drops to DOM, should be easy to 
> implement
...
> RESOLUTION: close ISSUE-23 with advice that tools should support mixed as 
> a DOM node, but YMMV

Mixed is an area where we are weak.  But it is an area where we would like 
to do better.  Therefore, if this route is taken we would prefer not to be 
limited to dropping to DOM if possible so that we can go beyond that.

Also, are you meaning pure DOM as in the API, or DOM-like?  i.e. lots of DOM 
wouldn't be required for this use-case (such as attributes etc).  A binding 
tool could potentially do a lot better on the typing of the data than DOM 
typically allows etc.

> ISSUE-20 - Extension of collections
>
> dezell: it's horrible, no matter what you do
>
> pault: sounds like it comes from ASN.1 - cites pattern of extending at the 
> end

(Irrelevant comment: I'm amused that because it's horrible, it must be 
something to do with ASN.1 :-) )

> pault: can't we suggest extension at end in a different namespace

I think this weakens to notion of what extensions go where (unless you 
pre-define unique namespaces for each possible extension point).

> pauld: do you mean identify extension for v2 in v1 structures?
> ... coming to the conclusion we won't have a basic pattern for versioning, 
> but we'll have to show our working

My personal feeling is that, horrible (even embarrassing) that it is, 
versioning should be considered in the basic profile.  Implementers 
typically want versioning, there are ways to do it, and the tools can cope 
with it.  I think those conditions would typically mean that it's included 
in the basic profile.  Just because it is horrible shouldn't be a reason for 
excluding it!  In fact, because it's not obvious, it's even more of a reason 
for including it.

> ISSUE-15 - ASN.1 null datatype

I thought this issue was basically about standardising on an idiom for the 
empty type, which is effectively what ASN.1's NULL type is.


Pete.
--
=============================================
Pete Cordell
Tech-Know-Ware Ltd
                         for XML to C++ data binding visit
                         http://www.tech-know-ware.com/lmx
                         (or http://www.xml2cpp.com)
=============================================

Received on Monday, 13 March 2006 13:50:42 UTC