Re: ISSUE-19: Advice against using the \'all\' model group

If anyone is monitoring what the Schema WG is doing, (s)he will notice 
that we are trying to improve <all> groups. One of the reasons for doing 
this, which I always use, is to help data-binding.

As you mentioned, <all> feels very natural in modelling programming 
language constructs. If there are particular restrictions on it that 
prevent its meaningful usage, maybe the 2 working groups should try to 
work together to resolve them, before we tell the users to give up. (Some 
of you will be meeting in Mandelieu soon, which sounds like a great 
opportunity for such discussion.)

Thanks,
Sandy Gao
XML Parser Development, IBM Canada
(1-905) 413-3255
sandygao@ca.ibm.com


public-xsd-databinding-request@w3.org wrote on 02/21/2006 09:00:42 AM:

> 
> 
> ISSUE-19: Advice against using the 'all' model group
> 
> http://www.w3.org/2005/06/tracker/databinding/issues/19
> 
> Raised by: Paul Downey
> On product: Basic
> 
> The input document advises for a collection: 
> 
> """
> The all model group may appear attractive given programming language
> techniques 
> such as introspection or reflection often return items in a random 
order. 
> However, there are significant restrictions placed upon all types, 
> not least an 
> element cannot have a maxOccurs value greater than 1. 
> The Unique Particle Attribution (UPA) constraint prevents a model group 
of all
>  from being extended, either by containing an any element wildcard, 
being 
> incorporated in a substitution group or derived using extension or 
> restriction.
> """
> 
> Should we carry this advice into our Basic Patterns?
> 
> 
>