RE: Substitution Groups Reconsidered

[Hal Lockhart <hal.lockhart@entegrity.com>]

A couple of clarifications:

> Kohsuke is a (very well respected) colleague of mine.  His 
> article is quite 
> provocative, given that many users of XML Schema precisely 
> want the complex 
> type functionality; in our case, it's one of our main 
> motivations in our 
> schema design.

I took his comments to say essentially "Don't use these if you don't know
what you are doing." Since I assume we know what we are doing, I was not too
worried about his suggestions. I included the pointer mainly to indicate
that the lack of tool support may persist.

> On the matter of data binding: You're no doubt right that 
> there's little 
> support for the substitution group type hierarchy in data binding 
> software.

It is my understanding that the problem is not just that the substitution
groups cannot be referenced, but that we cannot use the tool at all if the
schema contains them.

This is a significant issue to us and I assume others. Our current
workaround is to manually recode the schema into an equivalent one without
substitition groups and run tool over that. This is obviously undesirable
for a number of reasons.

> I think Hal's basically right, 

I was trying hard not to express an opinion. (I guess I am just not good at
that. ;-) It seemed to me that the lack of tool support was an important
piece of information that was not presented as one of the pros and cons of
using substitution groups.

I strongly agree with Phill about the need to retain as much flexibility as
possible and the need to close issues and leave them closed. However, it was
my understanding that the ability to use a tool to create the code to
marshall and unmarshall XML was one of the major benefits of using XML
Schema.

Hal