Re: Redefinition of group, who's correct

On Mar 25, 2013, at 2:49 AM, Pierre Attar wrote:

>=20
>=20
> When trying to write this schema, my assumption was a little bit DTD =
like interpretation : the first defininition is the right one.


That's true for entity and attribute declarations in DTDs, but not for=20=

element declarations (which lead to errors when duplicated) or=20
attribute-list declarations (which accumulate).

An idea very similar to your operating assumption was in fact=20
proposed (a sort of schema-component search path), but I am
ashamed to relate that it was not taken very seriously and=20
was not adopted.  The XSD rules for include and import might=20
have been much easier to understand (and thus easier to write=20
without contradictions) has the WG adopted it.  But the=20
majority of the WG thought that it would be too error-prone.

--=20
****************************************************************
* C. M. Sperberg-McQueen, Black Mesa Technologies LLC
* http://www.blackmesatech.com=20
* http://cmsmcq.com/mib                =20
* http://balisage.net
****************************************************************