On Mon, 12 Aug 2002 12:11:35 -0400 (EDT)
John Cowan <> wrote:
> > - Type-assignement should be separated from validation (It should be
> > possible to do type-assignment without performing validation, and
> > vice versa)
>
> Does this mean that it is possible to type-assign even schema-invalid
> documents?
I think that validation is too expensive for simple applications. Thus,
I would like to separate validation and type-assignment. Another reason
is modularity of processes.
If type-assignment is separated from validation, type-assignment is happy
to handle invalid documents. The result may be almost correct or may be
completely meaningless.
--
MURATA Makoto (FAMILY Given) <>
