W3C home > Mailing lists > Public > www-tag@w3.org > January 2006

Re: speech grammar spec recommends xsi:schemaLocation [namespaceDocument-8, namespaceState-48]

From: <noah_mendelsohn@us.ibm.com>
Date: Mon, 9 Jan 2006 18:13:10 -0500
To: Jon Hanna <jon@hackcraft.net>
Cc: www-tag@w3.org
Message-ID: <OFD8459019.64BFC9C8-ON852570F1.007EC050-852570F1.007F8D0C@lotus.com>

Jon Hanna writes:

> As either a document recipient, creator, publisher or anything 
> else I want the document to be valid against a shared schema 
> deemed useful by all parties (whether explicitly, or through 
> the schema being "well-known").

In framing things this way, I think you've ruled out many of the most 
interesting use cases that the Schema WG has been considering regarding 
versioning [1].  XML is in part useful for communication between loosely 
coupled organizations, and experience suggests that author and consumer 
may evolve their software at different times.  The trick with versioning 
is not to have a "shared schema deemed useful", but an ability to detect 
the degree to which interoperation is safe between two parties who may 
have somewhat differing views of the schemas.   Maybe or maybe not 
schemaLocation is a helpful or appropriate hook, but I strongly object to 
the suggestion that in the general case there is one schema to which both 
parties must agree.    On the contrary, they must reliably discover in the 
face of potentially conflicting schemas which parts, if any, of the 
communication can be reliably interpreted, in the sense that a) the 
receiver correctly determines what the sender has intended to transmit and 
b) that the receiver does not inadvertently miss something essential (I.e. 
everything you act on is correct, and you don't fail to notice something 
that makes proceeding unsafe.)

Even if the latest schema is available directly or indirectly from the 
namespace URI, that doesn't guarantee interoperation of such software 
written at different times.  There is likely to be a lot of logic in a 
program that doesn't update automatically when a new schema is received. 
At best such a program might retrieve the schema and discover: oops, looks 
like someone's changed the schema since my code was written.  Even then, 
I'll be curious whether the person who wrote this message knew about the 
latest changes.

Noah

[1] http://www.w3.org/2005/05/xsd-versioning-resources.html
[2] http://www.w3.org/XML/2005/xsd-versioning-use-cases/

--------------------------------------
Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------
Received on Monday, 9 January 2006 23:13:16 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:56:10 UTC