Re: [office-metadata] Re: preserving metadata (was deadlines?)

[marbux <marbux@gmail.com>]

On 5/17/07, Patrick Durusau <patrick@durusau.net> wrote:

> I don't think there is a problem.
>
> The XML standard *does not* require that any specification based upon it
> to use RDC 2119 definitions.
>
> Those are cited solely for the purpose of interpretation of XML 1.0 (or
> any other standard that cites them).
>

I'm still under the gun for time, so won't be able to address this
much before the weekend.

I am sorry that I missed the switch from RFC 2119 and the ISO
definitions between ODF 1.0 and ODF 1.1. I think that rather than
focusing on the conformance issue (ODF is an eXtension of XML, not a
mere derivative), the shortcut answer is TC Charter requirements 2 and
4:

>>>

2. it ***must*** be compatible with the W3C Extensible Markup Language
(XML) v1.0 and W3C Namespaces in XML v1.0  specifications,

4 it ***must*** be friendly to transformations using XSLT or similar
XML-based languages or tools,

<<<

As I read those requirements, the change from RFC 2119 definitions in
ODF 1.0 to the ISO definitions in ODF 1.1 violated the quoted Charter
requirements and required their amendment before the switch was made.
I.e., the Charter and ODF 1.0 required interoperability and ODF 1.1
departed from both, breaking interoperability.

I've checked with Gary and he recalls no TC discussion of the change
whatsoever. I did not have time to determine who was responsible and
whether there was any relevant discussion on the TC, so will not
further discuss that issue right now.

To argue that the Charter requirements do not require the RFC 2119
definitions would require that at least the following words to be
interpreted as synonyms:

* must = may
* compatible = incompatible
* friendly = unfriendly
* intereoperability = non-interoperability

In summary, the Charter requirements forbid ODF 1.2 from retaining the
ISO definitions and require restoration of the XML 1.0/RFC 2119
definitions. Unless the TC Charter is first amended, ODF 1.2 "must"
use the definitions provided by RFC 2119 rather than the more
permissive ISO definitions.

Am I wrong?

Best regards,

Marbux