[office] Open Office Specification Draft 9

[Michael Brauer <michael.brauer@sun.com>]

Dear TC members,

I've copied a new specification draft into the TC's document section:

http://www.oasis-open.org/apps/org/workgroup/office/download.php/5692/office-spec-1.0-draft-9.sxw


I've also copied three Relax-NG schemas, that all were extracted from 
the spcification document:

1. office-spec-1.0-draft-9.rng is the full schema. It now allows any 
content within <office:meta> and the formatting properties elements.
2. office-spec-1.0-draft-9-strict.rng allows only the predefined 
elements and attributes within <office:meta> and the formatting 
properties elements. It is a rather short schema that includes the full 
schema and simply replaces some definitions.
3. office-spec-1.0-draft-9-manifest.rng is the schema for package 
manifest files.

The first two schemas solve the validation issue we discussed last Monday.

The specification document is complete. There are only two open issues 
that we should discuss on Monday:

1. The usage of the words MUST, SHOULD, etc.
2. An "TODO (TC)"  on page 569 regarding the style:overflow-behavior 
attribute.

I suggest that do a review of the specification until Monday, February 
the 8th. If any additional TC decisions are required, we may vote about 
them at Monday, the 8th, so that we can integrate them into the spec in 
the same week. At the same time, we may remove all "Notes to the TC" 
from the spec, so that we get the draft that we use for voting about a 
committee specification. The voting might take place in the con call at 
Monday, the 15th, or by an e-mail voting that starts in the week of 
Monday, the 8th, and closes in the week of Monday, the 15th.

Regarding 1: We have been very carefull using these words as keywords. 
We have introduced them to describe what documents conforming 
applications must be able to read and what they should do with unknown 
content. In these chapters, we use must, etc. as keywords already, or 
may check this again.

Beside these requirements, we only have litte requirements regarding 
conforming applictions. This means that the words here probably in most 
situation are used to express what the schema formally describes (i.e. 
an required attribute MUST be present, an optional one MAY be present). 
For this reason, it seems to be a valid option to me to leave the 
specification as it is, or to remove the usage of these words as 
keywords ouside the chapters where we define conformance.


Best regards

Michael