Wednesday, September 05, 2007

Document Format Wars

Microsoft has proposed to fast track its Open Office XML (OOXML) definition for documents produced in Office 2007 as an ECMA and ISO standard. Their intention is to capture the XML schema that defines the structure of Office 2007 documents including backward-compatible binary components that provide support for documents produced by earlier versions of the office product. These binary component definitions are specific to Microsoft products and are not capable of being produced with or without the OOXML definition.

The Open Document Format (ODF) for Office Applications (OpenDocument) is supported and promulgated by the Organization for the Advancement of Structured Information Standards (OASIS) and is enthusiastically supported by a growing band of followers. The reason for the strong following and broad-based support is that ODF is a well-designed and open definition that can be implemented by any party without the need for any additional information apart from that in other open definitions and international standards, like Scalable Vector Graphics (SVG).

OOXML, on the other hand, replicates the equivalent behaviour of numerous open format definitions and existing international standards. Among many others individuals, I submitted technical objections to my national standards body (Standards Australia) of the proposed standard on the basis that OOXML is a long and overly complicated definition of a document format that can only be implemented in full by Microsoft.

To: michael standards.au
cc: daniel systec
Subject: DRAFT INTERNATIONAL STANDARD ISO/IEC 29500, Information technology ?
Office Open XML file formats.

Dear Michael,

I humbly submit a few comments about the proposed DRAFT INTERNATIONAL STANDARD ISO/IEC 29500, Information technology ? Office Open XML file formats.

Please contact me if I can be of any further assistance.

Regards,
Daniel
--

Daniel M. Berinson BE PhD MIEAust MIEEE AFAIM GAICD
Managing Director, Systems and Software Architect
Systec Engineering Pty Ltd and Systec IT
daniel@systec.com.au
040 888 0278 (m)

Template for comments and observations

Date: Due to Standards Australia by COB 21 Aug 2007

Document: DRAFT INTERNATIONAL STANDARD ISO/IEC 29500
Information technology — Office Open XML file formats


1

2

(3)

4

5

(6)

(7)

MB1

Clause No./
Subclause No./
Annex
(e.g. 3.1)

Paragraph/
Figure/Table/Note
(e.g. Table 1)

Type of com-ment2

Comment (justification for change)

Proposed change

Observations
on each comment submitted



AU

1


ge

The document is overly long (over 6,000 pages) and complicated to be an effective specification for implementers.

Shorten the document for this standard by either: 1) partition the supporting material into several other standard proposals; or 2) amend with reference to appropriate standards.


AU

1


te

The document refers indirectly to binary implementation details that are not part of this proposed standard nor are they part of other existing or proposed standards.

Delete references to binary materials and delete references to other materials that are not part of an existing or proposed standard.


AU

2.15.1.28


te

Conflicts with ISO 10118-3 and W3C XML-ENC by defining nonstandard hashing and cryptographic algorithms (likely to be insecure).

Amend with reference to standards that specify hashing and cryptographic algorithms that are know to be secure.


AU

2.15.3.6


te

Example (one of many) of implementer being required to clone unspecified behaviour.

Specify explicitly the required behaviour; otherwise delete references to unspecified behaviour.


AU

2.18.52


te

Conflicts with ISO 639 by requiring the use of a fixed list of numeric language codes rather than the already existing set provided by ISO 639.

Amend with reference to appropriate standard.


AU

2.18.105


te

Refers to nonstandard twips.

Amend with reference to standard units.


AU

3.17.4.1


te

Conflicts with the Gregorian calendar in the calculation of dates by requiring spreadsheet implementations to incorrectly treat the year 1900 as a leap year.

Amend to conform with Gregorian calendar.


AU

4.4


te

Conflicts with W3C SMIL by defining nonstandard multimedia features.

Amend to conform with appropriate standard.


AU

5.1.10.45


te

Example (one of many) of inconsistent and poorly named XML elements.

Revise XML element names and type design (i.e. well-designed complex and simple types).


AU

5.1.12.42


te

Refers to nonstandard English Metric Units (EMU).

Amend with reference to standard units.


AU

6.2.3.17


te

Refers to Windows Metafiles or Enhanced Metafiles instead of using ISO/IEC 8632 or W3C SVG.

Amend with reference to appropriate standards; otherwise delete references to proprietary materials.


AU

6.2.3.23


te

Refers to Microsoft namespace (urn:schemas.microsoft.com:office:office).

Amend to appropriate open namespace.


AU

7.1


te

Conflicts with W3C MathML by defining nonstandard format for mathematical expressions.

Amend with reference to appropriate standard.


AU

8.6.2


te

Conflicts with W3C SVG by requiring support for VML drawing format (not a standard).

Amend with reference to appropriate standard.



1 MB = Member body (enter the ISO 3166 two-letter country code, e.g. CN for China; comments from the ISO/CS editing unit are identified by **)

2 Type of comment: ge = general te = technical ed = editorial

NOTE Columns 1, 2, 4, 5 are compulsory.

page 1 of 2

ISO electronic balloting commenting template/version 2001-10



Grokdoc has pages of contacts, objections and Microsoft's history of dirty tricks in opposing development that supports open standards to the detriment of proprietary standards. As it turned out, the fast-track proposal failed the vote so the final outcome has been deferred - however this state-of-affairs is a cop-out due to the late elevation of formerly uninterested national standards organisations from observer to participating (voting) member status. I regret to say that even Standards Australia abstained, as an unfortunate example of bureaucratic nonperformance.

Worse still is that the problem has not gone away since the fallout from the debacle is that standards work in the relevant working group has ground to a halt. The situation is an embarrassing debacle the blame for which we can lay at the feet of Microsoft and many other organisations that are either corrupt, naive or inept.