Universal Interoperability Council

The Universal Interoperability Council's September 22, 2008 comments on the Draft Document as Basis for European Interoperability Framework v. 2 may be downloaded in PDF format from this page.

For those looking for the statistics relevant to under-specification of OASIS OpenDocument v. 1.1, they are in Appendix A, "Signs of under-specification in OpenDocument v. 1.1," together with the methodology employed and appropriate caveats.

Those statistics might be usefully compared to an email post by XML guru Dave Pawson, who inspected the entire ODF v. 1.1 specification for testability, paragraph-by-paragraph:

Just ran some rough test counts.

[ODF] 1.1 Has 2578 'testable paragraphs'
[ODF] 1.2 currently has 3002.

Quite a task this TC has on its plate.1

Juxtaposing the 227 maximum conformance requirements statistic shown in the attached document with Pawson's report, one may divine that no more than 227 of the 2,578 testable paragraphs he could identify are conformance requirements, or restated, a maximum possible 8.8 per cent of testable paragraphs state conformance requirements. The rest are apparently mere options or recommendations.

For the those not numbered among the data format cognoscenti, options and recommendations in standards mask hard-coded "must" or "must not" programming decisions in implementing applications and in fact represent a standard's dependencies on the application with the largest market share, whose hard-coded programming decisions are concealed behind the facade of may, optional, and should clauses in the relevant standard.

Because options and recommendations represent variables in a data format, the barriers to interoperability among different implementations of a standard increase exponentially as the number of variables increases. In other words, there is a reason why non-lossy interoperability of ODF implementations has been demonstrated only between applications that share the same code base, such as OpenOffice.org and its various clones.

It is fair to question whether OpenDocument deserves the title of an open international standard in light of its gross under-specification. The governing ISO/IEC Directives provide that international standards are:

... to specify clearly and unambiguously the conformity requirements that are essential to achieve the interoperability. Complexity and the number of options should be kept to a minimum and the implementability of the standards should be demonstrable.2

Likewise, the Agreement on Technical Barriers to Trade requires that international standards specify [i] all characteristics [ii] of an identifiable product or group of products [iii] only in mandatory "must" and "must not" terms.3

"Quite a task" to clean up that mess, indeed.

1. 1.

   Dave Pawson, Test Count, post to oiic-formation-discuss mailing list (6 July 2008), http://lists.oasis-open.org/archives/oiic-formation-discuss/200807/msg00103.html. See also Rob Weir of IBM's reply the next day at http://lists.oasis-open.org/archives/oiic-formation-discuss/200807/msg00105.html. ("This is less work than I expected. I was predicting 10,000 test cases.")

2. 2.

   ISO/IEC JTC 1 Directives, (5th Ed., v. 3.0, 5 April 2007) pg. 145 (PDF), http://www.jtc1sc34.org/repository/0856rev.pdf.

3. 3.

   WTDS 135 EC - Asbestos, (World Trade Organization Appellate Body; March 12, 2001; HTML version), para. 66-70, http://www.wto.org/english/tratop_e/dispu_e/cases_e/ds135_e.htm; reaffirmed and further explained, WTDS 231 EC - Sardines, pp. 41-51 (World Trade Organization Appellate Body; 26 September 2002), pp. 41-51, http://www.wto.org/english/tratop_e/dispu_e/cases_e/ds231_e.htm.