DIS 29500 proposed dispositions

Today's announcement from Ecma that written responses have been provided for all 3,522 national body comments in the DIS 29500 standards process is a major milestone for all involved. Brian Jones has a few details on his blog as well.

Looking at the magnitude of what has been accomplished, I can't help thinking of the comparison between these two tasks:

  1. Create a standard specification of a few thousand pages, for a document format that's designed to provide compatibility with an existing document format, while working closely with the creators of that existing format.
  2. Write responses to thousands of comments, concerns, questions and complaints about the same standard from hundreds of people worldwide, including competing vendors, ideological opponents, and people ranging from deep experts in the subject to, well, those with a little less understanding of the details.

Which sounds more difficult? I think most people would say the second task sounds much more difficult, and yet Ecma TC45, which has pulled off both of these tasks, did the first one in about a year, and the second one in just four months. To say that they really stepped up their game is an understatement. They've had long hours of phone calls, meetings, and other activities, and have managed to create thousands of pages of high-quality written responses, with some of the individual responses totalling dozens of pages.

As an INCITS V1 member in the US, I have access to the responses, and looking around at them this afternoon I'm amazed at how much has been accomplished. I can't talk about specific details currently due to ISO regulations, but at a high level I'd divide the proposed dispositions into a few buckets:

  • Addition of useful information for developers, such as the thorough documentation of compat settings. Want to know what it means to "autospace like Word 95" or "truncate font heights like WP6"? That's all spelled out now, so that any developer can implement these behaviors.
  • New flexibility in the formats, such as extensible page borders, support for new types of content, and new options for date handling. Want to use ISO 8601 dates in an Open XML spreadsheet? Now you can.
  • Standards support. Dozens of international standards are normatively referenced in the proposed changes, making DIS 29500 a well-socialized and well-connected member of the international standards family. A good example is the use of ISO/IEC 14977:1996 (Syntactic metalanguage – Extended BNF) notation for spreadsheet formulas and fields.
  • Structural changes to allow for selective re-use of specific portions of the standard. One of the proposed changes would make OPC (Open Packaging Convention) and MCE (Markup Compatibility and Extensibility) separate parts, so that other standards can normatively reference these useful technologies separately from the rest of DIS 29500.
  • Clarification of numerous details, including conformance requirements, algorithms, syntactical details, and much more.
  • Correction of errors and typos that have made some of the details confusing in the past.

The resulting specification, after all of these proposed changes are made, will be the most thoroughly reviewed document format standard ever. And the level of detail, clarity and flexibility in the spec will be greatly improved by the process that has taken place. This is standards development in action, and the system is working, for the benefit of all.

Sure, there will be people who disagree with some of the changes, and those debates will take place from now through the BRM (Ballot Resolution Meeting) in late February. Many of those disagreements will be resolved, and some of them probably won't. Some commenters want things that most people don't want, and like any consensus-based process, the opinion of the majority will drive the outcome of those issues. When we're talking about a typo, technical error, or omission, there's not a lot to debate, but sometimes there isn't a solution that can satisfy every single party involved.

For example, I've noticed a few cases in the comments where two different comments from two different countries are mutually exclusive. As one specific example, I noticed a situation where one country says "please explain X in more detail," and another says "X has no place in a standard of this type." Resolving that situation to the satisfaction of all parties will require a delicate combination of technical expertise, communication skills, and respect for all points of view. But based on everything I've seen to date, I think Rex Jaeschke (the project editor) and Alex Brown (the BRM convenor) are up to the task.

Congratulations to Rex and the members of TC45 on this huge accomplishment over the last few months!