Hangout for experimental confirmation and demonstration of software, computing, and networking. The exercises don't always work out. The professor is a bumbler and the laboratory assistant is a skanky dufus.
I am not tracking every little move that happens in March 2008 as we determine whether or not the edited-per-the-BRM DIS 29500 will advance to an ISO/IEC IS 29500 and promulgation as an international standard. I have other work to do and other things to blog about. I shall return to my normal erratic mingling of topics. I promise.
I'm posting this article because of two interesting factors:
OOXML Maintenance might arise at ISO/IEC JTC1 Either Way
There is some interest in the US National Body for immediately creating a new JTC1 SC34 work item using an accelerated timeframe. The new project would produce an OOXML that has the BRM-recommended changes and also resolve more deliberatively all other BRM technical comments and more. This would be a new OOXML specification (revising IS 29500 if there is one of those).
The projected work is a little different depending on whether the edited DIS 29500 is approved or not. There's a suggestion that it could take 9 months to two years, depending on the situation and the quality of the edited DIS 29500. I have not found the relevant JTC1 procedures to understand what these accelerated timeframe processes are. I also have no statistics on the degree of under-estimation that happens with "accelerated timeframe" projects, although there is apparently strong oversight (section 5).
I raise this here as something to keep our eye on. We have the prospect of a moving target for a few more years. The legacy solution may require legacy handling.
What's the US National Body?
There is some confusion about who these folks are and the interests they represent. Here's a scorecard for those playing along at home.
United States Votes to Approve Open XML. Jason Matusow reports that the US National Body has concluded a letter ballot that basically affirms its original approval of DIS 29500 in the September, 2007, ISO/IEC JTC1 balloting. This is the official result. There will be a resolution meeting for this, but I don't think there is any expectation that the outcome will change.
As Matusow points out, the vote for approval is 11-4-1 (yes-no-abstain) with one "not yet voted" (perhaps like voting "present"?). Either way, the 11 affirmative votes ensure a 2/3 majority.
Whatever adjustments are made, this is the US National Body Executive Board determination.
This is different than the V1 technical committee recommendation. This is the response to a letter ballot that was conducted following the V1 technical committee recommendation. This is the Executive Board determination.
The V1 technical committee recommended approval. There was a lengthy conference call on Friday March 7, 2008, where the BRM outcome was discussed and the committee then voted to recommend to the Executive Board that the original approval of DIS 29500 be affirmed.
The US Delegation to the Ballot Resolution Meeting voted "no" on some resolutions there. This was a small delegation led by Frank Farance. These were votes on editing of DIS 29500 and not on DIS 29500's approval or rejection at ISO/IEC JTC1. The most famous delegation "no" vote was a blanket vote on over 800 proposed responses (and although selected line-item voting was allowed, the US did not do that).
In the United States, the promulgation of standards is accomplished by non-governmental bodies. There is a complex of such organizations. Keeping track of them is further complicated by the different ways they are associated with international activities, such as ISO/IEC JTC1 and other technical committees of ISO.
The pinnacle US standards organization is ANSI, the American National Standards Institute. ANSI accredits other organizations to develop standards. There are over 200 of those. Some of them, such as Underwriters Laboratory, are audited designators, but most standards developers don't certify processes or products. My favorite, after tallying the current list, is the Hydrogen Executive Leadership Panel. (By the way, OASIS is not one of these, nor are the W3C and IETF. Ecma International is not a U.S. organization, of course.)
INCITS, the InterNational Committee for Information Technology Standards, "is the primary U.S. focus of standardization in the field of Information and Communications Technologies (ICT), encompassing storage, processing, transfer, display, management, organization, and retrieval of information. As such, INCITS also serves as ANSI's Technical Advisory Group (TAG) for ISO/IEC Joint Technical Committee 1." With respect to ISO/IEC JTC1, INCITS is the US/TAG and acts as the US National Body in that capacity. It is the INCITS Executive Board that just held the letter ballot that affirms the US approval of DIS 29500.
INCITS has a pot full of technical committees (and they may have their own subcommittees and working groups of various kinds). Technical Committee V1 is on Office and Publishing Systems. They are a Technical Advisory Group for ISO/IEC JTC1 SC34. We see similar mirroring in other National Bodies. The German DIN NIA 34 has a similar function. V1 was once a tiny little group. Now there are 22 members and much is being said about that. Whatever the composition signifies, V1 has a broader charge than just dealing with DIS 29500:
"Since the scope of V1 is quite broad, it is necessary to create closer liaisons with other technical committees and special groups in order to establish good relations to avoid duplication of efforts. A number of new or extended areas are gaining interest in V1 such as Voice Messaging, Icons, Data Architecture, Font Services, MHS/X.400, Publishing Applications, Models and others related to Office and Publishing Systems."
If you wonder where INCITS came from and how it operates, here is more context:
"The InterNational Committee for Information Technology Standards (INCITS) is the forum of choice for information technology developers, producers and users for the creation and maintenance of formal dejure IT standards. INCITS is accredited by, and operates under rules approved by, the American National Standards Institute (ANSI). These rules are designed to ensure that voluntary standards are developed by the consensus of directly and materially affected interests.
"INCITS is sponsored by the Information Technology Industry Council (ITI), a trade association representing the leading U.S. providers of information technology products and services. ITI members employ more than one million people in the United States and in 2000, their revenues exceeded $668 billion worldwide.
"INCITS was founded as Accredited Standards Committee X3 in 1961. The last accreditation was April 19, 2001 under the name INCITS."
When I was last directly involved in these activities, around 1970, it was still X3 and the technical committees had designations like X3.4.2, X3J3 and X3V1 (if there had been a V1 then). At that time, ITI was known as CBEMA, the Computers and Business Equipment Manufacturers Association.
In addition to standards development being carried out by non-governmental bodies in the United States, the US also operates under a voluntary standards regime. For the most part, there is no mandating of standards and no one, not even participants in standards development, are obligated to adhere to a standard just because it exists.
Of course, adherence to specific standards can be required in procurement activities, and requirement of particular certification of standards-conformance might also be required. This is all pretty-much outside the purview of the standards developers in the voluntary model. (Some standards are given regulatory force, incorporated in building, electrical, food-processing, and other codes, but this is not under the authority of standards developers as standards developers in the ANSI regime.)
Here's the reminder that appeared at the bottom of the just-conducted letter ballot:
"Voluntary Standards are developed with the intention and expectation that the standards will be suitable for wide application. As their use is likewise voluntary, an affirmative vote does not commit an organization or group represented on the committee to the use of the voluntary standard under consideration. If you find that you cannot vote YES and wish to vote NO or ABSTAIN, please state this and explain the reasons for your position in the places provided."
Of these, Farance is not a household word. Farance, Incorporated, is a technology firm that is heavily engaged in the development of standards, with some current emphasis on metadata and learning technologies. Frank Farance was the Head of Delegation to the DIS 29500 Ballot Resolution Meeting. As part of the August, 2007, deliberations on a US response to the DIS 29500 ballot, Farance cast this ballot recommending conditional approval (that is, "No, with comments").
A common theme is the impossibility of conducting adequate review of a 6,000 page specification under the fast-track procedure. IBM and Oracle both refer to "procedural irregularities" with IBM recording an extensive challenge to the validity of the INCITS process.
The Farance objection asserts that "A position of 'approval' is completely unacceptable to us" so it is a procedural appeal too, I suppose. The objection goes on to propose a remedy after enumerating the deficiencies. More about that below (section 5).
EMC, self-identified as "the world leader in information management and storage." EMC provides solutions for mainframe, IBM DB2, Microsoft, Oracle, and SAP. Documentum is now their product. Oh. I need to get out more.
GS1 US, another non-profit accredited standards developer that focuses on global supply-chain solutions. If you need a Universal Product Code to stick on your product packages, you get it from GS1 US.
The last three "yes" votes are of an entirely different breed:
U.S. National Institute for Science and Technology (NIST). This is a Federal agency under the U.S. Department of Commerce. Formerly know as the National Bureau of Standards, NIST is not in charge of standardization in the United States. NIST/ITL (the Information Technology Laboratory) is an accredited standards developer, however. NIST has, in the past, drafted Federal Information Processing Standards (FIPS) that apply to agencies of the U.S. Federal Government. These are often existing public standards adapted for governmental use. There is an extensive report on NIST participation in INCITS with respect to ODF, OOXML, and the DIS 29500 process (PDF text). There is also a NIST press release on the matter.
I'm intrigued by the following statement in the NIST ballot:
"With respect to the 'Additional INCITS/V1 Recommendations Not Related to this Vote,' included in INCITS Letter Ballot 2558, NIST interprets these recommendations to be that the USNB should support one of the following options:
"1) Should DIS 29500 fail to achieve the criteria for approval as an ISO/IEC standard, DIS 29500, as modified by the BRM, should be submitted to JTC 1/SC 34 for fast processing, via the JTC 1 combined NP [New Work Item Proposal] and CD [Committee Draft] ballots process, and progressed under the JTC 1 accelerated time frame.
"2) Should DIS 29500, as modified by the BRM, achieve the criteria for approval as an ISO/IEC standard, to correct remaining issues from the BRM, SC 34 should approve an NP for a revision of IS 29500 using the JTC 1 5-stage process under the accelerated time frame."
No "Additional INCITS/V1 Recommendations" appear as part of the ballot form. There is something about them in the Farance ballot though:
"Although we disapprove of the DIS 29500 at its present state, Farance Inc. is supportive of the 29500 technologies becoming an ISO/IEC standard. We want the work done right and done quickly to address industry and international needs. Farance Inc. has proposed in INCITS/V1 two New Work Item Proposals [NPs] for SC34: one NP if DIS 29500 succeeds and a different NP if DIS 29500 fails. These NP proposals were acknowledged in the INCITS EB ballot wording. We hope these NPs will be a starting point to trying to fix/finish issues within the 29500 text, and we hope these immediate needs can be addressed over the next 9-14 months."
It is not clear that NIST is throwing its support behind that idea or is simply clarifying its understanding. I have failed to find the particular JTC1 procedures that are referred to as "the JTC1 5-stage process under the accelerated time frame." I'm still looking.
[Update: There are three different schedule templates for JTC1 New Work Item Proposals: accelerated (24 months to published standard), default (36 months), and extended (48 months) on the lines of [1: section 6.3]:
Committee Draft (CD)
Approval Draft (DIS)
6 months (enquiry draft)
The entire section on Programme of Work is important for an understanding of the process [1: section 6].
The JTC1 5-stage procedure has 6 stages (0 to 5), [1: section 12.1]:
Preliminary: A study period is underway [probably brief for OOXML]
Proposal Stage: the NP is under consideration (requires balloting and formulation of the work, assurance of participation, and other important formalities, [could be brief for post-IS 29500, might be contentious otherwise])
Preparatory Stage: Working Draft (WD) under consideration [probably using either IS 29500 or updated ECMA-376 as initial WD]
Committee Stage: A Committee Draft is under consideration [something tells me there will need to be something like that either way, using normal processes]
Approval Stage: An FDIS is under consideration [more-or-less where the Fast-Track DIS 29500 process started and where DIS 29500 is right now]
Publication Stage: An IS is being prepared for publication [a process yet to be completed if DIS 29500 is approved]
My personal reality check suggests that the odds of this getting off on the right foot (starting work around September-October 2008) is if there is a published IS 29500 as a concrete starting point. Whatever process is put in place, there must be room for feedback from implementation experience and any working agreements that are formulated as part of document interoperability efforts. Putting the necessary technical muscle into an SC34 working group and coordinating with NB counterparts (such as INCITS/V1 and others) will be daunting and time-consuming. Some way of providing intermediate guidance and (provisional) interpretations may also be important. I'm not betting on the accelerated timeframe even under the best of conditions.]
Here's what I still think. OOXML is the new COBOL. Really. That's a good thing (with a 40-year life span, for starters), if the energy and pace can be achieved with the requisite quality and care.
[update 2008-03-17T19:45Z: I received a very helpful e-mail through the NIST Media Relations Office, pointing me to the relevant sections of the JTC1 directives. The selections, summaries, and interpretation are all mine.
update 2008-03-16T15:45Z: I learned from an Eric Lai Computerworld article that AIM Global and Qualcomm are apparently new members of the INCITS Executive Board. I updated my observation about that and who are voting members accordingly.]