The conspiracy to save OOXML from being so crappy

What is sauce for the goose is sauce for the gander

By Rick Jelliffe
June 19, 2009 | Comments: 4

[UPDATE: My focus when I wrote this was that by commenting on scope for improvement in ODF, I was not suggesting that OOXML was immune from criticism or improvement. However, a side issue in the second sentence seems to have derailed this, so I am moving it to the end.]

I have been lurking on the teleconferences for the working group maintaining OOXML: SC34 WG4. ... Maintenance is very important for standards, indeed it is a good (but not the only) benchmark for gauging the approach a particular standards body takes to things. I recall what I wrote in The Perpetual Ballot Resolution Meeting a year ago

But the idea that this is it, this is Microsoft's only accountability moment where they get a pass or fail is propaganda, not the ISO process. It is completely true that the maintenance procedure needs continued interest and continued pressure, but it is not true that this is the last chance to improve the standard as if it will be frozen for all time....
At ISO/IEC JTC1, the rule is that the Editor has to handle defect reports "promptly". ("Promptly" needs to be measured in quarters of years, it won't be weeks. But it won't be years or decades, which is how long some bugs have persisted in Office without the circuit-breaking of National Body scrutiny.)

Indeed, WG4 publishes a graphic of the response rates for OOXML defect handling to help project management. At the moment, there are no issues that have not been looked at within 60 days of being registered, with a certain chunk of them being put on the stack for further consideration. (The current plans are that issues are being placed either for a corregendum document, which —someone correct me if I have this wrong— is for minor editorial errors or where a NB says the BRM's requirements were followed too conservatively or something was missed, for voting soon, or placed in an amendment document for more substantive changes. Amendment documents have a more elaborate procedure, and result in a new year number for the standard.)

The teleconferences are once a fortnight, at 11pm Sydney time, for about 2 hours, which is not so bad. My landline phone is unworkable due to ADSL hum, so I have been using LiveMeeting. The first was OK (featuring video of Alex Brown's living room) except I couldn't track down my mic. The next one I tried to use a swanky wireless headset from work, but my laptop couldn't track down the drivers, and the LiveWire couldn't connect on audio. See how it goes this time: SNAFU.

It seems like the group is making all its mail-list archives public, which is a good thing. Denmark, through Jesper Lund Stocholm, in particular has pushed for more openness.

People interested in the recent composition of WG4 should look at the minutes of the last meeting (April 2009) in Prague (PDF) On my count they show

  • Users/governments/standards bodies: 17

  • Microsoft: 8

  • Other vendors: 5

So according to my balance principle, I would say that SC34 WG1 needs more participation from (non-MS) vendors to get a good balance: it is currently tipped in favour of users/governments/standards bodies.

Another interesting way to slice that cake is to look at the balance between US/Europe and the rest of the world--about:

  • Europe: 11

  • North America: 8

  • CN, JA, KR, ZA: 12

It is interesting to see the ODF and OOXML standardization efforts side-by-side, since they exhibit complimentary failings: IS 29500 (OOXML) is regarded as too big and randomly specific, while IS 26300 (ODF as updated at OASIS) is regarded as too small and randomly vague. OASIS ODF TC needs more user participation; SC34 WG4 needs more vendor participation. In the spirit of fairness, and because I recognize that what is in one's mind is not always fairly mirrored by what is in one's blogs taken collectively, let me repeat the call for participation I made about ODF (note to self):

I would also issue a challenge to the FOSS people Office community who spent so much energy complaining about DIS2956300 only to find that, miracle of miracles and horror of horrorswith due credit to all involved, many of their complaints werewill be taken on board to improve DIS29500ODF 1.2 (and continue to be taken on board for ODF next gen): For every hour you have spent on OOXMLODF review, spend an hour on ODFOOXML review. Join the conspiracy!


Update: The original second sentence was: Actually, I have joined it as an invited expert (which is an honour) a month ago, and trying to get up to speed on it. I should have been clearer that "joining" meant the teleconference not the WG proper, and I apologize for giving the wrong impression: I need to be more attentive to this. In my mind was the "conspiracy" that is the topic of the blog: the need and opportunities for broader community engagement and participation through whatever avenues are available.

The invitation to me was to subscribe to the mail-list and to be available to occasionally participate in the teleconferences (I presume if they have a schema question or if I made a suggestion on the mail list they wanted to explore.) The effort by WG4 to try to get broader community engagement within the constraints of the JTC1 Directives is laudable. My own interest in monitoring the teleconference and list is to see whether the maintenance effort is credible (which is not to say that it would turn a sow's ear into a silk purse.)

For comments on WG1 and my unsatisfactory status as an unattached editor, see the comment thread below.

My themes that balance in standards groups is important and that observers/monitors should be allowed where possible are not new: see this XML-DEV post from 1998 noting that the "WG4" in that is what we now call SC34.

You might also be interested in:


Rick, I'm fascinated by this "invited expert" participation level in an ISO/IEC JTC1 SC WG. Can you tell me more about this, such as who makes these invitations, and how one applies for this?

Also, JTC1 Directives says "All participants at each WG meeting must be authorised by their NB or appropriate liaison organisation." There is nothing said about any "invited expert" category.

What am I missing here? Has Standards Australia nominated you? In that case, you would just be a normal NB representative, like most WG participants. But the most recent meeting minutes has you listed as "Invited Expert" with no stated NB affiliation. This is quite irregular. How do you explain this?

I can imagine the fascination!

My understanding is that observing is not participation, and that participation is not the same as attendance.

Invited expert observers are a long-standing practice at JTC1. A Google of "invited expert jtc1" will bring up tons of examples over the years. I believe this is the "guest" category in for example.

I certainly agree that the JTC1 directives could be explicit on the status of observers, experts and editors. (Where an issue is not dealt with explicitly in the Directives, it is left to the discretion of the chairs, secretariats and so on up, who will take into account the normal practice.)

For example, for WG1, I am an editor appointed by SC34. There is no requirement in the JTC1 Directives that an editor be a NB or liaison body delegate. And editors agree when taking on the role to be available for the long haul. An external editor can of course be instructed by correspondence from the WG (indeed, most of my interactions is by correspondence, I miss many more WG1 meetings than I am able to attend) but it would be strange to say he/she could not attend a WG meeting in person to be asked questions and given instruction, if invited.

If push came to shove so that this needed to be made explicit in the JTC1 Directives, I expect the result would be along the lines that an editor is an officer of the parent body (the SC) therefore a proper person to attend on request at WGs and to answer ad hoc question and take instruction. But not 'participating' in the sense that their opinion was given any weight for ascertaining the consensus of the WG, etc. or on unrelated standards.

In recent years, there has been a strong push to make sure that unattached editors etc get a proper home: Australia has not had a mirror committee for most of this decade, so we have looked at the SGML/XML Guild and the option of me getting a BSI invitation (I am eligible for a British passport.)

A invited expert observer is pointless. Why would you invite an expert who can only observe and not participate? What would be the value of their expertise? And why would you be honored merely to observe but not share your expertise?

And note that the word "observer" is not mentioned at all in JTC1 Directives. In fact 7.7 is quite explicit and restrictive: "Only delegates officially nominated by the NBs and the representatives of other TCs and organisations in liaison may attend meetings." The use of 'only' does not give you much wiggle room.

This looks like a problem, Rick, regardless of what others may have done in the past. Why don't you just get SA to nominate you. Presumably they are aware of your relevant expertise in this area? And they are a P-member? So following the rules should not be particularly onerous in this case. Heck, if SA won't nominate you, get Ecma to. They'll allow anyone.

Well I certainly will take it up with the Secretariat next week. I don't have any wish to cause them trouble.

News Topics

Recommended for You

Got a Question?