As many of you are aware, Alex Brown will be the "Convenor" of the OOXML Ballot Resolution Meeting (BRM) that will run from February 25 through 29 in Geneva, Switzerland. Alex has a variety of unenviable tasks, including:
Trying to interpret various standing Directives and other ISO/IEC JTC1 rules and practices that were created for what might be described as kinder, gentler times (not to mention for shorter specifications).
Figuring out how to process c. 1,000 comments (after elimination of duplicates) during a 35 hour meeting week, without the currently contemplated possibility of an extension.
Herding 120 cats, some of which will have strong opinions on individual points, others of which will have alternating suggestions on how to resolve a given point, and many of whom may be just plain bewildered, due to the lack of time to be fully prepared.
For better or worse, the rules that Alex will be interpreting and applying are not as comprehensive, and certainly not as detailed, as the situation might demand to put everyone on exactly the same page regarding what should (or at least could) be done at many points in time. As a result, knowing how Alex's thoughts are shaping up is both interesting and important. To his credit, he has been generous about sharing those thoughts, and often how he arrived at them, at his blog, which can be found here.
While I've often linked to Alex's blog and have had a permanent link in the "Blogs I Read" category for some time, I'd like to point to Alex's latest entry, which covers several important points that others have recently blogged on. In many cases, Alex comes out differently than some others that have stated firm opinions, and since Alex has the gavel, his opinion will be the one that counts.
Alex’s latest blog entry was posted yesterday, and is titled Tracking OOXML issues, and here are some of the things that I found instructive.
For example, last week I posted a blog entry where I noted the difficulty that delegates would have preparing for the BRM, given the number of comments and the lengthy proposed dispositions document (2,300 pages). Rick Jelliffe took issue with that, posting the following in a comment:
Third, a reviewer for a national body will be primarily interested in the comments from that body. In most cases, the NBs comments are much less than 100 comments.
Fourth, within each National Body, different members of the committees (and delegations) have different interests and strengths, so of course there is a division of labour. There will be touchstone issues for each reviewer that they will often check thoroughly, and flip through the responses to issues that don’t interest them or which are out of their expertise. There will also be some die-hard reviewers who will want to read all the responses, but they would not be the majority, for a large standard.
Where did you get the idea that every person needs to understand every part of every response? That is not the way things work in large standards, it is committee work. Why don’t you join an international standards committee, so you can enhance your excellent prose with experience? I am sure everyone would benefit.
Among other things I noted this is hardly a normal standards situation, that some delegations will consist of only one individual, who will therefore need to cover all comments, that many National Bodies (NBs) were indeed interested in the comments of other NBs, and that the dispositions of comments can cause new issues, which can be of concern (something Rob Weir wrote about a few days ago here).
So there you have the view of two pundits. What’s the real story? Here’s what the man with the gavel has to say:
The work however does not end there as the UK must finalise its view on other NBs’ comments too. As the JTC 1 Directives explicitly state, the reason why all NB comments are distributed is to allow all NBs to form an opinion on all of them:
Upon receipt of the ballot results, and any comments, the SC Secretariat shall distribute this material to the SC NBs […] The NBs shall be requested to consider the comments and to form opinions on their acceptability. (13.6)
By extension, of course, NBs shall naturally be considering Ecma’s responses to these comments too. It is this considered national position that delegations will be taking to Geneva:
NBs […] shall appoint to the ballot resolution group one or more representatives who are well aware of the NB’s position. (13.7)
So, NBs need to do their homework so that delegations arriving at the BRM in Geneva are fully briefed. The delegation should ideally know their national position on all 1,000 or so distinct comment/responses that could be discussed. It is the responsibility of the delegation to faithfully represent their national position (not individual divergent delegate views), and to be prepared to respond to any fresh issues that arise in line with guidance their NB has given them.
The point of the above isn’t to show that I’m right and Rick is wrong, but rather that it doesn’t matter – what matters is what Alex thinks, and also what the Directives say, and not what people may be used to in previous, less contentious settings where having a copy of the Directives at your elbow, and then acting in strict compliance with them (or at least to the best of one’s good faith interpretation) is not as important. And that’s why I’m going out of my way to point you to Alex’s blog, so that you can keep current with how his planning is progressing for the BRM.
Here are some other significant items from Alex’s latest blog entry:
Given the five day time limit of the BRM, a frequently asked question is: how can 1,000 issues be addressed in the time, even if NBs already know what their position is? The answer, I think, must lie in paper voting. I am sure that the overwhelming majority of meeting resolutions will be decided by voting (as allowed for by the JTC 1 Directives), and delegations will be given lengthy voting papers allowing them to approve, abstain, or disapprove for any proposed resolution. The voting papers are likely to have three kinds of proposed resolution listed on them:
Verbatim responses from Ecma’s proposed disposition of comments (as contained in the document published by SC 34 as N 980)
Ecma responses that have been amended by the BRM
Fresh responses arising from BRM discussion
for the latter two types, consensus might well be reached during in-session discussion, in which case there is obviously no need to put the proposed resolution to the additional test of a redundant vote.
Anyone who has had to bother with paper voting in a crowd will know that even this will be quite time consuming. Hence, the definition of what "consensus" means will likely become quite important. Here’s what Alex has to say on that point:
In ISO (and as adopted by JTC 1), the word “consensus” has a specific meaning:
[…] general agreement, characterised by the absence of sustained opposition to substantial issues by any important part of the concerned interests and by a process that involves seeking to take into account the views of all parties concerned and to reconcile any conflicting arguments. Consensus need not imply unanimity.
Different meeting chairs take different approaches to determining consensus. In general, if the existence of consensus is not beyond doubt on any issue at the BRM, it will be deferred to paper balloting alongside the undiscussed issues.
So what happens when the clock runs out, especially if not all 1,000 comments have been addressed to the satisfaction of the NB delegates?
I asked Alex that question in a comment to his preceding blog entry, which he answered as follows:
I’m curious how the decision was made to take one week, without provision for an extension? Clearly, no one wants to spend forever discussing one standard, but other than the practical reality that five days is the space between two weekends, there doesn’t seem to be any magic to that amount of time.
[Alex] The decision to schedule the BRM strictly as a five-day meeting was taken at a very high level within ISO/IEC. I am informed that such timetabling has happened before.
Certainly, having a deadline will help people focus, but deciding in advance the amount of time that can be spent doesn’t seem to serve anyone very well otherwise. It seems that with this much to cover, you’re as like as not not going to be able to cover everything to everyone’s satisfaction.
[Alex] You’re right in that some people from both "sides" consider 5 days too little. However, on the other hand some standards veterans think this DIS is taking quite enough time already, thank you very much!
If that’s the case, some people are going to be unhappy, no matter which way the final vote comes out – either proponents will be unhappy, if DIS 29500 fails because some NB’s think it’s not "done," or opponents (and neutrals) will be unhappy, because they think that the standard is now adopted, even though it’s not "done." And the process itself is shown to be inadequate to the challenge as well.
[Alex] Ecma chose to fast track the DIS, knowing perhaps more than anyone about both the spirit and the letter of the Fast Track process. In the end, it’s up to the NBs whether the DIS gets passed or not, and a consideration of whether the process is adequate is a perfectly legitimate question they can ask themselves in making their decision, as this FAQ item makes clear.
I found Alex’s last comment particularly interesting from a strategic point of view. As I’ve repeatedly noted in a variety of prior blog entries over the past two years, Microsoft has adopted a high risk strategy by pushing OOXML so aggressively through the Ecma, and then the ISO/IEC JTC1 process. Already, it’s received one set back, in that its failure to gain approval in the first voting period has resulted in much bad press, and a seven month delay (through the expiration of the second consideration period, which will end on March 30).
If those extra months had been invested in a voluntary extension of the Fast Track period, perhaps more comments could have been resolved to everyone’s satisfaction prior to the BRM. On March 30, we’ll find out whether this pedal to the floor strategy succeeds or backfires, because if not enough votes change to the plus column, than any reconsideration of OOXML will take a very long time under JTC1 rules.
Until then, Alex’s blog is the place to stay in touch with What Happens Next. Hopefully, he will still be in as good cheer at the end of the day on February 29 as he was when he closed his latest blog entry, as follows:
This will be no love-in: I am expecting some hard work and high-quality technical discussion!
I’ll give odds he’s right on the first point. It will be interesting to see if his hopes are fulfilled on the second.
For further blog entries on , click here
The rational thing to do in the face of so many comments is to cancel "Fast Track" and put the proposed standard into "Slow Track" — "Very Slow Track."
A five day BRM without extension doesn’t seem to be a problem to me. If you can’t handle a BRM in five days, it means the standard is nowhere near ready. If by the end of the BRM the job is not done, NBs should be able to draw correct conclusions.
It sounds like Alex Brown is an even-handed man. Much like what happened in Mass., I’d guess that there will be tremendous pressure to have him removed. I also can’t imagine that they could get through 1000 "issues" in five days. That’s over 25/hour, or one every 2 minutes, 24 seconds. Good luck with that! "Unenviable task," indeed!
The Australia National Body requested that comments from the community not duplicate the known comments of other National Bodies.
If Rick’s argument were proceeded with then the Australian community would be disenfranchised.
Glad to read that this won’t be the case.
The lack of formal process for the NB to determine how its representatives are briefed is a concern. Especially as Australia abstained due to a "lack of consensus". That lack of consensus still continues, so just what can the Australia National Body say when briefing its representatives? The lack of transparency in this process is astonishing, especially when compared to the IETF’s and IEEE’s standards-making processes.
Glen Turner.
I am glad to see the pointer to Alex Brown’s blog, it is very good indeed.
But please note Alex says The delegation should ideally know their national position on all 1,000 or so distinct comment/responses while the paragraphs of mine you quote were responding to your curious evocation of "one man" reviewing all the dispositions (i.e. prior to the national body setting its position) as some kind of typical case. Not the same thing, and I apologize that I didn’t make my comment clear enough to withstand transplanting into a different context like that.
Don’t forget that the national position could be "We have no interest or strong feelings on this issue". For example, an NB might say "We support deprecating and removing VML, so we have no particular interest in reviewing comments relating to particular details: the delegation is free to agree to anything that seems sensible or abstain at their discretion." A delegation might be there to observe or to make sure that its pet issues were dealt with. A delegation from a OOXML-sceptical National Body might be there to make sure that the spec is as least-bad (from their POV) as it could be, and still be making a positive contribution to things.
It certainly will be more work for one-man delegations, if they want to take individual positions on each issue (rather than, for example, "abstain" or "vote with Freedonia" or "do what is consistent with our general goals") and if they are also the sole reviewers for their national bodies and if they are out of contact with experts from home, but if my math serves me correctly, with 35 NBs attending, 120 person limit (I dare not say "seats"), and a limit of 4 per delegation, there could only be 6 single-person national delegations out of the 35 at a maximum (e.g. 28x4person, 1x2person, 6x1person) i.e. 1 in 6 maximum.
Cheers, and keep up the good work
Rick Jelliffe
Rick,
I think we’re mostly looking at the same elephant, although each of us is more concerned with some parts of its anatomy more than others. One place we do see it differently, though, is on the size of the delgations. The maximum allowable size is 6, rather than 4 (the US, for example, is sending the full 6; I’m not sure how many other NBs are as well). Going back to the same elephant, I’m not sure that a delegation of 2 is going to be a whole lot less stressed than a delegation of 1, and I exepct that there may be a number of two person delegations, purely for cost reasons.
That said, I certainly agree that different delegations will have differing length lists of comments with which they have genuine concern, and I expect that some will have short lists, although they will have to pay close attention to all comments to see which head in directions beyond the proposed disposition, to be sure that no new dominoes are proposed that might fall in a direction that leads to a concern.
I also think that the fact remains that to vet even the 1,000 dispositions, and then decide which are of concern before the meeting, is quite a task. And the task of following the dialogue, voting and changes through the week in Geneva will be another. No matter how you slice it, it’s a very big job, and not necessarily one that you would design from a fresh sheet of paper for a spec that’s this long.
The bottom line, from my point of view, is that this may be a process that worked well in the past in less contentious settings and working with shorter specifications, but that it has been severely tested from start to finish in the current case. That seems to me like just a bad way to run a railroad, no matter what your stand on the standard itself may be.
Or, to continue the zoological metaphor, we shouldn’t expect a system created to design a mouse to be able to process an Elephant in the same period of time. This may, however, help explain the final appearance of the platypus.
– Andy
Andy: I see the platypus has just last week been discovered to be <a href="http://www.smh.com.au/news/science/platypus-fossil-as-old-as-the-dinosaurs/2008/01/22/1200764228042.html">as old as the dinosaurs</a>. It is the elephant that has been fast-tracked 🙂