Alex,

I'm a little surprised by your blog entry for two reasons. First, it doesn't have anything to say on the merits, which are important for those who filed the appeals, for others who have concerns, and for the integrity of the process.

But second, in your prior posts and comments at my blog, I recall that you've stated (as criticisms) that NBs ought to get together and talk about things when they have problems with technical issues. I believe that you may even have said that in connection with appeals, although my memory isn't that good.

Either way, this seems like the type of situation where it makes perfect sense for different NBs that are aware that they share common concerns to talk to each other, and even share drafts. Why not? Wouldn't that make things easier for all concerned in the process, if they can reach consensus on their concerns before they request action? I'm not sure why you would see anything inappropriate here, and your take seems inconsistent with prior statements you've made.

For my thoughts on the recommendation, see: http://www.consortiuminfo.org/standards ... 9060030380

Best,

Andy

@Andy

> I'm a little surprised by your blog entry for two reasons. First, it doesn't have
> anything to say on the merits, which are important for those who filed the appeals,
> for others who have concerns, and for the integrity of the process.

It's really not my place to comment on that. I think the document speaks for itself - very clearly.

> But second, in your prior posts and comments at my blog, I recall that you've stated
> (as criticisms) that NBs ought to get together and talk about things when they have
> problems with technical issues. I believe that you may even have said that in
> connection with appeals, although my memory isn't that good.

I sincerely hope note, since this was a very sensitive topic in the run up to the BRM: for late-stage documents the Directives clearly state: "NB positions on the document under ballot are not to be formally discussed at any working level." One of my proposed plans (inviting NBs to share their responses to Ecma's proposed disposition) foundered because of that very rule. Of course, an appeal is not a "document under ballot" so there is nothing in the rules to prohibit coordinated appealing.

(I happen to think this is a reductive rule, but that's another question).

> Either way, this seems like the type of situation where it makes perfect sense for
> different NBs that are aware that they share common concerns to talk to each other,
> and even share drafts. Why not?

Well, like for form-letters, their content is likely to get proportionately discounted when considered.

Do you think Venezuela and South Africa shared drafts?

> Wouldn't that make things easier for all concerned in the process, if they can reach
> consensus on their concerns before they request action? I'm not sure why you would
> see anything inappropriate here, and your take seems inconsistent with prior statements
> you've made.

The problem is that this is meant to be an international process, not a process decided by opaque alliances. As I say, I don't think this is inconsistent with anything I've written previously. At least I hope not!

> For my thoughts on the recommendation, see: http://www.consortiuminfo.org/standards ... 9060030380

I read it, of course, as soon as it hit my feed reader ...


- Alex.

If I recall it right the South African appeal was published early and the appeal of Venezuela came a bit late. I would however suggest that there is a common source, a source protest letter A, a position paper of a party with contacts to both NBs and maybe more.

The insightful message is that the ISO secretariat does not only know a default "Yes" but also a default "No".

After reading the pdf document, my feeling is the system is in a mess.

Nobody, ISO, NBs, joe public, Andy Updegrove, me etc, seems to know where the responsibility for different aspects of the process lies. This is serious, as it means parties can manipulate the process to suit their purpose. One example is the Microsoft spreading the view that there ISO had determined, before BRM, that there is no contradiction with ODF, but the report says that NB must take into account contradiction when casting the final vote.

Some of the recommendation seems to be in contradiction with some written rules, such as P member/O member voting. It is strange to see the report conclude that Brazil, an O member, have no rights to appeal but says voting by O member in BRM is fine (which some parties says is actually against the rule)

One interesting thing which I think we need your view is the fact I recall a few delegates seems to be under the impression that they are there to do their best to reconcile the problems. I have a feeling that a lot of them voted for the "bundling of 800+ comments" in a vote because of this instruction, regardless of how they feel about the bundling. Now it seems it comes back to bite them in the butt when the report quote the voting results as a reason why bundling of 800+ comments are fine.

I will go as far as saying that the appeal is now bigger than OOXML approval debate. I hope the four NB continues the appeal: One need to get ISO to write down the precise "rules of engagement" for future reference.

@Wu MingShi

Interested commercial entities are obviously not to be taken as authoritative sources for information on standards procedure. Obviously.

The P-member and O-member voting was all carried out correctly, as the ISO document confirms. Again, it should be obvious that such things were carefully decided in advance of the meeting. I accept the procedural description in the Directives is unnecessarily arcane here - it was my problem too.

The voting at the BRM was the result of attending NB delegations considering all available options and choosing what they thought was the best one. The NBs knew what they were doing, and appreciated the constaints of the situation. If any NB thought the BRM was inadequate they could subsequently disapprove the DIS (as they were repeatledly informed). Overall, they did not.

This is not to say there is not a requirement for process improvement - there most certainly is. But ultimately for this project the (no doubt carefully crafted) recommendation of this leaked document sums the situation up very well.

- Alex.

Alex,

>I sincerely hope note,

We've certainly traded enough blog posts, comments and email that my memory may be wide of the mark here, but what I recall best (or think I do) was a comment that people shouldn't have shown up at the BRM and expected to talk through issues that they could have addressed among themselves before hand and suggested something. But it doesn't really matter, as it only related to my puzzlement rather than substance.

>...the Directives clearly state: "NB positions on the document under ballot are not to be formally discussed at any working level."

I guess I am once again puzzled by Directives language. What does "formally discussed" mean? And what is a "working level?" And while I understand that someone needs to be able to call the coin toss, how does tha help anyone trying to operate in good faith under the Directives?

>Of course, an appeal is not a "document under ballot" so there is nothing in the rules to prohibit coordinated appealing.

So I guess I'm puzzled again why it would sound inappropropriate?

>Well, like for form-letters, their content is likely to get proportionately discounted when considered.

I have to say that I really don't get this at all. A bunch of people sit in a room (the BRM) voice displeasure at the time, chat in the hallways (probably with you as well), and then are not allowed to talk about their joint displeasure after they leave Geneva? I just really don't see what the concern would be here. My honest reaction is that such a reaction ("discounted") would be highly inappropriate and an example of arbitrary and subjective practice on either the CEO's or the ITB's part.

>Do you think Venezuela and South Africa shared drafts?

I have no information at all on this other than what I read in your original blog entry. While I know folks in South Africa, I have seen nothing to indicate that this happened, and I don't know anyone, nor have I been in communication with anyone in Venezueala.

>The problem is that this is meant to be an international process, not a process decided by opaque alliances.

I see the theoretical concern, but to me this just seems like an over reaction. It seems like not only is there no way for a genuinely upset National Body to have its concerns addressed ("sorry - if you were so upset about the BRM, you should have turned around and gone home again. Otherwise, we'll count your continued participation as concurrence"). And now its appeal letter can even be discounted if it sounds to similar to another. Alex, I'm sorry, but I lose more respect for ISO/IEC by the day.

Finally, I can't resist noting this from your reply to Wu Mingh Shi:

>Interested commercial entities are obviously not to be taken as authoritative sources for information on standards procedure. Obviously.

I'd suggest that interested parties (i.e., the CEOs of ISO and IEC, who I expect were part of the decision making under review) not to be taken as authoritative sources for recommendations, either. Obviously.

Best,

- Andy

The name change request is particularly odd, because clearly IS29500 has absolutely nothing to do with the *process* of converting from the binary to the new format, it is only about representing the same information.

Now, I am happy if there is a technical report on this subject. And I am happy if the MS binaries are somehow standardized, such as through a technical report. And I understand that there are stakeholders who see that the central problem that needs to be addressed is of substituting Office for other products and that the binaries were more interesting from that POV. However, it just isn't the scope of IS29500.

Was the thought process something like this: we don't want a standard on this, we want a standard on that, therefore if we rename the standard for this as "A standard for That" we can then point out that it does not in fact match its name and get it withdrawn? A cunning plan, Baldrick...

A request for a name change would be just as perplexing a request to change the scope of ISO SQL to being "standard for converting from System R data into relational data." It just isn't, even though such a standard may be a good thing.

Now I do think the name and the scope are incorrect, but because they should describe the technical rationale (e.g. linear work processing documents, etc.) not the provenance.

There are stakeholders, integrators in particular, who don't have any particular interest in product substitutability: they are presented with data in office formats as a fait accompli and they want adequate documentation in an XML format. When ODF grows to provide that, IS29500 can fade away, mission accomplished.

(On the issue of the ISO discussion paper, I'd also like to point out that it seems to be about the issue of fast-tracked amendments to standards.)

You are right Alex, that's exactly like the form letters fuss in the early stages of "The project" I can even read the ‘Type standards organization name here’ parts of them...


In all seriousness, have you gone so low that are blaming the four countries of that? Under what foundation? Oh, similarities... Those similarities would be sooo suspicious if it wasn't for the fact the four letters are about the same topic!

First time I'm posting here,

I'm a little bit surprised that Mr Brown is playing judge and party. But of course it's his blog, and perhaps he's acting on the precedent set by Microsoft who have been playing this as well by stuffing all major national bodies around the world as well (with their own employees and close partners). So I guess we have to deal with hypocrites anyway.

Second, I find it a little disappointing the lack of humility of every single commenter here. None of those people have implemented OOXML in some way. I'm not talking just 50% of it, which would in and of itself would leave enough room for being unable to make clear decisions on whether the OOXML proposal achieves its goal or not. No, those guys have implemented absolutely nothing of OOXML. And yet, we can hear them not only on their blogs, but in comment areas of all relevant blogs, in national bodies, and who knows how many people behind the scene. They are saying that the ISO proposal is genuine. On what grounds exactly?

To say that it is ludicrous is a euphemism.

Actual implementers of OOXML should be given a weight at least 10 times more than all those commenters combined.

What did Microsoft and their crownies choose to do?

One of those commenters has the gal to comment on the bad title of the ISO proposal, saying that calling it "Converting legacy Microsoft documents to Office Open XML" would be somehow inappropriate.

Well, anyone who actually DID read this paper knows that it defines a format based on another format without ever making public the mapping tables so that people in the know can verify whether the claims are right or not. In other words, the claims are absolutely not verifiable. How can you push a proposal like this?

See, Microsoft FUD people, the bloggers, started a translation project whose goal is precisely to convert so-called legacy files to new files. Well, take a look at the status of this project : http://b2xtranslator.svn.sourceforge.ne ... tor/trunk/

(in case you don't get the point, this project is dead.)


It's nice to see so many non-Microsoft people helping that little company that really needed their help, because you know Microsoft has not much influence in IT circles.


Stephane Rodriguez
Independent vendor
diffopc+, xlsgen, ...

Hi Stephane,

Your sentiment that "implementers of OOXML should be given a weight at least 10 times more ..." (which I generally agree with) should mitigate some of your concerns about Microsoft partners participating in the process.

In the case of INCITS V1, for example, we had input from Microsoft partners who have developed large-scale OOXML-based solutions for their customers, and they brought that experience to bear in their assessment of DIS29500.

I believe Rick and Alex also have hands-on experience with OOXML. It's good to see so many informed voices in this discussion, in my opinion.

- Doug

I'm a little hesitant to add my comments to those of the heavyweights above, since I'm mainly just an interested spectator in this process. (obviously, though, I'm not _that_ hesitant, since I'm going to do just that.)

First, I seem to recall the anti-ooxml crowd being upset because Microsoft had many of it's "associates" send in form letters to the NBs and TCs to support the passage of OOXML through the standards. In most of those cases, the Microsoft "associates" didn't take the time to do anything but cut and paste the text onto their letterhead, and put a stamp on it. Honestly, I think those can (and should) be "discounted" as Mr. Brown suggests in his blog post.

However, to then suggest that because the complaints and recommendations on how to resolve the complaints against the passing of DIS 29500 all converge to the same point, and have the same wording, regardless of the other thousands of non-identical words of text, should result in the same loss of credibility as the Microsoft provided form letters is disingenuous.

Or are we supposed to believe that all of Microsoft's "associates" all had the chance to discuss and agree on their support of the OOXML proposal before sending in the form letters?

Second, I don't really find the name they proposed odd. Wasn't the whole stated point of OOXML to be able to capture all of the information in legacy documents into an XML form without loss? We were told so many times that ODF can't accomplish this, and thus we need OOXML in some form. Frankly, if that's not it's purpose, someone should be telling that to the ISO. We already have an accepted document format for editable documents, and a second would conflict with that. So yes, calling this format "Converting legacy Microsoft documents to Office Open XML" makes perfect sense.

@Andy

Additional points of appeal that reiterate mistaken points that also occur in other submissions simply do not add to the weight of the appeal. It's a parallel situation to how the 4,000 comments on the original ballot turned out to be 1,000 after de-duping. So if Vulgaria (say) appealed and submitted an appeal text that was word-for-word identical to something that had already been considered, there would be no need to consider it separately. It would be discounted.

Venezuela, BTW, were not at the BRM (I think they approved the DIS in the September 2007 ballot) so the picture you paint of their displeasure in the Geneva corridors simmering over doesn't quite ring true ...

You did not comment on the "common source" possibility for the identical wording; as with the September 2007 letter ballot that is another possibility.


@Andre

You suggest there was "a position paper of a party with contacts to both NBs and maybe more". Is this just a wild guess or do you know more ... ?


@Rick

The name change suggestion *is* bizarre (not that the current name is any good!)


@Vexorian

It helps if you read the entry before commenting.


@Stephane Rodriguez

Thank you for supplying the humility that you noted was lacking before

Implementers should indeed be listened to, but on a standards note it was interesting that NBs were (slightly to my surprise) not that bothered about the claim that a goal of OOXML is to represent the legacy MS corpus. By and large, they seemed to accept that this was merely stated as "a goal", and not a "fact" in the text of the standard.

I'd be interested in your opinion on whether Office 2007 itself is a good converter of legacy content into Ecma 376 format.


@Doug

My experience with Office XML is more with the 2003 flavour than with OOXML (and we have the scars to prove it). By and large, in this standards project my input was far more on the admin side, than on the technical.


@Anthony

As Rick suggests, the proposed name is wrong because the text itself does not describe a *process*, it describes a *format*. Maybe something like "Office Document Format for Legacy Content" would express better what I think was intended. However, the NBs themselves voted on a scope statement for the text which made plain it should look forward as well as back, so I'd be susicious of names changes like this as being a sort of reverse-branding exercise.

- Alex.

@Doug Mahugh

"Your sentiment that "implementers of OOXML should be given a weight at least 10 times more ..." (which I generally agree with) should mitigate some of your concerns about Microsoft partners participating in the process."

Hmm. You probably remember you guys were giving me high praise (you were evangelist at the time, and there was absolutely no such thing as community support for OOXML, still none by the way) towards me and my products, until I decided to start a blog where I would post some of troubles implementing OOXML and related stuff. That's when you guys decided to go the opposite way.
Fair?

I'm not sure about your comment. Can you name a single full non-Microsoft implementation of OOXML? Remember I asked this question on your blog about two years ago, and you fail to answer it even now. Many Microsoft partners were here to just do what they were asked for or face troubles with Microsoft. I saw that in Afnor/FRANCE for instance. Some of these include vendors who would simply associate file extensions like ".docx" with "Office Word 2007 document". Yes, that is the kind of thing you call implementation of OOXML. Don't laugh, we are not done yet.

ISO's Alan Bryden wrote in his recent paper that national bodies had the chance to vote for or against the proposal. Well, when so many national bodies are stuffed with a majority of Microsoft employees and/or partners, or when the comments are outright ignored only because they are not in favor of Microsoft, I wonder how "national bodies had the chance to vote for or against it".

I wonder how you call this transparent when so much obviously happened behind the scene. Now that you are an Indian citizen, or may be that was to influence the Indian national body, perhaps you can tell us a bit more. Aren't international standards meant to benefit everyone? If so, why isn't Microsoft going through the normal standards procedure only to make sure their proposal meets the stated goals?

"Open Standard"? Hmm. The paper that were made available earlier this month mention that they are protected by patents. How do you reconcile this? (by the way, the fact that Microsoft finally posted earlier this month what I have been asking they should publish is fine. It simply acknowledges what I have been saying was right. But then, due to the giant gap still present, the only logical conclusion is that the proposal should be brought back to the drawing board so that Microsoft gets a chance to come up with a much better OOXML. With all the help provided by non-Microsoft people already, that leaves a lot of the cost to others. But of course, perhaps Microsoft is just not interested in having a usable OOXML.)


"In the case of INCITS V1, for example, we had input from Microsoft partners who have developed large-scale OOXML-based solutions for their customers, and they brought that experience to bear in their assessment of DIS29500."

I have followed this very well. You never answered Patrick Durausau and Jon Bozak concerns about the gap between the ISO proposal title and the actual contents. This is all readable even now since Bozak has a public archive of the entire thing (well at least what was posted through emails). How do you reconcile this? Again, defining a new format solely on another one without making the mapping tables public is ludicrous.

Large scale implementations? Hmm. I think I know what you are talking about. These implementations have little to do with OOXML and more to do with trivial stitching of bits and parts, most often template based scenarios. You seem to never mention what matters, or you conveniently avoid it, that a really significant implementation needs to understand the actual bits and parts, needs to provide some kind of rendering/calculation of the bits and parts otherwise those solutions are simply little addons to stuff that gets ALWAYS viewed in paid licenses of Office 2007. The kind of thing OOXML is supposed to help avoid, unless I got the memo wrong obviously.
When you get to really implement bits of OOXML, you get half a dozen variants of vector graphics markup in Word alone. Which means you are in it for a decade, only because Microsoft chose to not streamline the legacy stuff into a consistent and factored thing.
You know why you guys did not do so? Because, by preserving all the old mess, you ensure that any competitor has to implement everything. Very simple to understand. ISO standards, on the other hand, rely on other standards to help factor in all the work. This is precisely what you refused to do. Morally speaking, OOXML is at complete odds with existing ISO standards.
Any non trivial stitching of bits and parts would conflict with the fact that, for stated performance reasons, Microsoft chose to have indexes and other references at a much granular level than the part level. Meaning that implementers are caught in a trap.

"I believe Rick and Alex also have hands-on experience with OOXML. It's good to see so many informed voices in this discussion, in my opinion."

I directly asked Rick, he answered "none". I haven't asked Alex Brown, but what he posts suggests the same.


@Alex Brown

"Implementers should indeed be listened to, but on a standards note it was interesting that NBs were (slightly to my surprise) not that bothered about the claim that a goal of OOXML is to represent the legacy MS corpus."

This does not seem to reflect what I have seen posted from people in national bodies. For instance, at complete odds with what was discussed in Afnor/FRANCE, which initially voted no.

Of course, if your reference national body is one of those who became P-members just days before the first vote, and now no more, well I guess it speaks by itself...



"I'd be interested in your opinion on whether Office 2007 itself is a good converter of legacy content into Ecma 376 format."

It does not convert into ECMA376. It converts back and forth between BIFF12/DOC12/PPT12 and Office 2007 DOCX/DOCM/XLSX/XLSM/PPTX/PPTM (there are a couple more variants, but you get the idea).

Why, you ask? Because ECMA 376 describes a fraction of the parts whereas BIFF12/DOC12/PPT12 contain bits that are converted into binary parts, not described in ECMA 376.

In fact, preserving binary parts even in a so-called fully XML file format was part of the Microsoft agenda :
- easy to convert back and forth. That's why we have half a dozen vector graphics formats in Office 2007 documents, not just one. A software engineer should be embarassed reading this. But of course I can only speculate the quality tools and methodologies used by Microsoft employees. And what to say about implementers out there, who have to catch up with all of this?
- easy means very quickly done. No need to think about a file format really thought to address 21st century problems.
- binary parts means patents brought in Office 2007 documents. Microsoft cleared the way by saying ECMA 376 is royalty-free. But of course since Office 2007 documents is a superset of ECMA 376, and that in real life people manipulate Office 2007 documents anyway, this does not mean much.
- No "Save as ECMA 376" option in Office 2007.

In fact, if you read my blog OOXML is defective by design, where I explain a little bit the problems when it comes to implementing OOXML, you would have had those answers already.


Stephane Rodriguez
Independent vendor
diffopc+, xlsgen, ...

Andy,

Groklaw has posted a link that suggests Brazil is a P-member, but according to this decision they're not and I don't recall them voting on 29500 as such before... Could you clarify for me?

-Gerben

@Gerben

I believe Brazil is a P-Member of SC 34, not a P-Member of JTC 1. To appeal the result of a JTC 1 ballot (as was the DIS 29500 one), you need to be a P-Member of JTC 1.

- Alex.

One would have thought any self-respecting standards body would not brazenly crib the text of an appeal (an appeal!) from outside its own walls ...

Well, are you suggesting, that the NBs are restricted from working together and making some parts of their appeals common? Or maybe you try to suggest that it's IP issue and one of them should now sue the other for blatantly copying the other?

a truly ridiculous excuse for an argument on your part

arty

Alex,

>Additional points of appeal that reiterate mistaken points that also occur in other submissions simply do not add to the weight of the appeal.

This seems to me to be the opposite conclusion to be drawn. If only one NB makes a point, it may of course be mistaken. If more than one are concerned in the same point, and then echo it, this seems to me to be an indication that the point is more likely to be of real concern, and to be given greater presumption of validity.

What you are doing seems to be to try to undercut the point by implying the type of action that you have deplored so many times in the past. In other words, that someone must have bought or bullied the standards bodies of four separate, sovereign nations into colluding to push a spurious point. Or, in the alternative, that the NBs are somehow acting in bad faith if they discuss and agree upon a common concern. I do not know what is motivating you to do this, but I expect that you are not making friends in any of these countries among the standards professionals that felt strongly enough to take the time, and the heat, to file appeals.

>It's a parallel situation to how the 4,000 comments on the original ballot turned out to be 1,000 after de-duping. So if Vulgaria (say) appealed and submitted an appeal text that was word-for-word identical to something that had already been considered, there would be no need to consider it separately. It would be discounted.

"Vulgaria." Interesting choice of words. Please see prior point.

>Venezuela, BTW, were not at the BRM (I think they approved the DIS in the September 2007 ballot) so the picture you paint of their displeasure in the Geneva corridors simmering over doesn't quite ring true ...

If I was wrong on that point, then let me try again: apparently they were so distressed over what happened, then notwithstanding their prior vote they decided to appeal. The point remains.

>You did not comment on the "common source" possibility for the identical wording; as with the September 2007 letter ballot that is another possibility.

I thought I had, but let me try again. I don't think that your blog entry has any premise to begin with, other than to impugn the motives of National Bodies acting in good faith. You've already admitted above that there was no rule to prevent NBs talking among themselves. I don't know if that happened or not, but if so, so what? What's your point? Let's say two or more NBs checked in with each other to say, "Hey, am I off base on this? What do you think?" Why shouldn't they share common wording among themselves rather than try to disguise a discussion? Would that be preferable, and if so, why?

Alex, I have always felt that you are more satisfied with the process than you should be. Having been put in the position of managing the BRM this is understandable. But this post seems more like evidence of bias, even if unconscious. Sure, there are people who post comments at blogs that may be poorly informed. But these are the National Bodies that own the right to be happy or unhappy with the process, and they deserve to be heard, and not demeaned.

Rules that would prevent, or impugn, NB consultation are very convenient to entrench management, and serve no other useful purpose in a situation like this. It would be convenient, for example, for England* to have a rule that would have prevented its 13 American colonies from sharing grievances before raising them. I expect they did have such a rule, and called it "treason." That didn't end well for management, and I expect that this might not either.

In this case, I think that Patrick Durusau and I are very close to each other in our conclusions. He recently wrote: http://www.durusau.net/publications/appealprocess.pdf

"The only difference [after the appeals] is the loss of at least 240 days (approximately 8 months) we could [have spent] re-casting the Directives to: 1) empower National Bodies; 2) elect a JTC 1 Chair and Secretariat who serve the National Bodies (not the other way around); 3) eliminate both Fast-track and [publicly-available specification] submissions; 4) encourage meaningful relationships with other standards organizations; 5) empower [standards committees] to decide on submissions and their processing; 6) eliminate the vague, confused and contradictory Directives we have today. It is time that national bodies shook themselves out of their lethargy and took their rightful places as being in charge of JTC 1. They have nothing to lose but their chains."

I expect that those in charge of the NBs of the four Vulgarias that filed the appeals are more inclined to agree with Patrick after reading your blog entry and subsequent comments than they were before.

- Andy

*Note to Andrew Sayers: Yes, this time I do mean "England" and not Great Britain - I think a Scot would prefer that characterization in 1776 as well.

> This seems to me to be the opposite conclusion to be drawn. If only one NB makes a
> point, it may of course be mistaken. If more than one are concerned in the same point,
> and then echo it, this seems to me to be an indication that the point is more likely
> to be of real concern, and to be given greater presumption of validity.

I disagree. A point of appeal is either valid or invalid, no matter how it is made.

> What you are doing seems to be to try to undercut the point by implying the type of
> action that you have deplored so many times in the past. In other words, that someone
> must have bought or bullied the standards bodies of four separate, sovereign

I never mentioned buying or bullying, and mentioned the similarity between only 2 appeals. *You* (like Groklaw) are raising the number (incorrectly) to 4. Groklaw wrote that the basis of my blog is that "the four [appeals] have a sentence here and there that is similar". Please don't similarly misrepresent what I put. What I have pointed out is that two appeals have portions of text that are identical or near-identical.

> nations into colluding to push a spurious point. Or, in the alternative, that the
> NBs are somehow acting in bad faith if they discuss and agree upon a common concern.

Bad faith is not a possibility in that scenario.

> "Vulgaria." Interesting choice of words. Please see prior point.

Ha! you're over-reading things here, Andy. Vulgaria is just a an often-used fictional country name (have you not seen Chitty Chitty Bang Bang?)

> If I was wrong on that point, then let me try again: apparently they were so
> distressed over what happened, then notwithstanding their prior vote they decided
> to appeal. The point remains.

I also notice they made the same mistake as SA in complaining about the lack of a BRM report, despite the fact it was circulated as a public document after the BRM (it was/is public on the SC 34 site).

> I thought I had, but let me try again. I don't think that your blog entry has any
> premise to begin with, other than to impugn the motives of National Bodies acting
> in good faith. You've already admitted above that there was no rule to prevent
> NBs talking among themselves. I don't know if that happened or not, but if so, so what?
> What's your point? Let's say two or more NBs checked in with each other to say, "Hey,
> am I off base on this? What do you think?" Why shouldn't they share common wording among
> themselves rather than try to disguise a discussion? Would that be preferable, and if so, why?

No, by common source I'm referring to what Andre mentions, whereby a "position paper" (his words) or some such has been submitted by a third party outisde the NBs.

> Alex, I have always felt that you are more satisfied with the process than you should be.
> Having been put in the position of managing the BRM this is understandable. But this post
> seems more like evidence of bias, even if unconscious. Sure, there are people who post
> comments at blogs that may be poorly informed. But these are the National Bodies that own
> the right to be happy or unhappy with the process, and they deserve to be heard, and not demeaned.

I posted some copied text and asked some questions about how that might have come about. It is an interesting question.

> Rules that would prevent, or impugn, NB consultation are very convenient to entrench management,
> and serve no other useful purpose in a situation like this. It would be convenient, for example,
> for England* to have a rule that would have prevented its 13 American colonies from sharing
> grievances before raising them. I expect they did have such a rule, and called it "treason."

I don't think you'd have needed to share treasonable grievances to commit treason; so I don't see the analogy.

> In this case, I think that Patrick Durusau and I are very close to each other in
> our conclusions. He recently wrote: http://www.durusau.net/publications/appealprocess.pdf

PD: "Appealing ISO/IEC 29500 will not benefit anyone, no matter how the appeal turns out."

Exactly.

The trouble is (as Patrick points out) that the appeals have put any question of reform on ice. Dicussion of Fast Tracks, etc., have been removed from agendas while the appeals are ongoing.

It is partly for that reason that any appeal which is less than fully considered is to be particularly regretted.

- Alex.

Stephane bandies about a thread we had last year or more, and he repeats it regularly, but he never admits his definition of "implementer" is one that excludes most of who I see the target users of OOXML as, in particular users like me.

This is people who do file format conversion. Now indeed I do now have experience with OOXML on this in a few real commercial projects now, as I have almost two decade's experience in data conversions betweem scores of other languages. There is nothing much in OOXML (at least WordProcessingML) that is different from what has been found in other simplistic markup languages: you only have to go to the Rainbow DTD in the early 1990s to see that. And if you complain about complexity but haven't had to deal with HyTime or XBRL, you don't know.

I happily admit I have never implemented a full office suite as a user application. (I have implemented large text editor and validation application, such as the Topologi Markup Editor, so I am actually fairly familiar with desktop application development.) But so what? If OOXML is not specified enough for every need, then the people who have need for extra info should stop whining and participate and extract the information they need by participating.

I am more than happy to allow Stephane his opinion that the parts he is interested in in OOXML are not adequate for his requirements; I hope he will have the humility to admit that there may be other people for whom the parts they are interested in are adequate and useful. Stephane, your technical points are often really good, however they are utterly undermined by your attitude that "my needs are more important than your needs."

Cheers
Rick Jelliffe

Alex,

<Ha! you're over-reading things here, Andy. Vulgaria is just a an often-used fictional country name (have you not seen Chitty Chitty Bang Bang?)

Er, I think I'm missing your point again. Here's a few outtakes from the Wikipedia about Vulgaria:

"After the picnic, the children ask Caractacus to tell them a story, and he begins to tell a tale of the villainous Baron Bomburst...the ruler of Vulgaria, who is out to steal Chitty Chitty Bang Bang....
Baron Bonburst sends two Vulgarian spies to steal the car, but all their schemes are thwarted. Eventually, they mistakenly kidnap Grandpa Potts, thinking he is the inventor, and take him to Vulgaria....[and so on]" http://en.wikipedia.org/wiki/Chitty_Chitty_Bang_Bang

So an NB should feel good about being compared to Vulgaria...why?

>The trouble is (as Patrick points out) that the appeals have put any question of reform on ice. Dicussion of Fast Tracks, etc., have been removed from agendas while the appeals are ongoing.

This, of course, is another decision that need not turn out this way. There is no reason why discussions relating to reform could not occur concurrently. The appeals instead are evidence of the fact that they are more rather than less urgent. New rules will surely not be applied retroactively, so - as you have pointed out more than once - whether the process should be reformed going forward is not relevant to decisions made under existing rules.

- Andy

@Andy

Sorry, you've over-reading again - "Vulgaria" is equivalent to me saying "Country XYZ" - and I didn't "compare" any country to Vulgaria in any case.

Andy, you have a tendency to get spooked by phantoms of your own imagining!

@Andy

BTW, if you haven't seen it, Chitty Chitty Bang Bang is a great film!

- Alex.

Alex,

>Sorry, you've over-reading again - "Vulgaria" is equivalent to me saying "Country XYZ" - and I didn't "compare" any country to Vulgaria in any case.

>Andy, you have a tendency to get spooked by phantoms of your own imagining!

I've learned my lesson on this too many times to agree with you, Alex. I once used the word "Balkanization" in a blog entry once, and got the predictable response from someone in Croatia. A long time ago I used the word "pontificate" in a letter to a Catholic friend, who became extremely upset with me.

Once we release words, we are responsible for how they are heard, whether intended or not, and especially where one has played a personal role in the substance of the appeals in question. I accept that you didn't mean to use "Vulgaria" in this. But let's just say that I think that it was an unfortunate word choice to use in responding to a comment that a blog entry was impugning the permissible activities of National Bodies: "One would have thought any self-respecting standards body would not brazenly crib the text of an appeal (an appeal!) from outside its own walls ..."

- Andy

Holy smokes! So Alex Brown can now put himself *above four sovereign nations of the world*. All of the lowly people who argued against this blatant and unashamed corruption-- who are *they* to question sovereign nations, eh? They're just not Alex Brown, who apparently has that power.

Hypocrite to the core, and don't you dare try to deny it.

@Rick

"I am more than happy to allow Stephane his opinion that the parts he is interested in in OOXML are not adequate for his requirements;"

Interesting statement. What you are saying is that OOXML would be fine it it just meets a couple ad hoc cases that make people like you happy. You are advocating an OOXML specification which does not specify the general case, which is a full implementation.

Can you make an even more ridiculous statement than that?

With a logic like that, when it comes to string encoding handling, you would be happy with ASCII only since you guys don't need accents.



"This is people who do file format conversion."

If you really did (I suspect you have only done the trivial stitching I mentioned before), you would have realized that everywhere you've got to understand and make decisions that are not specified in the specification. Those decisions include mapping values, introducing new ones, hiding others. In the general case, this is not trivial. Worse, those are decisions that are the choosing of every single implementer, therefore every single implementer will have their own OOXML.

Alex,

Without wanting to get in the way of the most adorable flame war I've ever seen, there is an important point to draw from Andy's confusion. For a long time, he I, and others have been trying to explain that there is something not quite right about the way you present your case, but have consistently failed to find a form of words that would make you understand. That sort of misunderstanding is very common in life, and can normally be ignored, but it becomes a serious handicap for someone such as yourself when the cumulative effect makes it hard for moderate people like Andy Updegrove to assume neutrality on your part. Since I don't have the words to say what I'm trying to say, is there anything you would like to ask about the way you come across, that would help inform your future approach?

- Andrew

Alex, if you meant to say Country XYZ you should have said Country XYZ. Using Vuklgaria as a random country is a very graphic word, it brings up all kinds of negative images. Even more so, if you, like me, have seen Chitty Chitty Bang Bang. I personally own a copy of it.

I'd still like to know how do I space like Word 95?

I am an independent consultant and have and do build automated document processors. I won't be implementing OOXML and will recommend to all my customers not to do so either. It would be pointless, since the only reason to do so would be to support Word documents. But, since Word doesn't implement or even support OOXML it would be a failed experiment.

I thought ISO had lost all credibility but you continue to educate me that you still haven't hit bottom.

Hopefully some other organization will step in and do the job that ISO should have done.

Furthermore your statement that the ISO rules were followed is ludicrous. Clearly the Directives do not allow for O member voting in the particular instance of which is being discussed. How can a standards body be relevant when they don't even comply with their own standards and rules?
Oh, how the mighty have fallen.

BTW, I make most of my money from supporting Microsoft formats. But, I am so small potatoes, that I have nothing to fear from Microsoft. Although, internally we use truly open formats like ODF.

Two or more countries wish to appeal an issue for "similar" reasons. These countries can either write "unique" appeals or, with obvious coordination, jointly create one appeal which the countries believe accurately reflects the similar reasons for which the countries are appealing.

If the countries submit unique appeals, the body ruling on the appeals has to read and rule on both of them. Due to the differences in wording, the ruling body may rule differently on the two appeals despite the original premise that the appeals are similar. If the appeals are unique, some of them might even ask for mutually exclusive remedies.

For example, two countries may disagree with somethings name (call it Name A). One country may propose changing Name A to Name B. Another country may propose changing Name A to Name C. Depending on the details of the appealed issue, the ruling body may let Name A stand because of lack of consensus about what is a proper name. A truly independent ruling body may do so only reluctantly, but a ruling body predisposed against appellants or the appeal may welcome the opportunities such confusion brings.

If the countries submit a common appeal, the ruling body only has to consider that one appeal, including which (if any) requested remedies to apply.

If each country has a unique appeal, it is obvious why each country should submit its own appeal. But why should both countries submit the common appeal if the ruling body will just "discount" the duplicates? Shouldn't a point of order be valid or not regardless of how many countries submit appeals?

In an appeals system that does not allow countries to withdraw appeals once submitted, and requires the ruling body to issue a ruling regardless of whether or not the submitting country changes its mind, the duplicate submissions may seem frivolous (although simply as a matter of national pride, both countries might want to submit appeals so that it doesn't look like one country is active and the other passive).

National pride aside, if the appeals system allows countries to withdraw appeals, and only one country submitted the appeal, that country could withdraw its appeal and end the issue regardless of what its "conspirator" countries wanted. With only one country submitting an appeal, any parties that would stand to lose something with a successful appeal could better focus its efforts and resources by changing only one country's mind.

In short, countries with similar issues act most rationally when they work together to develop common language for those issues, and each country acts most rationally when it submits an appeal with that language instead relying on an unaccountable third party (e.g., other countries) to submit the appeal.

Alaric Fox

Independent Software Engineer

Watch out! Just today a member of the NB of South Africa and the NB of Brazil jointly submitted a digital signature proposal to the OASIS ODF TC. This is additional evidence that these colonials are talking among themselves, collaborating, and working effectively on standards. If we do not do something about this soon, who knows what dire consequences might result.

To arms, to arms! Ancient privilege must be defended!

Hello, long-time reader, first-time poster.

This is not to say there is not a requirement for process improvement - there most certainly is.

I'd like to hear your perspective about where the problems are in the process. What do you think went wrong? How badly do you think the final spec suffered from it?

This is a moot debate. It is obvious that the ISO intends to charge ahead with DIS 29500, regardless of whatever merit the appeals may have, and regardless of whatever damage it does to its reputation in the process.

The real action is with the EU Competition Commission. It is performing its own independent investiagation of the process, and I think there is little doubt that it will conclude that DIS 29500 is not a legitimate standard, and recomend that EU governments not impliment it.

Interview with the director of the ISO on DIS 29500 and other matters:

http://www.iso.org/iso/pressrelease/int ... l_iso29500

faq on DIS 29500

http://www.iso.org/iso/pressrelease/faq ... c29500.htm

@Andrew Sayers

> Without wanting to get in the way of the most adorable flame war I've ever seen

Is it truly scrumptious?

> the cumulative effect makes it hard for moderate people like Andy Updegrove to
> assume neutrality on your part

I think Andy often has more of a problem with what I don't write, than what I do. Also different people naturally have different ideas about what "neutrality" is and who "moderate" people are!

As to miscommunication -- *shrug* it happens, in electronic communication quite a lot: all one can do is explain more and move on.


@Brian Densmore

If you really want to do the autospacelikeWord95 thing (etc) that stuff is all in the "transitional" part of OOXML.

Many small independents would rather not work with OOXML and would prefer to work with ODF (my company falls into that category), but the reality is that for the overwhelming majority of commercial users MS Office is the office suite in use -- so one has to work with what it produces for these clients. Having the formats properly documented, maintained and stabilised is an advantage that standardisation brings. Anyobody who has been through the RTF/WordProcessingML mills is surely going to see the win there ...

As to O-Member voting I think "clearly" is the last word I'd use to describe the Directives - but you are wrong in your assertion that anything was amiss in the process in this respect.


@Alaric

I completely agree that "a point of order [should] be valid or not regardless of how many countries submit appeals".

You have some interesting thoughts about how duplicate text might come about. As my original entry says, I do not know how this happened - and would be interested to!


@Rob

What with your "desert island nations" and now "colonials" this is all getting rather exotic. Opining about a nation's document format standards work purely because of its size, history or location would be pretty dumb.


@MikeF

What went wrong? Mainly, that the processes admitted texts like OOXML, and ODF before it, onto accelerated standards tracks where they got insufficient scrutiny - as I have written many times before. The result is that JTC 1 has ended up with technically poor standards at the point of publication.

Jan van den Beld (of Ecma fame) famously maintains that it is better to have a good standard today, than a perfect one tomorrow. His view (that timeliness trumps quality) is certainly one among many arguable ones - but the trouble with PAS and Fast Track is that this view is rather forced upon nations.


@eduardo

"ISO" doesn't really have any power (see the interview with the Secretary General you link to in your other comment) to "charge ahead". Ultimately, the decisions are made by NBs - and that is how this appeal is being handled.

As to the EU investigation, I'm sure the duplicated text phenomenon (more particularly in the letter ballot) is one of the many things they'll want to look into.

- Alex.

'What went wrong? Mainly, that the processes admitted texts like OOXML, and ODF before it, onto accelerated standards tracks where they got insufficient scrutiny - as I have written many times before.'

Unless i'm misunderstanding what you are saying there, Alex, you are totally wrong, ODF was never on any kind of fast track process, and to pretend otherwise is plainly a lie.

First time i posted, but read a lot here.

@Alex,

So are you saying that if the commission determines that the appeals were duplicated for some illegitimate reason, but correct in content of their complaints, the commission will go ahead and declare DIS 29500 a legitimate standard? That would be very odd.

Words, Words, Words...

Will all of you get a life; please?

We all know that OOXML will be approved, but who gives a toss anyway?

It will be of little or no importance to anyone. It's a dead duck before the shell is even broken.

Nobody believes it was an "honest" process. No body believes that Microshaft didn't screw the process. No body believes that Doug 'Mawho' is Vice President of IASA Malaysia. Nobody believes that Azerbaijan, Côte-d’Ivoire, Jamaica, Kazakhstan, Kenya, Lebanon, Malta, Pakistan, Saudi Arabia, Uruguay and others were bought and paid for. And EVERY one believes Martin Bryant when he said so publicly:

"The disparity of rules for PAS, Fast-Track and ISO committee generated standards is fast making ISO a laughing stock in IT circles. The days of open standards development are fast disappearing. Instead we are getting “standardization by corporation"."

Who are you trying to kid Alex? What do you gain? A few nice juicy contracts and some speaking engagements at M$'s 'special rates'?

Move along. Nothing more to see here...

@illbite101

ODF was standardised in ISO/IEC using the PAS submission process, which is an accelerated standardisation process very similar to the Fast Track.


@eduardo

As I mentioned, my guess was that the duplication in the letter ballot would be of more interest.


@Open Sourcerer

Isn't it rather reductive (not to mention cynical) to reduce everything to personal greed?

Perhaps we all need a holiday ...

Though not in Vulgaria

Alex: I use "Freedonia" (as in Duck Soup) and "Laputa" and other Swiftian NBs (though Laputo is clearly based on Japan.) I quite like "Vulgaria" because it has nice connotations of "the common people" as well as de Tocquevillean ideas of the disenfranchisement of talent: so it works for democracies as well as mob-ruled nations.

That someone nutty gets offended by a turn of phrase is no reason in itself not to use it, if it is the clearest way to say what you mean and if what you mean to say is not nasty. (Of course, if you are writing in some official capacity, that it different.) Some people are just too sensitive and censorous: it is all part of the same fault-finding mentality perhaps. Honi soit qui mal y pense.

In particular, "Balkanization" is a historical technical term with no real equivalent: that some people will even get upset about it as a general term is rather the point!

"It is partly for that reason that any appeal which is less than fully considered is to be particularly regretted."

Imho, the same holds true for any specification that is less than fully implementable.

@Stephane

"Can you name a single full non-Microsoft implementation of OOXML?"

And what would be the point of a company making it part of their business plan to be a "full non-Microsoft implementation"? Maybe, what you are really asking is if there are partners who read/write every sentence defined in the format spec(s)? Again, what would be the point of this?

We read/write the formats, according to spec, and support the full spec(s). We also work with the document formats and conversions between the formats and prior versions of office; and the formats lend themselves to us going between formats very well. We have a lot of experience here, I am curious, Stephane, why you talk as though you are an expert in this area, when it is clear that you have not gotten your hands dirty enough to really know. (BTW: The partners exist, they just would rather build strong business relationships with people who care, rather than have their names thrown around on blogs for yours (and others) entertainment; maybe that is the real reason why you have not received an answer to your question)

Having observed a lot of companies/partners try to work with the formats over the last couple years; they generally simply lack correct information about the spec(s), and/or lack having spent enough time to learn them. I personally have helped answer many of thier questions; and they have been able to solve real world business problems for thier companies that have been well received, because these companies have been able to take advantage of the technologies that they have already invested in, rather than having to buy/learn new ones.

"When you get to really implement bits of OOXML, you get half a dozen variants of vector graphics markup in Word alone. Which means you are in it for a decade..."

Stephane, have you ever built a software package as big as would be required to implement the format definitions? (be careful here, may I remind you of our discussions last year where you point out just how big the implementation would have to be). Your complaint here is a bit hypocritical (sound familiar); I would encourage you to try and define something of this magnitude in less than a decade and actually be successful at it.

"Meaning that implementers are caught in a trap"

That is a bold statement; we’ve found Microsoft willing to be helpful whenever we’ve had questions; and have yet to run into a scenario where we did not get the support we needed. You complain about parts of the spec that are optimized for performance; what software or spec doesn't take this into consideration? Are we (engineers) really complaining about a technology that makes considerations for performance???

I thought we were past these petty issues and on to working on 1) improving the spec and 2) helping companies, globally, to understand and implement IS29500 (BTW: IS, not DIS).

-Stephen
Co-Founder & CTO
Xinnovation, Inc.

@Stephen Perront

To present yourself as someone from XInnovation is a misqualification. You sided with Microsoft during the INCITS V1 review, provided them appalling defense (never an argument, just basic stone throwing at critics), you've been at numerous tech-ed conferences paid (directly or not) by Microsoft, you've been mentioned quite a few times by Microsoft evangelists about OOXML during the time that OOXML was being evangelized by Microsoft and a couple of their (paid) partners such as Novell. There is probably more than that.

In a nutshell, you are part of their bitch ring. Make sure to remind this to readers here.

When you go so far as to pretend I probably lack the experience in this area, I guess this pretty much disqualifies anything you say.

Let me just give you a bone. Take a dictionary and learn the meaning of things such as "rendering" and "calculation". Perhaps, sometimes in the future, you will be able to figure out that it may have a meaning in the context of implementing OOXML, and that perhaps read/write angle brackets arbitrarily is not what OOXML is trying to solve (by the way, read/write angle brackets arbitrarily is a matter of 30 lines of code on top of an XML package).

@Stephane

Apparently you have resorted to name calling (again); rather than dealing with the issues at hand.

(from your first post)
“Actual implementers of OOXML should be given a weight at least 10 times more than all those commenters combined.”

(from most recent post)
"When you go so far as to pretend I probably lack the experience in this area, I guess this pretty much disqualifies anything you say."

I would hope that we would not disqualify anything that people have to say; especially when your earlier comment seems to contradict your most recent comment (seeing as we are implementers). My point is simply that the topic of Alex's article is about the text submitted from various appeals; not "rendering" and "calculation"; the comments that you made were off topic, inaccurate and dirty the healthy conversation going on here… I am not pretending to be anything (I actually do represent Xinnovation); my comments were simply based on the inaccuracies of your own text (which is the only thing we have to work with in a blog).

I guess the reality is that you are not going to get back on topic unless we address your concerns; so here goes; I will address one. Maybe then, we can get back to the topic of the original blog post.

(in an earlier post, you say)
“Those decisions include mapping values, introducing new ones, hiding others. In the general case, this is not trivial. Worse, those are decisions that are the choosing of every single implementer, therefore every single implementer will have their own OOXML”

Implementing any specification is something that is left up to interpretation and implementation of the implementer. The synergy of various implementations comes through collaboration and provides interoperability. The difference in various implementations comes from distinction and provides diversity and choice. The goal is to provide enough collaboration that we don’t run into the issue that you mention here, but allow enough variation that users are given distinct choices in what implementation they choose to use. Because it is just not realistic to expect every entity in the world to conform to one way and one implementation of, well…. Anything!

While I agree with you that this spec, as almost any other spec, requires various implementers to work together (i.e. collaborate) on certain pieces to ensure solid interoperability; we disagree with your approach of attacking the spec. We believe that working together to improve the spec. is the right approach. If you believe that this is the only specification that has flaws; just look at the various implementations of PDF and the headaches that causes. And for the record, we are also implementers of PDF and believe that the right approach with that specification is to also work to improve it. Look how great it is now compared to 3 versions ago!

If you knew anything about our company and how we help businesses, you would laugh at your own comments; because “rendering” and “calculation” are at the core of what we help our clients achieve through documents!

Get your facts straight, and join the world. Rather than ranting on and on about topics that are actively being addressed - help us to improve them! Quite frankly some of us are tired of hearing them repeated from those who are not willing to help to address them!

-Stephen Peront
Co-Founder & CTO
Xinnovation, Inc.

Maybe you could bring at least some clarity in this here as written up by Groklaw:

"Brazil has other issues. Its appeal says it tried to present some material that it considered an "important proposal regarding the legacy binary mapping" at the BRM, but it was told by Alex Brown to meet outside the BRM to discuss it with other NBs interested in the topic, and it did. It prepared a presentation and made it available to all, and the second day, it was introduced by the US delegate, and then Brown stepped in, Brazil recounts, and was told that there was "no time" to discuss it. Yet others, like Ecma, Brazil says, were allowed to speak for a half hour at a stretch, despite Ecma not being a member. Others besides Brazil protested Brazil not being allowed to present its proposal, but to no avail. They were told there was insufficient time to consider it. So Brazil objects on the basis that there was insufficient time and as a result, it considers the outcome inconclusive."

So, what is your side of the story or are you already away on holidays in Vulgaria?

Ah, Stephane is not banned under hun real name on this blog yet.
But as always he is trying hard.

Well I'm not so sure about Brazil's appeal but the gumph on Groklaw is typically inaccurate.

> "Brazil has other issues. Its appeal says it tried to present some material that
> it considered an "important proposal regarding the legacy binary mapping" at the BRM,
> but it was told by Alex Brown to meet outside the BRM to discuss it with other NBs
> interested in the topic, and it did.
> It prepared a presentation and made it available to all, and the second day,

The timestamp for the Brazilian multipart "presentation" is 29 Feb (the final day). On day 2 the USA raised a minor point about list enumeration, not about binary mapping (much to everyody's surprise).

> it was introduced by the US delegate, and then Brown stepped in, Brazil recounts,
> and was told that there was "no time" to discuss it.

Quite right. The Brazilian "proposal" is a 5-page Microsoft PowerPoint presentation. What NBs needed to be producing at that point was (in Tim Bray's words) "crisp fully-edited deltas" -- and they need to have submitted them to the BRM web site in enough time so that other NBs could at least had a chance to read them overnight.

A proposal raised on Friday afternoon that Ecma should "add binary mappings" was quite clearly totally unrealistic. How could Ecma produce that over the tea break? How could NBs review what would be thousands of dense technicalities in the same time? It was totally within my discretion and obviously the correct decision to nip this in the bud. Brazil had already raised (and had actioned) their first raised point about de-vendorising the specification of web browsers in 29500 (kudos to them for that!)

> Yet others, like Ecma, Brazil says, were allowed to speak for a half hour at a
> stretch, despite Ecma not being a member.

Ecma never spoke for half an hour - that is just wrong. Ecma were only allowed to speak when responding to a NB or reporting on their progress with some work an NB had instigated. As submitter one would expect them to do most of the talking, as in any BRM. If they wandered off topic they were called out of order (as happened once).

> Others besides Brazil protested Brazil not being allowed to present its proposal,
> but to no avail.

My sense (and I may be wrong) was that the OVERWHELMING mood of the BRM was to get on with pressing work that was realistically achievable, and backed my decision. Patience was beginning to wear thin with anything that impeded efficient progress.

> They were told there was insufficient time to consider it. So Brazil objects on the
> basis that there was insufficient time and as a result, it considers the outcome inconclusive."

That every attendee at a standards meeting doesn't get their way is hardly news, now is it?

> So, what is your side of the story or are you already away on holidays in Vulgaria?

Ha! "Holiday in Vulgaria" - where the people dress in ... crimson?

- Alex.