Geneva, Day One

Although it has been stated in this blog prominently on the top right of this and every page, I will need to reiterate that the following information is and always has been based on my personal opinion and it may not reflect any parties I represent nor belong to, especially at this point in time, the Malaysian Delegation's position on any matters arising at the BRM in Geneva. All activities within the walls of the CICG where the BRM is held is to be shrouded in secrecy and will not be revealed to the world to safeguard the protection of the development of open standards within transparent and open processes. Yeah.

Additionally all names have been altered to protect the innocent. No animals were hurt in the production of this blog entry.

Yes, we have been told not to broadcast the specifics of the BRM. Additionally we will be sued if we take and publish any pictures of certain delegates. It seems Switzerland has some super privacy laws. I guess thats why Swiss Banks are so popular.

It was great to hear the concerns raised by the various National Bodies. Some were detailed, some were brief, some measured, some heated. It was a great exchange of different ideas and very early on it was easy to identify countries which are working hard in making progress of this Draft Standard, and others who were clearly biased and were towing a certain corporate line.

A few really experienced people deliberated on matters in such depth that it was a huge task to keep up after many hours of deliberation. It is certainly an eye opener, and as a newbie volunteer as I am, it is quite intimidating to represent my nation's views in front of so many other luminaries.

What was interesting was the different solutions presented by many NBs whom have obviously thought about certain matters in detail. As they describe their counter proposals, it is often hard to grasp their concepts which they have evidently worked hard in considering all available options. It would have been great if there was more sharing and joint development work prior to the BRM like in a mailling list or something as interactive. With such large changes proposed at the 2nd Sept vote, the suitability of having counter proposals appearing within these 5 days is certainly stressing the "Fast Track" process.

The issue of deprecation is featured prominently in Ecma's disposition of comments since the 14th of January 2008. What is puzzling about the term deprecation is that it describes something that should be considered obsolete and should therefore be recommended to be avoided in future encoding of that file format. Some standards experts prefer deprecated items to be normative, while some insist that it should just be informative.

In my limited understanding I would have thought that if you are describing something which is deprecated (only there for legacy issues), Annexed (moved from the main text to a tack on section) it should therefore be informative (available for informational purposes only, as a developer is not required to implement it for full conformance). However Ecma's disposition (since the 14th Jan) has proposed that the deprecated items like VML will be Annexed in normative sections. This means that for conformance, these deprecated features should still be implemented.

This is confusing, and I hope that in the near future I can understand why this would be so.

Other issues which I will have to brush up on are the concerns of conformance. Ask 5 people what they mean by it, and you get 6 different answers on how to define and test it. As pointed out in the 2nd Sept comments, many NBs are worried about the conformance clauses within the text. Some have provided their own ways of measuring the spec; some is to split it into two parts, others 5 parts, some 9 criteria. Each one different but having the same goal: defining conformance of the spec. Hopefully this is something which can be defined in the next few days so that we can eventually work towards harmonization with other file formats.

My favourite item in adoption of ISO Dates exclusively. It is good that this item is getting a worthy review, and I look forward in the NBs considering the Greek proposal which was brought to light over 12 days ago.

So today went by in a flash. The system laid out by JTC1 is still to be stressed tested in terms of Fast Tracking this substantial spec. This will become more apparent in the next few days, and it will certainly be interesting to see how the people involved will handle this. I will not say how many items we have resolved yet, because that would be unfair on the workload we are working through. It's still too early to tell. Day Two may give an indication. I am hopefully optimistic.

yk.

Comments

[Daniel]

The process here occurs within an international organisation which has nothing to do with Swiss government, which is also quite distinct from the Swiss Banks. Swiss politics is exceptionally open and democratic, while Swiss Banks are strongly secretive private and actually multi-national corporations. What ECMA and other international organisations do is also a completely different matter.

So please do not confuse everything.

[Stephane Rodriguez]

"However Ecma's disposition (since the 14th Jan) has proposed that the deprecated items like VML will be Annexed in normative sections. "

The probably reason why is because I guess Microsoft will simply add a DrawingML namespace and wrap VML around it.
This is deception plain and clear, and keeping VML in the specs tries to buy them some defense for this move.

[Simon]

Talking about OOXML irregularities, here is the sad case of Greece,

http://elot.ece.ntua.gr/te48/ooxml/2008-02-21-meeting-en

[yoonkit]

Daniel,

I understand the differences between the three, and it was in jest. However you really should have a look at the composition of your own country's delegation, and how it got to become as such.

So please get in touch with that NB and find out the details.

Stephane,

Evident from the number of concerns raised against VML, it certainly is a topic which is being considered. The question of course comes to the definition of the term "deprecated". So yeah, this is something which we must make sure doesn't seep back in through the back door.

Simon,

Yes, the Greece story certainly is interesting. We must make sure that this is really the case, so please research more via the National Bodies and make a report on it if you feel strongly about these things.

Make sure you get verifiable evidence, or it's merely internet gossip. Check the sources, OK?

Sometimes the truth is stranger than fiction.

yk

[orlando]

yk

can you tell us ( poor final software users and open standards advocates that don't deserve to know nor observe the inner details of how ISO produce our wonderful standards ;-) how was the participation of the new brand ISO P-members in this discussions?

Cyprus island
Jamaica island
Malta island
Cote d Ivoire
Lebanon

Do they have any concern, have made any technical contribution to the analisis of DIS 29500 in the BRM, or they are silently approving anything?

Thanks in advance.

Orlando

[Todd Knarr]

As far as "deprecated" features being annexed in normative sections, I suspect there's a simple explanation. Those particular features are important and the specification becomes essentially useless without them, but they're in such a state that ECMA fears the NBs won't accept the standard if those features are officially included. Hence the odd classification, allowing the party proposing the specification to emphasize either the "deprecated" or the "normative" part depending on circumstances.

I'm of your mind: deprecated and normative should normally be mutually exclusive, except maybe in the case of something that was non-deprecated in a previous version of the standard and is being removed.

[esni]

I think in ISO terms an annex is a part of the specification (ie normative) as opposed to an appendix, which is supposed to contain additional information.

So the deprecated items should really be relegated to appendices.

/esni

[Stephane Rodriguez]

I think all this discussion about what's becoming really deprecated (the ECMA 376 paper says VML is deprecated already, the problem was that Office 2007 still used it) is just a diversion.
In fact, I suspect Office 2009 to be even more at odds with the specification.

Microsoft Office has a long record history of being unable to do one thing right in just one product release. It usually takes two, three sometimes more releases to get one thing done.
That is the case with VML. I think the introduction of DrawingML means that Microsoft wanted to get away with VML and move with an even more complex rendering language that is made of so many incompatible, redundant and weird coordinate systems that the chances that this gets implemented fully by a competitor is exactly zero.

Well, they rushed Office 2007 out the door (presumably to ship simultaenously with Windows Vista) and the consequence is that they couldn't manage to finish this transition in time.

Then Microsoft wrote this ECMA 376 draft, filled with all the details about this unfinished job (as if it was designed for the future ).

And we, reviewers, are having incredibly stupid discussions about the quality of VML.

Let's be clear, VML would not be part of this thing, should Microsoft have taken the time to finish their job before they shipped Office 2007.

That's ridiculous. This file format is entirely designed around the limitationd and quirks of a single vendor's product.

How can just a limited number of national bodies not see the truth for what it is ?

Don't forget to ask if you are participing to the BRM : ask Microsoft if there is any VML left in Office 2009 . (Office 2009 should be in beta at this point. I sense Microsoft is not willing to talk about it very much at the moment. Imagine that national bodies understand that all the concessions that Microsoft is ready to make are concessions that are, shear coincidence, exactly the implementation decisions they took in Office 2009).

[Simon]

yk

The article was written by Antonis Christofides. He is the person that wrote the ISO dates document that you quote in your previous post.

He might be in Geneva at the moment as part of the delegation. It is sad the person who understands OOXML does not represents the interests of the country.

[yoonkit]

Orlando,

I cannot say who voted what, or who said what in the BRM. It would be against the agreement. There is a list available online and it is close to accurate. So most of the new-P countries you list are not represented. You can read about that one more on other sites.

==

Deprecated is now considered a bad word.

==

Simon, if you think there has been some injustice, I would highly recommend you find out and publish this information as soon as possible. Antonis is a great guy.

yk.

[Jefrey]

It makes little difference in the real worlds whether or not they call the Annex of deprecated items normative or informative.
It is hardly worth discussing it.

Microsoft will just convert the VML to DrawingML (which is more extensive anyways) in their next version of MS Office and then accept that VML will be dropped for a next version of OOXML.
Boring.

The data issue is more interesting but only slightly.
Frankly the only relavant issue is whether or not the opponents of OOXML have penetrated enough NB committees of the P-members sufficiently to keep the final vote below 2/3 positive.

Are you doing a headcount already ?