a 'mooh' point

clearly an IBM drone

I missed you, Rob

So today was the last day of the two-day meeting in the, in Oslo created, Ad Hoc Group 1-committee (AHG1). It has been a couple of interesting days and also rather productive. We managed to cover quite a lot of ground and I am quite pleased with the outcome of the meeting. Note, that we have not made any decisions at the AHG1-meeting. All we did was to suggest a possible structure for future work on OOXML in a new Working Group under SC34, Working Group 4 (WG4). SC34 will have to decide themselves (or, ourselves) on what to do in the event that the appeals on OOXML-approval are overthrown.

There were a total of 18 people attending the meeting – of these were nine people of either ECMA or Microsoft. The rest was comprised of representatives from British Standards Institute (BSI) and Dansk Standard (DS) and a couple of “neutral” people, herein myself, Francis Cave (UK), a guy from NL-net Foundation in the Netherlands, Keld Simonsen (NO), a guy from IBM (HUN), the convener Dr. (*giggles*) Alex Brown (UK) and Murata Makato (JPN). The meeting took place in the pleasant surroundings of British Library near St. Pancras Station in London.

The meeting report is available from the SC34 website.

I think the content of most of the discussions of the meeting can be summed up in these three words: “Openness”, “transparency” and “participation” – with the latter perhaps being the most important.


Murata Makato is currently the convener of SC34 WG1 – the WG that currently holds responsibility of the 29500-project in SC34. He will therefore be acting convener of WG4 as well until SC34 formally points out who should hold the position. We have suggested to SC34 that Murata Makato be pointed convener of WG4, but as with all positions in JTC1, it is at the discretion of the NBs and I encourage each NB to think hard on whether they have someone who could fill the position (I should note that I personally think that Murata is an excellent choice as convener).  Since the work-load of WG4 quite possibly will be rather large, we have also suggested that WG4 should have a secretariat. It is at the discretion of the convener to point out who should run the secretariat – it could be an NB, but in theory just about anyone. ECMA has offered to run the secretariat for the W4.

We have suggested to SC34 that an editor of OOXML should be appointed to oversee and coordinate the overall process of work on OOXML. ECMA has offered to continue to fill out this position, but it will ultimately be up to SC34 to decide. We talked quite a lot about how to structure editing of the 29500-project. In ISO-terms, maintenance is defined as “revision, withdrawal, periodic review, correction of defects, amendment, and stabilization”. We discussed having multiple editors (possibly one for each of the IS 29500-parts (four in total)), we discussed multiple editors sharing responsibility of editing the whole text as well, but we ended with a suggestion to have a single editor for the project and assign an “editorial team” to him/her. We agreed that this was the most flexible way to structure this task. The editorial team should consist of SMEs (Subject Matter Experts) of either the NBs or ECMA or any other expert invited/nominated by an NB. This will allow WG4 to adapt the resource-level in terms of body-count to the specific work-load being thrown upon it.

Now, if you ask me, it is crucial that the NBs step up to the task and participate in the work on maintenance and future development of IS 29500 – should the appeals have a favorable outcome. If we don’t step up, SC34 – and WG4 - will in effect simply be a new place for ECMA to meet and work on OOXML. But it is a daunting task – at least compared to the normal work load of some WGs under the JTC1 umbrella. We talked a lot about the possible magnitude of work, but since none of us were able to predict the future, we have suggested the following as initial meeting schedule and work-load:

  • Weekly teleconferences
  • Quarterly face-to-face meetings
  • Overall communication via, possibly, email

The first face-to-face meeting should take place in early 2009.

If the work-load is not as big as we fear (or hope), the activity-level will quite possibly be adjusted really quickly to a more infrequent level.

But this work-load is rather big - at least if you ask me. Even if you don’t count the quarterly meetings in, we are talking about weekly teleconferences of maybe 2 hours each – and with preparations for them of at least a couple of hours for each, initially. This amounts to between a half and a whole work-day each week. Note that none of this is funded by ISO. This effectively means that you don’t want to join the WG4 (or the editorial team) unless you really mean it. I don’t think there is any way to sugar-coat this – participating in standards work, be that in OASIS, ECMA or ISO is serious business and it takes up a lot of time. That being said, we need the NB-participation in this – otherwise the whole ISO-process regarding OOXML becomes, well, ‘moot’.

We actively encourage the NBs of SC34 to participate in the WG4 [and on a personal note; this should be regardless of position on OOXML itself]. The 29500-project drew an enormous amount of attention throughout the last months, and especially the feedback from those opposing OOXML has been extremely valuable. It would be sad, if all those good resources chose not to participate.


Openness is here referred to as the ability of participating in the process. This was sadly one of those areas, where not much was changed – mostly due to JTC1 directives. The members of SC34 (and the subordinate working groups) are national bodies, so if you want to participate in maintenance of OOXML, you need to join a national body. In some countries there is a fee, in some not. In Denmark, as an example, it is free for NGOs and private persons to participate and there are discounts for SMBs if they should choose to join. For “regular” companies as CIBER or IBM, the annual cost is about €3000.

We talked about setting up an informal channel for feedback from the community, but we ended with a decision about a closed NB-website for submission of comments to WG4 – essentially an electronic edition of the “Defect report” form from ISO.

I am thinking about creating an open, informal channel for feedback and comments on IS 29500. It should allow everyone to submit comments to the site about the spec and allow “comments on comments” to facilitate discussions on the feedback. The channel (a website, really) will be REST-enabled and allow any NB to use the contents of the website in their own work on IS 29500. The idea is to create a single point of feedback to enable not only the community to provide feedback but also assist the NBs with their technical work on IS 29500. The website is not a direct channel to SC34 but an informal place to discuss issues with the text. If anyone would like to contribute to this work with either ideas, funding (website costs etc) or technical expertise, please let me know.


And what about transparency? Well, this will follow the rules of JTC1 which means that meeting reports and attendance-lists of face-to-face meetings will be posted on the SC34 website and be accessible for everyone. We will have to find a specific form in doing this, but I will do whatever I can to have the meeting reports be as much as possible like the meeting minutes from OASIS ODF TC. This means that not only will any decisions be recorded – details about the discussions around them should also be available. We will also likely be posting intermediate drafts of the specification for everyone to see – in exactly the same way OASIS ODF TC and ECMA TC45 has done until now. This will allow everyone to follow not only the work going on in WG4 – but also what the result of the work in the actual specification will be.

Participation – a final note

I missed a lot of people in London – the people and organizations opposing OOXML. I had expected a stronger representation from some of the big companies that have criticized DIS 29500 and I had also expected more of the opposing countries to attend. In effect, by not participating in the meeting, they contributed to the alleged “Microsoft stuffing”. I think it calls for a bit of after-thought on their part. They might not have had their cake (to eat) in Geneva, but not participating in the work is the sure road to an ECMA/Microsoft dominated WG in SC34. I will not begin to speculate (much) on possible reasons for not attending – maybe it’s just much easier to sit on one’s hands and claim “Microsoft stuffing” than actually attending the meeting. Just note, that it’s hardly “Microsoft-stuffing” when no one but Microsoft participates.

The happiness of solitude

Oh the lonelyness ... oh the solitude.

They say that parting is such sweet sorrow, but I beg to differ. Things have really, really cooled down in the otherwise warm and cozy OOXML/ODF-blogsphere. Rob and Arnoud seem to have gone back to their day-jobs and Brian has somehow completely dissapeared from the face of the Earth. Doug is mostly writing about what other people are writing about and Groklaw has gone back to their original angle - the SCO-Shenanigans. The only active blogger at the moment seems to be Rick, but even here, the normally so loyal Rick-bashers in the comment-threads seem to have gone AWOL.

Nothing seems to happen here in Denmark as well. The Danish NSB met about a week ago, and we decided to make the working documents public that formed the foundation of the arguments and decisions that took place in the last year. We formed a small technical sub-committee that did the technical work on first the responses to the Danish public hearing in Spring 2007 and later the responses from ISO to the Danish 168 comments to DIS 29500. The group consisted of CIBER Denmark, Ementor, IBM, Microsoft, ORACLE and the County of Aarhus. The technical group was an advisory group to the Danish SC34 mirror-committee. The working documents were made to allow us to keep up momentum and to document the progress we made. In short, for each meeting we made a list of the ISO editor responses that we could accept and those the we could not accept - and they were sent back to ISO editor for further processing. The documents are in Danish, but it still gives a good idea (regardless of native tongue) of what we did in the technical group and how we dealt with each issue. The documents are available at the Danish NSB website (last 7 documents at the bottom of the page in the section "Arbejdsgruppe-notater").

I have also more or less gotten back to my day-job as an Engineer with CIBER. I am currently investigating how to generate documents (ODF and OOXML) using .Net and is actually kind of fun. With that in mind I was interviewed for a video-cast by Microsoft for a small discussion about ODF and OOXML (they conveniently cut out the part where I said that I prefer the markup of ODF over the markup of OOXML but still prefer the tools for OOXML over the tools for ODF (for generating documents on e.g. a webserver or ERP-system), but what can you do?). One of the points I made in the interview was, that the tools were really important. If there are no good tools to create documents - it will slow down the adoption-rate of the particular file type. Regardless of SW-political view, the .Net-platform is rather large on a world-wide basis and the install-base of .Net-technology makes it a platform that should not be ignored (by size alone, if nothing more). And this puzzles me. If you look at the developer-hub of OOXML, you wil find libraries, scripts and tools for just about any operating system and programming language available. But if you want to generate an ODF-file using .Net technology - what do you do? Well, you will propably find that the only (OSS) library available is AODL, a project under the ODF Toolkit umbrella. Unfortunately, the project is not a priority of OpenOffice.org. I wrote an emails to the lead of the project (Dieter Loeschky from Sun) and he suggested that I joined the project as contributor. I have thought a bit about it, and I just might do so. I find it really important for the adoption of ODF that there are tools available for it, so if no one else will, I just might do it myself. I wonder if that will help everyone realize that I am a true ODF supporter.

And finally - the SC34 Ad Hoc Group 1 will convene in London in the end of July. We will meet and talk about what to do with both ODF and OOXML in the future. I am really looking forward to the meeting. The initial mail list reveales that there will be delegates from all over the world:

Austria  2
Canada  1
Chile  1
Czech Republic
Germany  2
Denmark  3
Finland  3
India  4
Japan  3
Republic of Korea  2
Malaysia  2
Norway  3
New Zealand
United Kingdom
United States


I hope we will have a couple of productive days in London. As Alex Brown wrote about after the Oslo plenary in April 2008, transparency of the process is a key point and any input from you, dear reader, to how this could be achieved would be appreciated.

And finally-finally, I seem to have been struck by a bad was of "YABS" - Yet Another Blog Syndrome. Within the next few weeks I will begin blogging on the best IT-website in Denmark, Version2.

Generation of ODF-files on the .Net-platform

Some time ago I wrote a couple of articles about how to generate ODF-files as well as OOXML-files using .Net technology (both articles are in Danish). For generation of OOXML-files I used the - at that time - new .Net 3.0 System.IO.Packaging assembly and for generation of ODF-files I used AODL - a part of ODF Toolkit.

I thought it was time to refresh my skills - and share them with you guys - since the OOXML/ODF-debate has cooled down to a more relaxing level.

A few weeks back Microsoft released the first production-code edition of their OpenXml SDK - version 1.0. I will dig into this a bit later.

I thought I'd kick this series off with a couple of articles about ODF-file generation on the .Net platform, but I was unpleasantly surprised to realize, that it might not be as easy as it sounded. First, I was told that AODL was a dead project. Surely, the latest addition of code was in April 2007 and it seems that nothing has happened since.  It looks as if the resources of ODF Toolkit is focused on ODFDOM - currently a Java-project. The problem is - AODL seems to be the only .Net-project available. I have stumpled across the ODF .Net project by IndependentSoft, but they sell an ODF library as Closed Source Software ... for (brace yourself) €999 a pop! Seriously - selling CSS-libraries is just sooo 2006 ...

And then I come to you, dear reader ... what the hell do I do? Do you know of other .Net libraries that allow me to create and manipulate ODF-files?



Beer - revisited

The beer-drinking was a huge success. We were only four of us, but we had a great afternoon and start of the evening. It was a lot of fun to meet offline and have a decent conversation on ODF and OOXML ... much more fruit-full than blogging.

Clock-wise, from bottom left, it is Henrik Stig Jørgensen, Eskild Nielsen, Jesper Lund Stocholm and Christian Nobel.

I am looking forward to doing this again, guys!


Should the appeals stop all work on IS 29500?

Well, you tell me. All kinds or rumours are circulating amongst SC34-members of what the consequences of the appeals of approval of IS 29500 could be. The latest rumours are dwelving on the possibe outcome that all work on IS 29500 will be suspended until appeals are sorted out. These rumours are quite possibly fed by the email from the JTC1 chair to the members of SC34. Amongst other things it said

At its recent plenary meeting in Norway, JTC 1/SC 34 established 2 ad hoc groups concerned with the collection of comments on and maintenance of ISO/IEC 29500. Adjudicating these appeals will necessarily delay ISO/IEC 29500, at a minimum, and this delay could have a significant effect on the work of these ad hoc groups. SC 34 and its ad hoc groups should take this into account when planning their future work.

I am a participant in AHG1 (maintenance of IS 29500) and the purpose of the group is to plan and define the future work onIS 29500 in SC34 in, quite possibly, SC34 Working Group 4. We will meet in London in the end of July 2007 to start our work. From the roster/mail list of the group, I see quite a lot of "big cahunas" of SC34 and indeed in the debates of the last year or so. All these people have signed up to contribute to the onwards development and maintenance of IS 29500.

Now, let's pause for a second to look at the possible outcome of the appeals and their impact on the work in e.g. AHG1:

Approval is overturned

This will mean that IS 29500 no longer exists and that the work in the two AHGs have been a waste of time. We will spend two days in London and these days could then have been used more productively on other matters - like contributing to development of ODF 1.2 .

Appeals are dismissed 

If the appeals are dismissed - we will already have started the work on IS 29500. We will already have an idea of what to do and a preliminary view of what to do in the near future.

The outcome of the appeals is currently blowing in the wind, and I would not be the one to predict the outcome. However, I will say, that I would much rather start the important work on IS 29500 maintenance as soon as possible. This is in fact the core of the worst-case scenario: we just loose two days of work. I think most will agree that IS 29500 needs proper maintenance and I would prefer to get that work started - even though the work might be wasted. If all work in the AHGs are suspended and the appeals are dismissed, we will not be able to start this important work until Spring 2009.

For those of you being worried that allowing the AHGs to proceed with their work could be interpreted as some sort of bias towards dismissing the appeals, I think you are wrong. Allowing us to start the work on a provisionary basis does not in any way impact the decision of JTC1/ITTF regarding the appeals. It simply allows us to get a head start on maintenance of IS 29500 ... nothing more.

So please, let all the distinguished people in Ad Hoc Group 1 and 2 get at chance to start the important work they have volunteered to participate in. 

Get the ball rollin'

The latest couple of weeks have been almost as quiet as the couple of weeks after the BRM in Geneva. Most bloggers have cut down on their posts regarding ODF and OOXML (including your's truly) and have apparently gone back to their day-jobs to do a bit of Summer cleaning before the Summer vacations. There has, however, been some interesting development in some of the corners of the ODF/OOXML blogsphere.


ODF TC has launched the preliminary work to form another ODF TC to focus on interoperability between ODF-enabled applications. I am one of the "lurkers" on the mail list, and I encourage everyone to either eavesdrop on the conversations taken place or actively participate in the debate. As noted before, conformance and interoperability is so much more than schema validation and the initiative is highly valuable. The topics being discussed are conformance, interoperability, IPR, AcidTests and much, much more.


Microsoft has joined ODF TC in OASIS. By Paul's message to the maillist of ODF OIIC  it seems that Microsoft has already been admitted into ODF TC. Microsoft is now listed as one of the "Sponsor Level"-members along with Adobe, Google Inc, IBM, Intel Corporation, Microsoft Corporation, Novell, and Sun Microsystems. (PS: Why is there an asterix after the names of Google and Novell?). This would mean that, as far as I get the legal mumbo-jumbo, that Microsoft will include ODF in its OSP. According to the content of the list as I write this, it has not yet been included. This also reminds me, that Microsoft said that they would modify the OSP for OOXML and specify that it also covers any future versions of OOXML that Microsoft participated in producing. We have still to see the modified text of the OSP, so Microsoft please "encourage" your legal team to finish up the wording.

and third,

(and by far, most importantly) 

Today is the day of the off-line beer drinking of the participants in the Danish debates around ODF/OOXML. We decided that it would be fun to have an off-line meeting between those of us that know each other by (blog)name only. It takes place at 17.00 this afternoon at BrewPub in Copenhagen. If you have a chance, feel free come by and share a beer with the rest of us.


Only fools rush in

In the past weeks the OOXML-battleground has been covered with buzz, rumours, and innuendo about the missing IS 29500. Rob kicked it off with him revealing the little secret that he was actually in posession of a preliminary edition of the not-yet published final text.

Today I was told that ITTF was frantically (my interpretation) rushing to get the text done and have it published.

That scares me a bit.

When ITTF/IEC publishes IS29500 - that will be the IS29500 we will use in the foreseeable future. Yes, there will likely be an errata-sheet, but that is the edtion we're gonna use. Rushing to finish it could impose errors in the text (yes, I am aware of the irony).

So dear ITTF, please take your time to finish putting the spec together - there is no rush. ODF was approved on May 8th 2006 and was not published until November 30th the same year. That is more than 6 months from approval to publication. Granted, there are differences between the ODF-case and the OOXML-case. ODF was not changed during the approval-phase of ISO whereas substantial changes were made to OOXML. Never the less, I would much rather wait another month to have it done right.

At least we are all equal here - it is to the advantage of no single perty to have publication delayed ... so shouldn't we wait until it is ready? 

No reason anymore to mandate anything but ODF?

Yesterday the news broke about Microsoft adding support for ODF in Microsoft Office 2007 SP2. Within minutes the news spread like fire on the hills of Malibu, California and blog-entries started to pop up everywhere - even Brian Jones has apparently returned from Winter hibernation and has made his first blog post in almost 6 weeks. Welcome back to the party, Brian.


The Denmark IT-news sphere was not hesitant on the keyboard as well and ComputerWorld Denmark posted an article yesterday evening and the competition on version2.dk followed up on the news this morning. I myself got the information from Luc Bollen in his comment in the article I wrote on document translation (and why it sucks). I was sitting under a maple-tree (or some other wooden artifact) having a beer with a friend after a fabulous sushi-dinner and could do absolutely nothing about it.


Well, the reactions to Microsoft's move has actually been surprisingly positive. Even the ADD-bunch at noooxml.org said "If this is an honest attempt to play nice, it is a very welcome move" and even IBM has been quite positive - prompting Bob Sutor to turn the axe on Apple saying: "Hey, Apple, what about you? Let’s see you do this in iWork!". Simply starting to beat on someone else reminds me of the John Wayne quote "A day without blood is like a day without sunshine".

But what is missing from the reactions?

OSP coverage of ODF

One of the side-effects of Microsoft joining OASIS ODF TC is that ODF will likely be included in the list of specifications covered by Microsoft's Open Specification Promise (OSP). The list of specificationshas not yet been updated, but I would expect it to be updated soon - or at least when they officially join the ODF TC. When you think about all the fuss around IPR in this Spring, it is quite surprising that noone has picked up on this. It rams a huge stick through the FUD about the OSP not being applicable for GPL-licensed software. Now the OSP covers ODF as well and thereby the native document format of OpenOffice.org [LGPL 3.0 license] and (I think) OpenOffice Novell Edition.

But why OOXML, then?

A lot of people are now spinning information about this move pulling the rug under OOXML and that ODF should be mandated everywhere - but nothing could be further from the truth. The reason why we approved OOXML still stands and the incompatible feature-sets of OOXML and ODF did not suddenly become compatible. There are still stuff in OOXML that cannot be persisted in ODF and vice versa. The backwards compatibility to the content in the existing corpus of binary documents is still a core value of OOXML and this incompatibility of ODF has not dissapeared. You will still loose information and functionality when you choose to persist an OOXML-file in ODF ... just as you would when persisting it to old WordPerfect formats. Insisting that having ODF-support in Microsoft Office (12 SP2) makes the need for OOXML go away is a moot point - since I am sure no one would argue to replace OOXML with TXT - simply because TXT is a supported format in Microsoft Office.

Microsoft steps up to the task at hand

Some quite extraordinary news emerged from the Redmond, WA, headquarters of Microsoft today. In summary, they announced that

  1. Microsoft will join OASIS ODF TC
  2. Microsoft will include ODF in their list of specifications covered by the Open Specification Promise (OSP)
  3. Microsoft will include full, native support for ODF 1.1 in Microsoft Office 14 and in Microsoft Office 12 SP2 - scheduled for Q2 2009. Microsoft Office 12 SP" will have built-in support for the three most widely used ISO-standards for document formats, e.g. OOXML, ODF and PDF.

My initial reaction when I heard it was "Wow . that's amazing". I am sure a lot of people will react "It's too little, too late", though, but let me use a couple of bytes to describe why I think it is a good move by Microsoft.

Microsoft joins OASIS ODF TC

Well, Microsoft has been widely criticised for not joining OASIS a few years ago. I think it is a bogus claim, but never the less; it has been on the minds of quite a lot of people. Novell has had a seat in both ECMA TC45 and OASIS ODF TC for some time now, and it is my firm belief that both consortia has benefited by this. The move by Microsoft to join OASIS ODF TC will likely have a similar effect. One of the most frequent requests in the standardisation of OOXML was to increase the feature-overlap of ODF and OOXML. This is quite difficult to accomplish (effectively) without knowing what the features of the other document format is (going to be). By Microsoft participating on both committees (and IBM will hopefully consider joining ECMA TC45) harmonization (or "enlargement of the feature-overlap") will likely occur at a quicker pace.

This also means that the worries some of us have had about Microsoft's future involvement in standardisation work around document formats has been toned down a bit. Microsoft is now actively participating in this work in ECMA, in ISO and also in OASIS. I think this is really good news. Not good news for Microsoft - but good news for those of us that are working with document formats every day.

Microsoft will cover ODF with OSP

One of the most difficult, non-technical, discussions during the standardisation of OOXML was legal aspects. It was discussions about different wordings in Sun's CNS, IBM's ISP and Microsoft's OSP (Jesus Christ, guys, pick ONE single acronym, already!) and the possible impact on implementers of ODF and OOXML. One of the aspects of the discussion that never really surfaced was that if IBM has software patents covering ODF - some of them quite possibly cover parts of OOXML as well. But the ISP of IBM does not mention OOXML - it only mentions ODF. This leaves me as a developer in quite a legal pickle, because by implementing OOXML I am covered by the OSP - but I am not covered by IBM's ISP (and vice versa). To me as a developer, Microsoft's coverage of ODF in their OSP is a good move, because it should remove all legal worries I might have around stepping into SW-patent covered territory.

ODF support in Microsoft Office

Microsoft will finally deliver on requests for native ODF-support for ODF in Microsoft Office. Microsoft will support ODF 1.1 in Microsoft Office 12 SP2 and also have built-in support for PDF and XPS (these are currently only available as a separate download).

Denmark is one of the countries where both ODF and OOXML have been approved for usage in the public sector. This is currently bringing quite a bit of complexity to the daily work of information workers since there are not many (if any) applications offering high fidelity, native support for both formats. They hence rely on translators like ODF-Converter or similar XSLT-based translators. It's a bad, but currently necessary, choice. The usage of translators for document conversion has been widely criticised, amongst others by Rob and I, and the built-in support for ODF in Microsoft Office is a great step in the right direction.

As with everything Microsoft does, we need a healthy amount of scepticism as to which extend they will deliver on their promises. However, I truly believe that the moves by Microsoft here are good news - regardless of the scepticism. An old proverb says "don't count your chickens before they hatch" - and this applies perfectly here. We will have to wait and see what will eventually happen - but so far . it looks good.

Beer, beer, beer ... bed, bed, bed ... (Danish)

I den danske del af ODF/OOXML-blogsfæren har vi et stykke tid talt om, at det kunne være skægt at mødes til en pilsner et eller andet sted i København. De fleste af os kender jo kun hinanden fra vores respektive navne og gravatars, og det er i hvert fald min erfaring, at man er lidt langsommere på aftrækkeren, når man har fået sat et ansigt på navnet.

Jeg indbyder derfor til en omgang øl (eller lignende, det kunne også være noget uden alkoholisk indhold) 

torsdag d. 12. juni 2008 kl. 17:00  

Jeg finder ud af et sted at være inden længe - men hvis du har et bud på et sted, så sig til. Jeg hælder selv til ScrollBar på ITU i Ørestaden.

Tilmelding sker i kommentartråden herunder.