decoration decoration
Stories

GROKLAW
When you want to know more...
decoration
For layout only
Home
Archives
Site Map
Search
About Groklaw
Awards
Legal Research
Timelines
ApplevSamsung
ApplevSamsung p.2
ArchiveExplorer
Autozone
Bilski
Cases
Cast: Lawyers
Comes v. MS
Contracts/Documents
Courts
DRM
Gordon v MS
GPL
Grokdoc
HTML How To
IPI v RH
IV v. Google
Legal Docs
Lodsys
MS Litigations
MSvB&N
News Picks
Novell v. MS
Novell-MS Deal
ODF/OOXML
OOXML Appeals
OraclevGoogle
Patents
ProjectMonterey
Psystar
Quote Database
Red Hat v SCO
Salus Book
SCEA v Hotz
SCO Appeals
SCO Bankruptcy
SCO Financials
SCO Overview
SCO v IBM
SCO v Novell
SCO:Soup2Nuts
SCOsource
Sean Daly
Software Patents
Switch to Linux
Transcripts
Unix Books

Gear

Groklaw Gear

Click here to send an email to the editor of this weblog.


You won't find me on Facebook


Donate

Donate Paypal


No Legal Advice

The information on Groklaw is not intended to constitute legal advice. While Mark is a lawyer and he has asked other lawyers and law students to contribute articles, all of these articles are offered to help educate, not to provide specific legal advice. They are not your lawyers.

Here's Groklaw's comments policy.


What's New

STORIES
No new stories

COMMENTS last 48 hrs
No new comments


Sponsors

Hosting:
hosted by ibiblio

On servers donated to ibiblio by AMD.

Webmaster
The AFNOR Affair: Interview with Frédéric Couchet, Ex. Dir. APRIL, on OOXML in France
Saturday, April 19 2008 @ 01:58 AM EDT

The AFNOR Affair: Interview with Frédéric Couchet of APRIL on OOXML in France
~ by Sean Daly

[Interview was conducted in French and translated by Daly]

Q: Could you tell us a little bit about APRIL and your role in it?

Frédéric Couchet: Founded in 1996, APRIL is the main French advocacy association devoted to promote and protect Free/Libre Software. APRIL, made up of over 2000 members -- individuals, companies, and organizations -- is a pioneer of Free Software in France. Since 1996, it has been a major player in the field of Free Software. Its aim: making Free Software more accessible for the general public, professionals and institutions, and thus more widespread. It also acts as a watchdog on digital liberties, warning the public about the dangers of private interests keeping an exclusive stranglehold on information and knowledge. Its goals are to promote free software and open standards towards individuals, institutions and companies in the French-speaking space, to protect free software users' and authors' rights, and to favor knowledge sharing. I am the founder of APRIL and since 2005 I am the executive director of APRIL.

Q: I believe you had a seat on the AFNOR commission which evaluated OOXML. The abstention from the vote surprised many observers; you mentioned on your blog that this was not a consensus decision and indeed, that abstention was ruled out at a meeting on March 25th. What do you think happened?

Frédéric Couchet: The Association française de normalisation (AFNOR) is the official French standards body. It is a member of the International Organization for Standards (ISO) in which it represents France. In June 2007, AFNOR created a national standards commission concerning the standardization proposal of the format ECMA-376 Office Open XML. APRIL has participated in the work of this commission since June 2007.

Twenty or so members of the commission were present at the March 25, 2008 meeting. Unsurprisingly, no consensus was reached at the meeting. Some members wanted a Yes vote from the commission, others -- the majority -- wanted a No vote from the commission. As indicated in our press release, "OOXML, The Standard Adopted In Advance", at the meeting of 25 March 2008 meant to finalize the commission's position, all members present were unequivocally in agreement over the fact that an abstention position was not an acceptable choice.

The representatives of the General Business Directorate (DGE) and of the General Directorate for State Modernization (DGME) were clearly in favor of No at this meeting. An article which appeared on Friday, March 28, in the French business newspaper Les Échos announced by the way that France would refuse to approve the Microsoft document format. In the late afternoon of Friday the 28th, the secretary of the commission informed us by e-mail that AFNOR was awaiting "new elements of information" capable of having an impact on the finalization of the French position concerning OOXML. These "new elements of information" arrived around 9PM. It was a letter [PDF (French)] from the president of Microsoft France, Eric Boustouller [ed: see this article in Les Échos, in French], the support of Hewlett-Packard as well as the position of Patrick Durusau [PDF].

At the time of the March 25th meeting, AFNOR had already read to us a letter from the president of Microsoft France. The document sent on March 28th restated the same arguments. Patrick Durusau's position was already known on March 25th. Only the position of HP was not necessarily known, but the support of HP was not propped up by any technical argument. Moreover the representative of HP France had never participated in the work of the AFNOR commission, despite being a member. The elements sent by Microsoft therefore did not bring any truly new element.

The change in the position of the DGE and the DGME cannot therefore be explained by the analysis of these elements. Note by the way that the two representatives of the DGE and the DGME did not participate in the debates on the commission's mailing list after the documents sent by Microsoft were received. We learned of the change of position of the DGE and the DGME in an e-mail sent March 31, 2008 by the secretary of the standardization commission. In this e-mail he said AFNOR would abstain. He mentioned also that the DGE and the DGME had changed their position without giving us a copy of the complete text of the new position of these two structures. We therefore do not know who communicated this position.

I asked AFNOR and the representatives of the DGE and the DGME to let us know the technical arguments explaining the change in position or failing that the complete copy of the messages AFNOR received. I have not as of this date received any response. Why this change of heart at the last minute? I cannot seriously imagine that the representatives of the DGE and the DGME changed their minds on the subject. I therefore consider that these two structures acted upon instructions, perhaps by the cabinet of the Ministry of Finance (of which the DGE and the DGME are part) or even the cabinet of the President of the Republic. It's the question I asked in the APRIL press release on OOXML:

"The representatives of the DGE and the DGME were clearly in favor of a No at the closing meeting of the Commission's work. Why this change of heart at the last minute? The documents sent by Microsoft presenting no new element, must one conclude that the DGME and the DGE were ordered to change their mind? On whose instructions?" asks Frédéric Couchet, executive director of APRIL.

A very complete press article [French] on the AFNOR Affair, "OpenXML is standardized... the process of the French vote, less so", leads to a response:

Within the administration, a rumor is circulating that the French reversal was dictated by the Elysian Palace, via the adviser specialized in new technologies, Franck Suplisson. "The administration's experts held to their negative position until the last moment, March 25th. After that, it's a black hole..." according to our source.

There is an error in the name of of the technical adviser for information and communication technologies. It should be Franck Supplisson with a double "p". The instructions to the DGE and DGME to change their position would therefore have come directly from the Elysian Palace [President's office, ndlr]. The article is signed Reynald Flechaux, previously editor-in-chief of the magazine "Le Monde Informatique" which was a weekly publication dedicated to business and IT company executives, one can reasonably suppose he has good sources. Aside from his serial exchanges with AFNOR, did Eric Boustouller, president of Microsoft France, write to Nicolas Sarkozy as well?

Q: The French newspaper Les Echos published a letter [PDF, French] from the president of Microsoft France sent to the general director of AFNOR for transmission to commission members some 24 hours before the OOXML vote deadline. In the letter, reference is made to telephone conversations that day and contact with Microsoft headquarters. Microsoft promised to promote interoperability and maintain OOXML as a standard. Why do you think AFNOR and some commission members believe what Microsoft says?

Frédéric Couchet: AFNOR hopes in effect that ECMA and Microsoft will respect their "commitments". Olivier Peyrat, AFNOR's General Director, had indicated at the AFNOR press conference: "We reserve the ability to alert the market in case of inadequate behavior of certain actors." Olivier Peyrat thinks that any after-the-fact change in Microsoft or ECMA's position would backfire on them in the future. This argument is just plain ridiculous. A company like Microsoft couldn't care less about this type of possible threat especially now that its format has the ISO label, which is the key thing for Microsoft.

Q: Last August there were some reports that the AFNOR commission meetings were heated. Can you tell us anything about that?

Frédéric Couchet: On August 29th 2007 the AFNOR standardization commission meeting took place with the objective of establishing the position of the commission and therefore consequently France's position. The exchanges were stormy at some points since Marc Mossé, head of Legal and Public Affairs at Microsoft France, did everything, I thought, he could to sabotage the meeting. Marc Mossé, judging from appearances, seemed to have the very clear assignment to obtain AFNOR's abstention. Absolutely not constructive, not very polite either, in particular with the representatives of the French administration, Marc Mossé seemed to have decided to ruin the meeting and heighten the pressure -- well-known tactic to block the arrival at a consensus. But he did too much, way too much. The end was pitiful enough, notably when he accused one of the State's representatives of serving a "banana republic". He claimed by the way to be representing local administrations against the central administration. The resume of Marc Mossé is online but strangely, his stint at the BSA, the Business Software Alliance, is not mentioned in it. The meeting of March 25th 2008 was much more calm and cordial, perhaps because of the absence of Marc Mossé.

Q: What can you tell us, if anything, concerning the European Commission investigation of irregularities during the OOXML ISO fast track process?

Frédéric Couchet: I saw the announcement about that investigation in a Reuters dispatch but I don't have more information. It seems that the European Commission had sent a letter to different standards bodies, but AFNOR refused to confirm or deny having received such a letter. On this subject, the journalist Bertrand Lemaire indicated in an article:

According to our sources, the European Commission sent a letter to these different bodies, asking them if they had experienced pressures. Although Frédéric Bon denied the existence of "particular" pressures, Olivier Peyrat for his part flatly refused to confirm or deny having received such a letter and, moreover, to make the slightest comment: "No comment. If there is an investigation, we will respond to the investigation. I won't say anything else."
In any case the members of the standardization commission are not aware of anything concerning this inquiry. AFNOR did not inform us about having received a letter from the European Commission.

Q: Microsoft says that users should have a choice of standards. According to you, is it important that there be only one ISO-approved editable document format?

Frédéric Couchet: Microsoft's position reveals a profound disagreement on the definition and the role of standards. For users, the benefit of a standard is to have compatibility between competing products in the market. As soon as there are two or several standards for the same applications, these are no longer standards, but mere technical specifications. If there had been only one ISO-approved format, it would have been a means to relaunch competition and innovation in the office productivity market. if there are several, the situation remains the same as in the past. The existence of a dual standard doesn't bring anything more than the absence of a standard.

Already, AFNOR and the DIN agency, the official German standardization body, are imagining harmonization projects between ODF and OOXML, and thus recognize implicitly that the coexistence of the two standards poses a problem.

Q: Hewlett-Packard announced that it supports both ODF and OOXML and echoes Microsoft's view that there should be a choice of more than one editable document format. Did they have a role in the AFNOR decision?

Frédéric Couchet: In reality, Hewlett-Packard did not actively participate in the OOXML adoption process, for which they had neither a direct interest nor technological competence. Hewlett-Packard, in contrast to its competitors IBM (publisher of the Lotus suite) and Sun (publisher of StarOffice and sponsor of OpenOffice.org), has no experience and no activity in the field of office productivity software. Hewlett-Packard's position was determined by the commercial links which that company maintains with Microsoft. This was particularly visible in France, in relation to AFNOR. In effect, the HP statement was brought to AFNOR by a member of the Microsoft delegation and not by an HP representative. This statement was in reality a simple link to the public page "HP Position Statement on Standardization of Office Document Formats", which offers no technical argument.

Q: AFNOR last year recommended that OOXML be merged with ODF and even proposed a roadmap [PDF, French] to do so. Of course, Microsoft has consistently refused to support ODF natively, so it would be difficult to imagine Microsoft wishing to accomodate the existing ISO standard. Is this harmonization project stillborn?

Frédéric Couchet: Technically, this harmonization project no longer makes any sense. When Microsoft and ECMA developed OOXML, they in no way took into account the acquired experience of ODF while ODF 1.0 was already an international standard, that ODF was already published, and that certain work issues were already known. It was nevertheless an excellent situation for preparing future harmonization. The refusal to take ODF into account in the development of OOXML is proof that Microsoft is not inclined to meet halfway. Most of all, OOXML is basically conceived according to an architecture which reflects the organization of Microsoft office suites; the specifications are very heterogeneous and their separation reflects that of the Office System environment.

OOXML is therefore not easier to harmonize with ODF than the old Microsoft binary formats. The recent harmonization projects, as well as the ODF/OOXML converters sponsored by Microsoft but external to Microsoft Office, are more revealing of a seduction campaign for OOXML than a real perspective of interoperability. By the way, for the market, the only useful thing is compatibility with Microsoft Office, which means that Microsoft's competitors will have to perform reverse engineering of the true Microsoft Office formats and won't be able to trust the OOXML specification which is too complex and too theoretical.

Q: The ISO says that national bodies have two months to appeal the standardization decision. Do you think AFNOR would find grounds for appealing?

Frédéric Couchet: AFNOR went from a negative vote to abstention under political influence the details of which we don't know yet. In order for there to be an appeal against the ISO decision, AFNOR would have to first be liberated from this influence and following that explain why it didn't maintain its negative vote of 2007. It is highly unlikely that all that will happen in the timeframe of two months after the decision.

Q: The General Directorate for State Modernisation (DGME) began work a year ago, I believe, on the General Reference for Interoperability (RGI), a framework of technical recommendations for French administrations not unlike the Massachusetts Enterprise Technical Reference Model. The draft version 0.90 [PDF, French] of the RGI specifically recommends ODF. Reports published yesterday [ed: April 17 here, here, and here] refer to an internal e-mail of the DGME which proposes that the section relative to office productivity document formats should include OOXML and that the RGI should be urgently finalized now that OOXML has been approved by the ISO. What is your reaction?

Frédéric Couchet: Small correction: The DGME had in fact started work in April 2006 with a call for public comments concerning the "General Reference for Interoperability" (RGI) brought about by the Act Nr. 2005-1516 of December 8, 2005 relating to electronic exchanges between users and administrative authorities and between administrations.

This text, from its initial draft in 2006, recommended the use of the ODF format for exchanges of office productivity documents in civil services. Microsoft led an intense battle against the validation of the RGI project and its enactment. The RGI was therefore blocked for several months (from October 12th 2007, date of the last meeting of the committee in charge of validating the RGI's enactment) without any official reason.

If the information published by the French press concerning the RGI were to be confirmed, this would be truly scandalous.

Following the ISO vote, Benoît Sibaud, president of APRIL, had already indicated that "the contents of the 'Standard adopted in advance' is today mostly undetermined. It will be necessary to wait several months (or more) for 'the OOXML standard' to be published so that Microsoft's competitors can hope to use it."

The ISO voting procedure has moreover been subject to irregularities in numerous countries, in particular the countries which changed their vote from No to Yes. The investigation into the procedure process, recently launched by the European Commission, should shed light on the serious irregularities, manipulations and scandals which marred the procedure. It seems urgent to me to wait before modifying the RGI text and validating it.

This new event reinforces, if it wasn't already necessary, the absolute need for clear explanations concerning the change of direction of the French position and the role played by various actors (in particular at ministry cabinet levels). Aside from that, it is rather bizarre that France rushes to reinforce the dominant position of Microsoft, a company censured many times for its practices.

On a wider scale it is inadmissible to see applied, these past weeks in France, purely political decisions which disregard all technical reality, and the associated distortion of competition.

Q: Do you think the ISO's reputation has been adversely affected by the events occurring during the fast-tracking of OOXML? What about AFNOR's?

Frédéric Couchet: ISO standards were traditionally the result of compromise between competing players. With OOXML, Microsoft obtained a standard which none of its competitors had adopted. The first ISO standard in this area, Open Document Architecture (ODA, ISO 8613) was never applied. The second standard, OpenDocument Format (ODF, ISO/IEC 26300), was applied but never accepted by Microsoft. The third standard, OOXML (ISO/IEC 29500), can help Microsoft commercially with governments but brings nothing to the market in terms of convergence between competing software products, and thus has no added value as a standard.

The ISO appears then as a vendor of heterogeneous and contradictory standards, without a global objective of coherence and quality. Moreover, the incidents during the OOXML adoption process showed that the ISO decision structure and the national standards bodies (NSBs ou NBs) were very vulnerable to pressures, and that the voting result did not faithfully represent the strategic orientations of governments. The reputation of the worldwide standardization system is thus gravely compromised.

This compromising has already been denounced by Martin Bryan, the previous Convenor of JTC1/SC34/WG1, in his most recent public report. It is therefore possible that certain industrial firms or governments no longer have confidence in the ISO, or that they insist upon drastic changes. It is also possible that the ISO will be dropped by certain major players in innovation to the benefit of other organizations.


L'Affaire AFNOR: Interview avec Frédéric Couchet d'APRIL sur OOXML en France

Q: Pouvez-vous nous parler un peu sur l'APRIL et votre role dans son sein ?

Frédéric Couchet: Créée en 1996, l'APRIL est la principale association francophone dédiée à la promotion et la défense du logiciel libre. Pionnière du logiciel libre en France, l'April, constituée de plus de 2 000 membres (individus, entreprises, associations et collectivités locales), est depuis 1996 un acteur majeur de la démocratisation et de la diffusion du logiciel libre et des standards ouverts auprès du grand public, des professionnels et des institutions. Elle agit en tant que sentinelle pour les libertés numériques et veille à sensibiliser l'opinion sur les dangers d'une appropriation exclusive de l'information et du savoir par des intérêts privés. L'April a pour objectifs de promouvoir le logiciel libre et les standards ouverts auprès du grand public, des professionnels et des institutions dans l'espace francophone, de protéger les droits des auteurs et utilisateurs de logiciel libre et de favoriser le partage du savoir et des connaissances. Je suis le fondateur de l'April et depuis 2005 j'en suis le délégué général.

Q: Je crois que vous aviez siegé dans la commission de l'AFNOR qui a évalué OOXML. L'abstention du vote a surpris de nombreux observateurs; vous avez indiqué dans votre blog que cette décision n'était pas achevé par consensus et meme que l'abstention était exclu lors d'une réunion le 25 mars. Qu'est-ce que vous pensez s'est arrivée ?

Frédéric Couchet: L'Association française de normalisation (AFNOR) est l'organisme officiel français de normalisation. Il est membre de l'Organisation internationale de normalisation (ISO) auprès duquel il représente la France. L'AFNOR a créé en juin 2007 une commission de normalisation nationale concernant le projet de normalisation du format ECMA-376 Office Open XML. L'APRIL participe aux travaux de cette commission depuis juin 2007.

Une vingtaine de membres de la commission étaient présents à la réunion du 25 mars 2008. Sans surprise, aucun consensus ne s'est dégagé lors de la réunion. Des membres souhaitaient un vote OUI de la commission, d'autres -- majoritaires -- souhaitaient un vote NON de la commission. Comme indiqué dans notre communiqué de presse, « OOXML - la norme adoptée d'avance », lors de la réunion du 25 mars 2008 destinée à finaliser la position de la commission, tous les membres présents étaient sans équivoque d'accord sur le fait qu'une position d'abstention n'était pas un choix acceptable.

Les représentants de la Direction Générale des Entreprises (DGE) et de la Direction générale de la modernisation de l'Etat (DGME) étaient clairement en faveur du non lors de cette réunion. Un article du journal économique Les Échos paru vendredi 28 mars annonce d'ailleurs que la France va refuser de normaliser le format de document de Microsoft. Vendredi 28 mars en fin d'après-midi le secrétaire de la commission nous a informés par courriel que l'AFNOR était en attente de « nouveaux éléments d'information » susceptibles d'avoir un impact sur la finalisation de la position française concernant OOXML. Ces « nouveaux éléments d'informations » sont arrivés vers 21h. Il s'agit d'une lettre [PDF] du président de Microsoft France, Eric Boustouller (voir l'article des Échos, ndlr), un soutien de la société Hewlett-Packard ainsi que la position de Patrick Durusau [PDF].

Lors de la réunion du 25 mars l'AFNOR nous avait déjà lu une lettre du président de Microsoft France. Le document envoyé le 28 mars reprenait les mêmes arguments. La position de Patrick Durusau était déjà connue le 25 mars. Seule la position d'HP n'était pas forcément connue, mais le soutien d'HP n'est étayé par aucun argument technique. En outre le représentant d'HP France n'a jamais participé aux travaux de la commission de l'AFNOR, bien qu'il en soit membre. Les éléments envoyés par Microsoft n'apportaient donc aucun élément réellement nouveau.

Le changement de position de la DGE et la DGME ne peut donc pas s'expliquer par l'analyse de ces éléments. Notons d'ailleurs que les deux représentants de la DGE et de la DGME ne sont pas intervenus dans les débats sur la liste de discussion de la commission après la réception des documents envoyés par Microsoft. Nous avons appris le changement de position de la DGE et de la DGME par un courriel du secrétaire de la commission de normalisation envoyé le 31 mars 2008. Dans ce courriel il indique que l'AFNOR va s'abstenir. Il nous fait également part des changements de position de la DGE et de la DGME mais sans nous donner copie du texte complet de la nouvelle position de ces deux structures. On ne sait donc pas qui a communiqué cette position.

J'ai demandé à l'AFNOR et aux représentants de la DGE et de la DGME de nous faire connaître les arguments techniques expliquant le changement de position ou à défaut la copie complète des des messages que l'AFNOR a reçu. Je n'ai à ce jour reçu aucune réponse. Pourquoi ce changement d'avis de dernière minute ? Je ne peux pas sérieusement imaginer que les représentants de la DGE et de la DGME ont changé d'avis sur le sujet. J'estime donc que ces deux structures ont agi sur instructions, peut-être par le cabinet du ministère des Finances (dont dépendent la DGE et la DGME) voire même du cabinet de la présidence de la République. C'est la question que je posais dans le communiqué de presse de l'APRIL sur OOXML :

« Les représentants de la DGE et de la DGME étaient clairement en faveur du non lors de la réunion de clôture des travaux de la commission. Pourquoi ce changement d'avis de dernière minute ? Les documents envoyés par Microsoft n'apportant aucun élément nouveau, doit-on en conclure que la DGME et la DGE ont changé d'avis sur instructions ? Sur instructions de qui ? » se demande Frédéric Couchet, délégué général de l'APRIL.
Un article très complet sur l'affaire AFNOR, « OpenXML est normalisé... le processus de vote de la France, lui, l'est moins », donne une piste de réponse :
Au sein de l'administration, court la rumeur que le revirement français aurait été dicté de l'Elysée, par le conseiller spécialisé dans les nouvelles technologies, Franck Suplisson. « Les experts de l'administration ont gardé leur position négative jusqu'au dernier moment, mardi 25. Après c'est le trou noir... », reprend notre interlocuteur.
Il y a une erreur dans le nom du conseiller technique technologies de l'information et de la communication (TIC), il s'agit de Franck Supplisson avec deux « p ». Les instructions de changement de position de la DGE et DGME seraient donc venues directement de l'Elysée. L'article est signé Reynald Flechaux, ancien rédacteur en chef du magazine « Le Monde Informatique », qui était un hebdomadaire dédié aux dirigeants d'affaires et d'IT, on peut donc supposer qu'il a de bonnes sources. Outre ses échanges épistolaires avec l'AFNOR, Eric Boustouller, président de Microsoft France, aurait-il écrit également à Nicolas Sarkozy ?

Q: Le journal français les Echos a publié un courrier [PDF] du president de Microsoft France envoyé au directeur général de l'AFNOR pour transmission aux membres du commission quelques 24 heures avant la date limite du vote OOXML. Dans cette lettre, on fait référence aux conversations téléphoniques ce jour-la et contact avec le sièege de Microsoft. Microsoft a promis de promouvoir l'interoperabilité et de maintenir OOXML en tant que norme. A votre avis, pourquoi AFNOR et plusieurs membres du commission croient ce qui dit Microsoft ?

Frédéric Couchet: L'AFNOR espère effectivement que l'ECMA et Microsoft vont respecter leurs « engagements ». Olivier Peyrat, le directeur général de l’AFNOR, aurait indiqué lors de la conférence de presse de l'AFNOR : « Nous gardons la capacité d'alerter le marché en cas de comportement inadéquat de certains acteurs ». Olivier Peyrat pense que tout changement ultérieur de position de Microsoft et l'ECMA « grillerait » les intéressés dans le futur. Cet argument est tout simplement ridicule. Une société comme Microsoft n’a que faire de ce genre de menace éventuelle surtout maintenant que son format a la label ISO, ce qui est l'essentiel pour Microsoft.

Q: En août dernier il y avait des indications que les réunions de la commission de l'AFNOR étaient chaudes, pouvez-vous nous parler a ce sujet ?

Frédéric Couchet: Le 29 août 2007 avait eu lieu la réunion de la commission de normalisation de l'AFNOR avec pour objectif d'établir la position de la commission et donc par conséquence celle de la France. Les échanges ont été assez houleux par moments car Marc Mossé, Directeur des Affaires Juridiques et Publiques de Microsoft France, avait tout fait pour saboter la réunion. Marc Mossé avait visiblement pour mission très claire d'obtenir une abstention de l'AFNOR. Absolument pas constructif, pas très poli non plus, notamment avec les représentants de l'administration française, Marc Mossé avait clairement décidé de pourrir la réunion et de faire monter la pression. Tactique connue pour empêcher d'arriver à un consensus. Mais il en faisait trop, beaucoup trop. Le final était assez pitoyable, notamment lorsqu'il a accusé l'un des représentants de l'Etat de servir une « république bananière ». Il prétendait d'ailleurs représenter les collectivités locales face à l'administration centrale.

Le CV de Marc Mossé est en ligne mais bizarrement son passage à la BSA, la Business Software Alliance, n'y est pas mentionné. La réunion du 25 mars 2008 a été beaucoup plus calme et cordiale. L'absence de Marc Mossé y est peut-être pour quelque chose.

Q: Que pouvez-vous nous dire concernant l'enquête de la Commission Européene sur des irregularités pendant le processus « fast track » de OOXML a l'ISO ?

Frédéric Couchet: J'ai vu l'annonce de cette enquête dans une dépêche Reuters mais je n'ai pas plus d'information. Il semblerait que la Commission Européenne ait envoyé un courrier à différents organismes de normalisation, mais l'AFNOR a refusé de confirmer ou d'infirmer la réception d'une telle lettre. Le journaliste Bertrand Lemaire indique à ce propos dans un article :

Selon nos informations, la Commission Européenne aurait envoyé un courrier à ces différents organismes pour leur demander s'ils avaient subi des pressions. Si Frédéric Bon a nié l'existence de pressions « particulières », Olivier Peyrat a, quant à lui, très sèchement refusé de confirmer ou d'infirmer la réception de cette lettre et, plus encore, d'apporter le moindre commentaire : « Aucun commentaire. S'il y a enquête, nous répondrons à l'enquête. Je n'en dirai pas plus ».

En tout cas les membres de la commission de normalisation se sont au courant de rien concernant cette enquête. L'AFNOR ne nous a pas informé avoir reçu un courrier de la commission européenne.

Q: Microsoft dit que les utilisateurs doivent disposer d'un choix de normes. D'après vous, est-il important qu'il existe qu'un seul format de saisie de documents approuvé par l'ISO ?

Frédéric Couchet: La position de Microsoft démontre un désaccord fondamental sur la définition et le rôle des normes. L'intérêt d'une norme pour les utilisateurs est de permettre la compatibilité entre les produits concurrents sur le marché. Dès lors qu'il y a deux ou plusieurs normes pour les mêmes applications, ce ne sont plus des normes, ce sont de simples spécifications techniques. S'il n'y avait qu'un seul format approuvé par l'ISO, ce serait un moyen de relancer la compétition et l'innovation sur le marché des logiciels bureautiques. S'il y en a plusieurs, la situation reste la même que par le passé. L'existence d'une double norme n'apporte rien de plus que l'absence de norme.

Déjà, les agences AFNOR et DIN, l'organisme officiel allemand de normalisation, sont en train d'imaginer des projets d'harmonisation entre ODF et OOXML, et reconnaissent donc implicitement que la cohabitation des deux normes pose un problème.

Q: Hewlett-Packard a annoncé qu'il soutient a la fois ODF et OOXML et répété le point de vue de Microsoft qu'il doit y avoir un choix de plus d'un format de saisie de documents. Ont-ils joué une role dans la décision de l'AFNOR ?

Frédéric Couchet: En réalité, Hewlett-Packard n'a pas activement participé au processus d'adoption d'OOXML, pour lequel il n'avait ni intérêt direct ni compétence technologique. Hewlett-Packard, à la différence de ses concurrents IBM (éditeur de la gamme Lotus) et Sun (éditeur de StarOffice et sponsor d'OpenOffice.org), n'a aucune expérience et aucune activité dans le domaine des logiciels bureautiques. La position de Hewlett-Packard a été déterminée par les liens commerciaux qu'entretient cette société avec Microsoft. Cela a été particulièrement visible en France, auprès de l'AFNOR. En effet, la déclaration de HP a été rapportée à l'AFNOR par un membre de la délégation de Microsoft et non par un représentant de HP. Cette déclaration était en réalité un simple lien vers la page publique « HP Position Statement on Standardization of Office Document Formats », qui ne présente aucun argument technique.

Q: L'année derniere, l'AFNOR a recommandé que OOXML soit fusionné avec ODF et a même proposé un carnet de route [PDF] pour ce faire. Bien entendu, Microsoft a systématiquement refusé de soutenir ODF en natif, ce sera donc difficile d'imaginer Microsoft voulant accomoder la norme ISO existant. Est-ce que ce projet d'harmonisation est mort-né ?

Frédéric Couchet: Techniquement, ce projet d'harmonisation n'a plus aucun sens. Lorsque Microsoft et l'ECMA ont développé OOXML, ils n'ont tenu aucun compte des acquis d'ODF alors que ODF 1.0 était déjà un standard international, que ODF 1.1 était déjà publié, et que certains travaux d'ODF 1.2 étaient déjà connus. C'était pourtant une excellente occasion de préparer une harmonisation future. Le refus de tenir compte d'ODF dans le développement d'OOXML est une preuve que Microsoft n'est pas disposé à faire une partie du chemin. Surtout, OOXML est conçu à la base selon une architecture qui reflète l'organisation des suites bureautiques Microsoft ; les spécifications sont très hétérogènes et leur découpage reflète celui de l'environnement Office System.

OOXML n'est donc pas plus facile à harmoniser avec ODF que les anciens formats binaires Microsoft. Les actuels projets d'harmonisation, de même que les convertisseurs ODF/OOXML sponsorisés par Microsoft mais extérieurs à Microsoft Office, révèlent plutôt une opération de séduction pour OOXML qu'une réelle perspective d'interopérabilité. D'ailleurs, pour le marché, seule la compatibilité avec Microsoft Office est utile, ce qui veut dire que les concurrents de Microsoft devront faire un reverse engineering des vrais formats Microsoft Office, et ne pourront pas faire confiance à la spécification OOXML, trop complexe et trop théorique.

Q: L'ISO dit que les organismes de normalisation ont deux mois pour faire appel a la décision de standardisation. Pensez-vous que l'AFNOR pourrait trouver des raisons pour loger un appel ?

Frédéric Couchet: L'AFNOR est passé du vote négatif à l'abstention par l'effet d'une influence politique dont nous ne connaissons pas encore tous les détails. Pour qu'il y ait un appel contre la décision de l'ISO, l'AFNOR devrait d'abord être libérée de cette influence et ensuite expliquer pourquoi elle n'a pas maintenu son vote négatif de 2007. Il est peu probable que tout cela se produise dans le délai de deux mois après la décision.

Q: La Direction Générale de la Modernisation de l'Etat (DGME) a démarré il y a un an je crois les travaux sur la référentiel général d'interopérabilité (RGI), un cadre de recommandations techniques pour les administrations françaises comparable à l'Enterprise Technical Reference Model de Massachusetts. La version 0.90 de la RGI [PDF] préconise ODF explicitement. Des articles parues hier [17 avril : voir ici, ici, et ici, ndlr] font référence à un courriel interne de la DGME qui propose que la section relatif aux formats de document bureautique doit inclure l'OOXML et que la RGI doit être finalisé rapidement maintenant que OOXML a été approuvé par l'ISO. Quelle est votre réaction ?

Frédéric Couchet: Petite précision : La DGME avait en fait démarré les travaux en avril 2006 par un appel à commentaires public concernant le « Référentiel Général d'Interopérabilité » (RGI) induit par l'Ordonnance nº 2005-1516 du 8 décembre 2005 relative aux échanges électroniques entre les usagers et les autorités administratives et entre les autorités administratives.

Ce texte, depuis sa rédaction initiale de 2006, préconisait l'emploi du format ODF pour les échanges de documents bureautiques dans les services publics. Microsoft a mené une bataille intensive pour que le projet de RGI ne soit pas validé et que celui-ci n'entre officiellement en vigueur. Le RGI était donc bloqué depuis plusieurs mois (depuis le 12 octobre 2007, date de la dernière réunion du comité chargé de statuer sur le RGI) sans raison officielle.

Si les informations, publiées par la presse française, concernant le RGI venaient a être confirmées ce serait proprement scandaleux.

Suite au vote de l'ISO, Benoît Sibaud, président de l'April avait déjà indiqué que « Le contenu de la "norme adoptée d'avance" est aujourd'hui largement indéterminé. Il faudra attendre quelques mois (ou plus) pour que "la norme OOXML" soit publiée et que des concurrents de Microsoft puissent espérer l'utiliser. »

La procédure de vote à l'ISO a en outre fait l'objet d'irrégularités dans de nombreux pays, notamment les pays ayant changé leur vote de non en oui. L'enquête sur le déroulement de la procédure, lancée récemment par la Commission Européenne, devra faire la lumière sur les sérieuses irrégularités, manipulations et scandales qui ont émaillé la procédure. Il me paraît donc urgent d'attendre avant de modifier le texte du RGI et de le valider.

Ce nouvel évènement renforce, si ce n'était nécessaire, l'absolu besoin d'explications claires sur le revirement de la position française et le rôle joué par les différents acteurs (notamment au niveau des cabinets ministériels). Il est d'ailleurs assez étrange que la France se précipite pour conforter la position dominante de la société Microsoft maintes fois condamnée pour ses pratiques.

Plus globalement il est inadmissible de voir s'appliquer depuis quelques semaines en France des décisions purement politiques, qui font abstraction de toute réalité technique et de la distortion de concurrence associée.

Q: Pensez-vous que la réputation de l'ISO a été impacté de façon négative par les événements pendant le « fast-tracking » de l'OOXML ? Et quid de l'AFNOR ?

Frédéric Couchet: Les standards de l'ISO résultaient traditionnellement de compromis entre des acteurs concurrents. Avec OOXML, Microsoft a obtenu un standard auquel aucun de ses concurrents n'a adhéré. Le premier standard ISO dans ce domaine, Open Document Architecture (ODA, ISO 8613) n'avait jamais été appliqué. Le second standard, OpenDocument Format (ODF, ISO/IEC 26300) a été appliqué mais jamais accepté par Microsoft. Le troisième standard, OOXML (ISO/IEC 29500), peut aider commercialement Microsoft auprès des gouvernements mais n'apporte rien au marché en termes de convergence entre logiciels concurrents, donc n'a pas de valeur ajoutée en tant que standard.

L'ISO apparaît donc comme un vendeur de standards hétérogènes et contradictoires, sans objectif global de cohérence et de qualité. De plus, les incidents du processus d'adoption d'OOXML ont montré que la structure de décision de l'ISO et des organismes nationaux de standardisation (NSBs ou NBs) était très vulnérable aux pressions, et que le résultat des votes ne réflétait pas fidèlement les orientations stratégiques des gouvernements. La réputation du système mondial de standardisation et de ses composantes nationales (dont l'AFNOR) est donc gravement compromise.

Cette compromission a déjà été dénoncée par Martin Bryan, ancien Convenor du JTC1/SC34/WG1, dans son dernier rapport public. Il est donc possible que certains industriels ou certains gouvernements n'accordent plus leur confiance à l'ISO, ou qu'ils exigent des changements drastiques. Il est possible également que l'ISO soit délaissé par certains acteurs majeurs de l'innovation au profit d'autres organisations.


  


The AFNOR Affair: Interview with Frédéric Couchet, Ex. Dir. APRIL, on OOXML in France | 160 comments | Create New Account
Comments belong to whoever posts them. Please notify us of inappropriate comments.
Corrections, if needed
Authored by: Anonymous on Saturday, April 19 2008 @ 02:06 AM EDT
.

---
--Bill P, not a lawyer. Question the answers, especially if I give some.

[ Reply to This | # ]

Off Topic
Authored by: Anonymous on Saturday, April 19 2008 @ 02:06 AM EDT
Please pay attention to the instructions on the Post a Comment page.

---
--Bill P, not a lawyer. Question the answers, especially if I give some.

[ Reply to This | # ]

News Picks Commentary
Authored by: Anonymous on Saturday, April 19 2008 @ 02:07 AM EDT
Please use the title of your comment to help us identify the News Pick.

---
--Bill P, not a lawyer. Question the answers, especially if I give some.

[ Reply to This | # ]

What OOXML means to me
Authored by: robobright on Saturday, April 19 2008 @ 03:43 AM EDT
O is for all the circular references found in the spec.
O is for all the round & round spin on what a wonderful spec it is!
X is for all times the shills, trolls and astroturfers cross their hearts hoping someone will buy their deceptions.
M is for all the different angles that ooxml is framed to win support. "It's backwards compatible all the way to stone tablets!"
L is for this: The spec is so Lousy that at the end of the day not even Microsoft will put it in their own products...


Cheers & good night...

[ Reply to This | # ]

Thank You Sean...
Authored by: Zarkov on Saturday, April 19 2008 @ 03:44 AM EDT
... your report is as enlightening as it is horrifying.

While I am not so nieve as to think that an international standards process
could be totally devoid of politics, it is sobering to see that the reality is
that there is more politics involved than good sense or tachnical capability.

[ Reply to This | # ]

Which governments remain sovereign to Microsoft?
Authored by: grouch on Saturday, April 19 2008 @ 04:22 AM EDT
We've seen Massachusetts lead in succumbing to Microsoft, long before the "standard" farce, which was a sort of prelude to other governments falling, such as Colombia , Russia , Norway (recall that prior to the vote that the Norwegian minister said closed formats, vendor lockin are unacceptable ), Philippines , Romania , Singapore , Germany , the UK , and, of course, the U.S.A. .

It's pretty clear that the governments of countries whose NBs (National Bodies) voted to disapprove MSOOXML are still sovereign governments, concerning Microsoft at least. Are any of the governments of the countries in which the NBs voted in favor of MSOOXML still sovereign to Microsoft? Normal statistics seem to fly right out the window where global subversion of otherwise technical processes appears rampant.

---
-- grouch

"People aren't as dumb as Microsoft needs them to be."
--PJ, May 2007

[ Reply to This | # ]

The AFNOR Affair: Interview with Frédéric Couchet, Ex. Dir. APRIL, on OOXML in France
Authored by: PolR on Saturday, April 19 2008 @ 04:46 AM EDT
Thank you Sean. This is good work.

The link in the last paragraph to Martin Bryan final report is a must read that is too easily lost in the wealth of information. Martin is the predecessor of Alex Brown as the Convenor of JTC1/SC34. This is his parting report you can read on the official ISO/IEC JTC1/SC34 site and a joy to read.

Unless ISO tightens up on its rules, and removes or demotes, P members who do not vote as required by ISO rules I would recommend my successor that it is perhaps time to pass WG1’s outstanding standards over to OASIS, where they can get approval in less than a year and then do a PAS submission to ISO, which will get a lot more attention and be approved much faster than standards currently can be within WG1. 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”, something I have been fighting against for the 20 years I have served on ISO committees. I am glad to be retiring before the situation becomes impossible. I wish my colleagues every success for their future efforts, which I sincerely hope will not prove to be as wasted as I fear they could be.

[ Reply to This | # ]

What I don't understand.
Authored by: Ian Al on Saturday, April 19 2008 @ 05:33 AM EDT
Don't be silly. I couldn't possibly give you a comprehensive list.

The particular thing I don't understand is how Microsoft wield such influence. I
can see how the NB ballot-stuffing works. The Microsoft partners get a
commercial advantage in return. Perhaps they are also concerned about the
potential loss of dominance of their chosen liege.

The ballot-stuffing moved them away from total disaster, but did not win the day
by any reasonable ruling (geddit? ISO rules? Giggedy, giggedy). However,
Microsoft have actually got their way by strong-arming senior bureaucrats in the
UK, Norway, France and many other countries, as well as in the ISO organisation,
itself.

This was traditionally achieved by beating them about the wallet with heavy
brown envelopes. Is that what happened here? Will we ever find out? Is it
possible that even Microsoft did not want to chance bribery and found another
way? Politics (and politics with a small 'p') is the only other way I can think
of and that still requires the bureaucrat to perceive a genuine political
return. They couldn't really care much about their users having choice or
compatibility or interoperability or freedom. It's all about personal reward.

So, how was it done? That's what I don't understand. Do any of you know enough
about political bribery to suggest any answers to me? The folk concerned can't
imagine they are going to get high-paid jobs from Microsoft after this is all
over.

---
Regards
Ian Al

When nothing else makes sense, use Linux.

[ Reply to This | # ]

Online petition to SIS to appeal OOXML.
Authored by: Anonymous on Saturday, April 19 2008 @ 06:02 AM EDT

There's an online petition (only swedish residents) to SIS to appeal OOXML since a few days back.

[ Reply to This | # ]

This affair is getting very political.
Authored by: Anonymous on Saturday, April 19 2008 @ 06:07 AM EDT
Will the "Comments Guidelines" need revising?

[ Reply to This | # ]

In comparrison, the ODF development process looks like organized by Ghandi, King, and Mandela
Authored by: Anonymous on Saturday, April 19 2008 @ 06:21 AM EDT

If we look at the MS OOXML political and financial bribery, we see Judas, Brutus, and Dzerzhinsky at work. And they hired Mohammed Saeed al-Sahaf.

Rob Wire's previous post gives a view how ODF treats suggestions for adaptation, correction, and additions.

Sugg esting ODF Enhancements
A real eye opener. Standardization CAN be done without guards at the door!

Winter (not logged in)

[ Reply to This | # ]

The AFNOR Affair: Interview with Frédéric Couchet, Ex. Dir. APRIL, on OOXML in France
Authored by: Anonymous on Saturday, April 19 2008 @ 06:39 AM EDT
It looks as if the probability of Microsoft implementing ISO OOXML are minimal,
and the ISO process has been thoroughly discredited. If the EU investigation
finds anything irregular, Microsoft gets another fine and passes the cost on to
its captive market as it has done with previous convictions.

So, few will give two hoots as to whether ODF is an ISO standard or not; it has
been made quite clear who pwns the industry and calls the shots. Microsoft's
mission accomplished.

Vik :v)

[ Reply to This | # ]

Collateral Damage: destruction of ISO's work
Authored by: barbacana on Saturday, April 19 2008 @ 08:42 AM EDT

The report by Martin Bryan, the previous Convenor of JTC1/SC34/WG1 (referred to in the article) is worth reading in full. It's clear, reading this, that Microsoft's ballot-stuffing has essentially destroyed a lot of ISO's work unrelated to MSOOXML. Here's an extract:

This year WG1 have had another major development that has made it almost impossible to continue with our work within ISO. The influx of P members whose only interest is the fast-tracking of ECMA 376 as ISO 29500 has led to the failure of a number of key ballots.

What he's saying is that when a country becomes a P-member, it's committing to providing input on a number of standards, and ISO relies on that. But the ballot-stuffers brought into the process by and for Microsoft aren't interested in standards, or in fulfilling the commitment they made; their sole purpose in signing up as P-members was to push through MSOOXML. So a number of other standards are held up, waiting for input that will never come.

[ Reply to This | # ]

Lots of evidence for Neelie Kroes
Authored by: Anonymous on Saturday, April 19 2008 @ 09:30 AM EDT

But will it matter?

How many Microsoft agents have infiltrated her staff?
Enough to block any meaningful action?

[ Reply to This | # ]

I review 'concensus' meaning...
Authored by: Anonymous on Saturday, April 19 2008 @ 11:22 AM EDT
From Wikipedia:

Consensus has two common meanings. One is a general agreement among the members
of a given group or community, each of which exercises some discretion in
decision making and follow-up action. The other is as a theory and practice of
getting such agreements (for information on the practice of achieving formal
consensus, see Consensus decision-making).

Achieving consensus requires serious treatment of every group member's
considered opinion. Once a decision is made it is important to trust in members'
discretion in follow-up action. In the ideal case, those who wish to take up
some action want to hear those who oppose it, because they count on the fact
that the ensuing debate will improve the consensus. In theory, action without
resolution of considered opposition will be rare and done with attention to
minimize damage to relationships.

If bill or steve sold used cars instead of soft(headed)ware, would you buy from
them?

[ Reply to This | # ]

So it's goodbye for ISO
Authored by: mikeprotts on Saturday, April 19 2008 @ 01:37 PM EDT
Now that ISO has been bought by Microsoft (or at least it seems that way) I
assume that the organisation will be confined to history. This is the fate of
any organisation that does business with Microsoft and hasn't sufficient
independent wealth to ride out the bad times (think IBM & OS/2 here).

So maybe we should forget about the old bribable standards bodies, and go with
what we know is the best method, a free and open standards system. Something
like RFC's?

Cheers
Mike

[ Reply to This | # ]

Boycott Microsoft's ballot stuffers and fixers?
Authored by: Anonymous on Saturday, April 19 2008 @ 08:40 PM EDT
I wonder if the names of Microsoft's ballot stuffers in the various ISO
committees and national boards can be listed, so we can boycott them.

If Microsoft can play the game of rewarding the ballot stuffers, then surely us
consumers and IT specifiers can un-reward them by boycotting them. This seems to
be the only way for us to deal with this kind of corruption of standards
organisations like ISO.

[ Reply to This | # ]

Classification of standards into Open and Proprietary Standards, and different file extensions.
Authored by: Anonymous on Saturday, April 19 2008 @ 09:10 PM EDT
Specifiers assume that mandating an ISO standard ensures that a particular
vendor or monopoly supplier's product is not mandated. With the passing of
OOXML, this is no longer valid.

I believe what is necessary now is to classify standards into two groups:

1) ISO Open standards - vendor neutral, fully documented standards that do not
require payment of royalties or impose any restrictions on use, and that have no
dependencies that are not vendor neutral, not fully documented, require
royalties to be paid, or are subject to restrictions.

2) ISO Proprietary standards - those that are not vendor neutral, require
royalties to be paid, have restrictions on use, not fully documented, or have
dependencies which have any of these.

This will allow users to make an informed decision on standards they specify
compliance with, rather than being misled by ISO approval.

There also needs to be a requirement in the ISO standard as well as one imposed
by anti-trust regulators that a different file extension/mimetype is used for
each of the above and for any ISO standard file format with proprietary
extensions. This is absolutely essential for preventing abuse of monopoly with
regard to standards by tying of files into proprietary tie-in/lock-in, and
passing off non-compliant files as standards compliant for market advantage.

[ Reply to This | # ]

Groklaw © Copyright 2003-2013 Pamela Jones.
All trademarks and copyrights on this page are owned by their respective owners.
Comments are owned by the individual posters.

PJ's articles are licensed under a Creative Commons License. ( Details )