 |
 |
 |
| |
|
|
India Rejects Microsoft's OOXML Format: Reloaded!
 |
|
| |
The BIS Committee has rejected the OOXML format.
|
|
|
|
|
|
Thursday, March 20, 2008:
The BIS LITD 15 Committee has rejected Microsoft's document format OOXML. According to sources, out of 19 members, five of them did not attend the meeting, one of them abstained, five voted in favour of OOXML and the rest voted against. The meeting took place today in Delhi at the BIS office.
|
|
Microsoft has released a statement which says, "While we are disappointed with the decision of the BIS LITD 15 committee, we are very encouraged by the support of IT industry players like NASSCOM, TCS, Wipro and Infosys who voted in favour of Open XML becoming an ISO standard. Further, the Indian government, industry stalwarts and trade bodies have supported multiple standards and technology neutrality. We will therefore continue to work with the government to address any concerns they may have and to achieve its stated goal of technology neutrality. We are committed to working towards what is best for the Indian IT industry."
The statement is contradictory in itself where Microsoft is talking about neutrality. The European Union (EU) has worked really hard to neutralise most of Microsoft's antics. Microsoft's openness seems less out of desire and more out of growing power of FOSS movement and legal pressure by the EU. Neelie Kroes, Competition Commissioner, EU said last month, "Today, the European Commission has imposed a substantial fine, to be precise 899 million ($1.35 billion), on Microsoft for its non-compliance up until 27 October 2007 with its obligations under the Commission's March 2004 decision to provide interoperability information on reasonable terms."
Total penalties against Microsoft now stand tall at around $2.5 billion.
In the statement, Microsoft further said, "The concerns raised by the LITD 15 Committee have been addressed by the ISO and Ecma International (the proposer of the ISO/IEC DIS 29500 Office Open XML) with a majority of the comments getting addressed at the recently concluded Ballot Resolution Meeting (BRM) at Geneva. We hoped that 98.73 per cent of the total 1027 comments from all National Bodies stood resolved at the BRM would be welcomed by the BIS, as it has been by the National Bodies of numerous countries."
This is what the ISO website says about the BRM, "No decision or vote on the document as a whole or any part of it was taken at the meeting, but only on proposed changes to it."
It further says, "Following the BRM, all 87 national member bodies who voted in the original fast-track ballot have 30 days until midnight CET on 29 March 2008 to examine the actions taken in response to the comments and to reconsider their vote if they wish. If the modifications proposed are such that national bodies then wish to withdraw their negative votes, or turn abstentions into positive votes, and the acceptance criteria are then met, the standard may proceed to publication. Otherwise, the proposal will have failed and this fast-track procedure will be terminated. This would not preclude subsequent re-submission under the normal IEC and ISO standards development rules."
India's dismissal of OOXML could be seen as first major blow to Microsoft's ambitions of making OOXML as an ISO standard. It might encourage other countries to follow the suite.
And everyone hopes this time we would not come across the infamous case reported in Sweden where some alleged Microsoft employees bribed to vote in OOXML's favour.
[Note:updated]
Swapnil Bhartiya, EFYTIMES News Network
|
|
| |
|
| |
| |
This is an interesting turn of events. My sources indicate that ISO approval of BRM-OOXML is all but certain, even if that approval is for the wrong reasons. It's good to see India's BIS standing tall, doing the right thing, and insisting that ISO standards fulfill ISO interoperability requirements, the interop demands of international trade agreements, and, the publics expectation of interchangeable document formats.
Allow me to first make some comments outlining my basic position. Then i
will comment in-line to the MS statement. Sadly, it's been some time since i read through the Indian comments. I will say this though; if India did not get a satisfactory response to all 82 of their comments, then they should vote against MS-OOXML approval. And, if India found the comments of other NB's to be of interest, such as the Brazilian requests for a fully
documented semantic mapping of the binaries to MS-OOXML, then that would be
reason enough to vote against approval. Note that i've also attached a
report prepared by the Universal Interoperability Council for the State of
New York.
For me, the most glaring deficiency of MS-OOXML is that it does not meet current ISO interoperability requirements. (Please refer to the attached UIC report for a definitive discussion regarding this important issue). One could spend a lifetime seeking out the minutia of specific deficiencies and inconsistencies, and wonder how it is that MS-OOXML ever made it through ISO fast tracking. Or, one could look skip a life time of research and go directly to the heart of the matter, asking, "Where's the interoperability framework. Where's the compliance clause that will drive interoperability across infinite implementations?"
While there were upwards of 3,000 references to specific interoperability
problems filed on September 2nd, 2007, my approach goes directly to the
specifications most fundamental design. IMHO, there is no need to paw
through some 6,000 pages when all one really has to examine is the
compliance conformance clause to see that there is no interoperability
framework whatsoever.
This is a killer issue. Interoperability is the reason for standards. It is the business of ISO. Lacking an interop framework, MS-OOXML has no business at ISO, and one has to wonder how it is that ISO has wandered so far from their core responsibility? Maybe it's because MSOffice has a 90% or more marketshare, and dominates all government workgroup and workflow business processes? Why else would ISO abandon their responsibilities other than to take up the pragmatic cause of a real world dilemma? I would agree that MS-OOXML is a better document format than MS-binaries. But without an
interoperability framework, there is no reason for ISO to be considering
MS-OOXML as an international standard of document interchange.
Microsoft makes a lot of noise about MS-OOXML interoperability, and has
promised us the moon. But what they offer us is an
application-platform-vendor specific interoperability guaranteed to
transition and lock business processes into an all Microsoft stack, all
Microsoft cloud for years to come. No doubt MS-OOXML interoperability within the Microsoft cloud will be extraordinary. As long as users are running MSOffice-Outlook-IE on the desktop connected to an MS server stack core of Exchange, SharePoint, and MS SQL Server, and, developers are writing to .NET-WPF-Silverlight-OOXML dependent API's, the interop should be extraordinary.
The problem with this vision is that the promised interop is
application-platform-vendor owned and operated. Hardly something ISO should
be considering stamping as an international standard.
So, in the context of an interchange document format specification, what is
an "*interoperability framework*"?
The wikipedia
definitionstates:
"An interoperability framework is defined by all the policies, standards, rules and recommendations a network of actors has adopted in order to reach the highest possible level of interoperability. It also defines the operational rules on analysis, choice, acceptance and update of each of its elements Most European countries have adopted an interoperability framework."
From the Universal Interoperability Councilwe have this
statement concerning an interoperability framework: "What
CDF has and both OpenDocument and Office Open XML lack is: [i] a proven
and fully specified interoperabilityframework;
[ii] developed from scratch as a vendor-neutral framework; and [iii]
interoperability conformity requirementsthat
fully satisfy the ISO/IEC
JTC 1 Directives (PDF)
requirement that:"
"*Standards designed to facilitate interoperability need to specify clearly
and unambiguously the conformity requirements that are essential to achieve
the interoperability."*
The CDF Conformance
clause is
perhaps the gold standard of interoperability framework statements, demonstrating in a very direct manner how interoperability can be guaranteed across multiple and varied implementations. What we see with
the CDF Conformance clause is a
guideline for implementations attempting to balance the
conflicting needs of end user demanded "*interoperability*" and, application vendor "*innovation*".
Both MS-OOXML and ODF allow application vendors and providers to endlessly"
*innovate*" with new features that can be based on proprietary and often
undocumented eXtensions, without worry that these eXtensions will throw them out of compliance. This of course shatters any notion of interoperability.
The Universal Interoperability Council describes this as "application-level
interoperability", stating that "*both ODF and OOXML allow conforming
implementations to use application-specific extensions whose functionality
is not included in those standards themselves. The result are
interoperability break points because the functionality of the extensions is unknown to other developers.*"
Now we're getting down to where the rubber meets the road, the balance
between the networking value of interoperability and the marketplace,
systems vendor driven need for differentiation through innovation.
For me, i chose the First Law of the Internet, which states;
"Interoperability trumps everything. Including innovation." I prefer to see
standards focused on interoperability first, wrapping innovation into the
tight constraints of the interoperability framework. If a vendor wants to
innovate, and still maintain their compliance, the eXtensions should be
fully documented and submitted to the governing standards body for
consideration. If a vendor wants to innovate with undocumented and
proprietary eXtensions, going outside the conformance clause and breaking
interoperability, they should be free to do so. Just don't call it a
conforming implementation.
Is there a conformance test suite for either MS-OOXML or ODF? Of course not.
ZERO interop is guaranteed. Application and stack specific innovation will
bloom, under the cover of ISO approval, but at the harsh expense of the
interoperability the public demands and expects.
And so it goes. ECMA and OASIS are both vendor driven consortia having
somehow gained direct access to ISO through the PAS and Fast-track channels.
Once one group got a standards proposal through lacking any semblance of an
interoperability framework, the door was open for the other group to
similarly push their crap through.
Recently i was having a chat with a long time friend who participated in the recent Geneva BRM. He made an interesting statement that i had never heard before or thought about even though i am among those who have long believed that MS-OOXML is the key to a Microsoft assault on the Internet. He said that W3C HTML presented Microsoft with a very difficult problem. They needed MSOffice to become a web ready editor, but they were not about to channel billion of workgroup business process oriented documents into a W3C format standard. Then along comes ODF! Problem solved. All they needed to do was create an XML version of MSOffice documents in the same way that Sun created an XML version of OpenOffice/StarOffice documents.
In fact, we have this statement from Chairman Bill himself in a 1998 a memo
to the Office product group[2]:
"One thing we have got to change in our strategy - allowing Office documents to be rendered very well by other peoples browsers is one of the most destructive things we could do to the company. We have to stop putting any effort into this and make sure that Office documents very well depends on PROPRIETARY IE capabilities."
One can always point to the limited simplicity of HTML as the rational for
moving to an application specific XML encoding of the MSOffice binary
documents. But this innovation heavy argument flies in the face of the
interoperability focused alternative of going to the W3C and working on
improving and enhancing the HTML XHTML CSS family of web formats. Which
is exactly what Chairman Bill was determined not to do.
So here we are today, with ISO considering the approval of an
application-platform-vendor specific format totally lacking in any kind of
interoperability framework, and already demonstrating the innovative
possibilities enabled by a conformance clause that allows for unlimited and
undocumented proprietary eXtensions.
But that's only one part of the story.
The other part can be seen two other developments; the December 2007 beta
release of the MSOffice SDK, and, the recent IE 8.0 browser release said to
comply with the ACID II W3C open standards interoperability test.
Keeping in mind the above 1998 quote from Chairman Bill, let's look at IE-8
first. What we see is that IE-8 passed the ACID II test using bits of HTML
5.0 and CSS 2.1. Great. This demonstrates a significant improvement in IE
compliance with W3C standards. But note well what's missing! IE-8 does not
support the XHTML mime type. Nor does it support advancing ECMAScript
JavaScript standards.
What we have here is a rather serious interoperability break set exactly
where the legacy of simplified HTML documents transitions to the highly
structured, complex and extremely rich documents possible with XHTML 2.0
CSS 3.0, JavaScript, XForms and SVG.
Okay. So we have what appears to be this gapping hole in the IE web future.
Let's now look at the MSOffice SDK beta. Here we find an interesting
conversion component making short work of converting MSOffice OOXML to XAML.
XAML is a WPF (Windows Presentation Foundation) technology capable of
expressing two format variations known as fixed and flow. Fixed is also
known as XPS, and would be useful as a Microsoft specific alternative to
Adobe's PDF. Flow is something else. Where fixed is a static representation
of OOXML documents, flow is an interactive, highly fluid representation of
the same. I see flow as a Microsoft specific alternative to the W3C's XHTML
CDF branch of advanced web formats.
With the MS-OOXML <> XAML converter, Microsoft has their rich but
proprietary web ready document format, which IE-8 supports through the
Silverlight plug-in. And, they've figured out how to limit the richness of
competing web ready systems by limiting IE-8 to HTML 5.0 CSS 2.1. Very
clever, but as serious a breach of Internet interoperability as we will
likely ever see. The salient point is that MSOffice dominates the world of
business documents and business process workflows. Microsoft was not about
to transition those desktop bound processes to the web unless and until they had figured out how to control the transition through proprietary formats bound to Microsoft specific applications and API's.
ISO approval of MS-OOXML provides a cover of legitimacy to this certain
breaking of the web. Approval establishes MSOffice as a standards compliant
editor for the Microsoft specific cloud. The only thing Microsoft might
concern themselves with if there is approval, is that ISO JTC-S1 doesn't
insert an interoperability framework demanding compliance based on the
implementation transparency of fully documented eXtensions.
Microsoft continues to insist that conversion of the MSOffice binaries
remain application specific. Brazil was right in demanding that the mapping
of the binaries to MS-OOXML be provided to ISO. If Microsoft wants to argue
that the primary purpose of having MS-OOXML standardized is the high
fidelity conversion of legacy binary documents, then this all important
mapping should be fully disclosed and documented. More importantly though is the marketplace significance of Microsoft's continual control of the
conversion process. The MSOffice Compatibility Pack converts binary
documents to MS-OOXML. The MSOffice SDK provides conversion of MS-OOXML to
XAML. The proprietary XAML-Silverlight combination knocks out the W3C,
Adobe, Firefox, and Apache technologies.
Further comments at:
-
"Standards by Corporation"
and "Microsoft's
OOXML limps through ISO meeting"
-
"OOXML: insufficient and
unnecessary!"
discussion log
COMMENTS ON MS PRESS RELEASE:
New Delhi, March 20, 2008: While we are disappointed with the decision
of the BIS LITD 15 committee, we are very encouraged by the support of
IT industry players like NASSCOM, TCS, Wipro and Infosys who voted in
favor of Open XML becoming an ISO standard. Further, the Indian
government, industry stalwarts and trade bodies have supported multiple
standards and technology neutrality. We will therefore continue to work
with the government to address any concerns they may have; and to
achieve its stated goal of technology neutrality. We are committed to
working towards what is best for the Indian IT industry.
Speaking of technical neutrality and multiple standards, why not require
MSOffice to fully support (X)HTML 5.0, CSS 3.0, SVG, and XForms, and do so
within the compliance requirements of the W3C's CDF. I would also suggest
using the .epub container instead of OPS. These are
application-platform-vendor independent formats important to the future of
the Web.
Open XML encourages choice, flexibility and interoperability for IT
industry and IT users - a fact underlined by leading Indian IT solution
providers increasingly using the format as a preferred technology
standard for applications addressing both the Indian and global
customers.
Leading Indian IT solution must use MS-OOXML if they are to fully integrate their solutions into the market dominating MSOffice desktop. Overall, the MSOffice desktop enjoys a 92% monopolist marketshare. Within businesses, the marketshare is much higher, leaving solution providers with little choice but to work through whatever interop channels Microsoft provides. How this encourages choice and flexibility is beyond me.
The interoperability claim is totally disingenuous. Because MS-OOXML lacks
an interoperability framework, we are left with application defined and
directed interoperability. That might be fine for solution providers working the MSOffice MS Server Stack MS Cloud space. But it falls far short of the universal interoperability the public demands and expects from open standards. Is it possible to create and exchange MS-OOXML documents without proprietary Microsoft components?
The concerns raised by the LITD 15 Committee have been addressed by the
ISO and Ecma International (the proposer of the ISO/IEC DIS 29500 Office
Open XML) with a majority of the comments getting addressed at the
recently concluded Ballot Resolution Meeting (BRM) at Geneva. We hoped
that 98.73% of the total 1027 comments from all National Bodies stood
resolved at the BRM would be welcomed by the BIS, as it has been by the
National Bodies of numerous countries.
No doubt the world will applaud and appreciate every effort by Microsoft to bring their technologies in line with the principles of universal
interoperability. Years of foot dragging and interop breaking *embrace and
extend* is killing everyone but the smiling Microsoft shareholders. So yeah, even the smallest steps towards interoperability are appreciated. But it doesn't mean we should compromise the meaning and purpose of ISO standards. The ISO interoperability requirements must be respected, and MS-OOXML brought into full compliance before approval as an international standard is granted.
Open XML was approved by Ecma International in December 2006, and then
submitted by them for standardization at the ISO. At the conclusion of
the Ballot process, the BIS voted a 'Disapprove with 82 Comments' in
September 2007 for Open XML. Between September 2007 and January 2008,
Ecma International worked on and submitted proposed dispositions to all
the comments, all of which were discussed during the Ballot Resolution
Meeting last month; and 1011 of them adopted by consensus. The ISO will
take a decision after March 29, 2008, the last date for receiving votes
from all member nations.
Did Microsoft respond to all 82 BIS disapproval comments? Sounds to me that if they did respond, it was not to the satisfaction of BIS. Microsoft's arguments that they did respond to the comments of other NB's is not the same thing as resolving all the problems with MS-OOXML. In particular, Microsoft totally blew off Brazil's request that the mapping of MS binaries to MS-OOXML be provided. Without that mapping i don't see anyway of ever separating MS-OOXML from the sprawling cascade of dependencies on Microsoft specific technologies. No doubt that this time next year NB's will be asking for the mapping of MS-OOXML to XAML.
Hope this helps, and thanks for the consideration,
~ge~
--
Gary Edwards
OpenStack Business Systems
Redwood City, CA USA 94063 |
| |
|
| - Gary Edwards, Redwood City, 3/21/2008 3:30:30 PM |
|
|
|
|
|
| |
| This seems to be the immediate coverage, based on reports. They have now updated the news! Lets appreciate them for being fast enough and not crib for everything! |
| |
|
| - Rakshit_LeeNux, Mumbai, 3/21/2008 8:57:09 AM |
|
|
|
|
|
| |
It is interesting to note that while the rest of the committee (academics, industry and government) has voted NO, the software export houses are the only ones who have voted in favor of OOXML despite numerous technical issues that have been pointed out. Some of the software exporters who have a holier-than-thou image should take a look at the serious governance issues with OOXML pointed out at www.noooxml.org. Have they voted against the national interest and purely in their financial interests?
You quote the Microsoft press release as saying, "The concerns raised by the LITD 15 Committee have been addressed by the ISO and Ecma International (the proposer of the ISO/IEC DIS 29500 Office Open XML) with a majority of the comments getting addressed at the recently concluded Ballot Resolution Meeting (BRM) at Geneva. " Now, is that for Microsoft to say or for the committee (of which Microsoft is a voting member) to say?" The committee is clearly not happy with OOXML as it stands today or it would have voted in favor of the proposal.
They also say, "We hoped that 98.73 per cent of the total 1027 comments from all National Bodies stood resolved at the BRM would be welcomed by the BIS." Microsoft can say whatever they want but EFY should at least have done some fact checking before carrying their press release as it is and pointed out how controversial the BRM was. This is basic information that is just a search away. |
| |
|
| - Raj Mathur, Delhi, 3/20/2008 7:20:27 PM |
|
|
|
|
|
| |
|
|
|
| |
| |
| |
|
|
 |
 |
 |
 |
|
|
|
|
|
|
|
|
FREE NEWSLETTER
Receive the latest reviews, how-to's, news & more.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Technology: The Match Winner! |
| Intencity, a Pune-based exclusive video game store, has expanded its business and boosted revenue by 40 per cent, using online store solutions.... |
|
|
| |
|
|
|
|
|
|
|
|
|
|