Open Malaysia

When Patrick Durusau wrote that Microsoft had no opposition to ODF, he was indefensibly and gloriously wrong. Those of us involved in national standards bodies and on government policy panels have had the dubious honor of having to defend against Microsoft's desperate attempts to kill any possible introduction of ODF as a voluntary standard. Perhaps even worse was having to witness first hand Microsoft making loud complaints against any internal government policies of the usage of ODF and any software which uses ODF (OpenOffice.org being the most prominent). And, of course, many of us have faced the intense lobbying campaign carried out by Microsoft against open source governmental policies.

So when Yasmin Mahmood, Microsoft Malaysia Managing Director, made the famous statement that "It's not about choosing, but about having a choice" with reference to ODF and OOXML, we assumed this was a new chapter in Microsoft history.

Then came along a recent blogpost by Harisfazillah Jamel, "Apabila Membuat Pilihan Dipersoalkan / When Choices are Questioned", on the right of government agencies to make choices. It's written in Malay, but it has been translated by Yoon Kit for the benefit of our international readers. It makes for extremely interesting reading on the pressure Microsoft Malaysia is placing on a government agency because of their choice to use OpenOffice.org/ODF over Microsoft Office:

Apabila Membuat Pilihan Dipersoalkan (When Our Choices Are Questioned)

Ada yang sebuah syarikat ICT yang kononnya mempunyai satu lagi standard bagi satu perkara adalah satu pilihan kepada pengguna. Sedangkan pengguna, mana-mana pengguna juga faham, standard, mempunyai standard bermaksud kita telah membuat pilihan, membuat satu pilihan daripada pelbagai pilihan yang terbaik untuk pengguna.

There is an ICT company which has one more standard for one more application for one more choice for consumers. While consumers, many users understand standards, having standards means we have made a choice, a choice which  is the best for users from many options.

Wakil mereka di Malaysia dalam satu muka surat sebuah akhbar menyatakan, kejayaan mereka mendapatkan iktiraf standard bagi satu format adalah satu pilihan kepada pengguna. Teruskan membaca, apabila membuat pilihan bukan kehendak mereka, adalah bukan pilihan mereka.

Their Malaysian representative, in a page of a newspaper stated that their success of having the standard as a format is yet another choice for users. Read it, if we make a choice not to their liking, it is not their choice.

Lalu sebuah agensi dalam kerajaan Malaysia telah membuat pilihan. Membuat pilihan bukan dalam kehendak syarikat ICT tersebut. Pilihan setelah panjang membuat penilaian. Pilihan atas dasar untuk memberi sokongan dan pilihan kepada agensi-agensi lain. Pilihan yang dibuat atas kehendak sendiri tanpa  tekanan mana-mana pihak. Pilihan yang dipesetujui diperingkat pengurusan tertinggi hinggalah keperingkat bawahan.

An agency in the government of Malaysia had already made a choice. The choice was not the choice of the said ICT company. The choice was made after a long study. The choice was based on principles to bring support and choice to other agencies The choice was made on its own needs without any pressures from any party. The choice was agreed by the highest level of management to the lowest level.

Sekarang timbul isu, syarikat ICT itu tidak senang dengan pilihan yang dibuat oleh agensi itu. Mereka merasakan agensi ini mahu menentukan dasar yang keras terhadap pilihan yang telah dibuat. Oh ya betul, mereka tidak senang dengan pilihan yang dibuat oleh sebuah agensi kerajaan Malaysia dan mula menjalankan kerja-kerja melobi untuk memaksa agensi ini, menurunkan maksud polisi penggunaan standard yang telah dibuat oleh agensi ini.

Now this issue has surfaced. The ICT company is not happy with the choice that was made by the agency. They felt that the agency wants to set a policy which is incompatible with the choice that they prefer. Oh that is true, they are not happy with the choice that was made by an agency of the Malaysian Government and have started work lobbying to force this agency to dampen the meaning of the policy of using standards that was announced by this agency.

Polisi yang telah disusun, dibincang dengan panjang lebar dan diluluskan dalam pengurusan tertinggi agensi ini mahu diubah oleh syarikat ICT tersebut kerana ia tidak menepati pilihan yang kehendak mereka. Bahawa diingatkan polisi ini adalah khusus hanya untuk agensi ini supaya memudahkan pelaksanaan atas satu standard. Itu maksud tujuan standard, memudahkan pelaksanaan kepada satu pilihan.

The policy that was prepared, was discussed widely and in detail, was approved by the highest level management of this agency, is to be modified by the said ICT company because it is not inline with their needs. Whereas it is reminded that this policy is specifically for this agency such that it  can easily implement one standard. That is the meaning and purpose of a standard, the ease of implementing one choice.

Yang buat saya marah, adalah teknik yang saya panggil, kilas tangan dengan kasar memaksa agensi ini tukarkan polisi agensi ini. Ya memaksa, walau apa pun cara yang digunakan, bagi saya ia masih panggil ia memaksa, agensi ini tukarkan polisi dari segi pelaksanaan.

What makes me angry is  what I call their heavy handed technique of forcing the said agency to change its policies. Yes force, by any means possible. I still call it force, to get the agency to change its policies on implementation.

Bagi saya, campur tangan dalam urusan sebuah agensi kerajaan Malaysia terutama melibatkan polisi, dan nyata sekali polisi ini adalah untuk penggunaan agensi tersebut oleh sebuah syarikat luar negara merupa satu campur tangan asing... Sebuah bentuk penjajahan ...

For me, their intervention in the business of an agency of the Government of Malaysia especially when it involves policies, and especially when this policy is stated for the use of this agency, by a foreign company represents ... a form of colonisation ...

Siapa mereka untuk tentukan dasar dan pentadbiran sebuah negara.  Saya sebagai rakyat Malaysia cukup marah dengan campur tangan sebegini. Perkara ini telah lama diperhatikan sejak daripada melobi penerimaan standard yang kononnya adalah satu pilihan.

Who are they to define the policies and the running of a sovereign country. I, as a Malaysian citizen, have enough with their interference of this nature. This behaviour has long been witnessed ever since the lobbying efforts of the standards which is claimed to be yet another "choice".

Saya tidak akan duduk diam dengan bentuk penjajahan cara baru ini. Saya tidak akan berhenti menulis dan berkempen untuk mendedahkan apa jua yang dilakukan oleh syarikat ICT ini untuk terus memaksa penggunaan produk mereka.

I am not going to sit quiet with this new form of colonisation tactic. I am not going to stop writing and campaigning and resist any action by this ICT company which forces the use of their products.

Banyak sudah duit rakyat, duit rakyat mengalir keluar daripada Malaysia, hanya untuk pembayaran lesen. Pembayaran hanya atas kertas, tiada hasil kepada kita rakyat Malaysia, hanya menjadikan syarikat luar negara atau individu luar negara lagi kaya.

Too much of our citizens money, money of the people has flowed out from Malaysia, only to purchase licenses. The payment is only for paper, without any value to us Malaysian citizens, only to make a foreign company or individual foreigners rich.

Sedangkan duit yang berjuta-juta itu boleh dialirkan kepada industri ICT negara kita dalam bentuk pelaksanaan sistem sokongan, penyelidikan dan pelaksanaan kepada aplikasi sumber terbuka atau Open Source Software.

If only the multi-millions of riggit could flow to our ICT industry in the form of system support, research and implementations in Open Source Software.

Mengapa perlu kayakan orang lain sedangkan rakyat ini yang membayar cukai tersebut perlu lihat sahaja duit itu keluarkan. Sedang sudah ada pilihan, pilihan yang setelah dikaji, boleh mendatangkan faedah yang banyak kepada rakyat Malaysia.

Why must we make other people wealthy while citizens have to pay the tax. Just see the expenditure. While there is already a choice, a choice that has been evaluated which can bring much benefits to the citizens of Malaysia.

Polisi dan dasar sudah ada. Mengapa tidak sokong? Soalan yang hendak tanya kepada ahli-ahli politik negara kita ...

The Policies are already defined. Why is there no support? These are the questions which need only be asked to the politicians of our country ...

Apa yang saya tulis ini adalah pendapat peribadi, akan tetapi saya mahu ia dipandang  serius oleh banyak pihak. Lama mana kita mahu dijajah ... ???

What I have written is just my personal opinion, but I want this viewed seriously by many. How long do we ever want to be colonised ... ???

What?

Yes. Try to understand that fantastic quote from Yasmin Mahmood, Microsoft Malaysia's (current) Managing Director. This was reported by Tech&U, which is currently leading the pack as the most reliable source of Microsoft propaganda. What's wonderful is that they quote Yasmin word for word without questioning what she really means. It makes hilarious reading.

I have transcribed the article below, as it is not available online yet.

What's interesting is that finally, we have documented proof that Microsoft is not interested in real choice for customers, but merely an illusion of choice. This is straight from the horse's mouth, as they say:

"The industry just wants to have the best innovation; they want to have the freedom of choice. The whole idea is not about choosing, it's about having a choice ... and that is what customers and partners want," she said.

It's about having the freedom of choice. Not really about making a firm choice, but at least we have some choice. I could go on about how fallacious this argument is but I think you readers are intelligent enough to know how this is so wrong.

But if Microsoft really has the interests of "choice" for their customers, why then don't they really, walk the talk, and provide choice of ODF in their products? Why are they restricting my choice as an end user who prefers the better ISO standard? Do they really practice what they preach? Why is there less choice in their products?

Why do they complain so loudly when their 'formats' are not considered and yet are so blatant about leaving out competing formats in their products? Why is Novell so much more capable in building OOXML support in OpenOffice.org and Microsoft, a vastly larger company with teams of developers having such difficulty writing ODF support in their products?

What really is rich about her comment about the astroturfing Microsoft Malaysia had seeded to create the illusion of OOXML adoption in Malaysia:

... some strategic projects in Malaysia have successfully leveraged on the format. Among them are the Malaysian Halal Hub Open XML System, RosettaNet Automated Enablement, Tradenex.com and World Congress on Information Technology 2008's Registration System.

I have already elaborated on how Microsoft funded the development of OOXML integration on "strategic projects" above (Halal Hub and RosettaNet). What is new however is the WCIT'08 Registration System. Who needs OOXML to register for a seminar? Don't we just need HTML/HTTP to do online registrations?

It's laughable that Microsoft Malaysia is trying to tag on OOXML to everything they sponsor.

Speaking of which, if you visit the WCIT website, at the bottom, you can see these logos:

It does not take a statistician to find a correlation between voting patterns on OOXML in Malaysia and "smart partnerships" with this WCIT circus. MDeC and PIKOM voted "Approve". MoSTI (Ministry of Science, Technology and Innovation) voted "Abstain" and its Minister ultimately overrulled the overwhelming 81% Disapproval position by TC4 and ISC-G, the technical committees overseeing the OOXML draft standard.

What has to be said is that the Microsoft Malaysia propaganda machine has to go into overdrive now because at the end of the day, while Malaysia voted "Approve" to ODF, in all intents and purposes, Malaysia strongly rejected OOXML in all sectors of the ICT ecosystem. At best, if spun properly, Microsoft can say that Malaysia is non-committal to OOXML as it merely "Abstained".

I wonder what other strategic projects Microsoft Malaysia can sponsor next? To them, the money train just left the first station and must keep going! We just have to sit back, and enjoy the spin. Eventually they will find that they cannot fund everything and buy support all the time.

yk

I don't know what the attraction is, but somehow we all love the morbid fascination of Zombies in action. First, Microsoft^H^H^H^H^H CompTIA hires Mr. Jan van der Beld, Ex-Ecma Secretary General, to fly all the way here to KL, for an event supposedly about "good multiple standards".  There he challenges us to find a better way to Fast Track large, immature vendor dependent specifications. The answer is of course: "Don't do it." Later on that same day, like a man possessed, he turns up at a PIKOM meeting only to rant and thump tables.

Then yesterday, our fantastic broadsheet turned tabloid "The New Straits Times" featured a "Comment" by our so called "cooler head" Datuk Dr Mohd Ariffin Aton entitled "Walking the Talk on neutrality policy". If you've forgotten about him, you may be forgiven, but he is or rather WAS the CEO if SIRIM Bhd. Yes, another Ex-Somebody. He was the one with the inane explanations on why TC4 was shut down last year:

"There has been unprofessional conduct and a lack of ethical standards among some members of the technical committee," Ariffin said. "This is the first time in my 11 years at Sirim where ethics have not been followed." He, however, declined to name the individual members involved, citing his own ethical reasons. Ariffin said some TC/G/4 members had taken to belittling other members who did not share their pro-ODF views, both during committee meetings and in personal blogs.

First he said that he did not want to name 'em, because of his ethics, then immediately indicates who they are. Isn't it bizarre that a CEO would shut down his own TC without telling the TC what was wrong in the first place, and then going to the press to talk about it? How shameful is it that you can't keep your own house in order?

So yesterday, on Page 25 in the Malaysian Daily, he writes some amazing comments, which would make any Microsoft shill orgasm with delight. For you 'cold-heads' who are actually using your brain, please wear some protective gear around your gray matter, to prevent your head from from exploding!

Here goes:

So, after Doug voluntarily excused himself from TC4 [as we did not 'throw him out' as he claims], and after I gave my presentation on the BRM, and after Ditesh gave his technical evaluation of the resolutions, the TC4 Chair opened the meeting to the floor. Mr Cheong Yuk Wai, from IASA Malaysia, volunteered. We expected him to delve into the technical matters which Microsoft should have braindumped him.

Instead, we got a rant.

He started off saying that back in the old days of e-commerce, he was in discussion with the board of major banks trying to work out the best way to do internet transactions. The Secure Electronic Transaction (SET) payment method was decided to be the best method, which protects both Banks and Merchants. It failed miserably. He didn''t elaborate why, but I'll answer that later.

The issue, was that SET was designed by people in the boardroom without the proper input of the people that mattered; the Customers. I tried to find a link to our discussion here, but the next point was that "That is why we shouldn't just have one standard as we may have made a wrong choice" ... in reference to ISO 26300 (ODF) I'd assume. He did after all have the same preamble last year when he tried to present the case for TC4 not to send contradiction comments to ISO. After a year and a bit of think about this, I still do not see the link. Ho hum.

Anyway, after that, he demonstrated his Google skillz. Or so we thought. In a snazzy pre-recorded vidcast, Google was brought up, and searched for "patrick odf". The first entry of course was Patrick Durusau's plead to adopt OOXML as an ISO standard in his open letter entitled "The Importance On Being Heard". Not sure why a simple URL to the file wouldn't have sufficed.

Now Mr Cheong said that Mr Patrick Durusau was the ODF Author, Ditesh interjected and said that that was not accurate. He is the ODF Editor. There is a big difference being and author and an editor. I also wanted to interject to say that Patrick was making his remarks in a personal capacity, and his official capacity should not have been brought into discussion, nor relevant. But Mr Intel, who was suppose to remain an 'Observer' decided to play the role of the Chair and asked me to remain silent to give Mr Cheong a chance to speak. I let that slide.

Mr Cheong then went on to highlight some sections of the open letter:

"That point of agreement is that everyone at the table was heard. That may not seem like a lot to an Oracle or IBM, but name the last time Microsoft was listening to everyone in a public and international forum? At a table where a standard for a future product was being debated by non-Microsoft groups?"

With this Mr Cheong said, and I have to paraphrase, because I too am trying to grasp the logic: "See, here is our chance as Malaysia to be AT the table! So that we can contribute to the development of this format!"

[I digress here, but I think this paragraph is worth mentioning:

"... name the last time Microsoft was listening to everyone in a public and international forum? At a table where a standard for a future product was being debated by non-Microsoft groups?"

I love this subtle trick which Mr Durusau managed to get ALL the Microsofties to openly admit that Microsoft has the bad reputation of NEVER listening to the public in the development of their products. I mean All of them, in their zeal to reprint the Durusau letter, had to include this paragraph in their blogs and Press Releases, and it just made them admit that they have been all the time customer insensitive. Tee Hee]

I'm tired. Really, I'm honestly physically exhausted. Working a full time job, moonlighting on improving the DIS 29500 (Microsoft OOXML) specification, buying a house, selling a car, seeing the apartment building go up in flames[1], hosting two Couch Surfers, dealing with an engagement (don't get excited, it's not mine), taking this really hawt and intelligent woman[2] out and wanting to spend time with her but simply not having time to spend, takes its toll, you see.

So I'm going to keep my response to Doug's recent provocative spin-doctoring and fact-fudging [3] to a numeric series of objective observations and subjective argumentation. Make what you will of it, I personally think Doug needs to spend more time in constructive efforts (perhaps improving the specification?), and less time building elaborate conspiracy theories and flaming his "buddies" (his words, not mine).

Here goes:

1. Doug, you are claiming that you wanted a technical debate with YK and yours truly. You never contacted us prior to the PIKOM meeting and never told me that you would be present at TC4. How can you claim that you wanted a technical debate when you never bothered to get in touch with the people you wanted to debate with? Your logic simply does not compute[4].
2. Your blog post says "they threw me out before the meeting started". Well, no shit sherlock. If you turn up at a meeting without giving due notice of your nomination as an alternate representative of IASA, you really can't be surprised if they weren't too happy, eh? There are proper processes to follow in Malaysia's standards body and IASA flagrantly violated them.
3. Microsoft and IBM were specifically not invited. Doug, you are a high profile Microsoft employee and you circumvented this ruling by turning up as a Malaysian Vice President of IASA. It just doesn't sound to me that you attempted to participate in good faith basis, but rather tried to twist the rules to your favor. If you got in, why shouldn't Google, Oracle, RedHat, IBM, Sun etc all get in as well? The rules are there for a reason, you know. You have to accept that this is a Malaysian technical body, and we want the Malaysian agenda to be discussed, not the agenda of multinationals.
4. I arrived 10 minutes late, not 30 minutes late. Wrong turning on the Federal Highway, you see. Nothing to get worked up over. I have no idea when YK arrived.
5. Hitting out at Madam Tan is a new low, certainly not something I would have expected out of you[5]. I'm going to echo YK's comment and suggest that you stop burning your bridges.
6. Kamarul was invited specifically as an observer by SIRIM in the past. He did not turn up uninvited. Kamarul actually got his facts wrong on this point and was corrected by the secretariat immediately during the meeting. Kamarul acceded to his mistake during the meeting. You may wish to update your blog post accordingly
7. DSM wasn't making up the rules as they went along. Just so you know, I had to get MNCC to formally submit my status as an alternate before they would allow my participation. I'm not sure how you do it in the US but in Malaysia, there are proper processes to follow and IASA shouldn't be an exception.
8. You seem keen in arguing that there was no technical material discussed and presented. Au contra ire[6], I presented Malaysia's dispositions, output from the BRM and gave my technical opinion on the issues. You see, I actually write code for a living, I know what I'm talking about and am willing to defend it too[7]. Mr. Cheong (from the infamous IASA) did not raise a single issue with my interpretation of the technical deficiencies at hand and instead started discussing non-issues ("if Malaysia votes NO, Microsoft will never open anything up ever again"[8]). In fact, Mr. Cheong specifically declined to discuss the technical issues. I see hypocrisy in you blogging that there was no technical discussion when there was indeed such an opportunity present and your organization (or more correctly, Microsoft's front) chose not to engage in it.
9. I think Hasan was in SIRIM to hit on a secretary. Female, and from what I hear, quite the MILF-ish. Really, you should have stayed, if not for seconds, then for the entertainment.

There you go. I'm not going to pursue this any further (but Doug, do feel free to write another scathing post, I've got /dev/null all prepped up and ready here). I just felt folk reading Doug's blog and OpenMalaysiaBlog should be able to see this from another perspective. Specifically mine. Either that or this is just another excuse for more sandiwara (translation: drama).

You know, we Malaysians just love our sandiwara, especially since the election-fever recently ended and no new gossip seems to be making its rounds. Oh well, lets hope Doug keeps us entertained for another week or so (planning to infiltrate other National Bodies? I hear the Singaporeans are bored).

[1] Well not really, but it sure did feel that way.
[2] Mmmmph!
[3] If this a job requirement at Microsoft? That may explain why I never got called in for an interview.
[4] Which is really not a bad thing. It busts the myth that Vulcans inhabit Redmond. I, for one, am f*&#ing glad. I must admit you guys scare me at times with the paranoia but as long as it's not backed up by sheer cold and calculating logic, I guess we're cool.
[5] Well, maybe I expected the stalker to do the dirty hatchet job, but definitely not you Doug.
[6] Slpelnig may be fubared, I couldn't give a shit. Read the first line of this blog post again.
[7] This is really why chicks dig me. That and the 200 dollars I'm willing to part with.
[8] I must admit that this is my favouritest bad argument, ever. Partly because it was quite unexpected (Mr. Cheong does strike me as a fairly intelligent chap) but mostly because the argument is so mindbogglingly stupid.

In a rather ill formed press release published in a Malaysian daily on the 26th of August 2007, the New Straits Times (New Sunday Times edition) carried this story "Ministry still undecided on Open XML". Beside the obvious factual blunders in the story, the subtext is rather interesting.

But first, let us get the facts straight.

In the article it states that

"The Open XML document software has been recognised by the private sector and endorsed by Prime Minister Datuk Seri Abdullah Ahmad Badawi."

Open XML is NOT supposed to be a "document software". However since the latest Microsoft Office 2007 is the ONLY "document software" available to produce and consume "Open XML" documents, the reporter is by accident, correct about "Open XML". That is why it should always be referred to as  "Microsoft Office Open XML" or MSOOXML.

MSOOXML is NOT recognised by the private sector. FMM (Federation of Malaysian Manufacturers), MNCC (Malaysian National Computer Confederation), MDeC (Multimedia Development Corporation) and PIKOM (Association of the Computer and Multimedia Industry of Malaysia) have voted against MSOOXML. There is not support for MSOOXML in the private sector except the support from Microsoft Business Partners, who obviously have a few licenses of MsOffice 2007 to flog off.

About the endorsement by Malaysia's own Prime Minister; that's just stretching the truth. Here's more information from the article:

"The Open XML system rose to prominence in May when Microsoft Malaysia, in partnership with the Halal Industry Development Corporation (HDC), undertook to develop a system for producers of halal goods and services to carry out their activities on the Internet.
Abdullah had witnessed the exchanging of documents for the project between HDC chief executive officer Datuk Jamal Bidin and Microsoft Malaysia managing director Yasmin Mahmood."

HDC signed a Memorandum of Understanding with Microsoft with Microsoft's intention to develop a portal system for HDC. Whether HDC was to fund for the development, we do not know, but considering that the name of the system which was labeled "Malaysian International Halal Hub Open XML System" does not seem to originate from HDC.

Additionally, the MoU was all about "defining the scope of the project and method of implementation before deciding on these matters." This begs the question: why decide on a system which "will be based on the Open XML document standard" BEFORE the scope and method of implementation was decided?

Basing a modern B2B e-commerce transactional application on a File Based back end is just so 1980's! It is like running e-bay on DBase ][!

So the name was inserted into the MoU by the Microsoft Marketing people, to gain "face value" of MSOOXML with our Prime Minister, and leverage opportunities such as this. Also to impress upon weak minded people.

Which do you think is more likely:

1. HDC signs up with Microsoft because they want their portal system to be based on an immature, file-based and vendor restricted File Format in MSOOXML, or
2. HDC signs up with Microsoft because Microsoft is willing to fund the development of this portal system in exchange for Marketing value?

Cynical or just plain practical?

Technically, MSOOXML is a poor choice for an Islamic Halal Hub. The NetWorkDays function in MSOOXML calculates the number of work days within a certain date period. Secular countries like Malaysia will not have a problem with this (except for states like Kelantan), but some Islamic countries base their "weekends" around Friday, their day of prayer, and their calculations will be inaccurate.

Their weekends are not "Saturday and Sunday" as the American creators of MSOOXML experience. Users in Tel-Aviv might have Friday and Saturday off, while in Cairo users have Thursday and Friday as their weekend.

MSOOXML is a very Western defined file format, which does not take into consideration the cultural adaptability of its users. It ignores the sensitivities of a billion Muslim users worldwide.

So what does this say of the "Malaysian International Halal Hub Open XML System" when it forces its users to use the American weekend system? How will users of different cultures interoperate?

Here's another sentence which needs elaborating:

"The Open XML system rose to prominence in May ..."

Actually it rose in "prominence" in February when Malaysia voted it down at the ISO contradiction period. It then rose in infamy in March when Malaysia's and 14 other National Bodies' concerns were not addressed by Ecma. Subsequent manipulation of National Bodies worldwide has exposed countries with weak standards governance. The ISO process is being put through a lot of pressure, and the results in the next few weeks will prove if it can weather a crisis of vendor manipulation.

Attack on Malaysia's Ministries
The fact that the article has some basic mistakes (for example the issue of awarding ISO 9001 status to Microsoft, instead of the Approval of ISO/IEC DIS 29500 by Malaysia) shows that the journalist is not an expert in this matter. Additionally it's not like a journalist to report on esoteric matters such as standardisations of a particular vendor's product. The press release must have come from a source who has the interest to get this matter in National News to apply pressure to certain parties to get a desired outcome.

It is not like the New Straits Times to be critical of the Government. So it is extremely abnormal to read sentences like this:

"Despite this, the Science, Technology and Innovation Ministry appears to be dragging its feet over an application to support the award of ISO 9001:2000 standard for the Open XML."

The newspaper is actually criticising the Ministry of Science Technology and Innovation (MoSTI) and directly, the Minister himself, Y.B. Dato' Seri Jamaludin Jarjis for non-performance. Not only that, the insinuation that the PM already endorses MSOOXML is a double blow to the reputation of the Minister in-charge.

Microsoft appears to be applying public pressure on this matter even though this issue is not being delayed. NB's have another week before submitting the vote. Malaysia's ISC-G (equivalent to America's INCITS) will be deliberating this week as pre-scheduled. The Technical Committee for E-Commerce (TC4: equivalent to America's INCITS-V1) had already voted a 9:1 vote against MSOOXML back in February.

TC4 was not called to make a subsequent vote this time, although since the spec has not changed, and the issues not resolved, the vote should be the same. This could make the work at ISC G (Industry Standards Committee G, the committee above TC4) more difficult. So, to MoSTI's defence, there is no "dragging of feet".

This pressure probably comes from the failure of last two weeks' peculiarities in Malaysia. It looks like Microsoft has exhausted the utility of its strategic partners and in its desperation to achieve its KPI's, they are now going all out and apparently burning bridges in the process. As many have said, the standardisation of MSOOXML is absolutely crucial for Microsoft and they are willing to achieve it at all costs!

Their worldwide campaign has been brutal with many Experts, Committee Members, National Bodies, Ministries, Ministers and Countries' reputations tarnished.

yk.

p.s. Yusseri has a lighthearted take on this issue too! "How low can you go?"

In today's online edition of the Daily Telegraph (UK), there is a headline and leader in the sports section which says:

'Miracle' of Istanbul

When Apple released iWork '08, the Microsoft bloggers immediately jumped in to comment on how this is proof that MSOOXML is easily implementable by third parties. First of all, we need to realise that Apple sits in the Technical Committee at Ecma which "developed" this so called standard. So we should not be surprised at all if they have support of this file format in their upcoming products including the iPhone.

What is not said however, and I have been waiting for days for a response from Microsoft's Stephen McGibbon, is why Apple does NOT really support MSOOXML.

What? iWork '08 DOESN'T support MSOOXML?!

Yes. If you have a careful look at the examples the Microsofties provide, they just demonstrate read only features of iWork. iWork does not WRITE to MSOOXML.  As a fellow member of the technical committee which actually developed the spec, you would have expected some form of read / write ability to perhaps 70% compliance, to complete the circle of "interoperability," wouldn't you?

Unless of course it is too hard to do?

Maybe Apple would rather spend time creating great software instead of implementing a backwards looking file format which its "partner" is steamrolling through the standardisation process? Maybe it has taken advice from Microsoft's own Macintosh team, and the difficulties they are facing in porting the MSOOXML file format on a new platform? Microsoft's Mac Business Unit claims that it would take a team of 5 people 44 weeks to implement a quarter of one product.

Andrew Shebanow (from Adobe) extrapolated that and calculated that it would take a "mere" 120 man years to implement MSOOXML. This large amount of investment just to read and write files for one vendor is something American countries would find daunting:

"How can competitors afford to make that level of investment? Novell says they will support import and export for Open XML with financial and technical help from Microsoft. Corel says they’ll do it too. Guess we’ll need to wait and see how successful they’ll be at maintaining fidelity and compatibility, though given what Rick has to say, I’m not super confident."

It is no wonder why Office for Mac has been delayed twice since, with the "projected" date of release in 2008. It had better be 2008, because the product name has been set to "Microsoft Office 2008 for Mac".

Now back to Apple. They are committee members. They allegedly contributed to the development of the specification. When I asked Stephen McGibbon what Apple contributed, [Updated 070825] Brian Jones said:

"Apple made huge contributions. They included improvements to the future versioning functionality; drawingML and VML; and big time focus on conformance."

I thought this was hugely ironic because MSOOXML suffers from huge versioning issues with the mixup of the legacy and supposedly deprecated but surprisingly prevalent VML and the "richer" DrawingML. As soldiers of aesthetics, one would have thought the Apple representatives would have balked at the idea of the two functionally equivalent specifications to be included in the same body of work. How un-minimal.

Pinning the monkey

I then asked "Can iWork save MSOOXML?" Stephen then started to divert the discussion with issues of ODF, KOffice, and OpenFormula, even my attitude, and continued to avoid that simple question. The reason why he cannot answer the question is because he entitled his blog post: "iWork '08 supports OpenXML"

It makes you wonder what "support" really means to Microsofties. Maybe it "supports" his view.

Instead, National Bodies and reviewers of the MSOOXML spec should ask these further questions:

• Why doesn't that "stamp of approval" from Apple translate to native file format support in its productivity application?
• Didn't Apple think MSOOXML was good enough for them?
• As owners of MSOOXML, why wasn't Microsoft curious why their paid to be business partners like Apple (and Novell) didn't choose MSOOXML over their own native file formats?
• Why didn't Microsoft resolve any issues relating to Apple's reluctance to adopt this supposedly universal file format within Ecma TC45, or even prior to the iWork launch?

Additionally:

• Could it be that Apple has been holding back on its contribution to TC45 and has kept secret iWork's new features so as to have a "one-up" on its fellow technical committee members?
• Or was it because Apple did not bother to contribute?
• Or was it because TC45's scope was so narrowly defined to just one vendor's (Microsoft's) line of products?

These are extremely important questions. I posted these questions on Stephen's blog on the 16th of August, and till today (25th of August), not a squeal. He tends to ignore you when he's trapped. It's happened quite a few times already.

If Apple is so keen on its own file format, and Microsoft loves the "choice-in-standards" argument, would Microsoft give a helping hand to Apple in standardising Apple's own file format through Ecma and ISO since Microsoft has laid the groundwork by investing so much in buying up technical committees, standards bodies and ministers around the world?

iWork XML vs ODF vs OOXML

And it gets more interesting. The Cybertech Rambler has taken some time out to review the file formats in iWork '08. What he notes is mighty interesting: "iWorks XML format vs ODF vs OOXML preliminary thoughts"

"Like ODF single file format, index.xml is self-contained. It actually resembles ODF more closely than OOXML in this and many other respects. In particular, we do not see abbreviated XML element name or attribute name except those that are extremely obvious and borrowed from HTML. This, of course, is the key to why I can decipher the file quickly. We do not see “run tags” (rPr tags for example) that MS insist it needs to fully  document the properties of the text.  Those property tags are attributes of the element it refers to, the way it is done in ODF and the way I think it should be."

He also confirms what the Microsofties refuse to confirm: "on closer reading of iWorks website and documentation, it appears that iWorks can only read OOXML file but cannot write it. That’s a pity."

What a pity indeed.

Fellow committee members have no faith in the file format they supposedly worked on for less than a year. I wonder why?

yk.

This is news making waves across the blogosphere: INCITS (an industry forum to recommendation US position on JTC1 ballots) could not reach a consensus to vote "APPROVE" on OOXML. This is indeed big news. There has been much fear that this far-reaching standard may be adopted at ISO despite the serious and grave technical deficiencies in the proposed standard, despite its vendor specific nature, despite troubling licensing concerns and most importantly, despite strong objections from technical bodies of 20 countries.

More troubling then the technical problems of OOXML is the extent to which the standardization process is being subverted to the favor of Microsoft. Rob Weir states in his blog post that:

An important factor in the V1 vote was the large number of members who joined very late in the process. At the start of the year, V1 had only 7 voting members. But by Friday's meeting V1 had 26 voting members. There was a clear pattern in the voting where the long-time V1 members voted for the "Disapproval, with comments" position as well as "Abstention, with comments" while the newer members voted overwhelmingly "Yes, with comments" and against "Abstention with comments." This is not surprising since the new members were largely Microsoft business partners.

We see today in a GrokLaw news pick the following item from a member of the standards committee in Portugal:

I was present on the meeting of the Technical Commission (CT) created to award the ISO standards in the area of structured documents (in Portugal)

A Technical Commission (CT) did not exist when ISO 26300 (Open Document) was submitted neither when there was a submission of OOXML (ECMA 376, potential ISO 29500) for the the fast track, and that was the reason why Portugal did not submit any opinion nor had any right to vote. We expect that now, with the pressure made and the CT created there would be right to vote.

The CT was created by the Computing Institute, in which  is delegated the responsability [sic] for the norms of the IT sector; a delegation granted by the Portuguese Institute of Quality (IPQ), the point of contact of ISO in Portugal. Its creation is motivated mainly by the pressures and availability of some people when the proposal for fast tracking of OOXML and a neccessity to avail now the OOXML as standard ISO and as a Portuguese National Standard

In the meeting they were present:

2 persons from II (Instituto de Informatica [Computing Institute])

1 person from the local government (Alentejo region)

1 person from Jurinfor [Jurinfor is a Microsoft partner]

2 persons from Microsoft

1 person from Primavera [Primavera is a Microsoft partner]

1 person from ISCTE  

2 persons from Assoft [reportedly, most members of ASSOFT are Microsoft partners]

1 person from the Inst. Informatica da Seg Social [Computing Institute of the Social Welfare Department]

1 person from the Inst. Tecn. Informacao da Justiça (eu) [Technical Institute Information of Justice (eu)]

... The 8 vocals will readily follow to the election of the president of the CT. There was 1 candidate in the place (Miguel Sales Dias, from Microsoft).

... The vote results were 7 votes in favor of Miguel Sales Dias, of Microsoft, who was designated to preside over the CT, and a (1) blank vote. It was decided to adopt consensus as the form of adoption of any proposed norm, following to majority vote in case there is no consensus in the CT and if there is a strong opposition to submit any norm.

To summarize the above, the meeting in Portugal to decide the fate of OOXML as an ISO standard and Portuguese National Standard was presided over by a presentative from Microsoft, was attended by Microsoft business partners and the decision reached in the meeting was to adopt consensus for any proposal even if there is a strong opposition!

But that's not all the news from Portugal. GrokLaw also notes (through Openxml.info and another source) that members from IBM and SUN were not allowed to vote in the technical committee deciding OOXML's fate!

None of this is new to us in the Malaysian Open Standards scene. When it was obvious that there was a strong technically oriented open standards lobby in Malaysia who may pose objections to OOXML due to its shortcomings, we suddenly saw an unprecedented influx of Microsoft allied partners in the various forums where OOXML was being discussed.

So, who's being unethical here?

Update (19th July 2007): There is another blog post detailing the standards process being subverted in Italy when OOXML was being considered (thanks yk and LinuxToday for the headsup):

The voting in Italy was scheduled to end the 13 of July, for members enrolled on or before 8th July. Strange things started to happen, not unlike other member bodies' situations abroad.  Up and until mid-may the members of the relevant Uninfo committee (JTC1) were five: IBM, Microsoft, CEDEO (Leonardo Chiariglione), the PLIO organization (Openoffice.org in Italy) and HP. Then new members started flocking. At the last count, voters were 83 [0].

Wow! a great number, indeed! If one considers that admission to JTC1 costs in excess of EUR 2000 (more than 2700 USD), it shows a great deal of interest in the standardization process (is the irony sufficiently evident?).

... Actually it is quite impressing seeing how the voting panel was formed. Particularly noteworthy is the fact that among those favouring the adoption of the standard without  reservation a large majority is made of business partners of the proposing entity, a law firm retained by the latter, the official certified business partners association of the proposing entity ... "Money can't buy me love" Beatles used to sing: perhaps neither a standard.

I do not claim to be an XML expert and my experience in the standards process is admittedly limited. I definitely will not even try to dispute the fact that Mr Rick Jelliffe, a well known developer in the XML world with valuable contributions in creating and promoting open standards would be an authority on what would constitute a good standard.

However what is puzzling me is that some of the myths he attempts to debunk as reported in this article about his latest visit to Bangkok does not dispell any myths at all. In fact it reinforces some myths and misrepresented the facts.

The article from the Bangkok Post is available here: "Open standards advocate comes out in favour of Microsoft", and syndicated in ZDNet Asia here.

On Dates

To demonstrate why I think Mr Jelliffe may be mistaken, my previous commentary entitled "Malaysia's History is ill-formed" has some information on how the Microsoft Office Open XML (MSOOXML) specification handles the encoding of dates; the very same subject which he addresses.

Here are his quotes:

"he [Jelliffe] pointed out that Microsoft had started work on developing an XML data format back in Office 2000, work that predates some of the ODF work."

"He said that accusations that Open XML contradicts other ISO standards can be explained and are not significant. One case is that Open XML stores dates as numbers, as has been the case in Microsoft Office all along."

Now let us examine some of the comments he has made with regards to dates. He states that Open XML "stores dates as numbers, and has been the case all along." This is an interesting statement due to the fact that its not true.

Microsoft has been working on XML data formats back in 2000 and drummed up huge marketing publicity of this support in Microsoft Office 2003. The then new file format however obviously was not open and good enough. Subsequently they had to change that format just one release later (in 2007). And you would have thought that XML stood for eXtensible.

However the Microsoft Office 2003 XML format (MSO2003XML) did at least one thing right; they had the dates encoded not in a serialized "number" format but was in fact encoded in an ISO standard format (ISO 8601) which was gave the format these features:

1. Human Readability ( i.e I should not need to refer to the specs to figure out what the number "means") and
   
2. It did not have to carry the application level baggage (bugs) of
1. leap year miscalculations ( where 1900 was mistaken as a leap year )
   
2. pre-1900 limitations ( where historical dates are hashed "####" out ) nor
3. 1900/1904 epoch confusions. ( where different platforms have different starting dates )
   

Here is my justification:
The XML code above was generated by Microsoft Office 2007 in the form of the Microsoft Office 2003 XML format.

This shows that MSO2003XML format encoded dates "the right way," (conforming to well known internationally accepted date encoding standards) while the latest format, MSOOXML takes a step backwards in terms of open standards by going back on their decision in adopting the obscure "serialized number" format.

So while it is true that Microsoft has been working on XML formats prior to the 2003 release of their productivity software, it is not true that the serial dates were used "all along," as asserted by Mr Jelliffe. They went back on their decision in adopting ISO 8601 date encoding to their obscure binary representation of serialised dates.

On Size and Speed

He then proceeds to spread more myths to the argument:

"ODF 1.10 has 760 pages. However, it refers to a lot of standards such as SVG, MathML, Open Formula, xlink, zip. These are not ISO standards, these are from the W3C. Once you add them, they are quite comparable in size," he said.

This is his and many Microsoft representative's justification in why MSOOXML was so large at 6000 pages and is no different to ODF and should therefore be ratified at ISO.

The argument put forward by the many (at least 14) National Bodies who spoke up against MSOOXML back in February 2007 was not because of the fact that MSOOXML is the largest proposed standard ever to be standardised at ISO. If fact many NBs responded positively to the comprehensive information. Some even claimed that this was still not complete, and requested for more information.

The issue was because MSOOXML is probably the largest proposed standard to be "Fast Tracked" through ISO. Many NBs queried why such a large specification should warrant a fast tracking.

There is a big difference between "Fast Tracking" and the regular submission of ISO standards. This becomes more apparent if your specification is 10 times the size of the average proposed standard!

Additionally, ODF was not ratified with SVG, MathML, XLink, Zip and other W3C standards all together at the same time. Instead the prior W3C standards were already well established and approved in their own right and in their own time with the relevant experts of their specific domains vetting it.

MSOOXML also incorporates proposed "standards" which failed in the marketplace and now is offered a "backdoor" to standardisation process by piggy backing this nebulous specification. (See VML vs SVG, and MathML vs Microsoft Office MathML)

So there is a myth being built that ODF and its constituent parts are just as large as MSOOXML, and therefore MSOOXML is OK. I for one would rather MSOOXML be even larger; to cater for unknown tags like "lineWrapLikeWord6" or a Macro specification. However what troubles me is that the special relationship between Ecma and ISO should be abused with the fast tracking of this large specification.

We must never mix these two issues up ("ODF size" and "Fast Tracking") to justify ISO ratification.

On "Full Fidelity Decoding" and Interoperability

The final issue I would like to address is this statement:

" if I wanted to make sure that all the data in the document opens up the same way, then I'd go for Open XML," he said."

If I wanted to make sure that all the  billions of legacy documents to be opened up the same way with full fidelity, then I would go for the current binary and proprietary closed file format by Microsoft. There is and always has been a worldwide demand for these file formats to be opened up for easy decoding. In fact if Microsoft really practiced interoperability as they so often preach, they should open up this specification freely, and we probably would not have any need of this discussion today.

Their proposed "solution" which is the mapping of the binary stream to XML in MSOOXML however is not the correct answer. This is a pseudo opening of the binary formats, where things are being left behind, left out and ignored. Why should the market go through another layer of abstraction just to get access to our binary documents? We are already seeing leaks in this abstraction with the difficulty in getting the Mac OS X port of the translators to work, and the exclusion of Macros to yet another file format for users to get confused by.

So our ideal request here is to release the binary information required to access our billions of user data created in Word, Excel and PowerPoint. Once we have that, then we can decide on what futuristic XML file format to base on.

Microsoft is attempting to roll these two separate issues (the decoding problem and future encoding format) into the supposed "magic bullet" called MSOOXML.

And once we solve the problem of the decoding of our data ( a problem only solvable by the vendor ), I and many National Bodies worldwide wholeheartedly agree with the illustrious Mr Rick Jelliffe:

 

"If you want something for interchange and if it is platform neutral, then I'd tend to ODF."

yk.

VietnamNet Bridge reports that:

"The government has approved the software industry development programme to 2010, in which priority will be given to the use of open source software in state-funded IT projects.

The state will encourage and assist organisations and businesses in providing services supporting the use of open source software.

According to this plan, the country aims to become one of the top 15 countries providing software outsourcing services in the world by 2010. Development of human resources in terms of quality and quantity is the key to the success of the software industry."

In the context of the many open source national initiatives worldwide along with the pilot OLPC deployments, it should be noted that Microsoft has announced a development initiative for the "billions" of people in developing countries to get exposed to ICT:

"Bringing the benefits of technology to the next five billion people will require new products that meet the needs of underserved communities; creative, new business approaches that make technology more relevant, accessible and affordable; and close collaboration between local governments, educational institutions and community organizations."

I'll leave it to the readers to draw their own conclusions on this latest turn of events.

In a surprising twist of events, Datuk Dr Mohamad Ariffin Aton, the SIRIM CEO himself has gone public in his decision to suspend SIRIM's own Technical Committee for E-Commerce (TC4). As a member of TC4, I  am surprised that I would learn about this issue in the press before a clear explanation was given internally.

I have heard rumours that he had negative views when ODF went on Public Comment back in November 2006, and now it is finally confirmed. Fortunately, his fears, uncertainty and doubt against ODF is fully articulated, and his concerns relatively easy to dispel.

It is now clear what issues are bothering him, and it makes our job extremely easy in addressing these issues rationally and factually as we always have been doing with the concerns raised in the Public Comments. I have been doing this for over 6 months now, and I think I'm getting quite good at reviewing the concerns, collating the facts and using logic and rational arguments to convince any good technical person willing to understand the truth. Truth is always the best decider.

So I for one, look forward in clearing these issues up with the CEO in allaying all his fears and I am confident that SIRIM are willing to accept these explanations in a constructive manner.

For the readers of OpenMalaysiaBlog, I would love to hear from you if there are any issues which he raises which are relevant and would require extra attention.

Please read the InTech article which should be in the newsstands later this morning.

http://star-techcentral.com/tech/story.php?file=/2007/4/4/technology/20070404125811&sec=technology

yk.

I took a piss at Microsoft.

No, really. I took a piss at the Microsoft Convention Centre in Redmond, WA.

As you would already know, Yusseri and I were invited to attend this annual, exclusive event at Microsoft Corp. It's cold out, 10°C daytime and 2°C nighttime. They feed you with lots of expresso, latte, mocha and a fridgeful of soda. I really had to take a piss.

We got in yesterday afternoon, after 24 hours in the air and in transit via Singapore and Tokyo. As luck would have it, US Customs and Border Patrol in Seattle has to reward me with a random check. They were interested in the HackInTheBox sticker I had on my notebook though, but the whole check was a mere formality.  Boy, am I glad though they don't do checkups like they do in Wuhan, China as the pic on the right shows.

I still had to take a piss after that though, 16 hours on an aircraft and all that.

Day Zero, which is what yesterday is, was quite uneventful if you can discount a deep discussion of what last city visited meant on our Customs forms and having the Hummer H3 we are driving being broken into and our stereo nearly getting jacked. Pretty uneventful. Oh yeah, I also bought an IPod at the Apple Store and took a close up view of the newly launched Apple TV media device.

Up and early in the morning for the 7.30am shuttle to the Microsoft campus, the coffee they poured down our very willing throats was abundant, resulting in aforementioned piss.

Ditesh's email of the scanned in article in Monday's Tech&U on the ODF thingy had me quoted as well. And lo and behold, keynoting the morning's session was Bill Hilf, who was also quoted in the same article. Bill's the General Manager of Platform Strategy at Microsoft.

Bill is a former OSS dude who worked on Apache, and today heads Microsoft's platform strategy. Microsoft's Linux Labs started off being a lonely Dell tower sitting under Bill's desk. His message was one of community and collaboration, and how Microsoft is learning to play nice with us open source type folk while still maintaining they're software licensing based business model. Make no mistake about it though, Microsoft is in it for profit, Bill stressed.

What got really interesting was when Yusseri raised the issue of OOXML and why didn't Microsoft just work on ODF in collaboration instead of creating a new, bloated standard. Bill's answer was quite surprising, as he clarified that the file format (OOXML) was a part of the software and that OOXML and the software (MS Office) are quite inseparable. Ergo, OOXML is an integral and inseparable part of MS Office. That's why they could not adopt ODF as the file format for subsequent versions of MS Office.

I don't really know if Bill realises this, but he's just illuminated that this whole OOXML, ECMA and ISO standards play is not about standards nor about Microsoft vs IBM but about legitimizing as a standard a specific technology from a single company developed in isolation from the general technology and user community worldwide.

It was an eye opener, and in my opinion, quite damning to Microsoft. Are they really trying to push through their software, in the form of the inseparable OOXML/MS Office as an international standard ? Is this all about ensuring that future generations of MS Office have a purpose on the planet in the face of competition from other office suites like OpenOffice.Org ?

Perhaps these questions should be asked at the ECMA, ISO and at the level of National Standards Bodies as well.

We'll be sitting with Sam Ramji and Bill in a couple of days to go over this issue in much more detail, given the time constraints at the summit proper.

Kevin Schofield's presentation on what Microsoft Research is doing was interesting, from a geeky perspective. Showing prototypes of image processing technology, visual recognition software as well as working prototypes of new ground broken in the Human-Computer Interface as a leadup to truly pervasive computing was illuminating. While I have seen different implementations of these before at the MIT Media Lab, one does trust that Microsoft will be bringing some of these to the average user in the years to come.

It's nice to be here at Redmond, if anything, to get frank discussions going with the folk behind the stuff Microsoft creates without having to cut through the veils and mazes thrown up by the marketroids. I am looking forward to that discussion with Bill and Sam, as well as meeting up with an old misguided buddy, Tan Loke Uei, who's now based in Redmond.

More on the summit in the days to come, but I need to go take a piss again.

Yoon Kit wrote an article on ODF for the well regarded weekly business newspaper The Edge which was published 19th March 2007. Titled "Why OpenDocument Format matters", the article explores the importance of open standards in the realm of electronic documents. The article can be viewed by clicking on the thumbnail below (excuse my poor Gimp kungfu):

Disclaimer: We declare our independence of opinions from our employers, institutions, associations and clients, past and present. Thoughts and expressions in the Open Malaysia blog are rightly each blogger's own and each of us stand by what we individually write. Views by readers who post comments and others whose writings we link to in this blog are theirs.

Those that have been involved in National Bodies worldwide in the latest standardization exercise involving  Microsoft OOXML may have been puzzled at first, then later amused, that Microsoft representatives have chosen to rewrite not only well accepted definition of contradiction but also portions of technological history.

We will debunk Microsoft's latest assertions in this blog post.

To start off, let us accept that by virtue of accepted historical evidence, this new definition on the term 'contradiction' has only been suggested by Microsoft and Microsoft only. We will also note that this new definition has not been approved by JTC 1.

In light of the many contradictions submitted by concerned members at various National Bodies worldwide, Microsoft representatives have, in an act worthy of a team of scheming lawyers, decided to change the playing field by changing the definition of contradiction so as to invalidate the contradictions submitted by concerned members.

Let us look at their definition of contradiction:

"Does not disrupt nor prevent the coexistence with any existing ISO standard"

"[Contradictions] are really technical issues that should be raised in the next phase of the process"

To illustrate this point, Microsoft picks on two sets of areas of standardisation: "Programming Languages" and "Document Formats" to "prove" that this strange definition is consistent. The following sections will explain why their examples of "contradictory" ISO standards are fallacious.

C, C and COBOL as examples of contradicting standards in ISO

The examples Microsoft has used for the evidence of coexistence of yet "competing" standards are the ISO standards for Programming Languages:

• ISO 9899 for C
• ISO 14882 for C
• ISO 1989 for COBOL

This is their "proof" that their version of "contradiction" means that:

• Coexistence of competing multiple standards are allowed
• Overlap of standards are allowed
• First standards to the door does not win

This form of argumentation is completely fallacious as it draws a similarity of purpose between these programming languages. It was not noted in Microsoft's presentations that all three of these languages were designed for completely different purposes and that they are used in completely different markets.

The C programming language is a general-purpose, procedural, imperative computer programming language. The C language is widely used for systems programming. By comparison, C is sufficiently different as it was designed around the concepts of object-oriented and generics programming with static typing. C is most popularly used in building graphical applications. Finally, COBOL was designed from the start for the business domain and shares little with either C or C in its structure.

As such, these standards do not overlap in the substance of their specifications.

Given the many allusions by Microsoft that ODF supporters are undermining MS OOXML, the Microsoft team was clearly told that the reasons for standardising C and C and COBOL were NOT because of an ISO battle between programming languages in the ISO level, but rather due to the original intent of standards which is to standardise the programming language to prevent fragmentation in the market.

There were originally many implementations of C and C as each vendor provided different features. Every single one of these vendors provide different compilers that worked differently enough to make a software developer aiming at portability to tear his or her hair out. To promote better use and interoperability, the need for standardisation of these languages was created.

In addition, contrary to what was purported by Microsoft representatives, the purpose for standardization was not because the C team wanted to do a one upmanship on the C team. The only purpose for standardisation was to retain consistency between the different implementations by different competing vendors.

Finally, C, C and COBOL programs are typically are not exchanged between end users in the sense office documents are exchanged. Given the increased degree of cross-user interactions with electronic documents, there is a clear need for a single clear standard in this area. Careful observers will notice that MS OOXML is geared towards fragmenting the current ISO standard on Office Documents. We cannot afford fragmentation in document standards simply because a document format today facilitates a large degree of interaction between users on many different systems. A single document format will facilitate interoperatibility across many systems, much in the way that TCP/IP provided for the foundation for the Internet. Thus, the comparison to C, C and COBOL is misleading, false and unsubstantiated.

PDF, HTML and ODF as examples of contradicting standards in ISO

In addition, Microsoft argued that given the existence of PDF, HTML and OpenDocument Files within ISO "proves" that competing formats of documents are acceptable:

• ISO 19005-1:2005 PDF
• ISO/IEC 15445:2000 HTML
• ISO/IEC 26300:2006 ODF

They argue that these three ISO standards show that there are already a history of multiple standards in the area of document standards. While no unreasonable person would be against the possibility of multiple standards if the standards in question are sufficiently different, it must be noted that PDF, HTML and ODF address sufficiently different needs in electronic documents.

PDF files are most appropriately used to preserve the exact look of a document in a device-independent way. HTML, on the other hand, is a lightweight markup used in web pages to describe content. Neither is used for describing and facilitating the editing of documents, spreadsheets and presentations! Nobody in their right mind would use PDF or HTML for office documents. That is the sole domain of ISO/IEC 26300:2006 (ie, the OpenDocument Format - ODF).

To claim that PDF and HTML are examples of "multiple standards in the same area" is a gross miscommunication of ideas and an insubstantial red herring. This deliberate misunderstanding is similar to the C/C "justification" and is being used to convince the National Body to believe that ISO allows contradictory standards which "do the same thing"

Direct Contradiction to ISO 26300

Finally, ISO/IEC 26300:2006 already exists as an ISO standard and MS OOXML is a significant contradiction to ISO's motto of

"one standard, one test, and one conformity assessment procedure accepted everywhere.”

Both ISO 26300 and MS OOXML standard have identical aims to provide the functionality of

"text documents, spreadsheets, drawings and presentations for office applications."

Microsoft has never explained nor bothered to convince "Why Ecma 376 does not contradict ISO 26300", even when pressed multiple times. Instead they resorted to adjusting the meaning of the term "contradiction".

Microsoft has also never explained satisfactorily why MS OOXML is superior nor the different requirements it facilitates for over ISO 26300.

First through the door wins?

Finally, Microsoft argues that there should not be competition between standards such that the first standard through the door wins. For once, we agree with Microsoft. The first standard through the door does not and should not "always win". If the first standard is proven insufficient, it will be discarded. However if it is extensible, then new incoming standards should look towards harmonizing with existing standards.

In this case, the first standard is ISO/IEC 26300:2006. Fortunately for future applications, it is extensible so "new" standards such as MS OOXML should work towards harmonizing with this existing standard to prevent proliferation and confusion in the Document landscape.

There is certainly precedence in this area. China's Uniform Office Format (UOF) standard for document format is already in the process of harmonizing with ODF within OASIS.

In Conclusion

National Bodies worldwide, and JTC1 themselves, should focus on the technical substance of the contradictions submitted. We should be discussing the contradictions, and not Microsoft's definition of contradictions. This is a red herring, a means of diverting attention from the actual technical issues with the proposed standard. We should not be waylaid by the latest attempt from Redmond. We have the responsibility to ensure that the due process is followed strictly in the interests of all parties concerned, not just Microsoft.

An Abiword Developer responded to the technical commentary I posted up addressing the BiDirectional issue with AbiWord which was 'capitalized' by a Vendor to suggest that ISO 26300 did not support international documents.

The AbiWord developer requested for the test file as he wanted to look into the problem highlighted by this concern. I sent it to him, and within an hour, he reported that he has fixed the bug in the Open Document Format file import filter for AbiWord. The first screen shot is the current release version v2.4.6 which shows the error, and the second screen shot is the developer version (not released yet) v2.4.7 which corrects this bug.

His explanation on why this occurred was:

“The problem was indeed with the OpenDocument import filter.  The importer was completely ignoring the "style:writing-mode" paragraph attribute, so all paragraphs were rendered as left-to-right, even those that were correctly placed on the right-hand side of the page. 

The OpenDocument filter is not complete, as you mentioned in the post, so I was a bit dismayed to see it being used to spread misinformation.”

It must also be noted that the "style:writing-mode" in ISO 26300 is fully documented in Section 15.5.36 which satisfies the requirements of the concern.

This is a fantastic demonstration on ISO 26300's published capabilities and also the Open Source development process where bugs such as these are fixed almost immediately to respond to end users requirements.

The Vendor should have contacted the AbiWord developers or lodged a public bug report instead of using this error as an example of a 'deficiency' in the ISO standard. That way, they would not have embarrassed themselves on raising this problem as a major technical criticism to ISO 26300 when instead it is a mere application file filter problem.

If this is the source of concern which caused some companies to raise the issue of “ODF does not support international documents” during the Public Comments, then we can now confidently say that their concerns have been overly hyped and they have been misinformed as this concern has been a non issue to start with.

I hope that it is now apparent to us all that there are no substantial technical concerns regarding the adoption of ISO 26300 as a Malaysian Standard, and we should not delay this process any further.

This post is an update to Bi-Directional Multilanguage Fidelity ODF Concerns

yk.

“ODF Version 1.0 lack Multilanguage support for BiDi language requirements”

Summary

The concern raised by a Vendor is yet another statement which is too broad and misleading. This technical commentary will prove conclusively that ODF v1.0 or ISO 26300 clearly supports BiDi (Bi-Directional)  Multilanguage fidelity mixed in a paragraph. The source of this concern was due to the fact that the Vendor selectively chose an application which is still in development in the support of the full specification of ISO 26300. We can demonstrate this unfair choice by using a more mature application and can show that true interoperability and fidelity is possible, proving that ISO 26300 has excellent capabilities for multi-lingual usage, and suitable for multi-lingual usage in countries such as Malaysia.

The Vendor's Objection

The Vendor's Evidence 1 stated that: “ODF Version 1.0 does not explicitly address BiDi requirements.” and goes on to say “even if we assume that the Unicode BiDi algorithm is used, there will be scenarios where the default behaviour needs to be overridden for text within a paragraph.”

Evidence 2 states that: “ISO 26300 does not explicitly address BiDi requirements. ODF v1.1 incorporates additional section that deals with this (Appendix F of ODF v1.1 specs)”

Evidence 3 then refers to the screenshots which is similar to the one I have recreated as follows:

From the comparison of the two applications above, AbiWord shows that the BiDi mixed sentence is in error as the arabic and english words are in the wrong order. This was the Vendor's evidence that ODF v1.0 does not support BiDi.

The counter evidence

All these “evidence” however is moot when we test true interoperability and fidelity using an office suite which has a better developed import filter for ODF. To illustrate the point, here is the same file opened in KWord, a word processing application from the KOffice productivity suite:

As you can see, the fidelity between the different applications which support ODF is excellent. The mix of BiDi words are recreated accurately, and no information is lost or misrepresented. Just for completeness, let us view the same file using OpenOffice.org on a different Operating Systems:

Excellent fidelity of the same document across different Operating Systems and Applications is clearly demonstrated.

So the issue at hand here is not any deficiency of ODF file format at all. It is once again, the deficiency on an application level, where AbiWord has failed to comply with the Unicode capabilities of ODF.

If you were to notice, AbiWord does support BiDi, but only on a limited scale. The L-R BiDi mix was accurate. It is probably due to the developers preference for L-R authoring which made this feature better implemented. It also goes to show that AbiWord will have the capability to handle accurate R-L BiDi in the near future when it finally conforms to ISO 26300.

Why was AbiWord chosen as a candidate to be in this Vendor's example? 

Probably because in the AbiWord website, the current status (as of v2.4.6)  of the ODF Filter for Read/Write is clearly stated at only 60% compliance. This makes it easy for any detractor to demonstrate “loss of fidelity” not because of the file format specification, but because the application import filter is not complete.

The proof is that KWord happens to have better compliance to ODF as demonstrated by its interpretation of R-L BiDi mixes, performs in complete fidelity with OpenOffice.org.

This example also shows the wonderful portability of ODF in that it can be represented in multiple Application implementations on multiple Operating Systems, each developed completely independently and created via the open standards of ISO 26300. The choice available to citizens of Malaysia is unprecedented.

ODF v1.0 clearly states that it is based on  ISO 10646 which corresponds with Unicode 4.0 or better. This will give all the specifications required for the BiDi algorithm. Appendix F of ODF v1.1 details the cases on how to handle BiDi scripts more fully in fringe cases. But as you can see in this example, these detailed improvements to the standard does not affect the case of most documents, as demonstrated by this example provided by the Vendor.

Conclusion

As such this unwarranted concern that ISO 26300 does not retain fidelity on BiDi R-L mixed language support is yet another misunderstanding between the specification of the standard, and the complete implementation of application level file filters. With a simple check of the evidence, the Vendor's example is yet again proven erroneous and misleading. This commentary has proven beyond any doubt that ISO 26300 supports International documents completely and it leaves little reason for Malaysia to further delay the decision to adopt ISO 26300 as a Malaysian Standard.

yk.

Update 11th January 2007: This bug in the ODF file import filter in AbiWord has been fixed!
More details here: AbiWord responds positively to the BiDi concerns

Yoon Kit's article, 'Technology neutrality' and what it means to us Malaysians got published in the New Straits Times Online  today.

Yoon Kit wrote the article afer the MOSSA Round Table on Software Neutrality and Openness on 7 December 2006. The online article is found here and also transcribed further below.

Excerpts:

"We do need to clarify the terms "neutral technology platform policy" or "Technology Neutrality". "Neutral" seems to suggest a passive attitude towards adopting the best solution. We do not believe that this is the true meaning of the direction. A "Neutral" position should mean that all procurement decisions should be based on a level and competitive playing field, with regards to triggers to mitigate risks of market manipulation through monopolies, results which are openly accessible standards with no encumbrances to patents and royalty, and the merits based on solid business rationale on functionality, cost effectiveness, interoperability, security, flexibility and performance.

This is true "Technology Neutrality" as it encourages competition, innovation, better service and a healthy local ICT ecosystem."

"So what "Technology Neutrality" means to Malaysians is more competition and more choice. Our current ICT ecosystem is severely lacking of choice, and hopefully the recent announcement by the Minister would mean a more pro-active effort in enforcing "Neutrality" in terms of policies, government spending and education. OSS will always be on the forefront in terms of features, interoperability, customisability and of course value."

Image: scanned article (click image for full size)

The New Straits Times newspaper in the Tech&U supplement yesterday (11 December 2006) covered the MOSSA (Malaysia Open Source Software Alliance) Round Table in response to the recent announcement of the Malaysian Government's policy on "technology neutrality". It is not online, so I've transcribed it here.

Impact from software neutrality
By Izwan Ismail

      The Government's adoption of a neutral technology platform policy must ensure that a more level playing field is established, which benefits the consumers in terms of quality service received and long-term gains.
      According to newly established Malaysian Open Source Software Alliance (Mossa), the policy should include triggers that look beyond the short-term goals and mitigate risks of market manipulation by companies or organisations that might offer more in return for market monopoly, hence imposing restrictive strategies and hindering the growth of the local information and communications technology (ICT) industry in the long run.
      Mossa member Dr Yusseri Md Yusoff said that the concept of neutrality should result in openly developed, maintained and accessible as well as royalty- and patent-free standards to mitigate the risks, which favours both the market and industry and gives true meaning to neutrality. "Software neutrality needs to include understanding that considerations when acquiring software are independent of whether he software comes from a proprietary or open source provider. In this case, the procurement decision should be based on solid business rationale which must consider functionality, cost-effectiveness, interoperability via open standards, security, flexibility and performance.
      "And additional considerations would include the effect on local economies, the adoption of open file formats and adaptibility to future technologies," he said during a roundtable discussion in Kuala Lumpur last week.
      Commenting on the effect the new policy would have on the Government's present open source software (OSS) initiative, Yusseri does not see any problems here.
      "From the start, the Government OSS initiative has always had a balanced approach to software deployment with no preference given to either open source or proprietary software," he said, adding that the Government has never issued a mandate that procurment of ICT solutions should be based on OSS or proprietary.
       Yusseri also said the OSS initiative is to give the Government more choices in terms of software solutions.
      "The advantage of OSS is that you will get the source code, not vendor locked in, and the standards are almost completed."
      Meanwhile, OSS advocate Nah Soo Hoe said in procuring ICT solutions, the Government needs to realise the many benefits of open source. "What needs to be understood is that the cost of the software is just the icing on the cake; the main thing is that it gives you the freedom to learn, innovate and invent."

The Edge in its last week's edition on 4 December 2006 had 2 articles related to the recent announcement of the Malaysian Government's policy on "technology neutrality" and they are now online, as below.

Much ado about nothing
by Lee Wei Lian
Excerpt:
Michael Mudd, director at computer industry association CompTIA in Hong Kong, said in a letter to netv@lue2.0: "CompTIA also believes that [Jamaludin's statements] now result in a level playing field and mark the start of software platform democracy in Malaysia. This is healthy for the Malaysian software ecosystem as it encourages innovation and competition for a technology-neutral approach."

Software democracy was always here
by Dinesh Nair
Excerpt:
The minister's statement in no way deviates from existing government procurement methodology as even the Mampu Open Source Masterplan clearly states that all procurement will be evaluated firmly and equally on technical and financial merit.
Therefore, this is not a swing nor a victory for any party. Elements who are claiming a victory of sorts are quite obviously not up to date with Malaysia's policy on the matter. Our technology procurement policies should be technology neutral, based on validated open standards and beneficial to the government and the people of Malaysia. Such a policy would provide choice, efficiency and economies of scale to ministries, government departments and agencies.